[MeFi Site Update] June 19th June 19, 2024 2:13 PM Subscribe
Hi there, MetaFilter!
This month’s Site Update includes a particularly important announcement. You can find the last site update here.
I’m looking forward to your feedback and questions!
SPECIAL ANNOUNCEMENT
Since Kirkaracha joined our team 4 months ago, I have been shifting our priorities more towards the technical side. So far we have reviewed the site infrastructure and tackled small side projects which have allowed him to get more familiar with the current state of the site’s back end. After a few months of going over feature requests from staff, members, and the BIPOC Board, we have made the decision to focus resources on rebuilding the entire site from scratch to make it easier to maintain and to iterate. Goodbye ColdFusion!
Kirk started working on this several weeks ago to determine how complex it would be and has made a lot of progress already.
What to expect?
- For now, we are trying to replicate the UI, flow and features of the site as they are, with some minor accessibility and design improvements. Once the new site is live, we can start thinking about new features/changes.
- Kirk will put the new beta site on a test server and the next step will be to start gradually giving members access to test it as soon as it is ready.
- I’ll make a separate post about this once we’re ready for testing and share screenshots of the progress with the community at large there.
Profit & Loss
– You can find this month's P&L report here. The previous P&L reports are here.
Admin/Moderation
– We’re going ahead with the Trans Members Initiative as outlined in the past site update.
– A separate Meta to crowd-source questions from Trans members for the survey will be posted next Monday.
– We are getting close to both MeFi’s 25th Anniversary and Fundraising month, depending on how timing goes with the new site, we would like to plug all 3 of these together. I’ll keep you posted about this.
Tech
– Repaired Archives and made sure that they update
– Fixed several recurring tasks that were not running.
– Updated infodump and made sure that it is reloading correctly.
– Fixes to make sure contact form e-mails are not dropped by gmail.
– Frimble is updating and testing the site backup process
BIPOC Advisory Board
– The last meeting was postponed due to last minute scheduling conflicts and we’ll resume as soon as possible. I’ll confirm once the pending Meeting Minutes have been approved and posted.
If you have any questions or feedback not related to this particular update, please Contact Us instead. If you want to discuss a particular subject not covered here with the community, you’re welcome to open a separate MetaTalk thread for it.
This month’s Site Update includes a particularly important announcement. You can find the last site update here.
I’m looking forward to your feedback and questions!
SPECIAL ANNOUNCEMENT
Since Kirkaracha joined our team 4 months ago, I have been shifting our priorities more towards the technical side. So far we have reviewed the site infrastructure and tackled small side projects which have allowed him to get more familiar with the current state of the site’s back end. After a few months of going over feature requests from staff, members, and the BIPOC Board, we have made the decision to focus resources on rebuilding the entire site from scratch to make it easier to maintain and to iterate. Goodbye ColdFusion!
Kirk started working on this several weeks ago to determine how complex it would be and has made a lot of progress already.
What to expect?
- For now, we are trying to replicate the UI, flow and features of the site as they are, with some minor accessibility and design improvements. Once the new site is live, we can start thinking about new features/changes.
- Kirk will put the new beta site on a test server and the next step will be to start gradually giving members access to test it as soon as it is ready.
- I’ll make a separate post about this once we’re ready for testing and share screenshots of the progress with the community at large there.
Profit & Loss
– You can find this month's P&L report here. The previous P&L reports are here.
Admin/Moderation
– We’re going ahead with the Trans Members Initiative as outlined in the past site update.
– A separate Meta to crowd-source questions from Trans members for the survey will be posted next Monday.
– We are getting close to both MeFi’s 25th Anniversary and Fundraising month, depending on how timing goes with the new site, we would like to plug all 3 of these together. I’ll keep you posted about this.
Tech
– Repaired Archives and made sure that they update
– Fixed several recurring tasks that were not running.
– Updated infodump and made sure that it is reloading correctly.
– Fixes to make sure contact form e-mails are not dropped by gmail.
– Frimble is updating and testing the site backup process
BIPOC Advisory Board
– The last meeting was postponed due to last minute scheduling conflicts and we’ll resume as soon as possible. I’ll confirm once the pending Meeting Minutes have been approved and posted.
If you have any questions or feedback not related to this particular update, please Contact Us instead. If you want to discuss a particular subject not covered here with the community, you’re welcome to open a separate MetaTalk thread for it.
What does rebuilding the site mean for the existing archive?
posted by plonkee at 3:36 PM on June 19 [7 favorites]
posted by plonkee at 3:36 PM on June 19 [7 favorites]
Mod note: wait, the site was still on coldfusion?
Yup, I had the same reaction back then.
I'm meeting with Kirk tomorrow so I'll drop by to answer questions related to the new site after that.
posted by loup (staff) at 3:40 PM on June 19 [3 favorites]
Yup, I had the same reaction back then.
I'm meeting with Kirk tomorrow so I'll drop by to answer questions related to the new site after that.
posted by loup (staff) at 3:40 PM on June 19 [3 favorites]
Huh, it's finally happening! This sounds like a massive project, is there a projected timeline for the new site going live? Is it going to be rolled out in waves or is it going to be a 'flip the switch' kind of deal?
posted by TwoWordReview at 4:32 PM on June 19 [2 favorites]
posted by TwoWordReview at 4:32 PM on June 19 [2 favorites]
Will this be a full migration or will there be an old site and a new site, which would explain the archive and backup items?
posted by one4themoment at 4:50 PM on June 19
posted by one4themoment at 4:50 PM on June 19
Awww… I kinda liked that MetaFilter was still running on ColdFusion.
This feels a bit like being an adult child whose parents are selling the car you grew up in. Yeah, it’s seen better days, and it lacks some of the things you expect in a car, or most of them, and there are stains on seats that you’d rather forget how they got there, but it’s still a link to your past. It just feels carlike in ways other cars just don’t.
posted by Kattullus at 4:54 PM on June 19 [53 favorites]
This feels a bit like being an adult child whose parents are selling the car you grew up in. Yeah, it’s seen better days, and it lacks some of the things you expect in a car, or most of them, and there are stains on seats that you’d rather forget how they got there, but it’s still a link to your past. It just feels carlike in ways other cars just don’t.
posted by Kattullus at 4:54 PM on June 19 [53 favorites]
This is exciting news! I am eager to see what develops.
posted by NotLost at 6:56 PM on June 19 [4 favorites]
posted by NotLost at 6:56 PM on June 19 [4 favorites]
It's always fun to find exceptions for "never rewrite from scratch" but if y'all can do it while keeping all the existing content and (general) features, I'm sure that will make everyone's life easier in the long run and lead to a better website.
Good luck!
posted by Nonsteroidal Anti-Inflammatory Drug at 7:05 PM on June 19 [10 favorites]
Good luck!
posted by Nonsteroidal Anti-Inflammatory Drug at 7:05 PM on June 19 [10 favorites]
What does rebuilding the site mean for the existing archive?
The plan is to migrate the existing archive.
This sounds like a massive project, is there a projected timeline for the new site going live? Is it going to be rolled out in waves or is it going to be a 'flip the switch' kind of deal?
We don't have a specific timeline because we haven't worked out the rollout details yet. Currently I'm working on a minimum viable product (MVP) of the main functionality of the major MetaFilter sites: logging in, adding posts, and making comments on posts.
Will this be a full migration or will there be an old site and a new site, which would explain the archive and backup items?
A full migration with everything on the new platform. Depending on how the summer goes, we might temporarily have some features on the old site and some on the new.
posted by kirkaracha (staff) at 7:12 PM on June 19 [25 favorites]
The plan is to migrate the existing archive.
This sounds like a massive project, is there a projected timeline for the new site going live? Is it going to be rolled out in waves or is it going to be a 'flip the switch' kind of deal?
We don't have a specific timeline because we haven't worked out the rollout details yet. Currently I'm working on a minimum viable product (MVP) of the main functionality of the major MetaFilter sites: logging in, adding posts, and making comments on posts.
Will this be a full migration or will there be an old site and a new site, which would explain the archive and backup items?
A full migration with everything on the new platform. Depending on how the summer goes, we might temporarily have some features on the old site and some on the new.
posted by kirkaracha (staff) at 7:12 PM on June 19 [25 favorites]
Will this make new and experimental features more possible in the future? Sounds like good possibility for growth.
posted by tiny frying pan at 7:17 PM on June 19 [4 favorites]
posted by tiny frying pan at 7:17 PM on June 19 [4 favorites]
This is a great development! I'm guessing it'll be a little bumpy at times, but in the end it will be worth it!
posted by Brandon Blatcher (staff) at 7:30 PM on June 19 [7 favorites]
posted by Brandon Blatcher (staff) at 7:30 PM on June 19 [7 favorites]
Mod note: Will this make new and experimental features more possible in the future? Sounds like good possibility for growth.
Yes, that's the main reason behind this decision.
posted by loup (staff) at 7:33 PM on June 19 [19 favorites]
Yes, that's the main reason behind this decision.
posted by loup (staff) at 7:33 PM on June 19 [19 favorites]
Kirkaracha needs a "staff" tag.
posted by NotLost at 7:36 PM on June 19 [28 favorites]
posted by NotLost at 7:36 PM on June 19 [28 favorites]
this is excellent news!
posted by Kybard at 8:12 PM on June 19 [1 favorite]
posted by Kybard at 8:12 PM on June 19 [1 favorite]
Great news! I’d be happy to be a beta tester if and when you need one - I’m great at finding novel ways of breaking things or doing things in ways that are unexpected.
posted by ashbury at 8:35 PM on June 19 [7 favorites]
posted by ashbury at 8:35 PM on June 19 [7 favorites]
I can volunteer some QA and front-end dev hours for sure! I have lots of accessibility experience.
posted by sixswitch at 8:55 PM on June 19 [4 favorites]
posted by sixswitch at 8:55 PM on June 19 [4 favorites]
Thank you, loup and kirkaracha, for the update!
While I share some of Kattullus's nostalgia, this is surely going to be a GREAT, great move.
Also, I think it's probably really good that you're aiming to pretty much replicate the existing site, rather than giving in to the temptation to redesign a lot of things right now. Moving to just about ANY other technology will make future changes and redesigns so much easier, and the sooner you're able to get that move completed, the sooner you'll be able to start thinking about what you MIGHT want to change.
You all are obviously doing a ton of work to nudge this creaky old site into better shape. Thank you so much for all your hard work, and for sharing this news with us.
posted by kristi at 9:10 PM on June 19 [18 favorites]
While I share some of Kattullus's nostalgia, this is surely going to be a GREAT, great move.
Also, I think it's probably really good that you're aiming to pretty much replicate the existing site, rather than giving in to the temptation to redesign a lot of things right now. Moving to just about ANY other technology will make future changes and redesigns so much easier, and the sooner you're able to get that move completed, the sooner you'll be able to start thinking about what you MIGHT want to change.
You all are obviously doing a ton of work to nudge this creaky old site into better shape. Thank you so much for all your hard work, and for sharing this news with us.
posted by kristi at 9:10 PM on June 19 [18 favorites]
This is both long overdue, exciting, and a little scary. On one hand, ColdFusion is a dinosaur, developers fluent in it are hard to come by, and the site could easily die if an asteroid hits Adobe unexpectedly EOLs it. On the other hand, I've seen so many sites derailed by gratuitous redesigns that turned off longtime members, even when the leadership put a lot of thought and care into them. I really hope this effort takes cues from the last MeFi redesign 10 years ago (!) and focuses on efficiency and reducing overhead while taking pains to make the new version as similar as possible to the current one both visually and functionally, without making too many radical UI changes or losing any notable features.
On a related note, I've been working on a proposal to incorporate some of the more popular and useful MeFi userscripts into the site code (including getting permission from the original authors). For example, plutor's handy MeFiQuote script that makes quoting super-easy even on mobile, or yourcelf's awesome GraphFi sidebar that makes navigating long threads a breeze (update here). Perhaps even something like Mute-a-Filter, to give people the option to de-emphasize specific users (rather than a complete block). While these various scripts used to be convenient add-ons for power users, they're now more obscure and harder to implement for the less technically-inclined and almost completely inaccessible to mobile devices. The idea was to implement the code for these scripts into the backend wholesale, so they'd cache locally and then run client-side automatically for everyone on all devices (with an option to toggle them off, ofc) without having to re-invent the wheel. This would also be an important step towards welcoming member contributions to the codebase. Depending on the extent of the redesign, though, it's possible most of these scripts will no longer work without a significant rewrite.
Rather than take a step backwards here for longtime users by breaking their favorite scripts, I wonder if it would be possible to target some for inclusion from the start, so that any misgivings about redesign changes can be counterbalanced by increased useful functionality for everyone (rather than just those who know how to get ancient userscripts working on modern devices). Can you talk more about how the proposed redesign might impact legacy scripts and whether including a handful of them as official features in this way is doable? (I can also post this as a separate MetaTalk thread if that works better).
posted by Rhaomi at 9:36 PM on June 19 [38 favorites]
On a related note, I've been working on a proposal to incorporate some of the more popular and useful MeFi userscripts into the site code (including getting permission from the original authors). For example, plutor's handy MeFiQuote script that makes quoting super-easy even on mobile, or yourcelf's awesome GraphFi sidebar that makes navigating long threads a breeze (update here). Perhaps even something like Mute-a-Filter, to give people the option to de-emphasize specific users (rather than a complete block). While these various scripts used to be convenient add-ons for power users, they're now more obscure and harder to implement for the less technically-inclined and almost completely inaccessible to mobile devices. The idea was to implement the code for these scripts into the backend wholesale, so they'd cache locally and then run client-side automatically for everyone on all devices (with an option to toggle them off, ofc) without having to re-invent the wheel. This would also be an important step towards welcoming member contributions to the codebase. Depending on the extent of the redesign, though, it's possible most of these scripts will no longer work without a significant rewrite.
Rather than take a step backwards here for longtime users by breaking their favorite scripts, I wonder if it would be possible to target some for inclusion from the start, so that any misgivings about redesign changes can be counterbalanced by increased useful functionality for everyone (rather than just those who know how to get ancient userscripts working on modern devices). Can you talk more about how the proposed redesign might impact legacy scripts and whether including a handful of them as official features in this way is doable? (I can also post this as a separate MetaTalk thread if that works better).
posted by Rhaomi at 9:36 PM on June 19 [38 favorites]
Thanks for the update.
I'm another that feels a bit nostalgic for the existing site architecture. To stretch Kattullus' analogy, MetaFilter is that old guy that refuses to update his car because it still does everything it needs to do, ignoring the crazy maintenance schedule it takes to keep it moving and that there are a hundred secret tricks you have to use to make it work at all.
I do hope the approach is to initially duplicate the existing UI and functionality, but in a more efficient way, rather than taking the opportunity to make wholesale changes during the re-write. I know that may take longer in the end, but it will allow for coding against a very clearly known and fixed set of expectations so there's no need to consult widely through the process. Assuming the re-write is done well, it should then be easy (relatively) to consult on and make changes without delaying the rollout of the new site. Anyway, that's what I'd do, but I'm totally unqualified to give advice on programming projects!
posted by dg at 9:44 PM on June 19 [5 favorites]
I'm another that feels a bit nostalgic for the existing site architecture. To stretch Kattullus' analogy, MetaFilter is that old guy that refuses to update his car because it still does everything it needs to do, ignoring the crazy maintenance schedule it takes to keep it moving and that there are a hundred secret tricks you have to use to make it work at all.
I do hope the approach is to initially duplicate the existing UI and functionality, but in a more efficient way, rather than taking the opportunity to make wholesale changes during the re-write. I know that may take longer in the end, but it will allow for coding against a very clearly known and fixed set of expectations so there's no need to consult widely through the process. Assuming the re-write is done well, it should then be easy (relatively) to consult on and make changes without delaying the rollout of the new site. Anyway, that's what I'd do, but I'm totally unqualified to give advice on programming projects!
posted by dg at 9:44 PM on June 19 [5 favorites]
Great news! I’d be happy to be a beta tester if and when you need one - I’m great at finding novel ways of breaking things or doing things in ways that are unexpected.
Same; please add me to the tester list.
posted by Hardcore Poser at 10:00 PM on June 19 [3 favorites]
Same; please add me to the tester list.
posted by Hardcore Poser at 10:00 PM on June 19 [3 favorites]
My development philosophy is that the site should be as easy to use as possible, both for members and admins. The goal is for admins to do a lot of site configuration and editing in regular web forms.
I do hope the approach is to initially duplicate the existing UI and functionality, but in a more efficient way, rather than taking the opportunity to make wholesale changes during the re-write.
That's it exactly. I'm trying to duplicate the existing sites' UI and functionality, but with cleaner code, more accessible code.
posted by kirkaracha (staff) at 10:02 PM on June 19 [24 favorites]
I do hope the approach is to initially duplicate the existing UI and functionality, but in a more efficient way, rather than taking the opportunity to make wholesale changes during the re-write.
That's it exactly. I'm trying to duplicate the existing sites' UI and functionality, but with cleaner code, more accessible code.
posted by kirkaracha (staff) at 10:02 PM on June 19 [24 favorites]
Hurrah! Matt, pb, and frimble got 25 years out of this, that is *epic*!
New stack deets please? Still in aws or migrating to gcp or azure? Multicloud tenancy? Will there be eastereggs about pancakes.
ETA: Is this refactor the news that couldn't be mentioned that Brandon mentioned earlier? Thanks!
posted by riverlife at 10:06 PM on June 19 [13 favorites]
New stack deets please? Still in aws or migrating to gcp or azure? Multicloud tenancy? Will there be eastereggs about pancakes.
ETA: Is this refactor the news that couldn't be mentioned that Brandon mentioned earlier? Thanks!
posted by riverlife at 10:06 PM on June 19 [13 favorites]
How exciting. Look forward to seeing it. Please preserve the infodump, and seek volunteers if needed.
Question: is there oversight from the new Board on this?
I'm sure they would want to input on the strategy, such as whether to build custom forum software from scratch, or start from one of the existing, open-source, community products and evolve it to the site's needs. Just to take an example raised by adrianhon a while back, Discourse is used by 20k communities, has 1000 contributors on GitHub, been around 10 years already, etc. Other options are available.
posted by Klipspringer at 11:46 PM on June 19 [10 favorites]
Question: is there oversight from the new Board on this?
I'm sure they would want to input on the strategy, such as whether to build custom forum software from scratch, or start from one of the existing, open-source, community products and evolve it to the site's needs. Just to take an example raised by adrianhon a while back, Discourse is used by 20k communities, has 1000 contributors on GitHub, been around 10 years already, etc. Other options are available.
posted by Klipspringer at 11:46 PM on June 19 [10 favorites]
we have made the decision to focus resources on rebuilding the entire site from scratch
As someone who has been writing software for over 25 years, this statement sent a cold shiver down by back. I absolutely understand why this decision was made, given that the site is built on a very old technology that almost no one uses anymore. Starting from scratch always seems like a good idea (No more technical debt! New efficiencies! Better tools! A modern tech stack!) Any yet: Here be dragons.
This is going to take much longer than you expect.
This is going to be much more complicated than you expect.
There will be many times during the process where you think you're done only to find that the interconnectedness of the architecture means that things are more complicated than you expected.
This is going to introduce many new bugs, because coders are human.
There will be hair loss and tears.
With all that said, if y'all are willing to keep the existing site up while developing the new one, and as long as there isn't a point at which you ask the community to move over to a Minimal Viable Product that is missing a ton of features in the interest of "moving forward", I say good luck and godspeed.
And we will throw you one hell of a party when it's done.
posted by gwint at 12:02 AM on June 20 [76 favorites]
As someone who has been writing software for over 25 years, this statement sent a cold shiver down by back. I absolutely understand why this decision was made, given that the site is built on a very old technology that almost no one uses anymore. Starting from scratch always seems like a good idea (No more technical debt! New efficiencies! Better tools! A modern tech stack!) Any yet: Here be dragons.
This is going to take much longer than you expect.
This is going to be much more complicated than you expect.
There will be many times during the process where you think you're done only to find that the interconnectedness of the architecture means that things are more complicated than you expected.
This is going to introduce many new bugs, because coders are human.
There will be hair loss and tears.
With all that said, if y'all are willing to keep the existing site up while developing the new one, and as long as there isn't a point at which you ask the community to move over to a Minimal Viable Product that is missing a ton of features in the interest of "moving forward", I say good luck and godspeed.
And we will throw you one hell of a party when it's done.
posted by gwint at 12:02 AM on June 20 [76 favorites]
On the other hand, I've seen so many sites derailed by gratuitous redesigns that turned off longtime members, even when the leadership put a lot of thought and care into them.
Yeah, but I think in this case we more or less know it has to be done, for the reasons you gave. Adobe could decide to kill CF any moment, and we absolutely know they'll do it, they care nothing for the health of the web in general, let alone our tiny corner of it. It's only a matter of time.
posted by JHarris at 1:50 AM on June 20 [22 favorites]
Yeah, but I think in this case we more or less know it has to be done, for the reasons you gave. Adobe could decide to kill CF any moment, and we absolutely know they'll do it, they care nothing for the health of the web in general, let alone our tiny corner of it. It's only a matter of time.
posted by JHarris at 1:50 AM on June 20 [22 favorites]
I'm trying to duplicate the existing sites' UI and functionalityWill this include the classic theme?
posted by Strutter Cane - United Planets Stilt Patrol at 2:07 AM on June 20 [25 favorites]
I assume the interim board is looking at the P&L but what percentage of the cash reserves does that -$17k represent?
Good luck w/ the technical project. I wondered about the Discourse option as well. I know that would be a harder option for the community but it might have a lot of long-term benefits.
posted by warriorqueen at 3:45 AM on June 20
Good luck w/ the technical project. I wondered about the Discourse option as well. I know that would be a harder option for the community but it might have a lot of long-term benefits.
posted by warriorqueen at 3:45 AM on June 20
we have made the decision to focus resources on rebuilding the entire site from scratch
Fucking hell. Big sentence.
{thinks}
{exhales slowly}
This will likely be the biggest technical project in MetaFilter's existence. Please, for the love of the ancient gods, keep expectations and timescales realistic (that's as much for the community as it is for the mods).
With the resources and people MetaFilter has, and how the current site - or sites, as each service is a different codebase - is set up, am not realistically expecting to see a stable, glitch-free and historical content-filled MetaFilter collection of sites on a different platform until some vague, impossible to define, point in 2025 or 2026. Even then, through necessity and constraints, a lot of the tweaks and refinements from the past quarter of a century, some passionately argued for here on MetaTalk, just aren't going to make it in.
Also, everything gwint said above.
posted by Wordshore at 3:59 AM on June 20 [12 favorites]
Fucking hell. Big sentence.
{thinks}
{exhales slowly}
This will likely be the biggest technical project in MetaFilter's existence. Please, for the love of the ancient gods, keep expectations and timescales realistic (that's as much for the community as it is for the mods).
With the resources and people MetaFilter has, and how the current site - or sites, as each service is a different codebase - is set up, am not realistically expecting to see a stable, glitch-free and historical content-filled MetaFilter collection of sites on a different platform until some vague, impossible to define, point in 2025 or 2026. Even then, through necessity and constraints, a lot of the tweaks and refinements from the past quarter of a century, some passionately argued for here on MetaTalk, just aren't going to make it in.
Also, everything gwint said above.
posted by Wordshore at 3:59 AM on June 20 [12 favorites]
Perhaps even something like Mute-a-Filter, to give people the option to de-emphasize specific users (rather than a complete block)
If I lose Mute A Filter, or the ability to gray out people, I may lose my ever loving mind LOL
posted by tiny frying pan at 4:56 AM on June 20 [2 favorites]
If I lose Mute A Filter, or the ability to gray out people, I may lose my ever loving mind LOL
posted by tiny frying pan at 4:56 AM on June 20 [2 favorites]
As a web dev with 15 YOE, I am quite surprised and concerned. Metafilter is a legacy product with a dedicated but shrinking userbase and extremely limited prospects for growth. A full rewrite is going to be dangerous and resource intensive for all the reasons gwint and Wordshore mention. Some parts of this process, which I would expect to extend over years, will be bumpy or broken despite everyone's best efforts, and will alienate additional portions of the existing userbase. I don't think there's any kind of existing metrics or plan to for growth that has a website built to a 2020s standard as a growth pillar. So, very ordinary bumps in the road on the journey really could be the final nail in the coffin for this website. It seems like an existential risk to me, and frankly, at odds with the decades long community spirit of this site to find out about it after it's already underway.
posted by Kwine at 5:14 AM on June 20 [20 favorites]
posted by Kwine at 5:14 AM on June 20 [20 favorites]
Is this the big news that Brandon was referring to in the previous thread?
posted by trig at 5:14 AM on June 20 [4 favorites]
posted by trig at 5:14 AM on June 20 [4 favorites]
we have made the decision to focus resources on rebuilding the entire site from scratch
#whatcouldpossiblygowrong
posted by sammyo at 5:15 AM on June 20 [1 favorite]
#whatcouldpossiblygowrong
posted by sammyo at 5:15 AM on June 20 [1 favorite]
Is this the big news that Brandon was referring to in the previous thread?
Yes.
posted by Brandon Blatcher (staff) at 5:18 AM on June 20 [12 favorites]
Yes.
posted by Brandon Blatcher (staff) at 5:18 AM on June 20 [12 favorites]
Normally I would be on the side of all the cautious devs saying a rewrite is a dangerous path, but this is Adobe we are talking about and given about all the recent press about their over-reach in claiming peoples data for training AI, the sooner we can dump them the better. I'm honestly surprised they haven't gotten around to making ColdFusion a monthly subscription at an eye watering price.
posted by Lanark at 5:23 AM on June 20 [23 favorites]
posted by Lanark at 5:23 AM on June 20 [23 favorites]
Exciting news! I don't know the tech is what has been keeping Metafilter back necessarily, but I'm positive that a redevelopment will generate lots of stimulating discussion and maybe even improve the site :)
I wish you a happy initial commit.
posted by dmh at 5:31 AM on June 20 [4 favorites]
I wish you a happy initial commit.
posted by dmh at 5:31 AM on June 20 [4 favorites]
Given how unpredictably long projects like this can take, what's the plan regarding any interim changes to functionality? For example, in the last big thread, I asked whether it would be possible to run a short experiment to implement a change that some users were asking for, where instead of deleting comments they would be hidden, with a mod comment applied. This would be potentially a very meaningful change, and might (again, potentially) help avert a lot of the current routes to moderation- or bad comment-related drama. It's something people have been asking for for a long time. There was even a suggestion about a relatively quick and dirty way of implementing it that might not be ideal as a final implementation, but could certainly work as a rough interim implementation. Especially for the purposes of trying this out to get a sense of whether it's actually worth pursuing seriously.
Brandon's answer was that it's not worth running that experiment "for the foreseeable future", because of a big upcoming site change. He said the soonest he expected this to be worth considering would be early next year (so at minimum 6 months).
6 months doesn't seem like a realistic amount of time to implement a site rewrite. Even if the two (part time!) devs pull it off (which would be amazing), that's still a long time to put any changes to site operation on hold. Especially since meaningful changes to site operation have already been on hold. For... years.
Basically: I can see why rewriting is important, and maybe it is the best way to go. But if it means that other important things are going to continue to be on hold for lord knows how many more years, that's a problem.
What is the plan for pulling this off while also investing real effort and energy on the current site?
posted by trig at 5:31 AM on June 20 [9 favorites]
Brandon's answer was that it's not worth running that experiment "for the foreseeable future", because of a big upcoming site change. He said the soonest he expected this to be worth considering would be early next year (so at minimum 6 months).
6 months doesn't seem like a realistic amount of time to implement a site rewrite. Even if the two (part time!) devs pull it off (which would be amazing), that's still a long time to put any changes to site operation on hold. Especially since meaningful changes to site operation have already been on hold. For... years.
Basically: I can see why rewriting is important, and maybe it is the best way to go. But if it means that other important things are going to continue to be on hold for lord knows how many more years, that's a problem.
What is the plan for pulling this off while also investing real effort and energy on the current site?
posted by trig at 5:31 AM on June 20 [9 favorites]
Hopefully, the new site will function well for those of us existing on older machines and OS’s. I’d hate to see MeFi head down the path of New-Shiny-But-Now-With-High-Sys-Reqs that so many sites seem to be turning to.
posted by Thorzdad at 5:35 AM on June 20 [6 favorites]
posted by Thorzdad at 5:35 AM on June 20 [6 favorites]
Given the massive scope of this project, the goal of transitioning MetaFilter into a new era of self-governance, and the desire to let go of things that are no longer serving the community well, I'd be very curious to hear what features and functionality will NOT be making it into the new build.
For there is no point is taking on unnecessary work that doesn't support the new direction and perpetuates the old ways and biases.
Put another way, after 25 years, MetaFilter is moving house! (Are we taking with us the hardly-used office, all the overgrown plants in broken pots, the spiral staircase with missing stairs?)
posted by iamkimiam at 6:05 AM on June 20 [14 favorites]
For there is no point is taking on unnecessary work that doesn't support the new direction and perpetuates the old ways and biases.
Put another way, after 25 years, MetaFilter is moving house! (Are we taking with us the hardly-used office, all the overgrown plants in broken pots, the spiral staircase with missing stairs?)
posted by iamkimiam at 6:05 AM on June 20 [14 favorites]
Metafilter is simple in concept but has a shitload of edge cases, especially after 25 years, so I expect the minimum viable product to turn up quickly, but a full implementation to be slow to come.
Ideally, the same database could be used with both the new and old codebases, so the new code could be improved incrementally and functionality can be added in a use case-priority way. In other words, use the new code when possible, use the old code when necessary. Users can switch code-bases at will.
At some point there will be no need left for the old code and it can just be switched off silently.
That's how I'd do it.
posted by seanmpuckett at 6:06 AM on June 20 [17 favorites]
Ideally, the same database could be used with both the new and old codebases, so the new code could be improved incrementally and functionality can be added in a use case-priority way. In other words, use the new code when possible, use the old code when necessary. Users can switch code-bases at will.
At some point there will be no need left for the old code and it can just be switched off silently.
That's how I'd do it.
posted by seanmpuckett at 6:06 AM on June 20 [17 favorites]
I'm resisting sharing specific opinions on how to do this work. Just want to reiterate what an ambitious and risky project this seems to be. I think it's worth doing!
posted by Nelson at 6:26 AM on June 20 [3 favorites]
posted by Nelson at 6:26 AM on June 20 [3 favorites]
What gwint and Nonsteroidal Anti-Inflammatory Drug said. Generally rewrites are a bad idea. But there are always exceptions, in this case it might be a good idea. It sounds like the ColdFusion hosting costs a bomb, so maybe there's a solid case for a rewrite.
But I'm a bit surprised such a big decision was taken before the change to non-profit.
Also there are going to be lots of small but heated questions about what features are going to be introduced, and what abandoned. I love "Recent Activity" but it seems like something that's a big performance load. The FanFare IMDB integration could be fiddly to redo. Not sure about the Music playlists.
Who's deciding all this stuff when there is no clear mechanism for making decisions anymore, with Jessamyn hands-off, no steering committee anymore, and no non-profit board? Is it the Interim Board?
posted by TheophileEscargot at 6:28 AM on June 20 [10 favorites]
But I'm a bit surprised such a big decision was taken before the change to non-profit.
Also there are going to be lots of small but heated questions about what features are going to be introduced, and what abandoned. I love "Recent Activity" but it seems like something that's a big performance load. The FanFare IMDB integration could be fiddly to redo. Not sure about the Music playlists.
Who's deciding all this stuff when there is no clear mechanism for making decisions anymore, with Jessamyn hands-off, no steering committee anymore, and no non-profit board? Is it the Interim Board?
posted by TheophileEscargot at 6:28 AM on June 20 [10 favorites]
The specific opinion that I am keen to share is that I'd be very happy to accept and would expect a feature freeze on the current website until it's up and running in the new framework, whatever that is.
That very much seems like the minimum necessary constraint to make sure that future development is possible and that there's always a safe platform for the site.
posted by ambrosen at 6:30 AM on June 20 [15 favorites]
That very much seems like the minimum necessary constraint to make sure that future development is possible and that there's always a safe platform for the site.
posted by ambrosen at 6:30 AM on June 20 [15 favorites]
loup: – We’re going ahead with the Trans Members Initiative as outlined in the past site update.
I'm looking forward to this, loup! Let me know how I can help.
posted by capricorn at 7:03 AM on June 20 [3 favorites]
I'm looking forward to this, loup! Let me know how I can help.
posted by capricorn at 7:03 AM on June 20 [3 favorites]
I am going to miss being able to horrify tech people by saying "It's still running ColdFusion!" like the punchline of a particularly spooky ghost story.
posted by restless_nomad (retired) at 7:12 AM on June 20 [50 favorites]
posted by restless_nomad (retired) at 7:12 AM on June 20 [50 favorites]
still running ColdFusion
Actually, it's kinda scarier than this! A biiiiig chunk of the site, from a code perspective, is actually Microsoft T-SQL. There was this short but widespread fad back then for putting "the code in the database" for what were perceived as performance reasons — even Perl and Tcl people were doing it to a certain extent — and the site is a victim of that. The web development community regained its sanity shortly after, but...
A rewrite into a modern architecture will improve things massively for development velocity. Normally, and in fact in about 99.99% of cases, I'd caution against ground-up rewrites/rearchitects as not paying back the effort going in, but this is an exception. And I say that as someone with a conflict of interest when it comes to ColdFusion.
So thank you Kirk for taking this on!
posted by majick at 7:52 AM on June 20 [29 favorites]
Actually, it's kinda scarier than this! A biiiiig chunk of the site, from a code perspective, is actually Microsoft T-SQL. There was this short but widespread fad back then for putting "the code in the database" for what were perceived as performance reasons — even Perl and Tcl people were doing it to a certain extent — and the site is a victim of that. The web development community regained its sanity shortly after, but...
A rewrite into a modern architecture will improve things massively for development velocity. Normally, and in fact in about 99.99% of cases, I'd caution against ground-up rewrites/rearchitects as not paying back the effort going in, but this is an exception. And I say that as someone with a conflict of interest when it comes to ColdFusion.
So thank you Kirk for taking this on!
posted by majick at 7:52 AM on June 20 [29 favorites]
Can I strongly suggest that the code for the rewrite be open-sourced? Maybe not right at the beginning, but at some point? There is certainly some risk of bikeshedding and more dev management that would come with that, but I'd imagine there are also a fair few developers who would be eager to contribute, including myself. I'm excited, but also this seems like a huge effort, especially given that the posting for the web developer specified that it was a 15 hour per week position.
posted by whir at 7:56 AM on June 20 [24 favorites]
posted by whir at 7:56 AM on June 20 [24 favorites]
Much of the cost of a rewrite usually lies in ensuring *identical* functionality, so it is well-worth it to figure out whether some weird edges can be sanded over rather than re-carved in loving detail.
(Also... doesn't it give anyone pause that Spolsky's essay is the one surefire link that's always included in a discussion of rewriting software? For a given anecdote it bears a lot of memetic weight. The conventional wisdom is so thoroughly established that there should be many more examples (and there obviously are, there must be), it's just weird that this is always always the specific writing that gets linked, as if we knew that there were a thriving population of purple-tailed parrot-thrashers out there, but every time we caught one, it was the same one we already banded.
Beyond anecdotes, there was a study that found that if more than 25% of the codebase needs to change, a rewrite is likely more economical than attempting incremental change. If you are inclined to dismiss that because it's just one study... it might be worth considering why one persuasive anecdotal essay by Spolsky is harder to dismiss than an actual study.)
posted by a faded photo of their beloved at 7:59 AM on June 20 [11 favorites]
(Also... doesn't it give anyone pause that Spolsky's essay is the one surefire link that's always included in a discussion of rewriting software? For a given anecdote it bears a lot of memetic weight. The conventional wisdom is so thoroughly established that there should be many more examples (and there obviously are, there must be), it's just weird that this is always always the specific writing that gets linked, as if we knew that there were a thriving population of purple-tailed parrot-thrashers out there, but every time we caught one, it was the same one we already banded.
Beyond anecdotes, there was a study that found that if more than 25% of the codebase needs to change, a rewrite is likely more economical than attempting incremental change. If you are inclined to dismiss that because it's just one study... it might be worth considering why one persuasive anecdotal essay by Spolsky is harder to dismiss than an actual study.)
posted by a faded photo of their beloved at 7:59 AM on June 20 [11 favorites]
We've talked about this before, so I'll drop my thoughts from last time. And I want to tell you both good luck. We're all counting on you.
posted by Tehhund at 8:04 AM on June 20 [12 favorites]
posted by Tehhund at 8:04 AM on June 20 [12 favorites]
IDK, I'm a software developer and Metafilter is just a message platform. I think some of the thoughts and concerns are way overblown, and a rewrite that doesn't take literally years to change a picture of a flag is due. Or remember the drama about the header?
I'm not saying that big rewrites are easy, but most legacy code- re-writes that are hard have edges cases like "we combined a message board with a payment engine and everyone's unique rateplan, usage history, and structure and comments need to be retained".
posted by The_Vegetables at 8:21 AM on June 20 [16 favorites]
I'm not saying that big rewrites are easy, but most legacy code- re-writes that are hard have edges cases like "we combined a message board with a payment engine and everyone's unique rateplan, usage history, and structure and comments need to be retained".
posted by The_Vegetables at 8:21 AM on June 20 [16 favorites]
Yeah, big rewrites are scary in general but 1. sometimes you gotta do scary things, and 2. this feels extremely necessary given the age of the codebase relative to our dev resources (we are not a bank that can afford to retain/train COBOL experts indefinitely, for example) and 3. A lot of Mefi functionality is relatively straightforward stuff that, while absolutely nontrivial, doesn't inspire fear in me as a PM. As a member, I'm glad to see we're mitigating the risks of being Still On Coldfusion (tm) and looking forward to seeing the work in progress!
posted by Tomorrowful at 8:32 AM on June 20 [10 favorites]
posted by Tomorrowful at 8:32 AM on June 20 [10 favorites]
What does "rebuilding the entire site from scratch" actually mean?
I was assuming that we would be reskinning and customizing some existing software like Discourse or phpBB, so we can just keep updating it with whatever security updates it needs.
Are we talking about writing a brand new messageboard system from scratch? Because then that has to be maintained in perpetuity by some developer on staff.
posted by TheophileEscargot at 8:43 AM on June 20 [4 favorites]
I was assuming that we would be reskinning and customizing some existing software like Discourse or phpBB, so we can just keep updating it with whatever security updates it needs.
Are we talking about writing a brand new messageboard system from scratch? Because then that has to be maintained in perpetuity by some developer on staff.
posted by TheophileEscargot at 8:43 AM on June 20 [4 favorites]
Will it be Agile?
posted by 922257033c4a0f3cecdbd819a46d626999d1af4a at 9:01 AM on June 20 [1 favorite]
posted by 922257033c4a0f3cecdbd819a46d626999d1af4a at 9:01 AM on June 20 [1 favorite]
Re: What does "rebuilding the entire site from scratch" actually mean?
Came in to express the same sentiment as TheophileEscargot a couple of comments upthread. I'll expand a little. Lots of commenters here have experience developing software and know that "rebuilding the entire site" can span a wide range of technical approaches. Think of is as a volume dial that goes from 1 to 10. At the "1" end, you are just using common libraries, migrating and cleaning all of the data, and rewriting the UI to work exactly like the way it does today. As you turn the dial, you get to other architectures like using existing framework(s), customizing existing message board software, or even configuring a SaaS platform. And as that dial turns up towards 10, the implementation time gets shorter because you are reusing more existing code, and the changes in how things work get greater because you are accepting someone else's UI/functionality choices - more user shock. I'm very curious about where on that dial Metafilter is heading.
posted by BlueTongueLizard at 9:07 AM on June 20 [5 favorites]
Came in to express the same sentiment as TheophileEscargot a couple of comments upthread. I'll expand a little. Lots of commenters here have experience developing software and know that "rebuilding the entire site" can span a wide range of technical approaches. Think of is as a volume dial that goes from 1 to 10. At the "1" end, you are just using common libraries, migrating and cleaning all of the data, and rewriting the UI to work exactly like the way it does today. As you turn the dial, you get to other architectures like using existing framework(s), customizing existing message board software, or even configuring a SaaS platform. And as that dial turns up towards 10, the implementation time gets shorter because you are reusing more existing code, and the changes in how things work get greater because you are accepting someone else's UI/functionality choices - more user shock. I'm very curious about where on that dial Metafilter is heading.
posted by BlueTongueLizard at 9:07 AM on June 20 [5 favorites]
Will it be Agile?
The used the term 'minimum viable product' above, which implies an 'agile' development cycle.
posted by The_Vegetables at 9:17 AM on June 20 [1 favorite]
The used the term 'minimum viable product' above, which implies an 'agile' development cycle.
posted by The_Vegetables at 9:17 AM on June 20 [1 favorite]
A biiiiig chunk of the site, from a code perspective, is actually Microsoft T-SQL.
Wait, like stored procedures? Or something weirder?
Sure, NVARCHAR(MAX) isn't the best way of all time to store all our comments and posts (if that's what we're doing, which is probably not the case), but, like, I am trying to wrap my head around how you get the data to put into an API response without some kind of stored proc.
posted by thecaddy at 9:29 AM on June 20
Wait, like stored procedures? Or something weirder?
Sure, NVARCHAR(MAX) isn't the best way of all time to store all our comments and posts (if that's what we're doing, which is probably not the case), but, like, I am trying to wrap my head around how you get the data to put into an API response without some kind of stored proc.
posted by thecaddy at 9:29 AM on June 20
This is such a huge thing to take on. For 15+ years people have been saying we need to do this but it hasn't felt possible to tackle, and it is inevitably going to lead to some problems, but also never-doing-it leads to problems. I hope people can support the effort and roll with it rather than knee jerk being like "aha, there's a problem, I knew this was a bad idea." We all want Mefi to survive and thrive, and we all know there are a lot of external headwinds the site faces (e.g. the march of time and technological change), that aren't magic-wand fixable by anything the site can realistically do, and which the site can't exempt itself from forever (despite the outside-of-time feeling some of us enjoy here). It's gonna be imperfect. It's a brave thing to attempt. I also hope/believe the goal will be to keep things the same as much as possible - but we all need to expect there will also probably be some tradeoffs etc and there is just no way to have everything good with nothing bad. Fingers crossed for luck. Viva Mefi.
posted by LobsterMitten (staff) at 9:45 AM on June 20 [50 favorites]
posted by LobsterMitten (staff) at 9:45 AM on June 20 [50 favorites]
Holy craptons, you guys! You certainly have a lot of interesting work ahead of you to ease Metafilter out of the ancient bony clutching hands of ColdFusion and into whatever new programming language you all decide to use. I recommend a Costco-sized bottle of aspirin to be added to the mod locker and recreational gummies for those who partake. Good luck!
posted by Lynsey at 9:47 AM on June 20 [2 favorites]
posted by Lynsey at 9:47 AM on June 20 [2 favorites]
Tehhund: "We've talked about this before, so I'll drop my thoughts from last time. And I want to tell you both good luck. We're all counting on you."
I was wondering about where the distinctive phrasing in the banner came from -- turns out it's a nod to the end of NotLost's post in your first link! Thanks.
posted by Rhaomi at 9:49 AM on June 20 [4 favorites]
I was wondering about where the distinctive phrasing in the banner came from -- turns out it's a nod to the end of NotLost's post in your first link! Thanks.
posted by Rhaomi at 9:49 AM on June 20 [4 favorites]
I just took a look at Kirk's LinkedIn profile which includes a bevy of web development experience ranging over 20 years.
I'm sure the well-actually comments are well intentioned, but I don't think this is a dude reading his first how-to PHP book armed with a copy of Notepad.
It will be challenging, I'm sure, and I wish you good luck. It will assuredly be a net positive.
posted by kbanas at 9:54 AM on June 20 [28 favorites]
I'm sure the well-actually comments are well intentioned, but I don't think this is a dude reading his first how-to PHP book armed with a copy of Notepad.
It will be challenging, I'm sure, and I wish you good luck. It will assuredly be a net positive.
posted by kbanas at 9:54 AM on June 20 [28 favorites]
Wait, like stored procedures?
Yep, it's a ton of business logic procs. Wire CF to what's returned by them. I have no idea why people got enamored of doing application code inside the database suddenly, even at that early stage of things where we weren't really good at understanding how to horizontally scale. I guess the idea was to put the code where the data was, instead of in an n-tier, because... something-something faster? Nobody was really thinking about maintainability or scalability in the olden days, just the new shiny architecture.
This kind of thing is why I'm massively in favor of a full, outright, rewrite even though to your average software developer it looks like risk and time cost; even though to your average project or product person it looks like feature cans getting kicked down the road without enough migration incrementalism.
This shit has got to go away. There's nearly no way to do anything rational or useful to this codebase without heroics. The fact that people have is a testament to their skill and fortitude, but it's a huge risk to the future of the site in the hands of the public.
posted by majick at 9:57 AM on June 20 [3 favorites]
Yep, it's a ton of business logic procs. Wire CF to what's returned by them. I have no idea why people got enamored of doing application code inside the database suddenly, even at that early stage of things where we weren't really good at understanding how to horizontally scale. I guess the idea was to put the code where the data was, instead of in an n-tier, because... something-something faster? Nobody was really thinking about maintainability or scalability in the olden days, just the new shiny architecture.
This kind of thing is why I'm massively in favor of a full, outright, rewrite even though to your average software developer it looks like risk and time cost; even though to your average project or product person it looks like feature cans getting kicked down the road without enough migration incrementalism.
This shit has got to go away. There's nearly no way to do anything rational or useful to this codebase without heroics. The fact that people have is a testament to their skill and fortitude, but it's a huge risk to the future of the site in the hands of the public.
posted by majick at 9:57 AM on June 20 [3 favorites]
I just glanced at the ColdFusion hashtag on Mastodon and discovered there's an open source implementation, Lucee, though it's built on Java, which, I don't trust Oracle as far as I could throw them. So there's a bit of an escape hatch if Adobe EOLs ColdFusion.
I've seen rewrites work. Some notes and suggestions:
* write tests.
* you can send a shadow copy of real traffic to the new server, and compare it in a few ways. I used GoReplay, though there are others.
* Microsoft T-SQL??!?
posted by Pronoiac at 9:59 AM on June 20 [3 favorites]
I've seen rewrites work. Some notes and suggestions:
* write tests.
* you can send a shadow copy of real traffic to the new server, and compare it in a few ways. I used GoReplay, though there are others.
* Microsoft T-SQL??!?
posted by Pronoiac at 9:59 AM on June 20 [3 favorites]
Happy to see MetaFilter finally moving to Active Server Pages 3.0.
posted by mbrubeck at 9:59 AM on June 20 [20 favorites]
posted by mbrubeck at 9:59 AM on June 20 [20 favorites]
though it's built on Java
So's the commercial ColdFusion; it's been a Java-based runtime for years (I don't know when it transitioned, I'd have to go look). With the way Oracle's been acting in the last couple of months, I wouldn't trust running a public site on any runtime on any flavor of any JRE from any vendor. While it's very likely almost everybody's in the clear, there's a non-zero risk of incurring significant legal expense.
CF as a language is kinda dead-end: as a niche language, there are fewer people who could send a PR, fewer people who could volunteer to help with the code, and fewer pieces of outside code that could be incorporated or adapted. There's only one vendor out there, although it's cool to hear there's an open source implementation!
I don't have an opinion on what to reimplement to, that's best left to the folks who have to do the work. I hope they work with a toolbox popular enough that they'll have a chance to get help.
'kay, time to go pay attention to my (not ColdFusion) product for a while. Best of luck to the team and, once again, huge props to Kirk for taking this on.
posted by majick at 10:10 AM on June 20 [3 favorites]
So's the commercial ColdFusion; it's been a Java-based runtime for years (I don't know when it transitioned, I'd have to go look). With the way Oracle's been acting in the last couple of months, I wouldn't trust running a public site on any runtime on any flavor of any JRE from any vendor. While it's very likely almost everybody's in the clear, there's a non-zero risk of incurring significant legal expense.
CF as a language is kinda dead-end: as a niche language, there are fewer people who could send a PR, fewer people who could volunteer to help with the code, and fewer pieces of outside code that could be incorporated or adapted. There's only one vendor out there, although it's cool to hear there's an open source implementation!
I don't have an opinion on what to reimplement to, that's best left to the folks who have to do the work. I hope they work with a toolbox popular enough that they'll have a chance to get help.
'kay, time to go pay attention to my (not ColdFusion) product for a while. Best of luck to the team and, once again, huge props to Kirk for taking this on.
posted by majick at 10:10 AM on June 20 [3 favorites]
* Microsoft T-SQL??!?
They're good databases, bront.
posted by ambrosen at 10:19 AM on June 20 [15 favorites]
They're good databases, bront.
posted by ambrosen at 10:19 AM on June 20 [15 favorites]
I hadn't noticed the "We can develop Mefi’s offspring" banner until Rhaomi pointed it out. "Offspring" is a terrible word choice! (MeFi isn't the thing that this new codebase is going to be eventually outliving.)
posted by nobody at 10:34 AM on June 20 [2 favorites]
posted by nobody at 10:34 AM on June 20 [2 favorites]
As a webdev I don't really see this as being that big of a dragon to slay. Metafilter is not that complicated: people post stuff, other people comment on it, there are tags and flags, mods can edit or delete posts and comments. All of these are solved problems with a metric ton of examples and best practices and full-blown libraries ready to go.
If one were to start from scratch, the UI/UX would probably be the hard part, but Kirk's said they're starting from what we aready have, so that's solved too.
Optimization will also be an issue, especially to keep costs down, but that's not going to kil the project.
There are a few edge cases, sure, but I don't think they'd have to be 100% ported over in the first release.
I'm not trying to minimize the work Kirk and others have ahead of them at all—in fact I'm hugely grateful to them for taking it on—I just don't think we need to be alarmists or doomsayers about it.
posted by signal at 10:41 AM on June 20 [9 favorites]
If one were to start from scratch, the UI/UX would probably be the hard part, but Kirk's said they're starting from what we aready have, so that's solved too.
Optimization will also be an issue, especially to keep costs down, but that's not going to kil the project.
There are a few edge cases, sure, but I don't think they'd have to be 100% ported over in the first release.
I'm not trying to minimize the work Kirk and others have ahead of them at all—in fact I'm hugely grateful to them for taking it on—I just don't think we need to be alarmists or doomsayers about it.
posted by signal at 10:41 AM on June 20 [9 favorites]
+1000 for open-sourcing the codebase, please. And right away, honestly, or at least right away after the nonprofit transition if you think there might be legal/volunteering issues beforehand. You don't have to accept patches from randoes, but i think it's important to actually involve the community here.
posted by adrienneleigh at 10:57 AM on June 20 [18 favorites]
posted by adrienneleigh at 10:57 AM on June 20 [18 favorites]
"Offspring" is a terrible word choice!
Perhaps a different metaphor is regeneration in Doctor Who. Except MetaFilter is attempting to regenerate straight from William Hartnell as the First Doctor, to Ncuti Gatwa as the Fifteenth Doctor.
posted by Wordshore at 11:33 AM on June 20 [5 favorites]
Perhaps a different metaphor is regeneration in Doctor Who. Except MetaFilter is attempting to regenerate straight from William Hartnell as the First Doctor, to Ncuti Gatwa as the Fifteenth Doctor.
posted by Wordshore at 11:33 AM on June 20 [5 favorites]
I'm very curious about where on that dial Metafilter is heading.
These go to eleven.
posted by chavenet at 11:37 AM on June 20 [2 favorites]
These go to eleven.
posted by chavenet at 11:37 AM on June 20 [2 favorites]
After the transition to the new codebase, should we start asking for horses instead of ponies?
posted by rikschell at 11:50 AM on June 20 [3 favorites]
posted by rikschell at 11:50 AM on June 20 [3 favorites]
Well then now that this place is a non-profit, the code should be fully open source. Get the effort onto github and let the local coders take a look and accept pull requests. We can at the very least contribute to a test suite. And there should be an extensive suite, reading Sqlite has 600 tests per line of code.
posted by sammyo at 11:59 AM on June 20 [8 favorites]
posted by sammyo at 11:59 AM on June 20 [8 favorites]
There was this short but widespread fad back then for putting "the code in the database"
Was that really “a short-lived fad?” I never really worked in that era but it’s my impression that it was the norm for a lot of things for some time. I guess maybe it’s a short part of the web era.
posted by atoxyl at 12:00 PM on June 20
Was that really “a short-lived fad?” I never really worked in that era but it’s my impression that it was the norm for a lot of things for some time. I guess maybe it’s a short part of the web era.
posted by atoxyl at 12:00 PM on June 20
I think I’m on record saying
a.) it wouldn’t be that hard to recreate something that works a lot like Metafilter in theory which makes it frustrating when the codebase is cited as a reason it’s hard to improve anything
b.) handling a project like this poorly could destroy the site. And migrating the old content is kind of “damned if you do, damned if you don’t,” because it makes the project a lot harder but not doing it also makes the new site significantly worse.
So, uh, good luck!
posted by atoxyl at 12:03 PM on June 20 [2 favorites]
a.) it wouldn’t be that hard to recreate something that works a lot like Metafilter in theory which makes it frustrating when the codebase is cited as a reason it’s hard to improve anything
b.) handling a project like this poorly could destroy the site. And migrating the old content is kind of “damned if you do, damned if you don’t,” because it makes the project a lot harder but not doing it also makes the new site significantly worse.
So, uh, good luck!
posted by atoxyl at 12:03 PM on June 20 [2 favorites]
After the transition to the new codebase, should we start asking for horses instead of ponies?
Przewalskis
posted by trig at 1:05 PM on June 20 [2 favorites]
Przewalskis
posted by trig at 1:05 PM on June 20 [2 favorites]
You don't have to accept patches from randoes, but i think it's important to actually involve the community here.
At least a bugtracker members can submit tickets to and search.
posted by trig at 1:06 PM on June 20 [7 favorites]
At least a bugtracker members can submit tickets to and search.
posted by trig at 1:06 PM on June 20 [7 favorites]
Interesting news. I think I think this is a good thing, and it seems extremely necessary for the long-term future of the site for it to have happened by some point (I just hope that point wasn't 5 years ago). But I've been around the block enough in software development for "let's rewrite it from scratch" to give me that "ghost walked over my grave" feeling. All I can say is that I hope you find ways to do this incrementally with as much synergistic reuse as possible. One reason for this, beyond obvious technical/project management ones, is that it doesn't seem entirely impossible that a future change in management structure would lead to at least a pause in effort on this project. That is, I imagine a hypothetical ED / board would want involvement with the decision-making around such a big (proportionally) project.
posted by advil at 1:11 PM on June 20 [2 favorites]
posted by advil at 1:11 PM on June 20 [2 favorites]
The timing of this surprises me, in that it is coming while we're still in that space between the old corporate structure and the new. We are, as I understand it, still trying to get things going under an Interim Board and the new entity is not yet operational and ownership of the site has not been transferred so the final version of the board doesn't exist to sign off on it. There is not yet an ED to supervise a project like this, and the current staff seems to be running on fumes at the best of times.
Is there a pressing and (relatively, since this is a long process) immediate need that this is meant to address or is there some other reason why this decision is being taken now?
posted by jacquilynne at 1:24 PM on June 20 [16 favorites]
Is there a pressing and (relatively, since this is a long process) immediate need that this is meant to address or is there some other reason why this decision is being taken now?
posted by jacquilynne at 1:24 PM on June 20 [16 favorites]
Special thanks to gwint, Wordshore, Nonsteroidal Anti-Inflammatory Drug, TheophileEscargot, seanmpuckett, majick, a faded photo of their beloved, Pronoiac, advil and many others for the advice provided. Thank you all for the offers to test as well! We’re very keen on getting testers involved sooner rather than later, hence why we are setting up the test server. I've been particularly interested in getting accessibility right since the beginning.
This project is at an early stage of development, so we can’t really say which weird edges can be sanded over rather than re-carved in loving detail yet, or address how themes will be handled, or many other what if scenarios. So far, enough progress has been made that the next clear step was to tell you all about it, and to let you know that we’ll have something to show soon.
For the time being, frimble will continue to maintain the old site as usual, kirkaracha will continue building the new site.
- what percentage of the cash reserves does that -$17k represent?
Nothing, we’re very steady with expenses and the -$17k in the P&L is because april’s payroll was sent on Apr 30 and posted on May 1st, and then May’s payroll was sent on the 30th and posted on the 31st. Jessamyn and I will keep looking at finances closely and we will let you know whenever there are any expenses out of the usual and what those are.
- Any yet: Here be dragons.
Yes, we’re very aware and I trust Kirkaracha’s ability to navigate uncharted territory for MeFi.
- Przewalskis
Yes, please.
- Can I strongly suggest that the code for the rewrite be open-sourced? Maybe not right at the beginning, but at some point?
Yes, that is the goal.
- is there some other reason why this decision is being taken now?
Yes, we finally have someone who has the expertise, right motivation and undivided attention to do it.
- Please, for the love of the ancient gods, keep expectations and timescales realistic
Yes, my main concern as well.
- Question: is there oversight from the new Board on this?
Not yet as they are busy. But I’d be happy to get them involved ASAP.
- Kirkaracha needs a "staff" tag.
On it.
posted by loup (staff) at 2:07 PM on June 20 [17 favorites]
This project is at an early stage of development, so we can’t really say which weird edges can be sanded over rather than re-carved in loving detail yet, or address how themes will be handled, or many other what if scenarios. So far, enough progress has been made that the next clear step was to tell you all about it, and to let you know that we’ll have something to show soon.
For the time being, frimble will continue to maintain the old site as usual, kirkaracha will continue building the new site.
- what percentage of the cash reserves does that -$17k represent?
Nothing, we’re very steady with expenses and the -$17k in the P&L is because april’s payroll was sent on Apr 30 and posted on May 1st, and then May’s payroll was sent on the 30th and posted on the 31st. Jessamyn and I will keep looking at finances closely and we will let you know whenever there are any expenses out of the usual and what those are.
- Any yet: Here be dragons.
Yes, we’re very aware and I trust Kirkaracha’s ability to navigate uncharted territory for MeFi.
- Przewalskis
Yes, please.
- Can I strongly suggest that the code for the rewrite be open-sourced? Maybe not right at the beginning, but at some point?
Yes, that is the goal.
- is there some other reason why this decision is being taken now?
Yes, we finally have someone who has the expertise, right motivation and undivided attention to do it.
- Please, for the love of the ancient gods, keep expectations and timescales realistic
Yes, my main concern as well.
- Question: is there oversight from the new Board on this?
Not yet as they are busy. But I’d be happy to get them involved ASAP.
- Kirkaracha needs a "staff" tag.
On it.
posted by loup (staff) at 2:07 PM on June 20 [17 favorites]
Wow, brave decision!
posted by theora55 at 2:07 PM on June 20 [4 favorites]
posted by theora55 at 2:07 PM on June 20 [4 favorites]
Re relying on one article from Spolsky: all smart programmers in my era had near their desk a copy of The Mythical Man-Month by Frederick Brooks, which eloquently described the "second-system effect" back in 1975.
People bring up these worries because, well, they're real. I've architected two complete rewrites on a company's main product and they went well... but I've also helped rescue two rewrites of other products which had become disasters.
And I worked for awhile at a company that did automatic grading of the Iowa Tests, and inevitably people came in, worked on the code for two years and declared it needed a complete rewrite. This was exactly the situation Spolsky warned about: millions of people take test grades very seriously, and the process is full of weird exceptions you don't even learn about in two years.
But hey, bulletin board software is not that. Bugs won't produce false test scores or hallucinating AIs or plane crashes. So good luck! Just remember, 90% of the bugs will be found in the x months allocated for development, and the remaining 90% will be found in the next x months.
posted by zompist at 2:25 PM on June 20 [14 favorites]
People bring up these worries because, well, they're real. I've architected two complete rewrites on a company's main product and they went well... but I've also helped rescue two rewrites of other products which had become disasters.
And I worked for awhile at a company that did automatic grading of the Iowa Tests, and inevitably people came in, worked on the code for two years and declared it needed a complete rewrite. This was exactly the situation Spolsky warned about: millions of people take test grades very seriously, and the process is full of weird exceptions you don't even learn about in two years.
But hey, bulletin board software is not that. Bugs won't produce false test scores or hallucinating AIs or plane crashes. So good luck! Just remember, 90% of the bugs will be found in the x months allocated for development, and the remaining 90% will be found in the next x months.
posted by zompist at 2:25 PM on June 20 [14 favorites]
Thanks loup. Could you speak to this question that was asked earlier?
Who's deciding all this stuff when there is no clear mechanism for making decisions anymore, with Jessamyn hands-off, no steering committee anymore, and no non-profit board? Is it the Interim Board?posted by trig at 2:31 PM on June 20 [14 favorites]
Are we taking with us the hardly-used office, all the overgrown plants in broken pots, the spiral staircase with missing stairs?
Maybe we could leave the missing stairs behind?
posted by dg at 2:44 PM on June 20 [5 favorites]
Maybe we could leave the missing stairs behind?
posted by dg at 2:44 PM on June 20 [5 favorites]
Thanks for making this decision, loup. It's good for this community. (A better-functioning site is an advantage in the coming ED search, too.)
"we finally have someone who has the expertise" hang on -
I'm one of the people who launched SportsFilter using the MetaFilter codebase in January 2002. Matt provided a ton of programming help before and after the launch and has always been willing to give us advice.
The MetaFilter codebase (at least the one we started from) is great at what it does, but we've found it's difficult to make changes. Matt added features to MetaFilter over time, so the code's not as easy to figure out as it would be if he'd designed and built it all at once, and there isn't any documentation. (I'm not complaining; the code is what I'd end up with if I started a site and added features here and there.) Also, the programmer types among us are more comfortable with PHP/MySQL than we are with ColdFusion, and we feel that ColdFusion limits our choices of ISPs and probably adds to our cost because ColdFusion is a commercial app and PHP/MySQL are free. We're switching to a PHP/MySQL solution this summer.
-- posted by kirkaracha at 10:12 PM on July 14, 2004
Thanks for sticking around, and for taking on this task now, kirkaracha.
posted by Iris Gambol at 2:48 PM on June 20 [34 favorites]
"we finally have someone who has the expertise" hang on -
I'm one of the people who launched SportsFilter using the MetaFilter codebase in January 2002. Matt provided a ton of programming help before and after the launch and has always been willing to give us advice.
The MetaFilter codebase (at least the one we started from) is great at what it does, but we've found it's difficult to make changes. Matt added features to MetaFilter over time, so the code's not as easy to figure out as it would be if he'd designed and built it all at once, and there isn't any documentation. (I'm not complaining; the code is what I'd end up with if I started a site and added features here and there.) Also, the programmer types among us are more comfortable with PHP/MySQL than we are with ColdFusion, and we feel that ColdFusion limits our choices of ISPs and probably adds to our cost because ColdFusion is a commercial app and PHP/MySQL are free. We're switching to a PHP/MySQL solution this summer.
-- posted by kirkaracha at 10:12 PM on July 14, 2004
Thanks for sticking around, and for taking on this task now, kirkaracha.
posted by Iris Gambol at 2:48 PM on June 20 [34 favorites]
I have also spent my whole career working on legacy codebases and managing rewrites, and all I can say is please, please, please stay away from react and other JavaScript based frameworks. I am currently attempting to land a disastrous react-based rewrite that I inherited (on month 26 of 3) in such a way that we can reasonably go back to the rails based system that we had before. I have NEVER seen a system that is fundamentally about delivering content work well in react or next.js or the framework of the day, but I have seen it fail many, many times.
posted by rockindata at 3:07 PM on June 20 [10 favorites]
posted by rockindata at 3:07 PM on June 20 [10 favorites]
Who's deciding all this stuff when there is no clear mechanism for making decisions anymore, with Jessamyn hands-off, no steering committee anymore, and no non-profit board? Is it the Interim Board?
Short answer: a combination of Jessamyn and I.
Long answer: I take care of decision making when it comes to daily operations, for more strategic things I will confer with Jessamyn and we'll come to a decision together.
posted by loup (staff) at 3:07 PM on June 20 [11 favorites]
Short answer: a combination of Jessamyn and I.
Long answer: I take care of decision making when it comes to daily operations, for more strategic things I will confer with Jessamyn and we'll come to a decision together.
posted by loup (staff) at 3:07 PM on June 20 [11 favorites]
Big news!
Weird question, off the top of my head: When migrating the archive, will deleted posts be migrated?
posted by eirias at 3:42 PM on June 20 [2 favorites]
Weird question, off the top of my head: When migrating the archive, will deleted posts be migrated?
posted by eirias at 3:42 PM on June 20 [2 favorites]
Wow, I'm curious to see this unfold. Also, can the mods please change the site banner to have less mystery meat in it? This is a huge announcement.
posted by twelve cent archie at 3:48 PM on June 20 [4 favorites]
posted by twelve cent archie at 3:48 PM on June 20 [4 favorites]
Boldly forward, y’all. Thank you for the update!
posted by cupcakeninja at 4:27 PM on June 20 [1 favorite]
posted by cupcakeninja at 4:27 PM on June 20 [1 favorite]
Good luck with this.
Since Kirkaracha joined our team 4 months ago, I have been shifting our priorities more towards the technical side.
Is the technical side really where the userbase would have voted to shift "our" priorities at this point in the site's history? Finances and membership seem to me obviously more important priorities, and I'm wondering a bit how it was decided by the site's leaders to shift priorities "more towards the technical side."
But ok, I guess this is what's happening now, and I'm assuming it will occupy much of the site leadership's time for the foreseeable future. I hope they'll still have time to think about gaining new members, and finding new ways to encourage existing members to post to the front page. Assigning a mod or user to make another LinkMe post on the blue right now would be nice, for one example, instead of waiting a month or whatever to try it again. LinkMe posts twice a month seem like a very useful idea.
and as long as there isn't a point at which you ask the community to move over to a Minimal Viable Product that is missing a ton of features in the interest of "moving forward"
*laughs* Yeah, that does seem to be how it's done these days. I hope any major changes to the site's UI that are made during this new process are given a chance for discussion before becoming mandatory. I know that ask is a huge pain in the ass for developers, but the site really can't afford to piss off any of its remaining users by dropping features or making big changes to the user experience without discussion in the community.
posted by mediareport at 6:39 PM on June 20 [4 favorites]
Since Kirkaracha joined our team 4 months ago, I have been shifting our priorities more towards the technical side.
Is the technical side really where the userbase would have voted to shift "our" priorities at this point in the site's history? Finances and membership seem to me obviously more important priorities, and I'm wondering a bit how it was decided by the site's leaders to shift priorities "more towards the technical side."
But ok, I guess this is what's happening now, and I'm assuming it will occupy much of the site leadership's time for the foreseeable future. I hope they'll still have time to think about gaining new members, and finding new ways to encourage existing members to post to the front page. Assigning a mod or user to make another LinkMe post on the blue right now would be nice, for one example, instead of waiting a month or whatever to try it again. LinkMe posts twice a month seem like a very useful idea.
and as long as there isn't a point at which you ask the community to move over to a Minimal Viable Product that is missing a ton of features in the interest of "moving forward"
*laughs* Yeah, that does seem to be how it's done these days. I hope any major changes to the site's UI that are made during this new process are given a chance for discussion before becoming mandatory. I know that ask is a huge pain in the ass for developers, but the site really can't afford to piss off any of its remaining users by dropping features or making big changes to the user experience without discussion in the community.
posted by mediareport at 6:39 PM on June 20 [4 favorites]
I'm excited about this update. Yes, it is going to be a large project and yes, there is all kinds of potential for things to go awry.
However, I think one of the biggest things that has held MeFi back, as a site, is the lack of ability to make changes easily. The sooner we move to a non-CF-based solution, especially if it's been designed with growth and such from the start, the sooner I think we can breathe a bit of new life into the old place.
Good luck to Kirkaracha and all those involved with the project! Also, nthing making it open source! I'm not a full-time developer, but I can chunks of stuff in PHP/MySQL, so if that's the stack we're moving to, I would love to be able to add PRs and such. Super exciting. :)
posted by juliebug at 7:15 PM on June 20 [6 favorites]
However, I think one of the biggest things that has held MeFi back, as a site, is the lack of ability to make changes easily. The sooner we move to a non-CF-based solution, especially if it's been designed with growth and such from the start, the sooner I think we can breathe a bit of new life into the old place.
Good luck to Kirkaracha and all those involved with the project! Also, nthing making it open source! I'm not a full-time developer, but I can chunks of stuff in PHP/MySQL, so if that's the stack we're moving to, I would love to be able to add PRs and such. Super exciting. :)
posted by juliebug at 7:15 PM on June 20 [6 favorites]
Mod note: Finances and membership seem to me obviously more important priorities
Yes, I agree. Managing healthier finances and handing over the reins of the site (in better shape than received) to the community have been our main focus these past months. Thanks to that, and everyone's support is that we can now budget for extra attention to the technical side.
posted by loup (staff) at 7:26 PM on June 20 [2 favorites]
Yes, I agree. Managing healthier finances and handing over the reins of the site (in better shape than received) to the community have been our main focus these past months. Thanks to that, and everyone's support is that we can now budget for extra attention to the technical side.
posted by loup (staff) at 7:26 PM on June 20 [2 favorites]
Since it didn't seem to get an answer (or maybe I missed it), I'd like to nth Strutter Cane's question about the classic theme, are there plans to keep that available? I know it's not likely the highest of priorities, but here I am, typing from the white background classic theme, hoping it'll still be here when all the hammering and sawing is done.
posted by Ghidorah at 8:22 PM on June 20 [4 favorites]
posted by Ghidorah at 8:22 PM on June 20 [4 favorites]
Yeah, same feeling here about the classic theme (though I use the one with the color backgrounds), but scrolling back it looks like loup did say "This project is at an early stage of development, so we can’t really say which weird edges can be sanded over [...] or address how themes will be handled, or many other what if scenarios," which seems like a fair response.
posted by nobody at 8:46 PM on June 20 [4 favorites]
posted by nobody at 8:46 PM on June 20 [4 favorites]
Metafilter: on month 26 of 3
posted by Greg_Ace at 8:57 PM on June 20 [2 favorites]
posted by Greg_Ace at 8:57 PM on June 20 [2 favorites]
Well, it seems that, no matter what tech platform MeFi ends up on, it can't be worse at handling themes than it currently is. As another classic theme user (colours all the way), I certainly hope it can be retained, but would merely whine quietly to myself if not.
I do wonder why nobody is saying what the platform is planned to be, but maybe that's because, well, this is MeFi and there are likely to be adverse reactions to pretty much anything ;-)
posted by dg at 11:04 PM on June 20 [3 favorites]
I do wonder why nobody is saying what the platform is planned to be, but maybe that's because, well, this is MeFi and there are likely to be adverse reactions to pretty much anything ;-)
posted by dg at 11:04 PM on June 20 [3 favorites]
The question as to why you would put application logic in the database: well, 25 years ago we were all reading Phil and Alex' Guide to Web Publishing where Phil Greenspun was doing exactly that with AolServer and Oracle. And it was cool! The alternative at that point was CGI (so clunky and resource intensive) or nascent and unfun application servers. If you came up by just hacking on shit rather than through some more structured approach that got you to think about architecture at a higher level, of COURSE you would do it this way. And even now, many things will never need the scalability that this architecture eventually impedes... anyway, don't judge harshly. The state of the art then was not now, and our references were different.
posted by i_am_joe's_spleen at 2:01 AM on June 21 [4 favorites]
posted by i_am_joe's_spleen at 2:01 AM on June 21 [4 favorites]
And slightly more on topic: as a literal greybeard, I also reflexively oppose wholesale rewrites, but it's different when the delta between existing implementation and modern practice is as large as it is now. Hopefully there might be a way to do it one layer at a time as seanmpuckett suggests, but if not (I imagine that database is very differently structured to what you'd do if you were starting now), tough shit I guess.
For my part, I would just caution that "like for like" replacements never end up that way. There are existing features no one cares about. There are features that seem unimportant until the old system is decommissioned and you realise what they did. And there are things that are time honoured but actually a bit shit. So let's not kid ourselves, the best outcome is "fairly similar and better in some respects." Which is probably more doable than exactly the same on a new technology stack anyway.
posted by i_am_joe's_spleen at 2:07 AM on June 21 [7 favorites]
For my part, I would just caution that "like for like" replacements never end up that way. There are existing features no one cares about. There are features that seem unimportant until the old system is decommissioned and you realise what they did. And there are things that are time honoured but actually a bit shit. So let's not kid ourselves, the best outcome is "fairly similar and better in some respects." Which is probably more doable than exactly the same on a new technology stack anyway.
posted by i_am_joe's_spleen at 2:07 AM on June 21 [7 favorites]
(did I say 25 years ago? more like 30. god i'm old.)
posted by i_am_joe's_spleen at 2:08 AM on June 21
posted by i_am_joe's_spleen at 2:08 AM on June 21
Thank you, thank you, thank you, from the bottom of my bottom, for not picking Discourse.
posted by flabdablet at 4:15 AM on June 21 [11 favorites]
posted by flabdablet at 4:15 AM on June 21 [11 favorites]
One thing I'll echo from others: It would be valuable for somebody to play the "bad cop" product manager here. "Build an MVP and then build it up" is absolutely the right approach - but, often, the last 10% of functionality takes 50% of the time. It is better to get a good metafilter over to new tech and re-evaluate from there than to wait for 100% of everybody's favorite ponies to make it over, if those ponies cause meaningful delay. Thou Shalt Not Let The Perfect be the Enemy of Good, so to speak, and if this work is worth doing, then "good" is "Metafilter is on a new platform and not on coldfusion," not "The new metafilter experience is 1:1 indistinguishable from the old."
posted by Tomorrowful at 5:20 AM on June 21 [8 favorites]
posted by Tomorrowful at 5:20 AM on June 21 [8 favorites]
I do wonder why nobody is saying what the platform is planned to be...
In a distant field, loup is tossing a coin while muttering “Heads it's Visual Basic, tails it's COBOL.”
The coin leaps into the air, arcs, falls, and lands on its edge, wedged between two rocks.
“Fuck."
loup sighs, gets out their Nokia 1100, calls Kirkaracha. “It's ALGOL 58. Write the whole goddamned thing from scratch in ALGOL 58. Do not read MetaTalk till 2030.”
As Kirkaracha starts to scream, loup hastily ends the call...
posted by Wordshore at 5:28 AM on June 21 [28 favorites]
In a distant field, loup is tossing a coin while muttering “Heads it's Visual Basic, tails it's COBOL.”
The coin leaps into the air, arcs, falls, and lands on its edge, wedged between two rocks.
“Fuck."
loup sighs, gets out their Nokia 1100, calls Kirkaracha. “It's ALGOL 58. Write the whole goddamned thing from scratch in ALGOL 58. Do not read MetaTalk till 2030.”
As Kirkaracha starts to scream, loup hastily ends the call...
posted by Wordshore at 5:28 AM on June 21 [28 favorites]
It seems like the obvious way to start a project like this would be for the non-profit board to consult with the community and the developer and then decide on the scope of the project and how it will be implemented. (Striking a balance of community involvement and buy-in and leveraging the relevant knowledge of the community, while having an empowered decision-making body that can choose the best course even though not everyone can get exactly what they want.)
Even with a great programming resource, decision-making, project management, and communication with the community are all critical to success. I would gently suggest that member opinion is divided on the current admin's track record in all of these areas. This seems like a significant project, why not invest up-front in the conditions that might make it more likely to succeed?
posted by snofoam at 6:08 AM on June 21 [24 favorites]
Even with a great programming resource, decision-making, project management, and communication with the community are all critical to success. I would gently suggest that member opinion is divided on the current admin's track record in all of these areas. This seems like a significant project, why not invest up-front in the conditions that might make it more likely to succeed?
posted by snofoam at 6:08 AM on June 21 [24 favorites]
not "The new metafilter experience is 1:1 indistinguishable from the old."
I think for some of us the ask isn't for 1:1 indistinguishability, but rather some process by which users are included in the decisions about which features make it into that first pass at the new site. There's not much faith left in management to make those decisions completely on their own, for some of us.
posted by mediareport at 6:54 AM on June 21 [6 favorites]
I think for some of us the ask isn't for 1:1 indistinguishability, but rather some process by which users are included in the decisions about which features make it into that first pass at the new site. There's not much faith left in management to make those decisions completely on their own, for some of us.
posted by mediareport at 6:54 AM on June 21 [6 favorites]
My concern is with the decision to maintain a custom-coded site, which will mean we continue to lag far behind current web standards for/what users expect to see for tools. Things like a wysiwyg post editor and being able to save a draft, autofill @mentions, decent mobile editing, etc. As a small site, custom development is probably never ever going to close this gap.
Basically, what is the site strategy? If it’s “keep all the current members happy as they gradually leave” then investing in a custom same-same site is fine. If it’s ever growth, the continued and growing gap in tech would be a bigger barrier and I think effort on that front would be better.
Listening to people about what they love now is important and if the board/ED/staff are committed to NOT growing, I think this project is fine. But if it is growth, I am not sure this move makes sense. It’s going to take a ton of effort and that will be, realistically, at least a year of focusing on “what people in MetaTalk care about” with still zero - ZERO!! - real effort to promote the site at all.
If this is a decision, I’m fine with that. But if it isn’t, that is incredibly frustrating.
Also, has there been any risk analysis? I have no doubt kirkaracha is The person to lead this effort, but what happens if he leaves for any reason at a halfway point?
posted by warriorqueen at 7:39 AM on June 21 [17 favorites]
Basically, what is the site strategy? If it’s “keep all the current members happy as they gradually leave” then investing in a custom same-same site is fine. If it’s ever growth, the continued and growing gap in tech would be a bigger barrier and I think effort on that front would be better.
Listening to people about what they love now is important and if the board/ED/staff are committed to NOT growing, I think this project is fine. But if it is growth, I am not sure this move makes sense. It’s going to take a ton of effort and that will be, realistically, at least a year of focusing on “what people in MetaTalk care about” with still zero - ZERO!! - real effort to promote the site at all.
If this is a decision, I’m fine with that. But if it isn’t, that is incredibly frustrating.
Also, has there been any risk analysis? I have no doubt kirkaracha is The person to lead this effort, but what happens if he leaves for any reason at a halfway point?
posted by warriorqueen at 7:39 AM on June 21 [17 favorites]
I think warriorqueen has done a good job of articulating the misgivings I was feeling when I asked my question of 'why now?'
'Because we have the technical resources to do it now' is a very fine answer to the question of 'why are we doing some minor fixes to the code base right now when we haven't done it in the past?' and not at all a good answer to the question 'why are we making a huge strategic decision in the absence of an actual strategy?'
The decision to just rebuild MetaFilter-that-is on a new tech stack is a massive strategic decision and 'because we can' is not strategic analysis.
posted by jacquilynne at 8:08 AM on June 21 [15 favorites]
'Because we have the technical resources to do it now' is a very fine answer to the question of 'why are we doing some minor fixes to the code base right now when we haven't done it in the past?' and not at all a good answer to the question 'why are we making a huge strategic decision in the absence of an actual strategy?'
The decision to just rebuild MetaFilter-that-is on a new tech stack is a massive strategic decision and 'because we can' is not strategic analysis.
posted by jacquilynne at 8:08 AM on June 21 [15 favorites]
I think this is a good decision. I think that it's good that current leadership is willing and able to make big decisions.
posted by 4th number at 8:16 AM on June 21 [4 favorites]
posted by 4th number at 8:16 AM on June 21 [4 favorites]
that will be, realistically, at least a year of focusing on “what people in MetaTalk care about” with still zero - ZERO!! - real effort to promote the site at all
That's what bothers/worries me most about this. Not the idea of upgrading tech-wise -- I'm willing to hope kirkaracha will do a better job managing and implementing this project than loup/frimble have done on previous ones.
But if this is going to be the only direction of action for the next significant chunk of time, and progress on membership and fundraising (and... basic responsiveness to the community) is going to keep getting the "sorry, not enough resources" treatment, then that's not good.
It is actually really refreshing to see action and energy and initiative. I appreciate that a lot! It's been sorely missed! But it looks like this is coming mostly from kirkaracha. Is initiative and action on any other front coming soon from anybody else on the staff, or is that all continuing to be on hold until some indefinite, hypothetical time in the future?
posted by trig at 8:17 AM on June 21 [8 favorites]
That's what bothers/worries me most about this. Not the idea of upgrading tech-wise -- I'm willing to hope kirkaracha will do a better job managing and implementing this project than loup/frimble have done on previous ones.
But if this is going to be the only direction of action for the next significant chunk of time, and progress on membership and fundraising (and... basic responsiveness to the community) is going to keep getting the "sorry, not enough resources" treatment, then that's not good.
It is actually really refreshing to see action and energy and initiative. I appreciate that a lot! It's been sorely missed! But it looks like this is coming mostly from kirkaracha. Is initiative and action on any other front coming soon from anybody else on the staff, or is that all continuing to be on hold until some indefinite, hypothetical time in the future?
posted by trig at 8:17 AM on June 21 [8 favorites]
Yes I definitely should have said, seeing enthusiasm and action is great. I just would like it to bear fruit.
posted by warriorqueen at 8:21 AM on June 21 [3 favorites]
posted by warriorqueen at 8:21 AM on June 21 [3 favorites]
I don't handle change well, so if there are too many noticeable differences (and believe me; I notice differences) in interface, design and/or function then I might have to finally say goodbye. But I also recognize the necessity for the sake of the site itself, so I wish you all the best. Good luck.
posted by Faint of Butt at 8:56 AM on June 21 [2 favorites]
posted by Faint of Butt at 8:56 AM on June 21 [2 favorites]
from scratch in ALGOL 58
I was thinking that ChatGPT could just whip up some perl scripts.
posted by paper chromatographologist at 9:04 AM on June 21 [3 favorites]
I was thinking that ChatGPT could just whip up some perl scripts.
posted by paper chromatographologist at 9:04 AM on June 21 [3 favorites]
One thing I'll echo from others: It would be valuable for somebody to play the "bad cop" product manager here. "Build an MVP and then build it up" is absolutely the right approach - but, often, the last 10% of functionality takes 50% of the time.
and
I think for some of us the ask isn't for 1:1 indistinguishability, but rather some process by which users are included in the decisions about which features make it into that first pass at the new site.
I think it would be great—assuming we are taking an agile approach to this effort—to try to put together the backlog of features and epics and share that out to the community to comment on. At the very least an epic-level roadmap would give a good indication of priorities. A Metatalk thread would probably be a disaster, but a survey-style collection of feedback could give a good idea of what the highest priority features are without turning up the heat too high.
posted by thecaddy at 9:20 AM on June 21 [7 favorites]
and
I think for some of us the ask isn't for 1:1 indistinguishability, but rather some process by which users are included in the decisions about which features make it into that first pass at the new site.
I think it would be great—assuming we are taking an agile approach to this effort—to try to put together the backlog of features and epics and share that out to the community to comment on. At the very least an epic-level roadmap would give a good indication of priorities. A Metatalk thread would probably be a disaster, but a survey-style collection of feedback could give a good idea of what the highest priority features are without turning up the heat too high.
posted by thecaddy at 9:20 AM on June 21 [7 favorites]
(epic: Agile-speak for "a body of work that can be broken down into specific tasks (called user stories) based on the needs/requests of customers or end-users")
posted by trig at 9:36 AM on June 21 [1 favorite]
posted by trig at 9:36 AM on June 21 [1 favorite]
It seems like the obvious way to start a project like this would be for the non-profit board to consult with the community and the developer and then decide on the scope of the project and how it will be implemented.
I will be pulling this thread from my RA because it is blessedly no longer my fight but I am sympathetic to "we have a million-item-long backlog of fixes requested and we finally have enough developer hours to address it; do we dig ourselves another year into tech debt by working on this codebase or do we anticipate future need and do the rebuild before it becomes a life-or-death situation even though the current state of management is in flux with no firm timeline?" Because the state of flux has existed for years now, one way or another, and we don't have an end date. Seems to me like there's a vast amount of prep work that's going to need doing before most of the community-facing decisions are made and it's probably better to just get started.
posted by restless_nomad (retired) at 9:37 AM on June 21 [35 favorites]
I will be pulling this thread from my RA because it is blessedly no longer my fight but I am sympathetic to "we have a million-item-long backlog of fixes requested and we finally have enough developer hours to address it; do we dig ourselves another year into tech debt by working on this codebase or do we anticipate future need and do the rebuild before it becomes a life-or-death situation even though the current state of management is in flux with no firm timeline?" Because the state of flux has existed for years now, one way or another, and we don't have an end date. Seems to me like there's a vast amount of prep work that's going to need doing before most of the community-facing decisions are made and it's probably better to just get started.
posted by restless_nomad (retired) at 9:37 AM on June 21 [35 favorites]
CGI (so clunky and resource intensive)
Weirdly enough, mod_perl and Apache::DBI existed for a couple of years before P&A came out, and if I remember properly — the mists of time haven't been kind to my ability to remember some of my earliest projects — both were in Debian main. In fact, I think that's back before Alex really monkeyed up the way Apache was configured on Debian, so it was easier to do for newbies.
Anyway, I definitely think there was some division in cultures between folks who followed along with dotcommy ACS-like stuff like Phil was doing and people who had to turn a UNIX box—or lots of them—into an application, then a bigger application, then a bigger application.
It's funny and interesting, how those early days played out. So many stories about how all the different circles and cultural bubbles and "scenes" did their thing. I'm eternally fascinated by these technical subcultures, going all the way back to the 8-bit days.
Anyway, more on topic: the rewrite is a big, important thing to do, and it's an important thing to do for all the reasons some of the semi-detractors are citing. The sooner we get this done, the sooner we can actually do the things that implement the community's needs. It's a glacial, difficult process right now because this hasn't been done. Just getting a banner or something up had a headwind. Getting BIPOC Board notes takes some doing, and that's on the "easy" CF side of the site.
Doing the kinds of things that will grow the site, that will help existing members want to stick around, that will enable the governance and advisory system (of which I'm a part!) to perform its work, that will perpetuate the existence of Metafilter, it's all gated by Metafilter itself being a thing that can be worked on quickly and easily.
posted by majick at 10:13 AM on June 21 [5 favorites]
Weirdly enough, mod_perl and Apache::DBI existed for a couple of years before P&A came out, and if I remember properly — the mists of time haven't been kind to my ability to remember some of my earliest projects — both were in Debian main. In fact, I think that's back before Alex really monkeyed up the way Apache was configured on Debian, so it was easier to do for newbies.
Anyway, I definitely think there was some division in cultures between folks who followed along with dotcommy ACS-like stuff like Phil was doing and people who had to turn a UNIX box—or lots of them—into an application, then a bigger application, then a bigger application.
It's funny and interesting, how those early days played out. So many stories about how all the different circles and cultural bubbles and "scenes" did their thing. I'm eternally fascinated by these technical subcultures, going all the way back to the 8-bit days.
Anyway, more on topic: the rewrite is a big, important thing to do, and it's an important thing to do for all the reasons some of the semi-detractors are citing. The sooner we get this done, the sooner we can actually do the things that implement the community's needs. It's a glacial, difficult process right now because this hasn't been done. Just getting a banner or something up had a headwind. Getting BIPOC Board notes takes some doing, and that's on the "easy" CF side of the site.
Doing the kinds of things that will grow the site, that will help existing members want to stick around, that will enable the governance and advisory system (of which I'm a part!) to perform its work, that will perpetuate the existence of Metafilter, it's all gated by Metafilter itself being a thing that can be worked on quickly and easily.
posted by majick at 10:13 AM on June 21 [5 favorites]
If the code will be open source I envisage a day when people fork the code and we have competing MetaFilters (hello MonkeyFilter!). Part of me wants to suggest federating MetaFilter so these new MetaFilters can all talk to one another ... and eventually defederate from one another and block and ban each other.
On a slightly more serious note, this is good news. And I wish kirkaracha all the best. I too can volunteer to test if more people are needed.
posted by terrapin at 10:16 AM on June 21 [5 favorites]
On a slightly more serious note, this is good news. And I wish kirkaracha all the best. I too can volunteer to test if more people are needed.
posted by terrapin at 10:16 AM on June 21 [5 favorites]
(epic: Agile-speak for "a body of work that can be broken down into specific tasks (called user stories) based on the needs/requests of customers or end-users")
Oof. That's what I get for writing a comment about this on my lunch break. At least I didn't write "spin up a user story."
But anyway, what I'm recommending is listing out the high-level features along the lines of "See the front page of Metafilter" or "Post a comment on an Ask Metafilter thread" or "Create a new post for Projects", so the community can see what the approach will be.
posted by thecaddy at 10:25 AM on June 21
Oof. That's what I get for writing a comment about this on my lunch break. At least I didn't write "spin up a user story."
But anyway, what I'm recommending is listing out the high-level features along the lines of "See the front page of Metafilter" or "Post a comment on an Ask Metafilter thread" or "Create a new post for Projects", so the community can see what the approach will be.
posted by thecaddy at 10:25 AM on June 21
The sooner we get this done, the sooner we can actually do the things that implement the community's needs. It's a glacial, difficult process right now because this hasn't been done. Just getting a banner or something up had a headwind. Getting BIPOC Board notes takes some doing
As one of the semi-detractors I want to say again that I do understand why this is important and don't actually disagree with the initiative. What I disagree with is the idea that while kirkaracha is working on this, and also while the transition to a nonprofit happens, nothing else can be accomplished.
Granted: I don't actually know that site development is going to be the only or primary priority. I'm worried about it because I haven't heard otherwise, and because this site has a years-long track record of dragging its feet on every fucking thing, including really vital things like fundraising and publicity. And including tiny things like posting BIPOC Board notes (current backlog: 6 fucking months) or posting a banner. Which - however much the processes for those could be more convenient than hand-updating some HTML... we're talking about shaving minutes off a minutes-long task. And yet even those minutes-long tasks inspired multiple Metas about why on earth they're weren't happening. Because neglecting even small tasks, and citing some minor process friction as an excuse - or citing a lack of resources but then not doing anything about that lack of resources - says that ultimately neglect is the desired outcome, not something that the people in charge are fighting against.
So I'm not worried about kirkaracha, I'm worried about the actual mod/admin stuff saying "welp, that's all we're going to focus on until the rewrite is done and the nonprofit transition happens. Someday." Even though the majority of the workload for both of those projects is (if I understand correctly) not on the plates of the mod/admin staff.
If I hear that fundraising, marketing, community responsiveness, and hell, some serious consideration of the extensive community feedback about moderation practices are going to happen in parallel to the rewrite, I'll be happy. Not necessarily optimistic that they'll be done well, but whatever, at this point I'll take initiative and energy! But we've been in stasis for too many years to just keep deferring and deferring this work until... someday.
posted by trig at 10:40 AM on June 21 [19 favorites]
As one of the semi-detractors I want to say again that I do understand why this is important and don't actually disagree with the initiative. What I disagree with is the idea that while kirkaracha is working on this, and also while the transition to a nonprofit happens, nothing else can be accomplished.
Granted: I don't actually know that site development is going to be the only or primary priority. I'm worried about it because I haven't heard otherwise, and because this site has a years-long track record of dragging its feet on every fucking thing, including really vital things like fundraising and publicity. And including tiny things like posting BIPOC Board notes (current backlog: 6 fucking months) or posting a banner. Which - however much the processes for those could be more convenient than hand-updating some HTML... we're talking about shaving minutes off a minutes-long task. And yet even those minutes-long tasks inspired multiple Metas about why on earth they're weren't happening. Because neglecting even small tasks, and citing some minor process friction as an excuse - or citing a lack of resources but then not doing anything about that lack of resources - says that ultimately neglect is the desired outcome, not something that the people in charge are fighting against.
So I'm not worried about kirkaracha, I'm worried about the actual mod/admin stuff saying "welp, that's all we're going to focus on until the rewrite is done and the nonprofit transition happens. Someday." Even though the majority of the workload for both of those projects is (if I understand correctly) not on the plates of the mod/admin staff.
If I hear that fundraising, marketing, community responsiveness, and hell, some serious consideration of the extensive community feedback about moderation practices are going to happen in parallel to the rewrite, I'll be happy. Not necessarily optimistic that they'll be done well, but whatever, at this point I'll take initiative and energy! But we've been in stasis for too many years to just keep deferring and deferring this work until... someday.
posted by trig at 10:40 AM on June 21 [19 favorites]
We have already been told that even small changes to mod behavior are out of the question for the next year.
posted by bowbeacon at 10:52 AM on June 21 [6 favorites]
posted by bowbeacon at 10:52 AM on June 21 [6 favorites]
Yup.
posted by trig at 10:53 AM on June 21 [3 favorites]
posted by trig at 10:53 AM on June 21 [3 favorites]
'Because we have the technical resources to do it now' is a very fine answer to the question of 'why are we doing some minor fixes to the code base right now when we haven't done it in the past?' and not at all a good answer to the question 'why are we making a huge strategic decision in the absence of an actual strategy?'
The decision to just rebuild MetaFilter-that-is on a new tech stack is a massive strategic decision and 'because we can' is not strategic analysis.
I mean - in the post itself, Loup says that they and the technical staff reviewed feature requests over a period of months from users and determined the best way to do any of that is to get off ColdFusion - which going back to Mathowie's days of owning the site has been cited as the reason why things are so hard to change here.
Going back to the steering committee days too, there was a lot of desire for volunteers to help with coding and the unruly codebase that breaks easily, lack of documentation, and generally creaky technology we have here was cited as a reason why that was not really feasible. That coupled with the fact Coldfusion may become EOL at any point is another strategic risk.
I also think if we are to get new members and growth, we have to consider having a platform that can accommodate media other than text, social media integration, and a whole host of other standard site features for communities in 2024. This site is not really marketable to a 2024 audience as-is.
All of that suggests that building a platform that can accommodate change is a very strategic decision based on the technical analysis provided by the people we pay to provide that. The change to how flagging looked on this site took months because the code is not optimized to be changed.
I think the decision to root the MVP in only recreating what we already have may not be a great one, but I would like Kirk to get a shot at actually developing that MVP and presenting it to the community before deciding it's a bad decision.
posted by openhearted at 10:54 AM on June 21 [3 favorites]
The decision to just rebuild MetaFilter-that-is on a new tech stack is a massive strategic decision and 'because we can' is not strategic analysis.
I mean - in the post itself, Loup says that they and the technical staff reviewed feature requests over a period of months from users and determined the best way to do any of that is to get off ColdFusion - which going back to Mathowie's days of owning the site has been cited as the reason why things are so hard to change here.
Going back to the steering committee days too, there was a lot of desire for volunteers to help with coding and the unruly codebase that breaks easily, lack of documentation, and generally creaky technology we have here was cited as a reason why that was not really feasible. That coupled with the fact Coldfusion may become EOL at any point is another strategic risk.
I also think if we are to get new members and growth, we have to consider having a platform that can accommodate media other than text, social media integration, and a whole host of other standard site features for communities in 2024. This site is not really marketable to a 2024 audience as-is.
All of that suggests that building a platform that can accommodate change is a very strategic decision based on the technical analysis provided by the people we pay to provide that. The change to how flagging looked on this site took months because the code is not optimized to be changed.
I think the decision to root the MVP in only recreating what we already have may not be a great one, but I would like Kirk to get a shot at actually developing that MVP and presenting it to the community before deciding it's a bad decision.
posted by openhearted at 10:54 AM on June 21 [3 favorites]
This site is not really marketable to a 2024 audience as-is.
I just want to push back against this and then I'm staying out of the thread for a while.
This site is plenty marketable. Sure, it could be better, but it won't be better for at least another few years if we're being realistic, so maybe let's at least start and find out.
This is a human-moderated, hate-rejecting, non-billionaire owned, community-funded, text-focused international discussion site covering an huge range of topics. We probably have the best all-purposes Q&A and advice site on the internet; no need to search out some specialized subreddit or forum, just ask and get useful answers about anything. We have a userbase that generally speaks in coherent sentences. Nobody's data is being tracked or sold. These are not small things. They're not for everybody - maybe they're for 0.02% of everybody - but there are a fuckton of people online we're not even trying to reach, and 0.02% of a fuckton is way more than we need.
I don't think the biggest problem is people coming here and bouncing off, I think the biggest problem is people having no idea this place exists and is worth checking out in the first place. I could be wrong about that! But ffs let's try finding out. Instead of waiting for that golden "someday" when we'll finally be worthy.
posted by trig at 11:09 AM on June 21 [61 favorites]
I just want to push back against this and then I'm staying out of the thread for a while.
This site is plenty marketable. Sure, it could be better, but it won't be better for at least another few years if we're being realistic, so maybe let's at least start and find out.
This is a human-moderated, hate-rejecting, non-billionaire owned, community-funded, text-focused international discussion site covering an huge range of topics. We probably have the best all-purposes Q&A and advice site on the internet; no need to search out some specialized subreddit or forum, just ask and get useful answers about anything. We have a userbase that generally speaks in coherent sentences. Nobody's data is being tracked or sold. These are not small things. They're not for everybody - maybe they're for 0.02% of everybody - but there are a fuckton of people online we're not even trying to reach, and 0.02% of a fuckton is way more than we need.
I don't think the biggest problem is people coming here and bouncing off, I think the biggest problem is people having no idea this place exists and is worth checking out in the first place. I could be wrong about that! But ffs let's try finding out. Instead of waiting for that golden "someday" when we'll finally be worthy.
posted by trig at 11:09 AM on June 21 [61 favorites]
This site is plenty marketable.
I firmly agree with this and the rest of this comment by trig. Like 110%, down to my core belief. That doesn’t mean any other site is worse, just different.
We should definitely work on a regular marketing schedule to let people know we’re human, especially in this era of ‘bots and AI.
posted by Brandon Blatcher (staff) at 11:30 AM on June 21 [22 favorites]
I firmly agree with this and the rest of this comment by trig. Like 110%, down to my core belief. That doesn’t mean any other site is worse, just different.
We should definitely work on a regular marketing schedule to let people know we’re human, especially in this era of ‘bots and AI.
posted by Brandon Blatcher (staff) at 11:30 AM on June 21 [22 favorites]
This is going to take much longer than you expect.
This is going to be much more complicated than you expect.
I understand where this is coming from but I remain hopeful because like, "build a web blog with comments" is the goto demo app for pretty much every web framework I can recall. They are designed to make this kind of site with ease to let engineers focus on features that differentiate versus alternatives.
posted by pwnguin at 11:48 AM on June 21 [5 favorites]
This is going to be much more complicated than you expect.
I understand where this is coming from but I remain hopeful because like, "build a web blog with comments" is the goto demo app for pretty much every web framework I can recall. They are designed to make this kind of site with ease to let engineers focus on features that differentiate versus alternatives.
posted by pwnguin at 11:48 AM on June 21 [5 favorites]
My last comment was a bit of a drive-by, so I'll highlight a more helpful suggestion from my old comments: “It is technically possible, though laborious, to move the site to modern tech without users experiencing any changes. You just set up an app server next to ColdFusion and replicate the existing site page-by-page in the new platform, and only expose the new platforms' pages when they're basically identical to CF's output. This takes time but avoids a risky large cutover.” I would add that you might put the new framework in front of CF like a proxy so the new framework has complete control rather than relying on a separate proxy, but that's just a detail. Also it's probably best to start migrating the read-only features first since those can do the least damage, and take lessons learned by implementing the read-only pages to the pages that write to the database. Sounds like Kirkaracha has a lot of knowledge already from working on SportsFilter but there’s always some new weirdness to find.
This isn't the only way to do things, and I defer to the staff on the best way to do things. But big cutovers have a lot of risks. They introduce a lot of bugs all at once instead of a few at a time. As a result they require buttloads of testing before they happen instead of testing incrementally. The cutover itself is a big deal — what if the new site doesn't come up? So you have to test the cutover process itself thoroughly. What if you have to revert the change — can you get the old site back up? So now you should test the reversion process. What if you find there's an error writing to the DB after cutover — do you revert to CF or go read-only so you can thoroughly test a fix, or just accept the bad data while you rush a fix in?
I realize the tech staff probably know all of this already, and obviously I'm highlighting all the risks when in reality only a couple of them would come into play. But I'm trying to highlight that while the function-by-function and page-by-page process might seem slow, laborious, and full of overhead, so is the big bang cutover process. The big bang cutover just adds all that overhead up front and raises the stakes.
I'm not the one doing the work so it's not my call, but I thought I'd put in a plug for incremental changes. After all, the agile manifesto says “Deliver working software frequently, from a couple of weeks to a couple of months, with a preference to the shorter timescale.”
posted by Tehhund at 11:48 AM on June 21 [3 favorites]
This isn't the only way to do things, and I defer to the staff on the best way to do things. But big cutovers have a lot of risks. They introduce a lot of bugs all at once instead of a few at a time. As a result they require buttloads of testing before they happen instead of testing incrementally. The cutover itself is a big deal — what if the new site doesn't come up? So you have to test the cutover process itself thoroughly. What if you have to revert the change — can you get the old site back up? So now you should test the reversion process. What if you find there's an error writing to the DB after cutover — do you revert to CF or go read-only so you can thoroughly test a fix, or just accept the bad data while you rush a fix in?
I realize the tech staff probably know all of this already, and obviously I'm highlighting all the risks when in reality only a couple of them would come into play. But I'm trying to highlight that while the function-by-function and page-by-page process might seem slow, laborious, and full of overhead, so is the big bang cutover process. The big bang cutover just adds all that overhead up front and raises the stakes.
I'm not the one doing the work so it's not my call, but I thought I'd put in a plug for incremental changes. After all, the agile manifesto says “Deliver working software frequently, from a couple of weeks to a couple of months, with a preference to the shorter timescale.”
posted by Tehhund at 11:48 AM on June 21 [3 favorites]
rockindata: all I can say is please, please, please stay away from react and other JavaScript based frameworksNthing this. MetaFilter is a perfect match for bog-standard HTML, CSS, and a small sprinkling of vanilla JS.
posted by Tehhund at 11:50 AM on June 21 [11 favorites]
I don't think the biggest problem is people coming here and bouncing off, I think the biggest problem is people having no idea this place exists and is worth checking out in the first place. I could be wrong about that! But ffs let's try finding out. Instead of waiting for that golden "someday" when we'll finally be worthy.
I mean - I just struggle with what "let's try finding out means." Paid advertising? Someone doing social media regularly? Like I just don't understand what Metafilter, with the budget it has, can do to find users for what we offer. Who are we targeting and where is this untapped pool of people who need it hanging out?
I think realistically it will yield a main thing - members doing organic outreach and sharing it with their own friends and social media networks - and there is literally nothing stopping us from trying that ourselves now. Starting today. I'm guessing some of us are trying that. I have tried that myself but to be honest it doesn't really land with anyone I know and I feel like I know a lot of people who would make great Metafilter members. I think I've had exactly one friend or family member join in the years I've been here. 90% of the people who are looking for online communities within my friend group go to Reddit.
The thing I have noted from social media sharing too is - it doesn't present well to a largely mobile phone user base alongside content that is more dynamic and visually appealing. You get a screen shot of mostly text in a couple of weird colours, and then you come here to a site that is demonstrably not mobile friendly and is just a wall of text. And understanding the value proposition this site provides requires people to spend some time here.
And so that is why I think the site is not marketable in its current form - obviously there's immense value I see because I've been here for most of my adult life in some form. But even explaining the value of this site to people I know is difficult, and the way we share now socially doesn't present Metafilter in a very appealing way, and so I struggle to understand how we are going to market any of this.
posted by openhearted at 11:54 AM on June 21 [4 favorites]
I mean - I just struggle with what "let's try finding out means." Paid advertising? Someone doing social media regularly? Like I just don't understand what Metafilter, with the budget it has, can do to find users for what we offer. Who are we targeting and where is this untapped pool of people who need it hanging out?
I think realistically it will yield a main thing - members doing organic outreach and sharing it with their own friends and social media networks - and there is literally nothing stopping us from trying that ourselves now. Starting today. I'm guessing some of us are trying that. I have tried that myself but to be honest it doesn't really land with anyone I know and I feel like I know a lot of people who would make great Metafilter members. I think I've had exactly one friend or family member join in the years I've been here. 90% of the people who are looking for online communities within my friend group go to Reddit.
The thing I have noted from social media sharing too is - it doesn't present well to a largely mobile phone user base alongside content that is more dynamic and visually appealing. You get a screen shot of mostly text in a couple of weird colours, and then you come here to a site that is demonstrably not mobile friendly and is just a wall of text. And understanding the value proposition this site provides requires people to spend some time here.
And so that is why I think the site is not marketable in its current form - obviously there's immense value I see because I've been here for most of my adult life in some form. But even explaining the value of this site to people I know is difficult, and the way we share now socially doesn't present Metafilter in a very appealing way, and so I struggle to understand how we are going to market any of this.
posted by openhearted at 11:54 AM on June 21 [4 favorites]
I'm with Trig. Small caveat that I've seen the usefulness of Ask tick down a notch as there have been fewer active users, and that we're a community with blind spots and full on black holes of missing knowledge on some interesting topics and perspectives. Still nothing beats Ask for Ask's purpose, to me. I think incremental increases in membership would slowly and steadily bring Ask to its former glory (whether actual or idealized). Like I don't know what kind of lift or how big a lift would bring in 100 new regular users, but even 100 would probably make a huge improvement to an already good platform. And it's my suspicion that changes in functionality or fixing site culture around flameouts or whatever aren't even necessary to find those hundred people.
posted by kensington314 at 11:59 AM on June 21 [5 favorites]
posted by kensington314 at 11:59 AM on June 21 [5 favorites]
I am going to miss being able to horrify tech people by saying "It's still running ColdFusion!" like the punchline of a particularly spooky ghost story.
Still haven't seen an answer to what the target tech stack is, so I'm just going to assume it's migrating into HyperCard.
(For serious, though: don't crowdsource the decision, or worry about making it open source from the start, or we'll still be arguing about design details in MeTa, in 2029.)
(For further serious, I'm a backend dev who's been doing this for a long time, and if there's any way I can be of use technically, I currently have a lot of free time on my hands, and this is one of maybe three places on the web that I care enough about that I'd be willing to volunteer to help)
posted by Mayor West at 11:59 AM on June 21 [6 favorites]
Still haven't seen an answer to what the target tech stack is, so I'm just going to assume it's migrating into HyperCard.
(For serious, though: don't crowdsource the decision, or worry about making it open source from the start, or we'll still be arguing about design details in MeTa, in 2029.)
(For further serious, I'm a backend dev who's been doing this for a long time, and if there's any way I can be of use technically, I currently have a lot of free time on my hands, and this is one of maybe three places on the web that I care enough about that I'd be willing to volunteer to help)
posted by Mayor West at 11:59 AM on June 21 [6 favorites]
I just spoke with Kirkaracha and he will share some specifics about the technical part of what we are doing, for now let me address a couple of comments.
– Assigning a mod or user to make another LinkMe post on the blue right now would be nice, for one example, instead of waiting a month or whatever to try it again. LinkMe posts twice a month seem like a very useful idea.
I love the idea! Checking with the team to make this happen every other week for now.
– I hope any major changes to the site's UI that are made during this new process are given a chance for discussion before becoming mandatory.
Yes! This is exactly why we are planning to start testing ASAP even though so far the MVP only includes the main functionality of the major subsites (log in, add posts, make comments). From there on we'll need channels and processes to report bugs, ask questions and make suggestions.
– We have already been told that even small changes to mod behavior are out of the question for the next year.
Not exactly, more on that next week.
posted by loup (staff) at 12:05 PM on June 21 [1 favorite]
– Assigning a mod or user to make another LinkMe post on the blue right now would be nice, for one example, instead of waiting a month or whatever to try it again. LinkMe posts twice a month seem like a very useful idea.
I love the idea! Checking with the team to make this happen every other week for now.
– I hope any major changes to the site's UI that are made during this new process are given a chance for discussion before becoming mandatory.
Yes! This is exactly why we are planning to start testing ASAP even though so far the MVP only includes the main functionality of the major subsites (log in, add posts, make comments). From there on we'll need channels and processes to report bugs, ask questions and make suggestions.
– We have already been told that even small changes to mod behavior are out of the question for the next year.
Not exactly, more on that next week.
posted by loup (staff) at 12:05 PM on June 21 [1 favorite]
i experimented a bit with a MeFi TikTok account previously. Not saying its good, but there’s plenty of social media channels we might want to focus on.
We had a podcast for a while, maybe a videocast or whatever the youths are calling such things these days.
There’s also the Best Of MetaFilter blog, in which highly favorited and generally positive MeFi posts are talked about.
Perhaps repurposing that content across social media platforms would be good.
posted by Brandon Blatcher (staff) at 12:08 PM on June 21 [4 favorites]
We had a podcast for a while, maybe a videocast or whatever the youths are calling such things these days.
There’s also the Best Of MetaFilter blog, in which highly favorited and generally positive MeFi posts are talked about.
Perhaps repurposing that content across social media platforms would be good.
posted by Brandon Blatcher (staff) at 12:08 PM on June 21 [4 favorites]
Will this include the classic theme?
Yes, but I'm prioritizing functionality so the classic theme will come later in the process.
New stack deets please?
PHP/MySQL, using the free, open source Laravel web application framework (techy/less-techy overview).
The admin section is in Filament (demo), which is a free Laravel package that makes it almost magically easy to create forms for managing database content.
Can I strongly suggest that the code for the rewrite be open-sourced? Maybe not right at the beginning, but at some point?
The code will be open source at some point.
Much of the cost of a rewrite usually lies in ensuring *identical* functionality, so it is well-worth it to figure out whether some weird edges can be sanded over rather than re-carved in loving detail.
I'm mostly copying existing functionality. If anything changes it will be because we think it's an improvement.
MetaFilter is a perfect match for bog-standard HTML, CSS, and a small sprinkling of vanilla JS.
This is the way.
posted by kirkaracha (staff) at 12:16 PM on June 21 [45 favorites]
Yes, but I'm prioritizing functionality so the classic theme will come later in the process.
New stack deets please?
PHP/MySQL, using the free, open source Laravel web application framework (techy/less-techy overview).
The admin section is in Filament (demo), which is a free Laravel package that makes it almost magically easy to create forms for managing database content.
Can I strongly suggest that the code for the rewrite be open-sourced? Maybe not right at the beginning, but at some point?
The code will be open source at some point.
Much of the cost of a rewrite usually lies in ensuring *identical* functionality, so it is well-worth it to figure out whether some weird edges can be sanded over rather than re-carved in loving detail.
I'm mostly copying existing functionality. If anything changes it will be because we think it's an improvement.
MetaFilter is a perfect match for bog-standard HTML, CSS, and a small sprinkling of vanilla JS.
This is the way.
posted by kirkaracha (staff) at 12:16 PM on June 21 [45 favorites]
Is the site willing to think about the comment-hiding approach for, say, one week, in order to decide whether to do a limited-time experiment using the
Details tag even though it's not ideal from a database perspective, in order to at least try out comment-hiding before committing to or against it? And, if after the experiment it seems potentially worth implementing permanently, make the necessary infrastructure changes a high-priority dev item? #pleaseanswer
We are not at this point or for the foreseeable future, which I'm defining/guesstamiting as not till early next year, considering that change*.
If that is:
Not exactly
then I have to admit I'm puzzled and confused.
posted by bowbeacon at 12:17 PM on June 21 [4 favorites]
Details tag even though it's not ideal from a database perspective, in order to at least try out comment-hiding before committing to or against it? And, if after the experiment it seems potentially worth implementing permanently, make the necessary infrastructure changes a high-priority dev item? #pleaseanswer
We are not at this point or for the foreseeable future, which I'm defining/guesstamiting as not till early next year, considering that change*.
If that is:
Not exactly
then I have to admit I'm puzzled and confused.
posted by bowbeacon at 12:17 PM on June 21 [4 favorites]
Good on ya, Kirk! Thank you for the details, seems like you've got an entirely reasonable plan.
I remain, as ever, entirely "WTF?!" at the management over here for the last several years, but I'm looking forward to seeing your MVP!
posted by Alterscape at 12:28 PM on June 21 [1 favorite]
I remain, as ever, entirely "WTF?!" at the management over here for the last several years, but I'm looking forward to seeing your MVP!
posted by Alterscape at 12:28 PM on June 21 [1 favorite]
then I have to admit I'm puzzled and confused.
Yes, totally understandable. That comment was referring to changes in code or addition of new features, not moderation policies and mod behaviour. I'm happy to go over that in detail next week(ish) when we go back to that thread.
posted by loup (staff) at 12:35 PM on June 21 [1 favorite]
Yes, totally understandable. That comment was referring to changes in code or addition of new features, not moderation policies and mod behaviour. I'm happy to go over that in detail next week(ish) when we go back to that thread.
posted by loup (staff) at 12:35 PM on June 21 [1 favorite]
I just struggle with what "let's try finding out means." Paid advertising? Someone doing social media regularly?
It means having a professionalized marketing and membership function that can set targets, prepare toolkits, and develop strategy. A committee, an officer, a hired manager, a leadership structure of some kind, not just random members posting stuff to their social channels. This is a normal role and function for a web-based community. It's been argued for here from a long time back.
I find this all pretty confusing, if not concerning. I don't think big decisions should be made while the site's in transition to a community-driven leadership structure that might want to make fundamentally different choices, and already in this thread people are leaping to assumptions about functionality that profoundly impact site experience, and yet won't be directed by those community representatives. It seems like the cart (the availability of dev hours) is leading the horse here, and we'll end up living with permanent commitments that the eventual leadership may not have chosen. I'm sorry the incoming board will be saddled with dealing with something they didn't get to make choices about. Expediency without direction is usually folly.
Meanwhile, there are a hell of a lot-of warm, live, user-facing issues that need to be addressed to support the healthy transition. I hope this doesn't crowd those out.
posted by Miko at 12:35 PM on June 21 [16 favorites]
It means having a professionalized marketing and membership function that can set targets, prepare toolkits, and develop strategy. A committee, an officer, a hired manager, a leadership structure of some kind, not just random members posting stuff to their social channels. This is a normal role and function for a web-based community. It's been argued for here from a long time back.
I find this all pretty confusing, if not concerning. I don't think big decisions should be made while the site's in transition to a community-driven leadership structure that might want to make fundamentally different choices, and already in this thread people are leaping to assumptions about functionality that profoundly impact site experience, and yet won't be directed by those community representatives. It seems like the cart (the availability of dev hours) is leading the horse here, and we'll end up living with permanent commitments that the eventual leadership may not have chosen. I'm sorry the incoming board will be saddled with dealing with something they didn't get to make choices about. Expediency without direction is usually folly.
Meanwhile, there are a hell of a lot-of warm, live, user-facing issues that need to be addressed to support the healthy transition. I hope this doesn't crowd those out.
posted by Miko at 12:35 PM on June 21 [16 favorites]
– I'm sorry the incoming board will be saddled with dealing with something they didn't get to make choices about.
This is definitely not the case. When I said we have made the decision to focus resources on rebuilding the entire site from scratch I mean exactly that kirkaracha's time will be focused on this because we think this is doable, necessary, and not as complicated as just maintaining the site as it is.
As it currently is, this project is very self contained (all the usual resources and allocations we've had for the past years remain the same) and we are at an early stage of development. If it proves to be fruitless, not viable, or simply not what the community wants it can be terminated without major losses at any time.
When the proposal to do this was put forward we had two options: have a mega thread about it for weeks to maybe-maybe-not-consider-doing-it, or give it a try, share it with the community as soon as possible and show them that this can be done.
posted by loup (staff) at 12:49 PM on June 21 [14 favorites]
This is definitely not the case. When I said we have made the decision to focus resources on rebuilding the entire site from scratch I mean exactly that kirkaracha's time will be focused on this because we think this is doable, necessary, and not as complicated as just maintaining the site as it is.
As it currently is, this project is very self contained (all the usual resources and allocations we've had for the past years remain the same) and we are at an early stage of development. If it proves to be fruitless, not viable, or simply not what the community wants it can be terminated without major losses at any time.
When the proposal to do this was put forward we had two options: have a mega thread about it for weeks to maybe-maybe-not-consider-doing-it, or give it a try, share it with the community as soon as possible and show them that this can be done.
posted by loup (staff) at 12:49 PM on June 21 [14 favorites]
If it proves to be fruitless, not viable, or simply not what the community wants it can be terminated without major losses at any time.
By what mechanism? What is the process by which the community will determine this?
You present a false binary, as well. There is at least one third option: you wait to take direction from the people who the userbase has as representatives.
posted by Miko at 1:00 PM on June 21 [9 favorites]
By what mechanism? What is the process by which the community will determine this?
You present a false binary, as well. There is at least one third option: you wait to take direction from the people who the userbase has as representatives.
posted by Miko at 1:00 PM on June 21 [9 favorites]
It means having a professionalized marketing and membership function that can set targets, prepare toolkits, and develop strategy. A committee, an officer, a hired manager, a leadership structure of some kind, not just random members posting stuff to their social channels. This is a normal role and function for a web-based community. It's been argued for here from a long time back.
Which commercial web community is doing this with this level of resource and structure in 2024?
I don't think big decisions should be made while the site's in transition to a community-driven leadership structure that might want to make fundamentally different choices
So, if we need new strategic roles and management and strategies for marketing but we should also not make big decisions while in transition…I don’t understand how this reconciles other than do very little for the undetermined amount of time it takes for this transition to complete.
posted by openhearted at 1:11 PM on June 21 [5 favorites]
Which commercial web community is doing this with this level of resource and structure in 2024?
I don't think big decisions should be made while the site's in transition to a community-driven leadership structure that might want to make fundamentally different choices
So, if we need new strategic roles and management and strategies for marketing but we should also not make big decisions while in transition…I don’t understand how this reconciles other than do very little for the undetermined amount of time it takes for this transition to complete.
posted by openhearted at 1:11 PM on June 21 [5 favorites]
Which commercial web community is doing this with this level of resource and structure in 2024?
I said web-based - as in, a membership community whose platform is on the web. And I work for one that does this and has that role as a staff position, supported by a board committee.
posted by Miko at 1:13 PM on June 21 [4 favorites]
I said web-based - as in, a membership community whose platform is on the web. And I work for one that does this and has that role as a staff position, supported by a board committee.
posted by Miko at 1:13 PM on June 21 [4 favorites]
There is at least one third option: you wait to take direction from the people who the userbase has as representatives.
Fair point. Waiting to decide the best way to put our new tech hire's skills, expertise and motivation to good use was a decision I actively decided not to make.
Instead, I made a case for it, consulted this with Jessamyn (the current site owner), we weighted the risk, opportunities to make this possible, and resources available and we made the decision to move forward.
I need to shift my focus for the rest of my day to other tasks so I will not be replying more for today but I'm happy to continue next week.
posted by loup (staff) at 1:25 PM on June 21 [10 favorites]
Fair point. Waiting to decide the best way to put our new tech hire's skills, expertise and motivation to good use was a decision I actively decided not to make.
Instead, I made a case for it, consulted this with Jessamyn (the current site owner), we weighted the risk, opportunities to make this possible, and resources available and we made the decision to move forward.
I need to shift my focus for the rest of my day to other tasks so I will not be replying more for today but I'm happy to continue next week.
posted by loup (staff) at 1:25 PM on June 21 [10 favorites]
MySQL
I trust you deeply, utterly, totally on this. You've been in and done things with this code that I don't even want to think about. But, uh, just a quick reminder to please write out the mysqldump (and binlogs, if you're feeling frisky!) to S3 instead of a snapshot? Y'know, because of... the thing?
Although I guess the new thing is also to address the thing?
posted by majick at 1:27 PM on June 21 [1 favorite]
I trust you deeply, utterly, totally on this. You've been in and done things with this code that I don't even want to think about. But, uh, just a quick reminder to please write out the mysqldump (and binlogs, if you're feeling frisky!) to S3 instead of a snapshot? Y'know, because of... the thing?
Although I guess the new thing is also to address the thing?
posted by majick at 1:27 PM on June 21 [1 favorite]
I'm skeptical of a lot of the professionalization talk, but not terribly averse to it if that's the direction the eventual Board eventually decides to go. (But a truly professional ED would cost more than half the current budget, no?)
The way I see it, the only value growth has, in the long run, is to balance out attrition, making it a pretty modest goal. Buying ads, for example, sounds kind of ridiculous to me. But an effort to bring in a not very large amount of new members by giving existing (donation supporting?) users a couple gift memberships to hand out (say, to people they might know from a subreddit or discord or whatever, if not in person) sounds like it might prove fruitful. Tie it to the fundraising month each year.
posted by nobody at 1:31 PM on June 21 [4 favorites]
The way I see it, the only value growth has, in the long run, is to balance out attrition, making it a pretty modest goal. Buying ads, for example, sounds kind of ridiculous to me. But an effort to bring in a not very large amount of new members by giving existing (donation supporting?) users a couple gift memberships to hand out (say, to people they might know from a subreddit or discord or whatever, if not in person) sounds like it might prove fruitful. Tie it to the fundraising month each year.
posted by nobody at 1:31 PM on June 21 [4 favorites]
the rewrite is a big, important thing to do, and it's an important thing to do for all the reasons some of the semi-detractors are citing. The sooner we get this done, the sooner we can actually do the things that implement the community's needs.
I guess I'm a semi-detractor here. I don't really want to be a detractor - I consider our staff resources precious and so I'm concerned about burning through some good people without great results. It's fair if the issue really is building new is easier than maintaining, on a 6-12 month timeline.
So - I'm not yelling. I'm concerned.
But when you say "do the things that implement the community's needs" - yeah, that's exactly it. Is the barrier to posting the BIPOC minutes really the site, because if so, link to a free Wordpress blog that the team controls the password for and be done because for fuck's sake inequity on the site or lack of joyful welcoming of BIPOC experience is not because of the fucking minutes. Err sorry.
For the banner, is it the banner that's the issue (pro marketing tip here - banners do not work; banners are for UX notification like 'your comment cannot be posted' or like, fires) or is it that a) the sign up process never let anyone opt into something like "tell me about site updates!" and b) the code if I remember correctly has no way ever of emailing members (plus we promised we never ever would) and so people never have the option of finding things out except if they randomly stop ignoring the banner one day.
I have no doubt that the codebase causes issues. I accept that this is the chosen way forward. But I don't think that it is going to help MetaFilter survive unless the code does what we NEED it to do, and I just don't think there's been a process to uncover that.
I am not a tech person, but as someone who has been tech-adjacent, I am just very afraid that we will spend a lot of development time and money and a lot of community goodwill and end up with a site that is still a barrier to itself, and a team that has focused on the technology rather than other stuff. I am glad to be wrong!
I do know about marketing so I will take that in a separate comment although it seems like a separate Meta.
posted by warriorqueen at 1:33 PM on June 21 [11 favorites]
I guess I'm a semi-detractor here. I don't really want to be a detractor - I consider our staff resources precious and so I'm concerned about burning through some good people without great results. It's fair if the issue really is building new is easier than maintaining, on a 6-12 month timeline.
So - I'm not yelling. I'm concerned.
But when you say "do the things that implement the community's needs" - yeah, that's exactly it. Is the barrier to posting the BIPOC minutes really the site, because if so, link to a free Wordpress blog that the team controls the password for and be done because for fuck's sake inequity on the site or lack of joyful welcoming of BIPOC experience is not because of the fucking minutes. Err sorry.
For the banner, is it the banner that's the issue (pro marketing tip here - banners do not work; banners are for UX notification like 'your comment cannot be posted' or like, fires) or is it that a) the sign up process never let anyone opt into something like "tell me about site updates!" and b) the code if I remember correctly has no way ever of emailing members (plus we promised we never ever would) and so people never have the option of finding things out except if they randomly stop ignoring the banner one day.
I have no doubt that the codebase causes issues. I accept that this is the chosen way forward. But I don't think that it is going to help MetaFilter survive unless the code does what we NEED it to do, and I just don't think there's been a process to uncover that.
I am not a tech person, but as someone who has been tech-adjacent, I am just very afraid that we will spend a lot of development time and money and a lot of community goodwill and end up with a site that is still a barrier to itself, and a team that has focused on the technology rather than other stuff. I am glad to be wrong!
I do know about marketing so I will take that in a separate comment although it seems like a separate Meta.
posted by warriorqueen at 1:33 PM on June 21 [11 favorites]
Yeah, how to market/membership is a separate discussion. But for those pointing out “no, it’s material to the code and platform because #NewSite has to enable X, Y, and Z membership functions,” well yeah, that’s exactly why the incoming board should have had a chance to influence it. At least a consult.
Anyway, good luck, we’re all counting on you.
posted by Miko at 1:39 PM on June 21 [5 favorites]
Anyway, good luck, we’re all counting on you.
posted by Miko at 1:39 PM on June 21 [5 favorites]
it is kind of surreal to me that we are several years in to a post-Cortex Metafilter and we still have folks at the top making unilateral decisions about site resources. i suppose at least this is one of the big three things people have been asking for but, to me, it would have been a distant third behind "pay someone to do your admin work for christ's sake" and "we need to seriously re-think the moderation on this site". i don't think our outdated tech hasn't directly contributed to dozens of users buttoning over the last few years the same way our other existential/culture issues have. but i guess we'll hear about that... next week sometime? even though this was the monthly site update thread and it's been up for three days?
idk y'all! best of luck to kirkaracha who seems like the right fit but i am feeling pretty cynical about this whole endeavor.
posted by nourishedbytime at 1:44 PM on June 21 [14 favorites]
idk y'all! best of luck to kirkaracha who seems like the right fit but i am feeling pretty cynical about this whole endeavor.
posted by nourishedbytime at 1:44 PM on June 21 [14 favorites]
I said web-based - as in, a membership community whose platform is on the web. And I work for one that does this and has that role as a staff position, supported by a board committee.
In some future day where this is a non-profit and can have a volunteer board committee then sure - although I've been Director/ED for three national level ones (coops, social enterprise, and agriculture) with budgets 3-6x what Metafilters is and at 3x we had a shared role member services coordinator with someeone who did comms and marketing because that's a more realistic reality for a barely-scraping-by non-profit that we will be. Especially at Metafilter's $18k a month in existing compensation. Like to me the way you would have a hired manager, leadership structure and a staff member/marketing position here is to cut moderation from part-time to barely-at-all, which has been mentioned is Metafilter's secret sauce.
But in reality today - we have I think 4 part-time mods and 2 part-time technical staff. Nobody was hired to manage the commercial entity that it is today. Nobody was hired with marketing experience. And nobody can volunteer to do this function for the commercial entity. So even if this was a reasonable expectation for a non-profit of Metafilter's size to operate like that, it's impossible for the org to do this right now. And so I again arrive at what right now we would be doing to market Metafilter given we have moderators and technical staff on payroll.
posted by openhearted at 1:52 PM on June 21 [6 favorites]
In some future day where this is a non-profit and can have a volunteer board committee then sure - although I've been Director/ED for three national level ones (coops, social enterprise, and agriculture) with budgets 3-6x what Metafilters is and at 3x we had a shared role member services coordinator with someeone who did comms and marketing because that's a more realistic reality for a barely-scraping-by non-profit that we will be. Especially at Metafilter's $18k a month in existing compensation. Like to me the way you would have a hired manager, leadership structure and a staff member/marketing position here is to cut moderation from part-time to barely-at-all, which has been mentioned is Metafilter's secret sauce.
But in reality today - we have I think 4 part-time mods and 2 part-time technical staff. Nobody was hired to manage the commercial entity that it is today. Nobody was hired with marketing experience. And nobody can volunteer to do this function for the commercial entity. So even if this was a reasonable expectation for a non-profit of Metafilter's size to operate like that, it's impossible for the org to do this right now. And so I again arrive at what right now we would be doing to market Metafilter given we have moderators and technical staff on payroll.
posted by openhearted at 1:52 PM on June 21 [6 favorites]
I’m speaking from significant experience with nonprofit organizations, at least equal to yours, in orgs whose budget ranges from $10,000 to $22 million. And there is a very large range of potential structures that can meet the membership needs of an organization whose core function is membership. The one I currently work for has four staff, one of whom is membership and marketing manager. But again, no point having this conversation here and now. It’s totally immaterial at this point and in this venue and under this current management..
posted by Miko at 2:02 PM on June 21 [6 favorites]
posted by Miko at 2:02 PM on June 21 [6 favorites]
We should definitely work on a regular marketing schedule to let people know we’re human, especially in this era of ‘bots and AI.
You don't need a marketing schedule (yet), you need a plan and a strategy and then implementation of tactics.
I have a lot of respect for people and I know this will sound snippy because I have like 10 minutes here but here goes anyway.
Creating a TikTok account or posting our posts on X/FB/Etc. is NOT EFFECTIVE MARKETING. Effective marketing is saying "who are our people?" "Where can we reach them?" "How can we reach them?"
I wrote a list of emergency grow-activity tactics for MetaFilter before the Transition Team decided its role was not to run things and some of it would probably raise some eyebrows, because the choice is a) money or b) partnerships. Posting stuff on social is not marketing unless you are a local artist or some other service, or you have a really big social channel already.
So by partnerships I mean "we'll post about your Substack and you give us a mention in your Substack" and yes, I used the DREAD SUBSTACK. Co-host fan events on FanFare with indy producers - oh yes, add games to FanFare first. Tag writers when their articles appear and offer them a free account if they want to go in and talk to people. invite authors to host a post about books and include their book in it and shout to their fans and maybe they gain some and we gain some.
These are a few of my guerilla ideas and they are frankly not great because I wrote them on a walk in what was it, 2021? but I'm putting them here because marketers FIND AUDIENCES, they don't just post shit on Instagram. No shade to Brandon, but that's the difference.
This can't be individual members. I cannot represent MetaFilter to Jenny Nicholson and invite her in, but someone official can. We could start a campaign for members and provide toolkits! But like I said I'm just trying to get in here that when I say we're not investing in growing the site we really are NOT doing any of this.
And to be clear, I'm not saying Do This. I'm expressing that there is a trade off continuing here.
It means having a professionalized marketing and membership function that can set targets, prepare toolkits, and develop strategy. A committee, an officer, a hired manager, a leadership structure of some kind, not just random members posting stuff to their social channels. This is a normal role and function for a web-based community. It's been argued for here from a long time back.
Which commercial web community is doing this with this level of resource and structure in 2024?
A lot of them. Individual creators are doing this. They co-create! They buy ads! They trade shout outs! They provide weird little downloads. They buy search terms and Reddit ads.
posted by warriorqueen at 2:05 PM on June 21 [25 favorites]
You don't need a marketing schedule (yet), you need a plan and a strategy and then implementation of tactics.
I have a lot of respect for people and I know this will sound snippy because I have like 10 minutes here but here goes anyway.
Creating a TikTok account or posting our posts on X/FB/Etc. is NOT EFFECTIVE MARKETING. Effective marketing is saying "who are our people?" "Where can we reach them?" "How can we reach them?"
I wrote a list of emergency grow-activity tactics for MetaFilter before the Transition Team decided its role was not to run things and some of it would probably raise some eyebrows, because the choice is a) money or b) partnerships. Posting stuff on social is not marketing unless you are a local artist or some other service, or you have a really big social channel already.
So by partnerships I mean "we'll post about your Substack and you give us a mention in your Substack" and yes, I used the DREAD SUBSTACK. Co-host fan events on FanFare with indy producers - oh yes, add games to FanFare first. Tag writers when their articles appear and offer them a free account if they want to go in and talk to people. invite authors to host a post about books and include their book in it and shout to their fans and maybe they gain some and we gain some.
These are a few of my guerilla ideas and they are frankly not great because I wrote them on a walk in what was it, 2021? but I'm putting them here because marketers FIND AUDIENCES, they don't just post shit on Instagram. No shade to Brandon, but that's the difference.
This can't be individual members. I cannot represent MetaFilter to Jenny Nicholson and invite her in, but someone official can. We could start a campaign for members and provide toolkits! But like I said I'm just trying to get in here that when I say we're not investing in growing the site we really are NOT doing any of this.
And to be clear, I'm not saying Do This. I'm expressing that there is a trade off continuing here.
It means having a professionalized marketing and membership function that can set targets, prepare toolkits, and develop strategy. A committee, an officer, a hired manager, a leadership structure of some kind, not just random members posting stuff to their social channels. This is a normal role and function for a web-based community. It's been argued for here from a long time back.
Which commercial web community is doing this with this level of resource and structure in 2024?
A lot of them. Individual creators are doing this. They co-create! They buy ads! They trade shout outs! They provide weird little downloads. They buy search terms and Reddit ads.
posted by warriorqueen at 2:05 PM on June 21 [25 favorites]
Long answer: I take care of decision making when it comes to daily operations, for more strategic things I will confer with Jessamyn and we'll come to a decision together.
I think it's weird for a community weblog that one person is making strategic decisions in consultation with the owner who is minimally involved at this point. It's even stranger that the one person leading strategy does not seem very involved in or familiar with the site and sometimes makes statements that are at odds with how the membership understand and use Metafilter. If the answer were "the mods discuss strategy and then I take their ideas to Jessamyn," even that would make more sense.
posted by CtrlAltD at 2:13 PM on June 21 [17 favorites]
I think it's weird for a community weblog that one person is making strategic decisions in consultation with the owner who is minimally involved at this point. It's even stranger that the one person leading strategy does not seem very involved in or familiar with the site and sometimes makes statements that are at odds with how the membership understand and use Metafilter. If the answer were "the mods discuss strategy and then I take their ideas to Jessamyn," even that would make more sense.
posted by CtrlAltD at 2:13 PM on June 21 [17 favorites]
It's even stranger that the one person leading strategy does not seem very involved in or familiar with the site and sometimes makes statements that are at odds with how the membership understand and use Metafilter.
Let’s be blunt: this one person (“An Admin”, I guess) has been noted by many users to behave in a way that comes off as somewhere between indifferent and contemptuous toward this community. Needed change is great, but the way this is happening is deeply odd, to say the least.
posted by not just everyday big moggies at 2:24 PM on June 21 [12 favorites]
Let’s be blunt: this one person (“An Admin”, I guess) has been noted by many users to behave in a way that comes off as somewhere between indifferent and contemptuous toward this community. Needed change is great, but the way this is happening is deeply odd, to say the least.
posted by not just everyday big moggies at 2:24 PM on June 21 [12 favorites]
It's really, really not.
posted by kbanas at 2:36 PM on June 21 [4 favorites]
posted by kbanas at 2:36 PM on June 21 [4 favorites]
Mod note:
Coming back to fix the banner and address a few comment while at it.
- Wow, I'm curious to see this unfold. Also, can the mods please change the site banner to have less mystery meat in it? This is a huge announcement.
Whoops, I almost forgot. Done!
- If the answer were "the mods discuss strategy and then I take their ideas to Jessamyn," even that would make more sense.
This is a lot more complex than that, of course there is constant communication and consultation going on with the team (which I'm happy to say it's very balanced with the team we currently have) and we even consult with both members and previous staff members depending on the subject.
- Let’s be blunt: this one person (“An Admin”, I guess) has been noted by many users to behave in a way that comes off as somewhere between indifferent and contemptuous toward this community.
That would be me and, to be completely honest: I detest MetaTalk. I detest it profoundly as it has been a huge source of burnout for the staff and the community, it's impossible to keep up with it while trying to do several other things at the same time and it's designed in a way where reaching consensus isn't easy.
That being said, indifference and contempt are definitely not the feelings I have towards this community, if that was the case I would not have agreed to be in the position I currently am, where my responsibilities and the hours of dedication have increased substantially with no incentive other than just the well-being of the community long term.
If you have frustrations with me personally, feel free to reach out direclty via email of MeFi Mail and I'm happy to talk them trough, to listen and to adapt. If you think this is beyond anything I can say, you can bring them to Jesamyn instead, but please, can we center the conversation in this tread around the site and it's future and not the admin who seriously wants to see better days for this community?
posted by loup (staff) at 2:57 PM on June 21 [40 favorites]
Coming back to fix the banner and address a few comment while at it.
- Wow, I'm curious to see this unfold. Also, can the mods please change the site banner to have less mystery meat in it? This is a huge announcement.
Whoops, I almost forgot. Done!
- If the answer were "the mods discuss strategy and then I take their ideas to Jessamyn," even that would make more sense.
This is a lot more complex than that, of course there is constant communication and consultation going on with the team (which I'm happy to say it's very balanced with the team we currently have) and we even consult with both members and previous staff members depending on the subject.
- Let’s be blunt: this one person (“An Admin”, I guess) has been noted by many users to behave in a way that comes off as somewhere between indifferent and contemptuous toward this community.
That would be me and, to be completely honest: I detest MetaTalk. I detest it profoundly as it has been a huge source of burnout for the staff and the community, it's impossible to keep up with it while trying to do several other things at the same time and it's designed in a way where reaching consensus isn't easy.
That being said, indifference and contempt are definitely not the feelings I have towards this community, if that was the case I would not have agreed to be in the position I currently am, where my responsibilities and the hours of dedication have increased substantially with no incentive other than just the well-being of the community long term.
If you have frustrations with me personally, feel free to reach out direclty via email of MeFi Mail and I'm happy to talk them trough, to listen and to adapt. If you think this is beyond anything I can say, you can bring them to Jesamyn instead, but please, can we center the conversation in this tread around the site and it's future and not the admin who seriously wants to see better days for this community?
posted by loup (staff) at 2:57 PM on June 21 [40 favorites]
I’ll happily withdraw from this conversation after this comment, but on the topic of:
If you have frustrations with me personally, feel free to reach out direclty via email of MeFi Mail and I'm happy to talk them trough, to listen and to adapt.
Sorry, no. After the last few big Metas, and the complete failure of the mods to clarify some extremely basic standards regarding mod communications and what’s ok to report back to the community, this absolutely isn’t going to happen. You can’t have it both ways if you’re unwilling to do the work.
posted by not just everyday big moggies at 3:02 PM on June 21 [22 favorites]
If you have frustrations with me personally, feel free to reach out direclty via email of MeFi Mail and I'm happy to talk them trough, to listen and to adapt.
Sorry, no. After the last few big Metas, and the complete failure of the mods to clarify some extremely basic standards regarding mod communications and what’s ok to report back to the community, this absolutely isn’t going to happen. You can’t have it both ways if you’re unwilling to do the work.
posted by not just everyday big moggies at 3:02 PM on June 21 [22 favorites]
I'm pretty excited about the prospect of a rewrite, mostly because even though I've seen plenty of rewrites go off the rails, I've seen a lot of time wasted working around crufty poorly-supported frameworks nobody dares/wants to touch. So best of luck to kirkaracha!! I hope you'll let us know if we can help! It seems like there are a lot of web developers in the comment section.
I'm quietly hoping that Laravel makes it easy to expose both fully rendered HTML and some kind of JSON/XML version of major endpoints so people can write their own reader/widgets/iOS app more easily. But it's also a Good Outcome if you just get a codebase that's easier and more enjoyable for the team to work with, even if it looks exactly the same from the outside.
posted by pulposus at 3:14 PM on June 21 [4 favorites]
I'm quietly hoping that Laravel makes it easy to expose both fully rendered HTML and some kind of JSON/XML version of major endpoints so people can write their own reader/widgets/iOS app more easily. But it's also a Good Outcome if you just get a codebase that's easier and more enjoyable for the team to work with, even if it looks exactly the same from the outside.
posted by pulposus at 3:14 PM on June 21 [4 favorites]
These are a few of my guerilla ideas and they are frankly not great because I wrote them on a walk in what was it, 2021? but I'm putting them here because marketers FIND AUDIENCES, they don't just post shit on Instagram. No shade to Brandon, but that's the difference.
This can't be individual members. I cannot represent MetaFilter to Jenny Nicholson and invite her in, but someone official can. We could start a campaign for members and provide toolkits! But like I said I'm just trying to get in here that when I say we're not investing in growing the site we really are NOT doing any of this.
For the record I deeply respect all the work you did on the committee and continue to do and your list of ideas. I REALLY wish June 2024 had the steering committee so deeply into directing and doing things from your list and not where we are! And I 100% agree with you that we need a plan to actually move the needle and someone to find audiences and ultimately convert to members. Like that end state is ideal.
But like I said - with the staff that we have now who are not marketers, and the fact we don't have a marketing plan, and the fact that we can't have volunteers develop one for the commercial entity, and the fact that it doesn't make sense for the commercial entity of Metafilter to contract out for a marketing strategy that the future non-profit board will inherit - we have to accept that we aren't going to have strategic marketing and membership growth as a core Metafilter function for a while without big decisions from Loup. That is the decision-making structure we have today whether we like it or not, and it seems like a lot of people don't like them making decisions. Jessamyn seems completely disinterested in being involved at all and for the weeks, or months, until the non-profit can actually exist and take over the assets of Metafilter - this is what we have.
And that's why I say the only thing I think Metafilter can do right now are some basic, ineffective, grassroots things. Loup doesn't really have the background or I would say the cache with the user base right now to lead a marketing charge and spend money on it. And it may be helpful in the future when we envision collaborations with other content creators to have a platform that can support at least inline images without keeling over, as I am guessing most collaborators will have an interest in somewhat dynamic content given the fast majority of the content creators are doing so with video at this point.
And lastly - as part of the revive scenario, hiring an additional technical resource was one of the deliverables from that. And the hiring of Kirk was celebrated as a step forward and in his technical expertise, after a few months of looking at things, he has suggested it makes more sense to move to a new platform vs. fix problems on this creaky, old one. Like the decision at this point is not new platform vs. marketing and getting new members, it's whether the technical resource we already hired should spend the next few months making incremental improvements or if it makes sense to investigate a big shift. And so I see this as something that the competent person we pay to give us technical advice telling us his time is better spent doing this than it is Loup making big decisions unilaterally.
I will shut up now though as I think I have enough words on the subject.
posted by openhearted at 3:23 PM on June 21 [8 favorites]
This can't be individual members. I cannot represent MetaFilter to Jenny Nicholson and invite her in, but someone official can. We could start a campaign for members and provide toolkits! But like I said I'm just trying to get in here that when I say we're not investing in growing the site we really are NOT doing any of this.
For the record I deeply respect all the work you did on the committee and continue to do and your list of ideas. I REALLY wish June 2024 had the steering committee so deeply into directing and doing things from your list and not where we are! And I 100% agree with you that we need a plan to actually move the needle and someone to find audiences and ultimately convert to members. Like that end state is ideal.
But like I said - with the staff that we have now who are not marketers, and the fact we don't have a marketing plan, and the fact that we can't have volunteers develop one for the commercial entity, and the fact that it doesn't make sense for the commercial entity of Metafilter to contract out for a marketing strategy that the future non-profit board will inherit - we have to accept that we aren't going to have strategic marketing and membership growth as a core Metafilter function for a while without big decisions from Loup. That is the decision-making structure we have today whether we like it or not, and it seems like a lot of people don't like them making decisions. Jessamyn seems completely disinterested in being involved at all and for the weeks, or months, until the non-profit can actually exist and take over the assets of Metafilter - this is what we have.
And that's why I say the only thing I think Metafilter can do right now are some basic, ineffective, grassroots things. Loup doesn't really have the background or I would say the cache with the user base right now to lead a marketing charge and spend money on it. And it may be helpful in the future when we envision collaborations with other content creators to have a platform that can support at least inline images without keeling over, as I am guessing most collaborators will have an interest in somewhat dynamic content given the fast majority of the content creators are doing so with video at this point.
And lastly - as part of the revive scenario, hiring an additional technical resource was one of the deliverables from that. And the hiring of Kirk was celebrated as a step forward and in his technical expertise, after a few months of looking at things, he has suggested it makes more sense to move to a new platform vs. fix problems on this creaky, old one. Like the decision at this point is not new platform vs. marketing and getting new members, it's whether the technical resource we already hired should spend the next few months making incremental improvements or if it makes sense to investigate a big shift. And so I see this as something that the competent person we pay to give us technical advice telling us his time is better spent doing this than it is Loup making big decisions unilaterally.
I will shut up now though as I think I have enough words on the subject.
posted by openhearted at 3:23 PM on June 21 [8 favorites]
And lastly - as part of the revive scenario, hiring an additional technical resource was one of the deliverables from that
And hiring another admin.
I get what you’re saying but “we can’t do anything unless loup agrees to do it” is the ultimate problem. The problem isn’t metatalk. User committees have been ignored pretty blatantly for years at this point.
I really like Kirk, I think he’s likely to do a great job and put a lot of effort into it, similar with Brandon. I worry that they are going to get massively burnt out not because everyone is so mean on metatalk but because they’re constantly throwing their efforts into a black hole.
posted by knobknosher at 3:46 PM on June 21 [15 favorites]
And hiring another admin.
I get what you’re saying but “we can’t do anything unless loup agrees to do it” is the ultimate problem. The problem isn’t metatalk. User committees have been ignored pretty blatantly for years at this point.
I really like Kirk, I think he’s likely to do a great job and put a lot of effort into it, similar with Brandon. I worry that they are going to get massively burnt out not because everyone is so mean on metatalk but because they’re constantly throwing their efforts into a black hole.
posted by knobknosher at 3:46 PM on June 21 [15 favorites]
loup: I detest MetaTalk.
MetaTalk used to be a fun place to hang out in.
I realize it’s hard to believe. But we used to share photos of our pets and babies, talk about meetups and play silly, nigh-on incomprehensible word games. But slowly the silliness and oversharing was discouraged, and what’s left is grar and grief. Which we had back in the day too, but the lighter stuff balanced it out.
I don’t know if MetaTalk could ever go back to being the site hangout spot, to be honest, but it would be nice if IRL was closed and meetups at least were back in here, and meetup photos. Also cat photos. Photos of lobsters wearing slightly soiled lederhosen. Anything to evoke any reaction other than a deep sigh when opening up MetaTalk.
I thought this thread would be a technical and somewhat fun discussion about database management, but somehow we’re back on the grar and grief.
posted by Kattullus at 4:28 PM on June 21 [39 favorites]
MetaTalk used to be a fun place to hang out in.
I realize it’s hard to believe. But we used to share photos of our pets and babies, talk about meetups and play silly, nigh-on incomprehensible word games. But slowly the silliness and oversharing was discouraged, and what’s left is grar and grief. Which we had back in the day too, but the lighter stuff balanced it out.
I don’t know if MetaTalk could ever go back to being the site hangout spot, to be honest, but it would be nice if IRL was closed and meetups at least were back in here, and meetup photos. Also cat photos. Photos of lobsters wearing slightly soiled lederhosen. Anything to evoke any reaction other than a deep sigh when opening up MetaTalk.
I thought this thread would be a technical and somewhat fun discussion about database management, but somehow we’re back on the grar and grief.
posted by Kattullus at 4:28 PM on June 21 [39 favorites]
wow. maybe this new mefi shouldn't have Metatalk, if only so that mods aren't put in a position of being a) assigned as the only person to talk to users while also b) utterly hating the mechanism they have to talk to users. I sympathize, but if it's such an ordeal to engage with, why is *that* not something worth addressing? Hell, there are still probably things from the last few locked threads that still haven't been responded to.
I know you say you don't feel anything negative toward the community, but it still feels like users are being blamed for... having questions? Requiring "hours of dedication" that are apparently unpaid?
And, as usual, users are being told that, whatever avenue they're using for the discussion, it should be moved somewhere else.
posted by sagc at 4:30 PM on June 21 [20 favorites]
I know you say you don't feel anything negative toward the community, but it still feels like users are being blamed for... having questions? Requiring "hours of dedication" that are apparently unpaid?
And, as usual, users are being told that, whatever avenue they're using for the discussion, it should be moved somewhere else.
posted by sagc at 4:30 PM on June 21 [20 favorites]
where my responsibilities and the hours of dedication have increased substantially with no incentive other than just the well-being of the community long term.
loup is being paid for this work, correct? And they’re just venting about how much they don’t like it, in response to commenters asserting that they really don’t seem to like being here? Honestly, every time they deign to contribute to a Meta I’m left with a brand new flavor of “wtf-is-actually-going-on-here?”.
posted by not just everyday big moggies at 4:35 PM on June 21 [20 favorites]
loup is being paid for this work, correct? And they’re just venting about how much they don’t like it, in response to commenters asserting that they really don’t seem to like being here? Honestly, every time they deign to contribute to a Meta I’m left with a brand new flavor of “wtf-is-actually-going-on-here?”.
posted by not just everyday big moggies at 4:35 PM on June 21 [20 favorites]
That being said, indifference and contempt are definitely not the feelings I have towards this community, if that was the case I would not have agreed to be in the position I currently am, where my responsibilities and the hours of dedication have increased substantially with no incentive other than just the well-being of the community long term.
i believe this to be true but this is also functionally the exact same situation we were in for like a decade when the last owner ran the site and i think a lot of us were really hoping for something different when he left. it sounds like you are burnt out for the same reasons he was burnt out, or at least some of them. we are obviously still in a transitional period but hearing this is dispiriting.
posted by nourishedbytime at 4:43 PM on June 21 [7 favorites]
i believe this to be true but this is also functionally the exact same situation we were in for like a decade when the last owner ran the site and i think a lot of us were really hoping for something different when he left. it sounds like you are burnt out for the same reasons he was burnt out, or at least some of them. we are obviously still in a transitional period but hearing this is dispiriting.
posted by nourishedbytime at 4:43 PM on June 21 [7 favorites]
I detest MetaTalk.
There, we have a problem.
The staff can't figure out how to relate productively with this community. So they make unilateral decisions and batten down the hatches. It ain't great.
Providing a platform for people to weigh in, and then a system for proceeding with decisionmaking even though some people will be happy and some unhappy, listening with sympathy and taking in the wisdom of the crowd, while shouldering the burden of the decision - that's leadership. It's not impossible/
Part of the reason this community has leaned toward suck is that members of the staff really dislike interacting with the community.
posted by Miko at 4:59 PM on June 21 [36 favorites]
There, we have a problem.
The staff can't figure out how to relate productively with this community. So they make unilateral decisions and batten down the hatches. It ain't great.
Providing a platform for people to weigh in, and then a system for proceeding with decisionmaking even though some people will be happy and some unhappy, listening with sympathy and taking in the wisdom of the crowd, while shouldering the burden of the decision - that's leadership. It's not impossible/
Part of the reason this community has leaned toward suck is that members of the staff really dislike interacting with the community.
posted by Miko at 4:59 PM on June 21 [36 favorites]
Just want to reiterate how profoundly weird it is that Mefi has boomeranged so rapidly back from the brief vision of community ownership via an elected board to the same old model of a single, mostly-unilateral decision-maker who dislikes and is already burn out from engaging with the community.
It's already been said, but this is exactly the kind of project that the elected board should be overseeing, in collaboration with an advisory board of community members who represent a variety of demographics and thus user concerns and experiences. It's not clear why loup is making these unilateral decisions so soon in advance of the shift to community governance. That said, it's also not clear why there isn't an elected board, or why board elections aren't underway yet. Or why there hasn't been an admin hire, when that was something the TT specifically budgeted for with the intention of it happening in 2023.
posted by knucklebones at 5:25 PM on June 21 [17 favorites]
It's already been said, but this is exactly the kind of project that the elected board should be overseeing, in collaboration with an advisory board of community members who represent a variety of demographics and thus user concerns and experiences. It's not clear why loup is making these unilateral decisions so soon in advance of the shift to community governance. That said, it's also not clear why there isn't an elected board, or why board elections aren't underway yet. Or why there hasn't been an admin hire, when that was something the TT specifically budgeted for with the intention of it happening in 2023.
posted by knucklebones at 5:25 PM on June 21 [17 favorites]
That would be me and, to be completely honest: I detest MetaTalk. I detest it profoundly as it has been a huge source of burnout for the staff and the community, it's impossible to keep up with it while trying to do several other things at the same time and it's designed in a way where reaching consensus isn't easy.
I'm sorry you hate metatalk that much. Reapectfully, perhaps other people should be assigned to respond to us here in metatalk. Moderators who like being here. Who do not see interacting with us as chore or punishment. Who are taught that one key to achieving community consensus is moderator friendliness,respect,trust.
Having healthy community requires community building & consistent moderator interaction with members. Issuing statements or radio silence for days is not healthy interaction.
In th past, every moderator was a member of metafilter before they were hired. They were like Brandon. A person who spent their spare time here, had friends here, was invested in being here. That helped. Overlaping friendly moderators be active in metatalk may not be financially feasible but that helped too.
Metatalk failed when josh and some moderators showed their contempt for us. When they showed they hate interacting with us and all but stopped.
posted by qi at 5:26 PM on June 21 [21 favorites]
I'm sorry you hate metatalk that much. Reapectfully, perhaps other people should be assigned to respond to us here in metatalk. Moderators who like being here. Who do not see interacting with us as chore or punishment. Who are taught that one key to achieving community consensus is moderator friendliness,respect,trust.
Having healthy community requires community building & consistent moderator interaction with members. Issuing statements or radio silence for days is not healthy interaction.
In th past, every moderator was a member of metafilter before they were hired. They were like Brandon. A person who spent their spare time here, had friends here, was invested in being here. That helped. Overlaping friendly moderators be active in metatalk may not be financially feasible but that helped too.
Metatalk failed when josh and some moderators showed their contempt for us. When they showed they hate interacting with us and all but stopped.
posted by qi at 5:26 PM on June 21 [21 favorites]
I detest MetaTalk
Oh honey, it’s ok. Really.
MetaTalk is the contemporaneous version of those old Roman stadiums where they used to feed prisoners to the lions for the amusement of the middle class.
Or the Kobayashi Maru? (Sorry, on my phone and can’t link to the Wikipedia article about it, it’s a Star Trek thing)
I know nothing about programming, this upcoming giant project sounds interesting and it’s great to see all the programmers pop in with their weird advice (“ please write out the mysqldump (and binlogs, if you're feeling frisky!“ haha what does that even mean?)
(please don’t tell me)
I for one welcome our new ant overlords. Congratulations on the decision!
posted by Vatnesine at 5:28 PM on June 21 [1 favorite]
Oh honey, it’s ok. Really.
MetaTalk is the contemporaneous version of those old Roman stadiums where they used to feed prisoners to the lions for the amusement of the middle class.
Or the Kobayashi Maru? (Sorry, on my phone and can’t link to the Wikipedia article about it, it’s a Star Trek thing)
I know nothing about programming, this upcoming giant project sounds interesting and it’s great to see all the programmers pop in with their weird advice (“ please write out the mysqldump (and binlogs, if you're feeling frisky!“ haha what does that even mean?)
(please don’t tell me)
I for one welcome our new ant overlords. Congratulations on the decision!
posted by Vatnesine at 5:28 PM on June 21 [1 favorite]
Incredible that one of the first breakthroughs we've had through loup's curtain of bland corporate soother speech is to learn that they hate us.
posted by phunniemee at 5:28 PM on June 21 [31 favorites]
posted by phunniemee at 5:28 PM on June 21 [31 favorites]
Unrelatedly, yes I agree MeTa is a terrible place for decision making. Since it's easier to try to solve a problem with technology than address the underlying issues, I'll recommend exploring Loomio, an excellent open-source tool, or Consider.it.
posted by knucklebones at 5:29 PM on June 21 [5 favorites]
posted by knucklebones at 5:29 PM on June 21 [5 favorites]
My God. Of course they hate MetaTalk.
A small but oh-so-vocal minority come into every single one of these threads and you're mean, you're demeaning, you're snide, and you're utterly cruel in the way you talk, and you just ceaselessly aim it unrelentingly in one direction, and then you have the temerity to step back and go, "Oh! So now you hate us?!"
If loup and jessamyn had come in here with a plan to crowd source discussion, you'd raise hell that they weren't taking more direct and immediate action.
If they came in with a plan to roll out new features to the old code base, you'd raise hell that we aren't investing in new infrastructure.
If they came in with a plan to market the site, you'd raise hell that we weren't rolling out new features.
You have a problem with a single person, and you drag that axe into every single one of these threads, over and over and over and over, and then look around and go, "How could they react like this!?"
You have no real commitment to growth or change, you just want to be angry. So be angry. You're really good at it.
posted by kbanas at 5:37 PM on June 21 [96 favorites]
A small but oh-so-vocal minority come into every single one of these threads and you're mean, you're demeaning, you're snide, and you're utterly cruel in the way you talk, and you just ceaselessly aim it unrelentingly in one direction, and then you have the temerity to step back and go, "Oh! So now you hate us?!"
If loup and jessamyn had come in here with a plan to crowd source discussion, you'd raise hell that they weren't taking more direct and immediate action.
If they came in with a plan to roll out new features to the old code base, you'd raise hell that we aren't investing in new infrastructure.
If they came in with a plan to market the site, you'd raise hell that we weren't rolling out new features.
You have a problem with a single person, and you drag that axe into every single one of these threads, over and over and over and over, and then look around and go, "How could they react like this!?"
You have no real commitment to growth or change, you just want to be angry. So be angry. You're really good at it.
posted by kbanas at 5:37 PM on June 21 [96 favorites]
I also detest what MetaTalk has become. It is no longer fit for purpose. When I was a mere lurker, I was impressed by the ability of this community to have really difficult discussions on MetaTalk that certainly weren't pleasant, but were clearly productive. This was around the time that the community was confronting a lot of explicit and implicit sexism, and while things got heated, sometimes nasty, they also led to actual change.
For the last several years it seems all that MeTa can achieve is unpleasantness without productivity. A lot of folks show up here in every thread ready for a fight, ready to continue whatever fight they have been having for years now, regardless of how relevant the topic of the thread is to the issue they want to fight about. And I get it, people are frustrated that things they want to see changed aren't changing. Sometimes that's because there is too much institutional inertia, or lack of resources, or maybe lack of will. Sometimes it's because the people calling for specific changes actually don't have any significant consensus behind them, but continue to insist that their perspective absolutely must be addressed every time.
There is a lot of site business that has to get done, and it simply isn't possible to do it when this is what happens every time we try to have a significant discussion. And quite frankly, a lot of it is the structure of MeTa that's the problem. A while back I posted a thread to discuss the idea of instituting something like a site ombudsman. I actually feel pretty good about how that thread went in the end, but trying to craft the post and help facilitate the conversation was exhausting. And in the end it didn't go anywhere, I think at least in part because MeTa isn't designed to be an issue tracker or idea manager or anything like that, it's just a forum, and when things drop off the front page they are quickly forgotten.
I hope that once we have a new codebase, it will be possible to start spinning up other structures for having these conversations, so that people don't feel they have to show up in every single thread to remind everyone that they're still not happy about that thing that happened three years ago, or whatever. People need to be able to express their frustration more productively than MeTa allows.
I'm glad Loup acknowleged that they hate MeTa. I think that probably aligns them with the majority of their fellow MeFites. Honestly I'm surprised it's even a controversial opinion.
posted by biogeo at 5:39 PM on June 21 [41 favorites]
For the last several years it seems all that MeTa can achieve is unpleasantness without productivity. A lot of folks show up here in every thread ready for a fight, ready to continue whatever fight they have been having for years now, regardless of how relevant the topic of the thread is to the issue they want to fight about. And I get it, people are frustrated that things they want to see changed aren't changing. Sometimes that's because there is too much institutional inertia, or lack of resources, or maybe lack of will. Sometimes it's because the people calling for specific changes actually don't have any significant consensus behind them, but continue to insist that their perspective absolutely must be addressed every time.
There is a lot of site business that has to get done, and it simply isn't possible to do it when this is what happens every time we try to have a significant discussion. And quite frankly, a lot of it is the structure of MeTa that's the problem. A while back I posted a thread to discuss the idea of instituting something like a site ombudsman. I actually feel pretty good about how that thread went in the end, but trying to craft the post and help facilitate the conversation was exhausting. And in the end it didn't go anywhere, I think at least in part because MeTa isn't designed to be an issue tracker or idea manager or anything like that, it's just a forum, and when things drop off the front page they are quickly forgotten.
I hope that once we have a new codebase, it will be possible to start spinning up other structures for having these conversations, so that people don't feel they have to show up in every single thread to remind everyone that they're still not happy about that thing that happened three years ago, or whatever. People need to be able to express their frustration more productively than MeTa allows.
I'm glad Loup acknowleged that they hate MeTa. I think that probably aligns them with the majority of their fellow MeFites. Honestly I'm surprised it's even a controversial opinion.
posted by biogeo at 5:39 PM on June 21 [41 favorites]
Give Loup a break. Better yet, imagine being in their position, sharing an update, and then dealing with the deluge of comments, most of which can be summarized as “You’re doing it wrong”.
I hate MetaTalk too. This is MetaFilter at its worst.
posted by SafetyPirate at 5:40 PM on June 21 [24 favorites]
I hate MetaTalk too. This is MetaFilter at its worst.
posted by SafetyPirate at 5:40 PM on June 21 [24 favorites]
I would hate MetaTalk too if I were loup. Hell, I hate about 90% of it as box.
For years, it’s been a place where no argument or criticism is too small, where nobody ever moves on, where you can count on getting bad-faith readings interpreted in the worst possible way, where consensus goes to die.
There’s a long history or burnout among the moderators, and we’re much better at demanding the world from them than asking how we’re complicit.
posted by box at 5:40 PM on June 21 [34 favorites]
For years, it’s been a place where no argument or criticism is too small, where nobody ever moves on, where you can count on getting bad-faith readings interpreted in the worst possible way, where consensus goes to die.
There’s a long history or burnout among the moderators, and we’re much better at demanding the world from them than asking how we’re complicit.
posted by box at 5:40 PM on June 21 [34 favorites]
Mod note: Or the Kobayashi Maru? (Sorry, on my phone and can’t link to the Wikipedia article about it, it’s a Star Trek thing)
Here's the link, it's a great piece of Star Trek lore and I say that as not a huge fan of the series.
As to loup's comment, I'mma gonna maybe overstep some bounds and say that having worked with them and being initially trained by them, I don't think they dislike the members or the community at all.
It's just that MetaTalk has become a bit rougher, for various reasons. We, meaning all of us, members and moderators and boards and owners, need to talk about that a bit (probably in another thread) and decide how this part of the site could be less fightly and more of place where we work together to discuss ideas and find solutions. That won't happen tonight and we're heading into the weekend where moderations gaps are longer.
So let's recognize we all essentially want what's best for the site, though we have different ideas how and when that should happen. Let's try and assume best intentions and view others as having the same, so we can have better discussions with each other.
posted by Brandon Blatcher (staff) at 5:41 PM on June 21 [32 favorites]
Here's the link, it's a great piece of Star Trek lore and I say that as not a huge fan of the series.
As to loup's comment, I'mma gonna maybe overstep some bounds and say that having worked with them and being initially trained by them, I don't think they dislike the members or the community at all.
It's just that MetaTalk has become a bit rougher, for various reasons. We, meaning all of us, members and moderators and boards and owners, need to talk about that a bit (probably in another thread) and decide how this part of the site could be less fightly and more of place where we work together to discuss ideas and find solutions. That won't happen tonight and we're heading into the weekend where moderations gaps are longer.
So let's recognize we all essentially want what's best for the site, though we have different ideas how and when that should happen. Let's try and assume best intentions and view others as having the same, so we can have better discussions with each other.
posted by Brandon Blatcher (staff) at 5:41 PM on June 21 [32 favorites]
Incredible that one of the first breakthroughs we've had through loup's curtain of bland corporate soother speech is to learn that they hate us.
Loup was quite clear in not saying that, and I think you should really think about whether this is the kind of contribution you want to be making here.
posted by biogeo at 5:42 PM on June 21 [41 favorites]
I think with careful study you will find that I do not think at all about the kind of contribution I make here.
posted by phunniemee at 5:44 PM on June 21 [15 favorites]
posted by phunniemee at 5:44 PM on June 21 [15 favorites]
Brandon, thank you for doing what you do here. I really appreciate the energy you've clearly devoted to communicating with members here, especially to helping us understand what level of attention we're likely to get at any given moment due to the realities of staffing and work/life balance. I also really appreciate that you've been clear with "during [this time] I can do [this thing]" and then following through on it, or communicating when circumstances have changed and it's no longer possible. I don't always agree with all the calls you've made, but I see and respect what you've been bringing to MeTa lately. Thank you.
posted by Alterscape at 5:45 PM on June 21 [39 favorites]
posted by Alterscape at 5:45 PM on June 21 [39 favorites]
And in the end it didn't go anywhere, I think at least in part because MeTa isn't designed to be an issue tracker or idea manager or anything like that, it's just a forum, and when things drop off the front page they are quickly forgotten.
And because the main person in charge of dealing with it pretty much hates dealing with it. That is a completely fine opinion for them to have but it doesn’t really make sense for them to do it. You personally put a ton of effort into doing Metatalk the right way (or what you see as the right way) and it was ignored. There might be a lesson here about why people interact with metatalk the way they currently do.
When moderators are responsive, respectful, and relatively open people here are actually quite nice to them. You can see it very easily with the way people interact with Brandon and have gone out of their way to make it clear that they’re not criticizing Kirk. Again, there might be a lesson to be drawn from this.
posted by knobknosher at 5:51 PM on June 21 [19 favorites]
And because the main person in charge of dealing with it pretty much hates dealing with it. That is a completely fine opinion for them to have but it doesn’t really make sense for them to do it. You personally put a ton of effort into doing Metatalk the right way (or what you see as the right way) and it was ignored. There might be a lesson here about why people interact with metatalk the way they currently do.
When moderators are responsive, respectful, and relatively open people here are actually quite nice to them. You can see it very easily with the way people interact with Brandon and have gone out of their way to make it clear that they’re not criticizing Kirk. Again, there might be a lesson to be drawn from this.
posted by knobknosher at 5:51 PM on June 21 [19 favorites]
Me too, I find it amazing that anyone would not think the staff hates Metatalk, this place gets extremely rancorous sometimes. I feel like they have to walk in eggshells every time they interact here, for fear that someone will take a statement the wrong way, either honestly or, once in a while, disingenuously.
I have never seen any staff member of Metafilter react in a way that didn't indicate that they were trying to do the very best that they could. In exchange for that, a series of leaders have been essentially chased off, including some valued members of the community.
Sometimes I think there's real value in making a site that's more like Memepool of old, a list of links and interesting finds with no message board aspects at all. In the end I think Metafilter still benefits, overall, by allowing people to comment, and in turn comment on the comments, but there are days....
posted by JHarris at 5:53 PM on June 21 [17 favorites]
I have never seen any staff member of Metafilter react in a way that didn't indicate that they were trying to do the very best that they could. In exchange for that, a series of leaders have been essentially chased off, including some valued members of the community.
Sometimes I think there's real value in making a site that's more like Memepool of old, a list of links and interesting finds with no message board aspects at all. In the end I think Metafilter still benefits, overall, by allowing people to comment, and in turn comment on the comments, but there are days....
posted by JHarris at 5:53 PM on June 21 [17 favorites]
This is ultimately just a message board. Users aren't signing up. Just make something new on proboards for those who want to be wokescolded by Gen-X shut-ins and archive the existing site as is.
posted by fluttering hellfire at 6:00 PM on June 21 [12 favorites]
posted by fluttering hellfire at 6:00 PM on June 21 [12 favorites]
phunniemee, I'd like to add that I say that from a position of really appreciating you and your contributions to this site. Your user name is one I associate with positive memories here. I hope you know that I'm not reacting from a place of rancor, but surprise.
posted by biogeo at 6:06 PM on June 21 [4 favorites]
posted by biogeo at 6:06 PM on June 21 [4 favorites]
phunniemee: "Incredible that one of the first breakthroughs we've had through loup's curtain of bland corporate soother speech is to learn that they hate us."
They said they detest *MetaTalk*, which -- no shit, it's not a great way to do community management on any kind of complex or contentious topic. And that's not a loup thing -- it's always had a reputation for exhausting fightiness and angst. One of my very first-ever comments here was an only somewhat tongue-in-cheek description based on initial impressions of MetaTalk as "a realm of utter hopelessness. Perspective and logic mean nothing here -- what once was simple and unimportant blossoms like a poisonous flower into hitherto unexplored realms of pettiness and venom. Nothing is safe. There is nowhere to hide. The innocent and damned alike are skewered for the sport of cackling demons." And that was in 2000-and-fuckin'-7. Bush was president! Blackberries! MYSPACE. That's 17 years at least of MetaTalk being a bit shit.
Things were better, for awhile, circa '09-'13 when mods were always available and the callouts and interminable arguments were leavened with lighter threads, news, random site questions, meetup planning, etc. But there were layoffs and coverage reductions, the lighter community threads have largely decamped for MeFi proper, IRL took over meetups (and then died of COVID), and most site policy discussion has consolidated into these prickly, unresolvable, only partially-moderated official updates/slugfests where a significant amount of the ire is turned on the mods themselves.
I'm hopeful that the new infra will afford opportunities to bolt some new features onto the grey that make it easier to use as a tool for voting, feedback, and otherwise channeling community sentiments in a constructive way.
posted by Rhaomi at 6:21 PM on June 21 [26 favorites]
They said they detest *MetaTalk*, which -- no shit, it's not a great way to do community management on any kind of complex or contentious topic. And that's not a loup thing -- it's always had a reputation for exhausting fightiness and angst. One of my very first-ever comments here was an only somewhat tongue-in-cheek description based on initial impressions of MetaTalk as "a realm of utter hopelessness. Perspective and logic mean nothing here -- what once was simple and unimportant blossoms like a poisonous flower into hitherto unexplored realms of pettiness and venom. Nothing is safe. There is nowhere to hide. The innocent and damned alike are skewered for the sport of cackling demons." And that was in 2000-and-fuckin'-7. Bush was president! Blackberries! MYSPACE. That's 17 years at least of MetaTalk being a bit shit.
Things were better, for awhile, circa '09-'13 when mods were always available and the callouts and interminable arguments were leavened with lighter threads, news, random site questions, meetup planning, etc. But there were layoffs and coverage reductions, the lighter community threads have largely decamped for MeFi proper, IRL took over meetups (and then died of COVID), and most site policy discussion has consolidated into these prickly, unresolvable, only partially-moderated official updates/slugfests where a significant amount of the ire is turned on the mods themselves.
I'm hopeful that the new infra will afford opportunities to bolt some new features onto the grey that make it easier to use as a tool for voting, feedback, and otherwise channeling community sentiments in a constructive way.
posted by Rhaomi at 6:21 PM on June 21 [26 favorites]
When moderators are responsive, respectful, and relatively open people here are actually quite nice to them. You can see it very easily with the way people interact with Brandon and have gone out of their way to make it clear that they’re not criticizing Kirk.
This, 100%. When mods and admins engage in good faith, folks here love it. When they do weird shit (eg actively refusing to move their thread update to the correct thread despite being well aware that it was posted in the wrong one for more than a week), there’s an understandable “wtf is the problem here” reaction. Like, if loup is essentially making calls about the future of Metafilter some scrutiny about what they’re doing and not doing is understandable.
posted by not just everyday big moggies at 6:22 PM on June 21 [22 favorites]
This, 100%. When mods and admins engage in good faith, folks here love it. When they do weird shit (eg actively refusing to move their thread update to the correct thread despite being well aware that it was posted in the wrong one for more than a week), there’s an understandable “wtf is the problem here” reaction. Like, if loup is essentially making calls about the future of Metafilter some scrutiny about what they’re doing and not doing is understandable.
posted by not just everyday big moggies at 6:22 PM on June 21 [22 favorites]
Or, we could give Loup the benefit of the doubt, and assume that they’re doing what makes sense for the site. We could acknowledge that we’re less-informed and just trust Loup. We could reflect on what we might decide to do if we were faced with a mountain of problems and had a single technical person who could work on something technical, but not on anything else. We could even go so far as to offer some encouragement because all of this seems like hard work, and is obviously thankless.
posted by SafetyPirate at 6:51 PM on June 21 [14 favorites]
posted by SafetyPirate at 6:51 PM on June 21 [14 favorites]
idk, we're going to have the same back and forth every time this comes up. is it the mods who created the toxic environment or is it the users? is it the chicken or the egg? who cares? we're all in here breathing the toxic air. in my opinion, people are frustrated because of a long legacy of decisions that go back to the previous owner and even the owner before that. it's definitely not loup's fault metatalk became this way, but the current way things are being handled clearly is not working either.
at the end of the day, only one group of people is being paid to be here. nobody else in these threads can really contribute anything BUT their voice - at least at this juncture.
You have no real commitment to growth or change, you just want to be angry. So be angry. You're really good at it
no disrespect meant, but you are making the mistake of conflating the motivations of many different users into one monolith. when the community came together a few years ago to raise tens of thousands of dollars for the site, many people who have been critical on metatalk for years contributed both effort and funds. and many of the people who are consistently critical in these threads have also been saying literally the exact same things for years. that's not out of spite - people care about this site! they get passionate about it! that passion doesn't have to be a bad thing, but when things continue in a holding pattern despite the vast amount of passion and energy that has been poured into this site, things boil over.
posted by nourishedbytime at 7:23 PM on June 21 [20 favorites]
at the end of the day, only one group of people is being paid to be here. nobody else in these threads can really contribute anything BUT their voice - at least at this juncture.
You have no real commitment to growth or change, you just want to be angry. So be angry. You're really good at it
no disrespect meant, but you are making the mistake of conflating the motivations of many different users into one monolith. when the community came together a few years ago to raise tens of thousands of dollars for the site, many people who have been critical on metatalk for years contributed both effort and funds. and many of the people who are consistently critical in these threads have also been saying literally the exact same things for years. that's not out of spite - people care about this site! they get passionate about it! that passion doesn't have to be a bad thing, but when things continue in a holding pattern despite the vast amount of passion and energy that has been poured into this site, things boil over.
posted by nourishedbytime at 7:23 PM on June 21 [20 favorites]
That would be me and, to be completely honest: I detest MetaTalk. I detest it profoundly as it has been a huge source of burnout for the staff and the community, it's impossible to keep up with it while trying to do several other things at the same time and it's designed in a way where reaching consensus isn't easy.
I am very glad that you said this and made it explicit. MetaTalk is an overloaded tool. It's been clear for a long time that there are users that want it to have a much bigger role than you've wanted it to have. My mental model of things (and, as a highly-experienced eggshell walker, I might be projecting a bit here (because that's the only way to stay safe!)) has been that you just didn't think of MeTa as a useful part of the site, but didn't want to say that outright because it would just garner more outrage, so you were mostly just waiting for expectations to shift. Well, now that you've made your attitude more explicit, there wasn't that much outrage... MetaTalk does kinda suck, and I think people pretty much know that.
That said, people want to be heard. The idea of MeTa going away (or being ignored!) without other mechanisms of communal feedback rubs people the wrong way, I think. Purpose-built mechanisms for site surveys, task tracking, and other currently-common uses for MeTa would probably reduce demand for MeTa quite a bit, and I have hope that the site rewrite will facilitate that.
posted by a faded photo of their beloved at 11:10 PM on June 21 [11 favorites]
I am very glad that you said this and made it explicit. MetaTalk is an overloaded tool. It's been clear for a long time that there are users that want it to have a much bigger role than you've wanted it to have. My mental model of things (and, as a highly-experienced eggshell walker, I might be projecting a bit here (because that's the only way to stay safe!)) has been that you just didn't think of MeTa as a useful part of the site, but didn't want to say that outright because it would just garner more outrage, so you were mostly just waiting for expectations to shift. Well, now that you've made your attitude more explicit, there wasn't that much outrage... MetaTalk does kinda suck, and I think people pretty much know that.
That said, people want to be heard. The idea of MeTa going away (or being ignored!) without other mechanisms of communal feedback rubs people the wrong way, I think. Purpose-built mechanisms for site surveys, task tracking, and other currently-common uses for MeTa would probably reduce demand for MeTa quite a bit, and I have hope that the site rewrite will facilitate that.
posted by a faded photo of their beloved at 11:10 PM on June 21 [11 favorites]
Holy shit. As mentioned above (thankfully) loup took care to say that their comment was about the forum, not the user base, but still gets raked over the coals. How would anyone feel if their only real venue for communication was the equivalent of being the main character on Twitter, not just for a day, but every single time they made any comment at all?
I’m not remotely going to claim I agree with loup on a lot, but shit, so many people here are primed to jump at anything they say, so no wonder they hate this part of the site. In this same thread, we’ve had people talking about trying to attract new users, and worrying about things that might turn them off, but we have, essentially, someone right here who didn’t go through the last twenty years of this site fighting things out in the grey, ready for mortal combat every time there’s a post, and they’re being honest, saying they hate this part of things. Rather than pondering ifs and maybes about hypothetical new users, maybe listen to someone forced to interact with us at our absolute worst anytime they need to make an announcement.
Are things going the way I want them to? Holy fuck, no. Am I seriously worried about this place, shit yes, I am, but taking things in the worst possible light, using the least possible generous way of reading things at every opportunity, let alone specifically ignoring the parts of what someone says that doesn’t gibe with the gotcha we’re excited to spring isn’t fucking helping.
posted by Ghidorah at 11:22 PM on June 21 [41 favorites]
I’m not remotely going to claim I agree with loup on a lot, but shit, so many people here are primed to jump at anything they say, so no wonder they hate this part of the site. In this same thread, we’ve had people talking about trying to attract new users, and worrying about things that might turn them off, but we have, essentially, someone right here who didn’t go through the last twenty years of this site fighting things out in the grey, ready for mortal combat every time there’s a post, and they’re being honest, saying they hate this part of things. Rather than pondering ifs and maybes about hypothetical new users, maybe listen to someone forced to interact with us at our absolute worst anytime they need to make an announcement.
Are things going the way I want them to? Holy fuck, no. Am I seriously worried about this place, shit yes, I am, but taking things in the worst possible light, using the least possible generous way of reading things at every opportunity, let alone specifically ignoring the parts of what someone says that doesn’t gibe with the gotcha we’re excited to spring isn’t fucking helping.
posted by Ghidorah at 11:22 PM on June 21 [41 favorites]
A small but oh-so-vocal minority come into every single one of these threads and you're mean, you're demeaning, you're snide, and you're utterly cruel in the way you talk, and you just ceaselessly aim it unrelentingly in one direction, and then you have the temerity to step back and go, "Oh! So now you hate us?!"
Cannot favourite that comment hard enough.
I think with careful study you will find that I do not think at all about the kind of contribution I make here.
The lights are growing dim, Otto. I know a life of snark has led me to this sorry fate, and yet, I blame the mod team. The mod team made me what I am.
posted by flabdablet at 11:31 PM on June 21 [12 favorites]
Cannot favourite that comment hard enough.
I think with careful study you will find that I do not think at all about the kind of contribution I make here.
The lights are growing dim, Otto. I know a life of snark has led me to this sorry fate, and yet, I blame the mod team. The mod team made me what I am.
posted by flabdablet at 11:31 PM on June 21 [12 favorites]
wow. maybe this new mefi shouldn't have Metatalk
When it comes to encouraging new users to sign up and join the community, I think that given we are rebuilding anyway, some thought should be given to the future of all the less popular MeFi subsites.
A new user unfamiliar with the site is going to click along the top navigation row to see what is there: Projects, Music, Jobs, IRL, all of them are empty wastelands with a few posts per month and then they find Meta which is full of arguments. They may not realise that 99% of the content is under WWW. and Ask, so it should be no surprise when they think 'well this place looks almost dead' and click away to go somewhere else.
If it were up to me, everything other than WWW. and Ask. would be behind the login wall and hidden from the Google index.
posted by Lanark at 1:27 AM on June 22 [16 favorites]
When it comes to encouraging new users to sign up and join the community, I think that given we are rebuilding anyway, some thought should be given to the future of all the less popular MeFi subsites.
A new user unfamiliar with the site is going to click along the top navigation row to see what is there: Projects, Music, Jobs, IRL, all of them are empty wastelands with a few posts per month and then they find Meta which is full of arguments. They may not realise that 99% of the content is under WWW. and Ask, so it should be no surprise when they think 'well this place looks almost dead' and click away to go somewhere else.
If it were up to me, everything other than WWW. and Ask. would be behind the login wall and hidden from the Google index.
posted by Lanark at 1:27 AM on June 22 [16 favorites]
I detest MetaTalk.
There, we have a problem.
The staff can't figure out how to relate productively with this community. So they make unilateral decisions and batten down the hatches. It ain't great.
Providing a platform for people to weigh in, and then a system for proceeding with decisionmaking even though some people will be happy and some unhappy, listening with sympathy and taking in the wisdom of the crowd, while shouldering the burden of the decision - that's leadership. It's not impossible/
Part of the reason this community has leaned toward suck is that members of the staff really dislike interacting with the community.
As I said in a previous Meta, maybe it's time to "spend more time with family" or "pursue other interests."
posted by jgirl at 4:14 AM on June 22 [3 favorites]
There, we have a problem.
The staff can't figure out how to relate productively with this community. So they make unilateral decisions and batten down the hatches. It ain't great.
Providing a platform for people to weigh in, and then a system for proceeding with decisionmaking even though some people will be happy and some unhappy, listening with sympathy and taking in the wisdom of the crowd, while shouldering the burden of the decision - that's leadership. It's not impossible/
Part of the reason this community has leaned toward suck is that members of the staff really dislike interacting with the community.
As I said in a previous Meta, maybe it's time to "spend more time with family" or "pursue other interests."
posted by jgirl at 4:14 AM on June 22 [3 favorites]
Incredible that one of the first breakthroughs we've had through loup's curtain of bland corporate soother speech is to learn that they hate us.
It is this kind of complete bad-faith reading that would make MetaTalk such a horrible part of this job - not remotely what they said.
It should be noted that all of the previous people who have worked here and interacted regularly in MetaTalk are completely gone and burnt out. Even Jessamyn who owns the site is hardly here anymore.
There's a small group here who seem primarily motivated with getting their glib, asshole-ish one-liners off on whoever the wind has turned on to get high fives. It's gross behaviour.
As I said in a previous Meta, maybe it's time to "spend more time with family" or "pursue other interests."
Imagine working at a place where people come in openly suggested you shouldn't work here anymore multiple times. For a part-time job that you put in unpaid hours to do. The very definition of creating a toxic workplace.
posted by openhearted at 4:29 AM on June 22 [34 favorites]
It is this kind of complete bad-faith reading that would make MetaTalk such a horrible part of this job - not remotely what they said.
It should be noted that all of the previous people who have worked here and interacted regularly in MetaTalk are completely gone and burnt out. Even Jessamyn who owns the site is hardly here anymore.
There's a small group here who seem primarily motivated with getting their glib, asshole-ish one-liners off on whoever the wind has turned on to get high fives. It's gross behaviour.
As I said in a previous Meta, maybe it's time to "spend more time with family" or "pursue other interests."
Imagine working at a place where people come in openly suggested you shouldn't work here anymore multiple times. For a part-time job that you put in unpaid hours to do. The very definition of creating a toxic workplace.
posted by openhearted at 4:29 AM on June 22 [34 favorites]
We need to get through this transition. It's moved much more slowly than anyone hoped but the wheels are still in motion. If anyone here has earned trust as both mod and user, it's jessamyn. Being mostly hands off is probably the right call, and as long as her trust in loup continues, I figure she knows more than I do.
I'm not sure that MetaTalk could be anything but toxic at this point. There aren't enough mod resources for it to be a good avenue of communication, and until we shift to the new org structure, there's no avenue for the promised community empowerment, so it's just frustrating for everyone.
I, too, miss the days when MeTa was like having coffee or a beer with the mods. I hope that when we finally do the handover to a volunteer board, it can return to being a lower-stakes place for interaction with site leadership, but it can't be that right now. Current site leadership is paid, and strained by the workload, and MeTa only adds to that workload.
We've painted ourselves into a corner with paid mods who can't really be good community members and the nonprofit is the way out of that corner. I know it's unsatisfying to just keep saying "wait until this happens and THEN we can move forward," but continuing to argue about things that can't happen yet isn't helping either.
I have no doubt we'll eventually get volunteer mods and a minimal paid staff, it seems like there's no other solution based on the budget we have. We'll get a marketing plan and new moderation policies. But until the nonprofit is up and running, we need to keep our expectations low of communication between staff and users. That sucks, but a lot has sucked since Google's algorithm broke MeFi so many years ago. It's taken a long time to come to terms with that, and it's fair that the community that has kept coming back here is hungry to take ownership. But it's happening slowly because everyone is trying not to fuck this up.
posted by rikschell at 4:43 AM on June 22 [8 favorites]
I'm not sure that MetaTalk could be anything but toxic at this point. There aren't enough mod resources for it to be a good avenue of communication, and until we shift to the new org structure, there's no avenue for the promised community empowerment, so it's just frustrating for everyone.
I, too, miss the days when MeTa was like having coffee or a beer with the mods. I hope that when we finally do the handover to a volunteer board, it can return to being a lower-stakes place for interaction with site leadership, but it can't be that right now. Current site leadership is paid, and strained by the workload, and MeTa only adds to that workload.
We've painted ourselves into a corner with paid mods who can't really be good community members and the nonprofit is the way out of that corner. I know it's unsatisfying to just keep saying "wait until this happens and THEN we can move forward," but continuing to argue about things that can't happen yet isn't helping either.
I have no doubt we'll eventually get volunteer mods and a minimal paid staff, it seems like there's no other solution based on the budget we have. We'll get a marketing plan and new moderation policies. But until the nonprofit is up and running, we need to keep our expectations low of communication between staff and users. That sucks, but a lot has sucked since Google's algorithm broke MeFi so many years ago. It's taken a long time to come to terms with that, and it's fair that the community that has kept coming back here is hungry to take ownership. But it's happening slowly because everyone is trying not to fuck this up.
posted by rikschell at 4:43 AM on June 22 [8 favorites]
MetaTalk is the only part of metafilter where the user community (that's us) talks directly with the mods. It is the only part of metafilter where mods communicate directly to the user community. It's the only part of metafilter to have discussions about metafilter.
It's entirely fair to read "I detest MetaTalk" as "I detest the aspects of my job that require me to engage directly with the metafilter community about metafilter related things."
I suppose it's possible to read that statement and decide that it stops at "I detest the specific mechanism for entering white text onto a gray background" but that's getting into a level of reductive semantics that makes me think uncharitable things about you.
Could MetaTalk be a better tool? Of course. But as literally thousands of comments in the last few years have shown, begging the mods for information, action, feedback, MetaTalk is the only place we have to have those community conversations. If you hate one, you hate the other. It is the only way I'm able to read that statement.
posted by phunniemee at 4:47 AM on June 22 [23 favorites]
It's entirely fair to read "I detest MetaTalk" as "I detest the aspects of my job that require me to engage directly with the metafilter community about metafilter related things."
I suppose it's possible to read that statement and decide that it stops at "I detest the specific mechanism for entering white text onto a gray background" but that's getting into a level of reductive semantics that makes me think uncharitable things about you.
Could MetaTalk be a better tool? Of course. But as literally thousands of comments in the last few years have shown, begging the mods for information, action, feedback, MetaTalk is the only place we have to have those community conversations. If you hate one, you hate the other. It is the only way I'm able to read that statement.
posted by phunniemee at 4:47 AM on June 22 [23 favorites]
Imagine working at a place where people come in openly suggested you shouldn't work here anymore multiple times. For a part-time job that you put in unpaid hours to do.
OK, I’m sure I’m going to get a lot of criticism for this, but it is what it is. I don’t think that loup should be employed in their specific role anymore. I wish them the best, they seem like a cool person, but I don’t think they’re very good at the particular job they have defined for themselves. They weren’t hired for it, they were hired to be a mod. Someone had to step up and they did and I respect that, but it has been long past being obvious that it’s not a great fit. I think they would actually be pretty good as a mod, but they are a disaster as the main person doing payroll and general admin tasks.
A significant amount of money was fundraised for an admin to help them. That hire was never made. I don’t think it was even ever advertised.
They never actually said they were working unpaid. You can say I’m being picky, but there is a significant difference between working more than you want to and working unpaid. In any event, they are the one in charge of scheduling, staffing, and payroll.
They are also the main person dealing with metatalk and they hate it. Fair enough. But at this point the only person able to remove them from that particular position is them. They spend a lot of time interfacing with committees and that has gone disastrously as well, from what I can tell.
I sympathize with them but I also really sympathize with the people who donate money to the site, the people who have worked really hard on the transition team and the steering committee to keep this place going, and the people want this site to keep going in a healthy way. It sucks to be in this position for everyone.
If the answer is to email all this to jessamyn instead of saying it here, fine. Frankly, I don’t think her being hands off is great for the community at this point in time. I realize she’s set up when no one else would, and I have an incredible amount of respect for that. At the same time, having multiple people in leadership who are either incapable or uninterested isn’t working for the site. I think someone needs to figure out why this isn’t a nonprofit. I think someone needs to figure out how we can get this site quickly to the point where it is able to use the volunteer labor it desperately needs. I don’t know why the assumption is that kicking things to a nonprofit structure is six months down the line.
People really want to go back to the cortex model of having one person dictating what’s going on with this site, ok. Ultimately that was a complete disaster for the site. We could’ve been a nonprofit several years ago. We could have a passionate board, similar to the steering committee, doing exciting fundraising and other work. We don’t have that in part because one or two people decided it couldn’t be done and impart because a bunch more people decided that anyone who disagreed with mods was wrong and/or an asshole. Turns out it can be done. Turns out having community leadership was awesome and energizing for the site.
I’m not saying that to relitigate that particular issue. But it’s an allegory for a why trusting one or two people and prioritizing their comfort over the well-being of the site has been shown to have potentially disastrous effects.
posted by knobknosher at 5:33 AM on June 22 [27 favorites]
OK, I’m sure I’m going to get a lot of criticism for this, but it is what it is. I don’t think that loup should be employed in their specific role anymore. I wish them the best, they seem like a cool person, but I don’t think they’re very good at the particular job they have defined for themselves. They weren’t hired for it, they were hired to be a mod. Someone had to step up and they did and I respect that, but it has been long past being obvious that it’s not a great fit. I think they would actually be pretty good as a mod, but they are a disaster as the main person doing payroll and general admin tasks.
A significant amount of money was fundraised for an admin to help them. That hire was never made. I don’t think it was even ever advertised.
They never actually said they were working unpaid. You can say I’m being picky, but there is a significant difference between working more than you want to and working unpaid. In any event, they are the one in charge of scheduling, staffing, and payroll.
They are also the main person dealing with metatalk and they hate it. Fair enough. But at this point the only person able to remove them from that particular position is them. They spend a lot of time interfacing with committees and that has gone disastrously as well, from what I can tell.
I sympathize with them but I also really sympathize with the people who donate money to the site, the people who have worked really hard on the transition team and the steering committee to keep this place going, and the people want this site to keep going in a healthy way. It sucks to be in this position for everyone.
If the answer is to email all this to jessamyn instead of saying it here, fine. Frankly, I don’t think her being hands off is great for the community at this point in time. I realize she’s set up when no one else would, and I have an incredible amount of respect for that. At the same time, having multiple people in leadership who are either incapable or uninterested isn’t working for the site. I think someone needs to figure out why this isn’t a nonprofit. I think someone needs to figure out how we can get this site quickly to the point where it is able to use the volunteer labor it desperately needs. I don’t know why the assumption is that kicking things to a nonprofit structure is six months down the line.
People really want to go back to the cortex model of having one person dictating what’s going on with this site, ok. Ultimately that was a complete disaster for the site. We could’ve been a nonprofit several years ago. We could have a passionate board, similar to the steering committee, doing exciting fundraising and other work. We don’t have that in part because one or two people decided it couldn’t be done and impart because a bunch more people decided that anyone who disagreed with mods was wrong and/or an asshole. Turns out it can be done. Turns out having community leadership was awesome and energizing for the site.
I’m not saying that to relitigate that particular issue. But it’s an allegory for a why trusting one or two people and prioritizing their comfort over the well-being of the site has been shown to have potentially disastrous effects.
posted by knobknosher at 5:33 AM on June 22 [27 favorites]
I've always seen Metatalk as a crucial part of the glue that keeps the community together. Even if moment-to-moment it can come across as a horrible place rife with horrible people and horrible opinions. And people that simply love to hear themselves talk, or love to complain, etc. It's also understandable that "the place where people constantly criticize me" is not a favorite place. I'm not denying any of that - and yet, still - in a lot of ways, it's a pressure valve that people here really need. To be heard, to be listened to. To have some assurance of transparency, and willingness to be forthcoming.
Part of the job of someone like Loup, or Brandon, or anyone running this place and directly interfacing with the community, is to simply be present and listen, and try not to get too upset - publicly, at least - when reading upsetting things. And to engage sincerely, in good faith, and firmly yet kindly even if you feel people are being uncharitable, unfair, or missing the point entirely. There's a balance that our best admin/mods (Jessamyn) managed expertly and it's part of the glue that makes everything work.
I think it's essentially required in order to run this place in an outward-facing way, like it or not.
But explicitly saying you're not interested in doing that, and feel contempt toward the process by which it happens - not just frustration or weariness but full-on hate - I'm not sure what to do with that. Other than to suggest maybe someone, or multiple someones, more receptive to doing that particular work day-to-day steps up to be the primary point person. But I've always figured it to be a challenging, yet necessary, aspect of working on/for the site and keeping the community operating and healthy, full stop.
posted by naju at 5:38 AM on June 22 [21 favorites]
Part of the job of someone like Loup, or Brandon, or anyone running this place and directly interfacing with the community, is to simply be present and listen, and try not to get too upset - publicly, at least - when reading upsetting things. And to engage sincerely, in good faith, and firmly yet kindly even if you feel people are being uncharitable, unfair, or missing the point entirely. There's a balance that our best admin/mods (Jessamyn) managed expertly and it's part of the glue that makes everything work.
I think it's essentially required in order to run this place in an outward-facing way, like it or not.
But explicitly saying you're not interested in doing that, and feel contempt toward the process by which it happens - not just frustration or weariness but full-on hate - I'm not sure what to do with that. Other than to suggest maybe someone, or multiple someones, more receptive to doing that particular work day-to-day steps up to be the primary point person. But I've always figured it to be a challenging, yet necessary, aspect of working on/for the site and keeping the community operating and healthy, full stop.
posted by naju at 5:38 AM on June 22 [21 favorites]
Oh no, why is this not still about the tech change.
And that's why I say the only thing I think Metafilter can do right now are some basic, ineffective, grassroots things.
Yes. My OG question was, have we made a decision that links up to longer-term goals. That is a thinking question, not a doing question. I don’t think I got an answer. I did elaborate. This was more from a “sober second thought” perspective than anything else.
I’m not questioning that the tech needs to change, just the order of operations. I think restless nomad and majick both said “doesn’t matter, things are on fire.” And like, that’s fair. I was still probing a bit.
In my ideal world I would have liked to do a bit of a 5 whys type thing on this decision before investing - do we know how many hours? (I assume loup does.) This might land us where we’re at but the community could understand the whys.
I am sure that kind of feedback can be frustrating; I’ve been frustrated by it myself. But I don’t really like it characterized as “detraction.” Maybe it is, at this stage. I dunno. I’m done with participating in this particular line of discussion. I did go off on the marketing tangent but I know that’s a later question, if there’s a later.
Just making it clear. I have no expectation those marketing activities will take place. In fact, having been through migrations, I’m sure when the team says they are putting all efforts towards a rebuild, that’s what is happening for the next year.
posted by warriorqueen at 5:56 AM on June 22 [4 favorites]
And that's why I say the only thing I think Metafilter can do right now are some basic, ineffective, grassroots things.
Yes. My OG question was, have we made a decision that links up to longer-term goals. That is a thinking question, not a doing question. I don’t think I got an answer. I did elaborate. This was more from a “sober second thought” perspective than anything else.
I’m not questioning that the tech needs to change, just the order of operations. I think restless nomad and majick both said “doesn’t matter, things are on fire.” And like, that’s fair. I was still probing a bit.
In my ideal world I would have liked to do a bit of a 5 whys type thing on this decision before investing - do we know how many hours? (I assume loup does.) This might land us where we’re at but the community could understand the whys.
I am sure that kind of feedback can be frustrating; I’ve been frustrated by it myself. But I don’t really like it characterized as “detraction.” Maybe it is, at this stage. I dunno. I’m done with participating in this particular line of discussion. I did go off on the marketing tangent but I know that’s a later question, if there’s a later.
Just making it clear. I have no expectation those marketing activities will take place. In fact, having been through migrations, I’m sure when the team says they are putting all efforts towards a rebuild, that’s what is happening for the next year.
posted by warriorqueen at 5:56 AM on June 22 [4 favorites]
I think it would help a lot if jessamyn would post what is going on with the transition, with a roadmap and timeline for what's left to be done to change over leadership. I don't think anyone is purposefully stalling, and there were a couple of false starts where it turned out we were doing things legally wrong and had to start again. Everyone is trying to avoid that. (I'm aware that some community members were in favor of powering ahead the legally wrong way under the hope that no one would get litigious, just to get it done. I have some sympathy for that point of view, but I understand that keeping site ownership safe and legal is a valid choice that was made.)
It might calm some folks down to have a sense of when the transfer to community ownership will happen, along with when the board election happens and the director gets hired. Or at least to have realistic goals for making those things happen. MeTa will likely be a nonstop gripefest till then, and I'm sure that's a source of frustration for loup, understandably. But it's also the ONLY PLACE for people to share ideas and get information about what's going on.
MeTa is where the idea for transition to a nonprofit was first floated (and where site staff and ownership swore it could never legally happen). So both the idea sharing and the pushback have a long and troubled history. But we need to prepare ourselves to make this a less fighty place, because if people come to MeTa with the same attitude when it truly IS community-run, it's going to set up a board-vs-users mentality that's not going to be very helpful.
Professionalizing MeFi was something that happened. Deprofessionalizing MeFi (to a large extent) is something that's going to happen. But let's try to stay together as a community while that happens, even if it takes longer than we'd like.
posted by rikschell at 7:30 AM on June 22 [9 favorites]
It might calm some folks down to have a sense of when the transfer to community ownership will happen, along with when the board election happens and the director gets hired. Or at least to have realistic goals for making those things happen. MeTa will likely be a nonstop gripefest till then, and I'm sure that's a source of frustration for loup, understandably. But it's also the ONLY PLACE for people to share ideas and get information about what's going on.
MeTa is where the idea for transition to a nonprofit was first floated (and where site staff and ownership swore it could never legally happen). So both the idea sharing and the pushback have a long and troubled history. But we need to prepare ourselves to make this a less fighty place, because if people come to MeTa with the same attitude when it truly IS community-run, it's going to set up a board-vs-users mentality that's not going to be very helpful.
Professionalizing MeFi was something that happened. Deprofessionalizing MeFi (to a large extent) is something that's going to happen. But let's try to stay together as a community while that happens, even if it takes longer than we'd like.
posted by rikschell at 7:30 AM on June 22 [9 favorites]
Mod note: Note about expectations:
It's the weekend, I'm the only mod schedule for today, Saturday June 22 ET, and there will be long gaps in coverage. It's doubtful there will be much mod input in this thread today.
Please be kind to each other and note that we're all in this boat together, thanks!
posted by Brandon Blatcher (staff) at 8:05 AM on June 22 [13 favorites]
It's the weekend, I'm the only mod schedule for today, Saturday June 22 ET, and there will be long gaps in coverage. It's doubtful there will be much mod input in this thread today.
Please be kind to each other and note that we're all in this boat together, thanks!
posted by Brandon Blatcher (staff) at 8:05 AM on June 22 [13 favorites]
I think it would help a lot if jessamyn would post what is going on with the transition, with a roadmap and timeline for what's left to be done to change over leadership.
The new chair of the interim board is more involved in the transition, and said an update is planned soon.
posted by NotLost at 8:24 AM on June 22 [7 favorites]
The new chair of the interim board is more involved in the transition, and said an update is planned soon.
posted by NotLost at 8:24 AM on June 22 [7 favorites]
Could MetaTalk be a better tool? Of course.
From another angle, I think it's also worth considering whether MetaTalk could be a worse tool. What would have to happen for this subsite to have more negative effects than it sometimes currently has? (That's a thought question: please don't answer.) It's pretty constantly a locus of conflict, and I don't too often hear people saying they love MeTa. Yes the affordances--direct coms with mods and site leadership, lightly regulated conversation--but rarely a full-throated "MetaTalk is my favorite part of this website, I visit it every day, and I'd be sad to see it go." If there are people who feel that way, I'm hoping they weigh in as the site is re-envisioned and rebuilt.
Lanark's point above about rethinking parts of the site makes me wonder: are there parts of the site that have been deprecated and removed over the years? If so, how did we handle that? (Actual question: please answer.) I joined in 2007, but I lurked a lot, wasn't consistently active, and barely looked at anything other than the blue and the green until a few years ago.
posted by cupcakeninja at 8:32 AM on June 22 [2 favorites]
From another angle, I think it's also worth considering whether MetaTalk could be a worse tool. What would have to happen for this subsite to have more negative effects than it sometimes currently has? (That's a thought question: please don't answer.) It's pretty constantly a locus of conflict, and I don't too often hear people saying they love MeTa. Yes the affordances--direct coms with mods and site leadership, lightly regulated conversation--but rarely a full-throated "MetaTalk is my favorite part of this website, I visit it every day, and I'd be sad to see it go." If there are people who feel that way, I'm hoping they weigh in as the site is re-envisioned and rebuilt.
Lanark's point above about rethinking parts of the site makes me wonder: are there parts of the site that have been deprecated and removed over the years? If so, how did we handle that? (Actual question: please answer.) I joined in 2007, but I lurked a lot, wasn't consistently active, and barely looked at anything other than the blue and the green until a few years ago.
posted by cupcakeninja at 8:32 AM on June 22 [2 favorites]
I suppose it's possible to read that statement and decide that it stops at "I detest the specific mechanism for entering white text onto a gray background"
It’s possible to read it as “I don’t think it works well as a venue for communications” but
It is the only part of metafilter where mods communicate directly to the user community. It's the only part of metafilter to have discussions about metafilter.
Yeah, it’s the only extant venue for the broad community of a site that is nominally transitioning to community governance to… discuss governance, so the fact that the people in a position to respond in any sort of official capacity seem to dread doing so - and I respect loup for admitting that, because it has shown for a while - doesn’t make it work better. This isn’t a shot at the rewrite project but if I were running the site I would certainly prioritize improving public communication, perhaps looking to hire somebody who is more comfortable in that role.
posted by atoxyl at 8:54 AM on June 22 [11 favorites]
It’s possible to read it as “I don’t think it works well as a venue for communications” but
It is the only part of metafilter where mods communicate directly to the user community. It's the only part of metafilter to have discussions about metafilter.
Yeah, it’s the only extant venue for the broad community of a site that is nominally transitioning to community governance to… discuss governance, so the fact that the people in a position to respond in any sort of official capacity seem to dread doing so - and I respect loup for admitting that, because it has shown for a while - doesn’t make it work better. This isn’t a shot at the rewrite project but if I were running the site I would certainly prioritize improving public communication, perhaps looking to hire somebody who is more comfortable in that role.
posted by atoxyl at 8:54 AM on June 22 [11 favorites]
It is this kind of complete bad-faith reading that would make MetaTalk such a horrible part of this job
Thank you for calling this out. This discussion is so incredibly unhealthy. I don't read MetaTalk very much because it's usually pretty bad. I followed this discussion for a day or so but removed it from my recents because the comments were so rude.
posted by Nelson at 9:03 AM on June 22 [13 favorites]
Thank you for calling this out. This discussion is so incredibly unhealthy. I don't read MetaTalk very much because it's usually pretty bad. I followed this discussion for a day or so but removed it from my recents because the comments were so rude.
posted by Nelson at 9:03 AM on June 22 [13 favorites]
Just popping in as a long-time user on her second incarnation (who buttoned on that first incarnation because of a MetaTalk thread) to say 1. I'm excited about retiring all the technical debt with a new build for Metafilter, even though that will mean changes (changes, we hates them); 2. loup, you are not alone in hating MetaTalk and wishing people would focus on issues to hand (tech talk) and not other users (x admin should get the sack because we don't like what they said); 3. I would like more community control and all that other good stuff but I also recognize that the wheels turn very slowly here and I accept that and certainly don't blame Jessamyn and loup for it not all going the way I want immediately.
Everybody needs a hug and I'm here to hand them out.
posted by gentlyepigrams at 9:29 AM on June 22 [19 favorites]
Everybody needs a hug and I'm here to hand them out.
posted by gentlyepigrams at 9:29 AM on June 22 [19 favorites]
I don’t like how MetaTalk, the Threads goes.
But I do very much like MetaTalk, the concept where the community does the labour of working out community norms as a group, and talks to the mods, in public, where people can learn from each other. I have learned lots here about how other members experience the site. Sometimes with great surprise, sometimes not.
I think it needs more norms itself, like staying on topic (guilty!) and a [list I don’t want to kick off of how to treat people vs. Ideas.]
But it is MetaTalk discussions that for me makes me invest more than a good subreddit or link-sharing newsletter or forum. I have said “burn it to the ground” but on reflection I actually think it is part of the reason I actually have a longer term care for this community that goes beyond individual posts.
posted by warriorqueen at 9:39 AM on June 22 [20 favorites]
But I do very much like MetaTalk, the concept where the community does the labour of working out community norms as a group, and talks to the mods, in public, where people can learn from each other. I have learned lots here about how other members experience the site. Sometimes with great surprise, sometimes not.
I think it needs more norms itself, like staying on topic (guilty!) and a [list I don’t want to kick off of how to treat people vs. Ideas.]
But it is MetaTalk discussions that for me makes me invest more than a good subreddit or link-sharing newsletter or forum. I have said “burn it to the ground” but on reflection I actually think it is part of the reason I actually have a longer term care for this community that goes beyond individual posts.
posted by warriorqueen at 9:39 AM on June 22 [20 favorites]
I should add, I don’t think MetaTalk is great for UX/design decisions. That should be a wider-ranging process because a metric ton of ppl don’t come into MT.
posted by warriorqueen at 9:49 AM on June 22 [5 favorites]
posted by warriorqueen at 9:49 AM on June 22 [5 favorites]
One thing I haven't heard anyone mention (and maybe it's just the usual MeTa descent into grievances making it hard to find) is MetaFilter, as we know it, is essentially the same codebase as it was in 1999-00. In the years since, features not dreamed of then are now standard, best practices have changed considerably, and there's a lot of prior art across many different tech and app stacks for handling all those things that never got built in ColdFusion. Deploying some of these long-running asks in a modern open-source framework is going to take days, not years.
I have been part of "what if we blew it up and started again" projects, and they are 100% nightmares. But sometimes, you don't have a choice, because the codebase that got you here isn't the codebase to get you to where you need to be, and getting to where you need to be is far, far easier by starting over.
And lastly, as a UX designer with way too many years in this field, this is definitely not the right place to have design convos. This is a subset of MeFi users that may not represent all the kinds of users and tasks MeFi serves.
posted by dw at 12:35 PM on June 22 [22 favorites]
I have been part of "what if we blew it up and started again" projects, and they are 100% nightmares. But sometimes, you don't have a choice, because the codebase that got you here isn't the codebase to get you to where you need to be, and getting to where you need to be is far, far easier by starting over.
And lastly, as a UX designer with way too many years in this field, this is definitely not the right place to have design convos. This is a subset of MeFi users that may not represent all the kinds of users and tasks MeFi serves.
posted by dw at 12:35 PM on June 22 [22 favorites]
If it were up to me, everything other than WWW. and Ask. would be behind the login wall and hidden from the Google index.
there's no reason to keep the music site behind a login wall and hidden from the google index - at least if i'm going to keep posting there
posted by pyramid termite at 1:31 PM on June 22 [2 favorites]
there's no reason to keep the music site behind a login wall and hidden from the google index - at least if i'm going to keep posting there
posted by pyramid termite at 1:31 PM on June 22 [2 favorites]
Just popped in to say I would be willing to be a tester. I've served as an alpha/beta tester on a large project before and am known for breaking things.
posted by kathrynm at 3:09 PM on June 22 [1 favorite]
posted by kathrynm at 3:09 PM on June 22 [1 favorite]
As someone who knocked together a basic MeFi clone in PHP/MySQL in, uh, the year 2001, I just want to say: finally. It sounds like the approach is sound, the technology is pretty boring and reasonable, and just re-skinning some other forum software would have been at least of much of a nightmare and would really kill what I like about this place. (YMMV.)
posted by jimw at 8:44 PM on June 22 [8 favorites]
posted by jimw at 8:44 PM on June 22 [8 favorites]
I would like to participate in the trans members initiative listening session / survey but I will be far away from internet connectivity during the first week and a bit of July. No specific date for the listening session or survey has been mentioned on the trans members initiative page so I'm not sure when they are occurring.
I hope I don't miss my chance to have a say.
posted by neonamber at 9:38 PM on June 22 [1 favorite]
I hope I don't miss my chance to have a say.
posted by neonamber at 9:38 PM on June 22 [1 favorite]
I'm not sure why we're expressing opinions on a decision that's been made, but I think it's bonkers to be writing a message board system from scratch in 2024. It's like deciding to write your own word processor from scratch: it might be a fun experiment but it's never going to get close to LibreOffice Writer or Microsoft Word.
The good thing is that hosting costs will go down. But at the end we'll still be stuck with a customized solution that's dependent on the goodwill of a coder with the right skillset for even basic security updates as well as new features.
Then at the end when people do start asking for features, e.g. mod tools or better accessibility or sharing to social media, each one will have to be coded from scratch when an existing package would likely just have it already.
Seems far better to just use an existing message board system and customize it to be as Metafilter like as possible. Then migrate the old site to static HTML and start using the new one. Maybe migrate FanFare posts and comments if people really want to be able to comment on old threads.
posted by TheophileEscargot at 3:00 AM on June 23 [9 favorites]
The good thing is that hosting costs will go down. But at the end we'll still be stuck with a customized solution that's dependent on the goodwill of a coder with the right skillset for even basic security updates as well as new features.
Then at the end when people do start asking for features, e.g. mod tools or better accessibility or sharing to social media, each one will have to be coded from scratch when an existing package would likely just have it already.
Seems far better to just use an existing message board system and customize it to be as Metafilter like as possible. Then migrate the old site to static HTML and start using the new one. Maybe migrate FanFare posts and comments if people really want to be able to comment on old threads.
posted by TheophileEscargot at 3:00 AM on June 23 [9 favorites]
Laravel comes with lots of bells and whistles and there are plenty of libraries or packages or whatever you want to call them. Here's one, for example https://www.teamteatime.net/docs/laravel-forum/6.x/general/
posted by karasu at 5:18 AM on June 23 [4 favorites]
posted by karasu at 5:18 AM on June 23 [4 favorites]
I'm not sure why we're expressing opinions on a decision that's been made, but I think it's bonkers to be writing a message board system from scratch in 2024. It's like deciding to write your own word processor from scratch: it might be a fun experiment but it's never going to get close to LibreOffice Writer or Microsoft Word.
But on the other hand, writing such things gets easier as time passes. We've now got Python and Ruby and other such languages, where substantial parts of your work have been done for you in prebuilt libraries.
posted by JHarris at 5:47 AM on June 23 [2 favorites]
But on the other hand, writing such things gets easier as time passes. We've now got Python and Ruby and other such languages, where substantial parts of your work have been done for you in prebuilt libraries.
posted by JHarris at 5:47 AM on June 23 [2 favorites]
Seems far better to just use an existing message board system and customize it to be as Metafilter like as possible. Then migrate the old site to static HTML and start using the new one
I think you're vastly underestimating how difficult this would be, unless the plan is just to ditch every user's existing history, favorites and posts and start everyone from scratch, which I'm guessing would not be popular with the existing userbase. MeFi has an existing database schema which will differ significantly from the schema of any existing software. Just writing the code to migrate between the two schemas is a ton of work, if the new software even has concepts like, say, favorites, and then you have another huge chunk of work in getting the message-board system to behave as much like present-day MetaFilter as possible, and then another chunk to make it look the same.
The thing is that at its heart, beneath the mountains of tech debt, MetaFilter is a pretty bog-standard CRUD app of the type that is often used as an example in programming exercises ("write a system with users, posts and comments, where there are 0..n comments on a given post"). I imagine this was a factor in kirkaracha making fast progress on the prototype of the rewrite, and it also means that once the code is open-sourced, it should be pretty easy for other developers to contribute, especially given kirkaracha's (wise) decision to use a very widely-used and well-supported framework as a base.
posted by whir at 8:28 AM on June 23 [11 favorites]
I think you're vastly underestimating how difficult this would be, unless the plan is just to ditch every user's existing history, favorites and posts and start everyone from scratch, which I'm guessing would not be popular with the existing userbase. MeFi has an existing database schema which will differ significantly from the schema of any existing software. Just writing the code to migrate between the two schemas is a ton of work, if the new software even has concepts like, say, favorites, and then you have another huge chunk of work in getting the message-board system to behave as much like present-day MetaFilter as possible, and then another chunk to make it look the same.
The thing is that at its heart, beneath the mountains of tech debt, MetaFilter is a pretty bog-standard CRUD app of the type that is often used as an example in programming exercises ("write a system with users, posts and comments, where there are 0..n comments on a given post"). I imagine this was a factor in kirkaracha making fast progress on the prototype of the rewrite, and it also means that once the code is open-sourced, it should be pretty easy for other developers to contribute, especially given kirkaracha's (wise) decision to use a very widely-used and well-supported framework as a base.
posted by whir at 8:28 AM on June 23 [11 favorites]
I've led a few "let's rebuild the whole thing from scratch" projects. If the functional requirements are simple, like they are in this case, it's far easier to make a new thing using a sane language, frameworks, and practices than it is to try to bend the existing thing into shape.
This has been succesful for me, for example, migrating from a Wordpress-based* site to a Django based one, and Wordpress is a lot less tech-debty than ColdFusion, to put it mildly.
I agree with rockindata about avoiding React and other front-end frameworks du jour. I also agree with not using a pre-existing app, as we would immediatley inherit a whole new exciting challenge of trying to bend it to do what we want.
* Not because WordPress is intrinsically bad**, but because people tend to use if for things it wasn't designed for and end up with a slow, almost-impossible to admin, chockful of vulnerabilities frankensite.
** Though I personally dislike it, but that's a whole 'nother topic.
posted by signal at 8:42 AM on June 23 [6 favorites]
This has been succesful for me, for example, migrating from a Wordpress-based* site to a Django based one, and Wordpress is a lot less tech-debty than ColdFusion, to put it mildly.
I agree with rockindata about avoiding React and other front-end frameworks du jour. I also agree with not using a pre-existing app, as we would immediatley inherit a whole new exciting challenge of trying to bend it to do what we want.
* Not because WordPress is intrinsically bad**, but because people tend to use if for things it wasn't designed for and end up with a slow, almost-impossible to admin, chockful of vulnerabilities frankensite.
** Though I personally dislike it, but that's a whole 'nother topic.
posted by signal at 8:42 AM on June 23 [6 favorites]
I think you're vastly underestimating how difficult this would be, unless the plan is just to ditch every user's existing history, favorites and posts and start everyone from scratch, which I'm guessing would not be popular with the existing userbase. MeFi has an existing database schema which will differ significantly from the schema of any existing software. Just writing the code to migrate between the two schemas is a ton of work
This way we would neither ditch the old data, nor do a migration. There are tools to crawl a site and save the HTML. The old site will remain as HTML, so each comment will still have your name and your favorite count next to it. The old site's "comments by Whir" and "favorited by Whir" HTML will remain so you can look back at your own comments and favorites if you like.
If you want to look back through everything you would need to do it twice though, once on the old site and once on the new.
The thing is that writing new code is fun. Customizing and writing plugins for existing software is boring. Maintaining legacy code is hell. So every developer who's faced with a choice between those options has a tendency to rationalize that the fun option is the best option.
"Never rewrite from scratch" isn't exactly a rule, because sometimes it really is the best option. But it's a good place to start, because it frames the decision as "give me really, really good reasons why a rewrite is the best option".
posted by TheophileEscargot at 11:03 AM on June 23 [1 favorite]
This way we would neither ditch the old data, nor do a migration. There are tools to crawl a site and save the HTML. The old site will remain as HTML, so each comment will still have your name and your favorite count next to it. The old site's "comments by Whir" and "favorited by Whir" HTML will remain so you can look back at your own comments and favorites if you like.
If you want to look back through everything you would need to do it twice though, once on the old site and once on the new.
The thing is that writing new code is fun. Customizing and writing plugins for existing software is boring. Maintaining legacy code is hell. So every developer who's faced with a choice between those options has a tendency to rationalize that the fun option is the best option.
"Never rewrite from scratch" isn't exactly a rule, because sometimes it really is the best option. But it's a good place to start, because it frames the decision as "give me really, really good reasons why a rewrite is the best option".
posted by TheophileEscargot at 11:03 AM on June 23 [1 favorite]
This way we would neither ditch the old data, nor do a migration. There are tools to crawl a site and save the HTML. The old site will remain as HTML, so each comment will still have your name and your favorite count next to it.
So what happens when someone asks to have their account wiped? Would tools need to be developed to edit the old archive? Keep the old code running somewhere so it can be wiped and the “static” copy regenerated?
If you want to look back through everything you would need to do it twice though, once on the old site and once on the new.
Is that an acceptable trade-off for the unidentified benefits of replacing MeFi with an off-the-shelf platform?
posted by jimw at 11:17 AM on June 23 [2 favorites]
So what happens when someone asks to have their account wiped? Would tools need to be developed to edit the old archive? Keep the old code running somewhere so it can be wiped and the “static” copy regenerated?
If you want to look back through everything you would need to do it twice though, once on the old site and once on the new.
Is that an acceptable trade-off for the unidentified benefits of replacing MeFi with an off-the-shelf platform?
posted by jimw at 11:17 AM on June 23 [2 favorites]
I have no doubt that the codebase causes issues. I accept that this is the chosen way forward. But I don't think that it is going to help MetaFilter survive unless the code does what we NEED it to do, and I just don't think there's been a process to uncover that.
There are lots of good things about MeTa, but building consensus on big issues is definitely not one of them. It seems obvious to me that, for a lot of the things that need to happen for MetaFilter to survive or even thrive, there needs to be a platform that allows those things to happen. For years now, people have suggested all sorts of changes, many of which make a lot of sense and the response almost always includes '... but we can't because of the way the site is coded'.
It seems like it's pretty clear to almost everyone participating here that one of the biggest hurdles to positive change in how things work is getting onto a platform that allows change and that allows the community to participate in developing and maintaining that platform. I haven't seen anyone disagree with that, although some hold a reasonable view that maybe it's not what we should focus on right now. From my perspective and in my opinion, I'm over the moon that someone (and I don't really care who) has finally stood up and decided to fucking do something about the ancient, creaking and risky platform we all rely on. I'll guarantee, if a MeTa thread was started asking 'should we rebuild the site code starting right now', the response would be largely this very thread. I'm glad jessamyn and loup went ahead and made this decision and I'm very glad to see this process get moving in a way that will allow the community to be involved long-term in looking after the platform. The idea of building the new platform to where it can duplicate the existing functions seems like an excellent idea and that is the time when the community should get to chime in with where and how things maybe should change. No doubt, that's what will happen.
But at least something is finally getting done. It's clearly not the One Very Important Thing that everyone wants to be done right now and that's OK and normal for a community. But, of all the things that could be done right now, I'm glad they picked this one. This is the one that has been stopping most of the changes people have been asking for from happening (or the excuse - I guess we'll find out). All of the work around becoming a non-profit is critical, of course, but there's no law where we have to do only one thing at a time.
There are days when I hate MetaTalk too, today being one of those days. Most people, even when they have very strong dissatisfaction with how things are going, are respectful but holy shit some people are entitled and borderline abusive.
posted by dg at 7:42 PM on June 23 [35 favorites]
There are lots of good things about MeTa, but building consensus on big issues is definitely not one of them. It seems obvious to me that, for a lot of the things that need to happen for MetaFilter to survive or even thrive, there needs to be a platform that allows those things to happen. For years now, people have suggested all sorts of changes, many of which make a lot of sense and the response almost always includes '... but we can't because of the way the site is coded'.
It seems like it's pretty clear to almost everyone participating here that one of the biggest hurdles to positive change in how things work is getting onto a platform that allows change and that allows the community to participate in developing and maintaining that platform. I haven't seen anyone disagree with that, although some hold a reasonable view that maybe it's not what we should focus on right now. From my perspective and in my opinion, I'm over the moon that someone (and I don't really care who) has finally stood up and decided to fucking do something about the ancient, creaking and risky platform we all rely on. I'll guarantee, if a MeTa thread was started asking 'should we rebuild the site code starting right now', the response would be largely this very thread. I'm glad jessamyn and loup went ahead and made this decision and I'm very glad to see this process get moving in a way that will allow the community to be involved long-term in looking after the platform. The idea of building the new platform to where it can duplicate the existing functions seems like an excellent idea and that is the time when the community should get to chime in with where and how things maybe should change. No doubt, that's what will happen.
But at least something is finally getting done. It's clearly not the One Very Important Thing that everyone wants to be done right now and that's OK and normal for a community. But, of all the things that could be done right now, I'm glad they picked this one. This is the one that has been stopping most of the changes people have been asking for from happening (or the excuse - I guess we'll find out). All of the work around becoming a non-profit is critical, of course, but there's no law where we have to do only one thing at a time.
There are days when I hate MetaTalk too, today being one of those days. Most people, even when they have very strong dissatisfaction with how things are going, are respectful but holy shit some people are entitled and borderline abusive.
posted by dg at 7:42 PM on June 23 [35 favorites]
Please just keep the headphone jack
posted by General Malaise at 7:30 AM on June 24 [21 favorites]
posted by General Malaise at 7:30 AM on June 24 [21 favorites]
One is pleased that The Cabal have ritually decided on Laravel for implementing MetaFilter 2.0, as this particular quietly googles free and open-source PHP-based web framework for building high-end web applications will enable us to uh leverage the synergies of the existing database out of the box. I think we can all agree this is a far better choice than popular web frameworks such as uh GeoCities or Hawk Tuah etc etc.
I have no fucking idea what I'm talking about and am still amazed I somehow got a Master's Degree in Data Engineering back in 1992. But it's MetaTalk, so.
posted by Wordshore at 8:50 AM on June 24 [14 favorites]
I have no fucking idea what I'm talking about and am still amazed I somehow got a Master's Degree in Data Engineering back in 1992. But it's MetaTalk, so.
posted by Wordshore at 8:50 AM on June 24 [14 favorites]
Can we still have the occasional (fake) "JRun Error"? As a treat?
posted by slater at 10:05 AM on June 24 [9 favorites]
posted by slater at 10:05 AM on June 24 [9 favorites]
mbrubeck > Happy to see MetaFilter finally moving to Active Server Pages 3.0.
And hey.. If you need someone experienced in DreamWeaver circa mid-2000's, just ask.
posted by Hardcore Poser at 10:16 AM on June 24 [4 favorites]
And hey.. If you need someone experienced in DreamWeaver circa mid-2000's, just ask.
posted by Hardcore Poser at 10:16 AM on June 24 [4 favorites]
Ignore the haters and doom-sayers. Getting away from ColdFusion is definitely worth it. Adding affordances to simplify future improvements is YES worth it. Adding back the IMG tag is totally table stakes. :7)
I wouldn't blame you all for fading from these MetaTalk threads: they always turn into a food fight. Write your spec, write your code, run your tests, and have a few people alpha- and beta-test it. Vaya con dios.
posted by wenestvedt at 11:19 AM on June 24 [11 favorites]
I wouldn't blame you all for fading from these MetaTalk threads: they always turn into a food fight. Write your spec, write your code, run your tests, and have a few people alpha- and beta-test it. Vaya con dios.
posted by wenestvedt at 11:19 AM on June 24 [11 favorites]
I'm just amazed I got to the bottom of this thread and nobody was rude about PHP/Laravel.
(I have no opinion on the matter)
Django forever!
posted by Sparx at 5:06 PM on June 24 [7 favorites]
(I have no opinion on the matter)
Django forever!
posted by Sparx at 5:06 PM on June 24 [7 favorites]
MetaTalk used to be a fun place to hang out in.
The thing is, I still think it is. For the most part. I became a lurker of the site back in 2003 (and, technically, I still am?), but back then I read mostly just the blue. Then, when I joined I found myself reading the blue less and less because it is hard for me to take the mod hat off. And, today, MetaTalk is (although, admittedly, second to FanFare) one of the subsites where I find myself bookmarking conversations to read when I’m not working. It’s more than fun, it's very human and quirky, sometimes I'd say enlightening...
on reflection I actually think it is part of the reason I actually have a longer term care for this community that goes beyond individual posts.
I agree, wholeheartedly.
it sounds like you are burnt out for the same reasons he was burnt out, or at least some of them. we are obviously still in a transitional period but hearing this is dispiriting.
Except, I’m not burned out, not yet though.
Hell, there are still probably things from the last few locked threads that still haven't been responded to.
Yes, and we’ll go back to them soon, hence the note I left when I closed it. Sometimes there are just competing priorities. Let me provide context (that I could not provide back then). These past two metas required attention, researching to get a clear timeline of what happened, where the miscommunication happened, catching up with a thread that exploded fast, conferring with other team members. All fine until then, but at the same time, we had been working on the ins and outs of the site rebuild and planning for this announcement and to move forward with the project for Trans and Nonbinary members and those two projects required attention as well. Then travelingthyme was in an accident. Details are at their GoFundMe (they gave consent to disclose their IRL identity).
Of, course, we will not be discussing their personal situation further but I hope that gives some context on why that thread was closed in favor of other competing priorities.
posted by loup (staff) at 6:17 PM on June 24 [15 favorites]
The thing is, I still think it is. For the most part. I became a lurker of the site back in 2003 (and, technically, I still am?), but back then I read mostly just the blue. Then, when I joined I found myself reading the blue less and less because it is hard for me to take the mod hat off. And, today, MetaTalk is (although, admittedly, second to FanFare) one of the subsites where I find myself bookmarking conversations to read when I’m not working. It’s more than fun, it's very human and quirky, sometimes I'd say enlightening...
on reflection I actually think it is part of the reason I actually have a longer term care for this community that goes beyond individual posts.
I agree, wholeheartedly.
it sounds like you are burnt out for the same reasons he was burnt out, or at least some of them. we are obviously still in a transitional period but hearing this is dispiriting.
Except, I’m not burned out, not yet though.
Hell, there are still probably things from the last few locked threads that still haven't been responded to.
Yes, and we’ll go back to them soon, hence the note I left when I closed it. Sometimes there are just competing priorities. Let me provide context (that I could not provide back then). These past two metas required attention, researching to get a clear timeline of what happened, where the miscommunication happened, catching up with a thread that exploded fast, conferring with other team members. All fine until then, but at the same time, we had been working on the ins and outs of the site rebuild and planning for this announcement and to move forward with the project for Trans and Nonbinary members and those two projects required attention as well. Then travelingthyme was in an accident. Details are at their GoFundMe (they gave consent to disclose their IRL identity).
Of, course, we will not be discussing their personal situation further but I hope that gives some context on why that thread was closed in favor of other competing priorities.
posted by loup (staff) at 6:17 PM on June 24 [15 favorites]
If I might request features? The early days of the site we never dreamed that we might have periodic free threads and LinkMe threads. It might be useful to have a header with links to the most recent of both things for ease of finding. Not at the very top of the page, where announcements go, but a couple of small links, maybe with dates, at the top of the page scroll?
Also, there is this thing some of you may have heard of called the "Fediverse," and some sites emit posts that show up there so that others can follow and comment. It might be interesting/useful to do that too. I'm not suggesting full Fediverse support, to the extent where non-members can comment in threads on the site, but maybe an automatic echoing of FPP content there. If it can be done with Bluesky too then sure, why not, but I think this kind of thing is more supported on Mastodon?
posted by JHarris at 7:27 PM on June 24 [6 favorites]
Also, there is this thing some of you may have heard of called the "Fediverse," and some sites emit posts that show up there so that others can follow and comment. It might be interesting/useful to do that too. I'm not suggesting full Fediverse support, to the extent where non-members can comment in threads on the site, but maybe an automatic echoing of FPP content there. If it can be done with Bluesky too then sure, why not, but I think this kind of thing is more supported on Mastodon?
posted by JHarris at 7:27 PM on June 24 [6 favorites]
Doesn't Metafilter emit RSS? Or used to? I think that it's supposed to be easy to make a Mastodon bot that echoes an RSS feed. Maybe that would be a good solution with current tech?
posted by biogeo at 8:07 PM on June 24 [1 favorite]
posted by biogeo at 8:07 PM on June 24 [1 favorite]
Geez, that is awful news for travelingthyme. Please wish them all the best from this Internet stranger.
posted by Paragon at 11:05 PM on June 24 [28 favorites]
posted by Paragon at 11:05 PM on June 24 [28 favorites]
I'm just amazed I got to the bottom of this thread and nobody was rude about PHP/Laravel.
I was going to be, partly because of past experiences having to maintain PHP applications where I work, but really, almost anything is a step up from ColdFusion.
FLASK 4 LIFE
posted by Mr. Bad Example at 2:02 AM on June 25 [1 favorite]
I was going to be, partly because of past experiences having to maintain PHP applications where I work, but really, almost anything is a step up from ColdFusion.
FLASK 4 LIFE
posted by Mr. Bad Example at 2:02 AM on June 25 [1 favorite]
Really sorry for travelingthyme and wishing them the best in recovery.
posted by eirias at 2:59 AM on June 25 [10 favorites]
posted by eirias at 2:59 AM on June 25 [10 favorites]
I'm so very sorry thyme is going through this right now. If you can, folks, please consider the GoFundMe, I'm sure it will help.
I'm just amazed I got to the bottom of this thread and nobody was rude about PHP/Laravel.
If Kirk says it's a good idea, it's a good idea. I trust.
posted by majick at 7:35 AM on June 25 [6 favorites]
I'm just amazed I got to the bottom of this thread and nobody was rude about PHP/Laravel.
If Kirk says it's a good idea, it's a good idea. I trust.
posted by majick at 7:35 AM on June 25 [6 favorites]
Thanks for letting us know how to help travelingthyme and family out. I'm so sorry to hear of their loss.
posted by Stacey at 7:45 AM on June 25 [6 favorites]
posted by Stacey at 7:45 AM on June 25 [6 favorites]
I'm so sorry about travelingthyme's situation!
I'm just amazed I got to the bottom of this thread and nobody was rude about PHP/Laravel.
I'm glad people have restrained themselves from language snobbery. A lot of people use PHP, so this choice may help increase the expected number of available open-source contributors when that becomes possible.
On the note of "when that becomes possible"- I want to note that MeFi does not have to be a non-profit to accept pull requests on an open-source repository. For example, the open-source project React is maintained by for-profit Facebook/Meta. (That doesn't mean it's reasonable for readers to expect immediate open-sourcing of the project! For that to happen, passwords/secrets and some configuration information need to be isolated from the open-source part. I just mean that if the non-profit legalities take longer than expected, that doesn't have to block open-sourcing the codebase.)
posted by a faded photo of their beloved at 8:10 AM on June 25 [5 favorites]
I'm just amazed I got to the bottom of this thread and nobody was rude about PHP/Laravel.
I'm glad people have restrained themselves from language snobbery. A lot of people use PHP, so this choice may help increase the expected number of available open-source contributors when that becomes possible.
On the note of "when that becomes possible"- I want to note that MeFi does not have to be a non-profit to accept pull requests on an open-source repository. For example, the open-source project React is maintained by for-profit Facebook/Meta. (That doesn't mean it's reasonable for readers to expect immediate open-sourcing of the project! For that to happen, passwords/secrets and some configuration information need to be isolated from the open-source part. I just mean that if the non-profit legalities take longer than expected, that doesn't have to block open-sourcing the codebase.)
posted by a faded photo of their beloved at 8:10 AM on June 25 [5 favorites]
loup: "All fine until then, but at the same time, we had been working on the ins and outs of the site rebuild and planning for this announcement and to move forward with the project for Trans and Nonbinary members and those two projects required attention as well. Then travelingthyme was in an accident. Details are at their GoFundMe (they gave consent to disclose their IRL identity)."
Wtf, this is horrible. TT, I am so sorry.
Loup, I know there's a general rule against posting fundraisers and a tendency to keep personal matters separate, but can we please make an exception for someone whose literal *job* it is to help this place run who's in such dire straits? At least something more visible than an offhand mention at the bottom of a week-old 200+-comment thread. (As long as TT is okay with it ofc.) I'm at work but can submit something here later tonight if nobody else gets to it first.
posted by Rhaomi at 1:58 PM on June 25 [19 favorites]
Wtf, this is horrible. TT, I am so sorry.
Loup, I know there's a general rule against posting fundraisers and a tendency to keep personal matters separate, but can we please make an exception for someone whose literal *job* it is to help this place run who's in such dire straits? At least something more visible than an offhand mention at the bottom of a week-old 200+-comment thread. (As long as TT is okay with it ofc.) I'm at work but can submit something here later tonight if nobody else gets to it first.
posted by Rhaomi at 1:58 PM on June 25 [19 favorites]
Sympathies to travelingthyme. So sorry for what you're going through. Donation sent.
posted by hydra77 at 1:58 PM on June 25 [1 favorite]
posted by hydra77 at 1:58 PM on June 25 [1 favorite]
I've also been thinking about the fundraiser visibility since last night. Exceptions to the "no fundraising" policy does risk opening a slippery slope, but if there's any good case for it, in my opinion this is it. This is a terrible situation, and I think this community has a specific reciprocal duty of care to travelingthyme that differs qualitatively from other worthy causes which we may also approve of but wouldn't normally link.
At the very least perhaps making sure the information is more widely disseminated via the Free Thread on the Blue if it hasn't already been mentioned there? I posted it on Mefi.social (Mastodon) and it clearly got some visibility/traction there, but of course the most invested group of people to reach for help is right here on MetaFilter.
posted by biogeo at 2:28 PM on June 25 [8 favorites]
At the very least perhaps making sure the information is more widely disseminated via the Free Thread on the Blue if it hasn't already been mentioned there? I posted it on Mefi.social (Mastodon) and it clearly got some visibility/traction there, but of course the most invested group of people to reach for help is right here on MetaFilter.
posted by biogeo at 2:28 PM on June 25 [8 favorites]
Whoa, agree that there should be MUCH more site visibility for travelingthyme's situation (if they're ok with it). Maybe a whole thread? So tragic - wishing them care and sympathy.
posted by nouvelle-personne at 2:37 PM on June 25 [4 favorites]
posted by nouvelle-personne at 2:37 PM on June 25 [4 favorites]
All metafilter sites except for metatalk are showing 403 forbidden errors for me right now.
posted by msbrauer at 3:24 PM on June 25 [1 favorite]
posted by msbrauer at 3:24 PM on June 25 [1 favorite]
Mod note: All metafilter sites except for metatalk are showing 403 forbidden errors for me right now.
I'm not seeing this. Reporting it now.
posted by loup (staff) at 3:24 PM on June 25
I'm not seeing this. Reporting it now.
posted by loup (staff) at 3:24 PM on June 25
Nevermind...seems to be an issue with my VPN. Switched off and on again and all are working. Sorry to raise the alarm bells!
posted by msbrauer at 3:30 PM on June 25 [1 favorite]
posted by msbrauer at 3:30 PM on June 25 [1 favorite]
Whoa, agree that there should be MUCH more site visibility for travelingthyme's situation (if they're ok with it).
Seconded. (And my deepest sympathies and healing wishes to travelingthyme)
posted by paper scissors sock at 5:33 PM on June 25 [1 favorite]
Seconded. (And my deepest sympathies and healing wishes to travelingthyme)
posted by paper scissors sock at 5:33 PM on June 25 [1 favorite]
Have we considered implementing Metafilter in a font?
posted by Admiral Viceroy at 3:43 AM on June 26 [7 favorites]
posted by Admiral Viceroy at 3:43 AM on June 26 [7 favorites]
I just want to say, does it seem like the front page is more active than usual? It's great to see so much posting. I wish I had the time to dig into all of them, but YAY TEAM. Let a thousand FPPs bloom!
posted by rikschell at 1:19 PM on June 27 [7 favorites]
posted by rikschell at 1:19 PM on June 27 [7 favorites]
Coming in very late to this thread to note that I had a dream the other day that the new version of Metafilter included a "Simple" option that loaded the site into Picotron. Ask Metafilter would then generate a pixel graphics scene based on the tags assigned to a question.
posted by ewok_academy at 12:06 PM on July 3 [2 favorites]
posted by ewok_academy at 12:06 PM on July 3 [2 favorites]
I dont come around here much anymore - mostly for the reasons identified by the details and tone of this thread.
I am surprised no one mentioned that the site lost almost 18k dollars in June. I dont understand how that is not a huge redflag lol.
posted by archivist at 9:24 AM on July 13
I am surprised no one mentioned that the site lost almost 18k dollars in June. I dont understand how that is not a huge redflag lol.
posted by archivist at 9:24 AM on July 13
It's a good concern to have raised, but that was answered upthread. (Two months of payroll happened to hit the books during this calendar month, but that doesn't represent any actual increase in spending since it'll even out between the preceding or following months.)
posted by nobody at 5:36 PM on July 14 [4 favorites]
posted by nobody at 5:36 PM on July 14 [4 favorites]
« Older Hello from your friendly neighbourhood interim... | Interviewees, outlets sought for PR pitch Newer »
You are not logged in, either login or create an account to post comments
posted by signal at 3:36 PM on June 19 [17 favorites]