[some appropriate and witty Hamilton post title here] October 14, 2016 7:09 AM   Subscribe

I have questions about the rate of increase in commenting over this year's U.S. presidential election period.

It's my understanding we're in unprecedented territory here with the frequent need to make new election posts because the then-current one has grown unwieldy with 3,000 or 4,000 comments overall. I have a vague awareness of the Infodump but instead of learning what it is and how to use it I figured I would just ask if anyone has already looked up the numbers in some of the following. I guess I'll use Comments Per Day (CPD) as a metric. Maybe comments per hour would be more appropriate, I don't know.

- How does the CPD during this election period compare with the CPD during the last election?
- How does the CPD during this election period compare with the CPD during the same period last year (a non-election time)?
- How has the CPD rate increased even during this election period?

Basically I like numbers and graphs and want concrete proof that the amount of people typing in the comments box is not only unprecedented but an astronomical increase from previous times, because that sure is how it feels and I want my feelings vindicated with data.
posted by komara to MetaFilter-Related at 7:09 AM (50 comments total) 2 users marked this as a favorite

It'd be interesting to see some analysis of that! The Comments Per Day metric is a good way of thinking about it, though there's knock-on effects that'd be interesting to look at: not just what the raw CPD has been like but also what the proportion of CPD belongs to stuff tagged with election2016 vs. not, what the parallel changes in posting volume have been, etc.

I don't know whether 2012 election stuff was as consistently labeled as this year's cycle's have been; that we've been so actively pushing for one thread at the time may have made it easier to be steady about that than it would have been with a more scattered approach. So there might be an opportunity to do some searching and backtagging there.

In any case, for the graphs and numbers front the thing I'm going to point to people to, as always, is the raw data in the Infodump, which continues to get updated weekly and has skeletal stats about posts and comments suited well to this kind of number-crunching.
posted by cortex (staff) at 7:15 AM on October 14, 2016 [1 favorite]


CPD is already occupied this season
Crap Party's Decline
posted by Namlit at 7:40 AM on October 14, 2016


"but also what the proportion of CPD belongs to stuff tagged with election2016 vs. not"

Good point. I forget about tagging being included. I was thinking only of raw CPD increase. Maybe an analysis of the last 10 years saying something like
2006: xxx CPD
2007: xxx CPD (1.5% increase over last year)
2008: xxx CPD (1.8% increase over last year)
...
2016: xxx CPD (5.3% increase over last year)
so a baseline of expected commentary inflation could be established and the increase in 2016 could be judged against that.

But tagging included gives an even bigger depth to explore. With that available then we could see if non-election 2016 has the expected rate of CPD increase and it's just the election threads that are nuts, or if all FPPs are experiencing increased CPD, etc.
posted by komara at 7:45 AM on October 14, 2016


[Talk less, smile more] would have worked.
posted by roomthreeseventeen at 9:22 AM on October 14, 2016 [8 favorites]


[Can We Get Back To Politics. Please?]
posted by Gorgik at 9:50 AM on October 14, 2016 [6 favorites]


[fake] and [real] tags remind me of [fark]. [florida]
posted by ZeusHumms at 10:23 AM on October 14, 2016 [1 favorite]


I love these threads: it's been like pouring starter fertilizer on my Favorites count!
posted by wenestvedt at 1:10 PM on October 14, 2016 [9 favorites]


My observation is that other posting/commenting has dropped off noticeably while everyone has been focusing on the election threads.
posted by Greg_Ace at 4:07 PM on October 14, 2016 [10 favorites]


♪♫... R-E-S-P-E-C-T, a little Respect,
   ♪♫...Take out the CPD,
    ♪♫...sock it to me! sock it to me! sock it to me!
posted by y2karl at 4:18 PM on October 14, 2016


Apparently, we're all way too busy with the election thread to do any kind of analysis of the infodump.
posted by zachlipton at 7:08 PM on October 14, 2016 [2 favorites]


[Talk less, smile more] would have worked.

I would also accept:
[Fools Who Run Their Mouths], [I gotta holler just to be heard], and most of Non-Stop.
posted by zamboni at 7:17 PM on October 14, 2016


What percentage of those comments are comments commenting about the number of comments commenters comment in those threads? It feels like a lot.
posted by Alvy Ampersand at 12:24 AM on October 15, 2016 [3 favorites]


I don't have cold, hard data that you can prise from my English hand; just anecdote and observation. Namely, that in long-gone innocent times (Spring of this year), one FPP on the election would saunter along jollily for a good week without it getting so browser-breakingly loaded with comments that a new one would be needed.

Now, we're at two or more posts per week and a thousand or so comments per (non weekend) day is roughly the norm. True, we are very near the election and all manner of stuff is happening. I do wonder where folks will put their commenting energies post-election, but whatever the result there's going to be a lot of upheaval and aftermath (much probably unpleasant) so am not expecting a drastic tail-off of commenting.

Two side-points:

- Currently, the top four "popular" posts on MetaFilter in the last week are all election posts. Huh.

- The election day post for 2012 - relatively quieter times - received 8,185 comments although many of these were submitted in days afterwards. It could be that, so people can adequately comment on their tech, the community needs more than one election post for that day alone (which, with the 1-FPP-in-24-hours limit on FPP creators may require a bit of co-ordination beforehand).
posted by Wordshore at 2:11 AM on October 15, 2016


My observation is that other posting/commenting has dropped off noticeably while everyone has been focusing on the election threads.

This is also my non-scientific but very strong observation. And the mods are making it worse by driving traffic that way with their sites wide banners. Add in the increasing regularity of election threads posted just because rather than about something interesting happening (something clearly sanctioned by the mods months ago) and it's only getting worse with still some weeks to go. I don't see things ever going back to how they were before now that USA politics had become such a central part of metafilter, this is just no longer the welcoming and international-facing website it used to be.

Which actually rather sucks to be honest.
posted by shelleycat at 4:04 AM on October 15, 2016 [2 favorites]


I'm not sure how much traffic-driving we're doing with the banner. We've had it up twice for election stuff, once for each of the two debates so far, for a few hours, mostly so that people can find the mefi chat channel and not fill up other threads with live-tweeting / live-blogging type comments, and so that by pointing to where people are already discussing we don't get a ton of new posts to delete.

We've also kept the election related posts to mostly one at a time until those get too big to load. And not that it's either here or there, but I doubt you'll find anyone happier when the US elections are over than every single Metafilter moderator.
posted by taz (staff) at 4:40 AM on October 15, 2016 [8 favorites]


...the increasing regularity of election threads posted just because rather than about something interesting happening...

"just because" is simply not true. A lot of MeFites are adding a lot of comments to election threads, making them increasingly unworkable and the need for new threads so that lots of MeFites can still interact here.

...now that USA politics had become such a central part of metafilter...

Taking the most recent complete month, September 2016, there were 587 posts on the blue. Of these, 6 (or 1.02%) are Election2016 posts. Or, 98.98% are not.

That, surely, is not central; at the very least in terms of visibility? And one of the things that keeps the mods busy is deleting FPPs whose content is a duplicate of, or can be absorbed into, an existing thread, keeping the number and visibility of Election2016 to a minimum.

There's a whole bunch of topics (it's a long list) that come up regularly or frequently on the blue - totaling far more than 1.02% of FPPs - which I have utterly zero interest in at all. I just ignore them and move on to those of interest, or create posts on things that interest me and may or may not interest a subset of other MeFites e.g. cheese issues.
posted by Wordshore at 4:50 AM on October 15, 2016 [7 favorites]


The problem with moving on to other things of interest is that the other stuff is dieing off, there is increasingly less other discussion to move on to.

And I can see the election threads, it's very clear they're being posted when the old thread gets to a certain size rather than when something truly notable happens. It's just that for a lot of people everything is notable because this is their be all and end all. Which is fine, your prerogative, if that's what metafilter is going to be now (and the mods decision to let this stuff stand says yes, it is) then that's what it is.

I'm still allowed to be sad about the loss of an international community that I used to feel part of. And I see no reason to not mention that occasionally just because some people here are so upset by the idea that outsiders exist and have different opinions than them. Just because some people are somehow threatened by my existence doesn't mean I have to pretend I'm not here.

We've also kept the election related posts to mostly one at a time until those get too big to load.

This was a conscious choice on your part because despite how many times people talk about "needing" a new thread there is no such thing. Election threads are not a requirement for the site to exist or for anyone posting in them, they are here because of an overwhelming want for them. You could make a different choice to reduce posting, you could even have no election threads at all. But since what we have is both officially sanctioned and increasing, I don't see how you can avoid either daily or more than daily threads very soon. There's always going to be an excuse for people to post them with the current low bar after all.
posted by shelleycat at 5:15 AM on October 15, 2016 [3 favorites]


And I can see the election threads, it's very clear they're being posted when the old thread gets to a certain size rather than when something truly notable happens.

It's not as simple as that, though: we have specifically been keeping folks from forking off new election threads when some New Thing happens, in order to avoid the situation of having two or three new US election threads on the front page daily. That sort of thing has been a point of frustration for a lot of people in previous cycles, where the sheer footprint from post after post about the US primaries and general became an outsized share of what you'd see while scrolling through the front page every single day.

So the model of "don't post until something new happens" doesn't work, and analyzing it in those terms is getting it backwards: we're explicitly saying, "don't just make a new post when something new happens, even though you otherwise normally would", and limiting the footprint on the front page intentionally.

With a ton of interest in the election, not least because it has been an insane and terrible election thanks to the ludicrous candidacy of Donald Trump, we can go wide or we can go tall with the threads, but we can't and won't just say "don't be interested in this, folks". We've gone wide in previous cycles, we're going tall this cycle. I can understand preferring one or the other, and I think both approaches have had their upsides and downsides, but "please can you keep there from being so many goddam US election threads" is one of the core things we have heard from folks in past cycles. This is in part an attempt to respect that by taking a different approach.

If the ultimate request is "please let there be less interest in discussing the US presidential election", I can feel you—boy howdy can I feel you—but it's not something over which we have any non-draconian control, and I'm pointing squarely to the insanity of the externalities we as a site and an international community are stuck with this year.
posted by cortex (staff) at 7:14 AM on October 15, 2016 [10 favorites]


I'm a pretty big fan of Pres O and Mitt made a worthy and evil opponent but the last two elections were dull dull dull compared to this round of wackiness, just all makes for good posting rant fodder!
posted by sammyo at 7:53 AM on October 15, 2016


I am loving the "tall" approach to this year's cycle as a person who is very interested in the election (only one thread to follow) and as a person who also wants to read not-election things (far fewer US election posts on the front page),
posted by lalex at 8:07 AM on October 15, 2016 [8 favorites]


I read the election threads but don't post in them. It's my feeling that too many people continue to post comments without any real content or ideas in them. I know this would never fly as a pony, but it'd be nice if there was a comment limit for election threads. 10-15 comments so you could still take part in discussion, but not overwhelm the thread. I say this purely as someone who would never approach this limit, so I fully acknowledge it would be a terrible idea in practice. It would be nice to scroll a bit less and skip fewer comments though.
posted by Roger Dodger at 8:12 AM on October 15, 2016 [1 favorite]


I read through the election threads and all I got was this stupid hyena laugh echoing through my head.
posted by duffell at 9:20 AM on October 15, 2016 [2 favorites]


we're explicitly saying, "don't just make a new post when something new happens, even though you otherwise normally would", and limiting the footprint on the front page intentionally.

So it mostly seems that we have very different ideas of when or how often something happens. When actual substantial issues like the sexual assault allegations against Trump are just more noise amongst the already constant noise then I find that a problem. But I see it all over these days, it's the media's fault and the majority of people around me seem to have bought in to or even enjoy this new paradigm so meh. Tough luck to me I guess.

Certainly more threads is not the answer and apparently we're never going to get less. So I just hope that metfilter somehow recovers from this mess in the future.
posted by shelleycat at 10:36 AM on October 15, 2016


I pop my head into one of the political threads every so often when someone I know has a comment with enough favorites to show up on my sidebar, but every time I do I shake my head at the frantic arguing and vehement opinions and I can't imagine voluntarily diving into such a heaving maelstrom of consuming election stress. Nor can I imagine such discussions make a whit of difference to anyone's opinion or the outcome of the election itself. It all strikes me as not much more than a very noisy circle-jerk. Then again, I feel the same way about discussing sports...

In any case it's obvious that there are those who do choose to so indulge, so live and let live I guess. I'll cope with the lower non-political activity for now, and I doubt that Metafilter itself is teetering on the brink just because of some political threads.
posted by Greg_Ace at 10:46 AM on October 15, 2016 [2 favorites]


So it mostly seems that we have very different ideas of when or how often something happens. When actual substantial issues like the sexual assault allegations against Trump are just more noise amongst the already constant noise then I find that a problem.

Insofar as it's a thing that in any normal month would be its own cleanly seperable-from-the-rest big news, I totally feel you. And one of the most remarkable and depressing things about this election cycle is the capacity of the Trump campaign to be such a non-stop wall of awfulness that even shockingly gross news is just The Thing That Happened Friday, or The Thing That Happened Tuesday, or so on.

Essentially: the sexual assault stuff is huge, and it's something that's been more or less constantly discussed since the news broke; it's one of the main things driving discussion in the last couple of election-related threads. And I'm seriously, genuinely sympathetic if the big election threads just aren't working well for you, I know they're a bear, but in them it's been literally one of the biggest topics of discussion on the site since the tape leaked a week ago and that's much of why they've been so busy even by this cycle's standards this last week.

This election has been a nightmare in a lot of ways. We're muddling through, and muddling is never going to be terribly satisfactory. I really, really look forward to mid-November.
posted by cortex (staff) at 11:06 AM on October 15, 2016 [4 favorites]


I started to ask if it would make sense to set up a separate subsite ("PoliFilter"?) next time, but then I realized the only "benefit" would be to make it a leetle less visible to the non-participants, without it really solving any other problems or being any less work for the moderators. So never mind.
posted by Greg_Ace at 11:36 AM on October 15, 2016


It could be that, so people can adequately comment on their tech, the community needs more than one election post for that day alone (which, with the 1-FPP-in-24-hours limit on FPP creators may require a bit of co-ordination beforehand).

Not sure what the mods would say about a two-post day, but there's already coordination in the previous Election Threads MetaTalk. Generally someone will start building the next post well before we get near the phone-breaking point.
posted by ChurchHatesTucker at 2:32 PM on October 15, 2016


As a decided non-fan of the election threads, I did some work in the infodumpster to try to actually answer komara's question. Specifically, I created a chart of the number of comments per month for all months since the beginning of the site. I present the resulting data and chart in this google spreadsheet. Some observations:

* As is well known since eligito's post last year, activity on all fronts (posts, comments) on all subsites has generally been declining since Metafilter's peak in 2011.
* This past year has been a partial exception to this pattern. Comments per month are marginally up this year from this time last year. For example, there were 41175 comments this September compared to 38654 last September, or an increase of about 6%. Overall number of comments through September this year is up 12% compared to Jan-Sept. 2015.
* We are still, however, substantially down from Metafilter's peak. The number of comments this September, for example, is still down from two years ago, let alone Sept. 2011.

I also ran the calculations to see comments per month excluding threads of greater than a thousand comments (i.e. election threads). I present that data and chart in this google spreadsheet. Some observations here:

* The two charts largely coincide until this year. Until this year both charts show a peak around 2011 and a steady decline since then.
* They diverge in this year; while the total number of comments per month has marginally increased this year and returned to roughly 2014 levels, the number of comments excluding big posts is continuing to decline and even accelerating its decline.
* For example, putting aside big threads, there were 23113 comments in Sept. 2016. We have not seen that few comments in non-big threads since the modern period of $5 signups began in Nov. 2004 (!).

I think it's fair to say, then, that the election threads are 'sucking the air' out of the room. Much more worryingly, I think, they are papering over Metafilter's continued decline. In my opinion, the heart of Metafilter lies in its diversity of different posts on different topics, which once attracted a fair bit of activity but much less so now. Once the election actually happens next month, what will happen? Will people go to other, non-election threads? Or will the number of comments on other topics remain about the same and we will be staring in the face of declining activity? An observation: at current rates of decline, we will reach 0 comments in non-big threads in about Sept. 2028.
posted by crazy with stars at 3:48 PM on October 15, 2016 [14 favorites]


That's really interesting, CWS. An alternative analysis about the decline in comments in non big threads, though, could be the continued mod push to keep threads on topic and to prune long back and forth arguments between a few commenters leading to people in general being more likely to comment only when they have something substantive to say. I'd be okay with fewer comments but higher quality ones.

Alternatively I know that my own commenting has declined over the past few years because I now mostly read on my phone and typing is a little bit annoying. So I often don't bother unless I think my comment is really adding something.

Big threads are ones I'm more likely to bookmark for later and then read at my actual computer, which means there's less friction to commenting then.
posted by lollusc at 4:20 PM on October 15, 2016


thanks for running those numbers, CWS. I agree with you about the "sucking the air out" aspect. When I finally finish catching up on the election post.... I don't have time for anything else.
posted by rebent at 7:43 PM on October 15, 2016 [1 favorite]


I often use the "popular comments" page to check in on interesting discussions that I might have otherwise missed and/or not had time for, and one extremely marked pattern in recent months is that the vast majority of heavily favorite comments are now on the election threads (like right now, there are literally zero comments on the Popular Favorites page that are not election thread comments). Another tool I use is keeping an eye on comments from my contacts that get past the 12 favorite threshold to show up on my sidebar feed--here again, it's almost all election thread comments these days. I don't now if that's the best way to judge readership and not just participation, but it seems to me a sign that not only active participation but more passive forms of participation (reading threads and favoriting comments) has shifted dramatically toward the election threads at the expense of the rest of The Blue.
posted by drlith at 12:21 PM on October 16, 2016 [1 favorite]


As is well known since eligito's post last year, activity on all fronts (posts, comments) on all subsites has generally been declining since Metafilter's peak in 2011.

For some value of "well known". Here's a nifty chart which may, or may not, tell us something about what's changed since then, but either way I spent an hour making it for some stupid reason.
posted by sfenders at 12:44 PM on October 16, 2016 [6 favorites]


sfenders: "As is well known since eligito's post last year, activity on all fronts (posts, comments) on all subsites has generally been declining since Metafilter's peak in 2011.

For some value of "well known".
"

Ah, and I even misspelled elgilito's name. This is the post I had in mind for the evidence of Metafilter's gradual decline.

Your chart does show a striking rise is big threads; most, if not all, of which are election threads, I imagine. Maybe Brexit, Scottish independence referendum, and other related political threads as well.
posted by crazy with stars at 1:05 PM on October 16, 2016


These charts feel kind of alarmist. The sharp decline in activity looks scary, but it's only because it excludes the single largest source of comments. I'm sure my HDD would look a lot roomier if I ignored every folder with more than 1GB in files.

I can see the argument that the 5000-comment election posts are dominating the conversation. But you can chalk that up to the crazy high-stakes insanity of this political season, along with the concentration of the resultant discussion into a series of monolithic weekly megathreads, whereas 2008 and 2012 saw it diffused across several dozen more narrowly-focused posts over the same time period.

What I would like to see is a breakdown of the participation in the megathreads. Are they drawing in a microcosm of the site's userbase, or are they dominated by a handful of highly active users? And how does that compare to the average thread? It would be more worrisome if the mammoth threads sucking up all the air were really the product of a small core of commenters.
posted by Rhaomi at 6:37 PM on October 16, 2016 [1 favorite]


It would be more worrisome if the mammoth threads sucking up all the air were really the product of a small core of commenters.

To at least some degree, they are. That doesn't mean that's the case for readers, several who have recently joined to express their commitment to the site particularly for this election.
posted by ChurchHatesTucker at 7:24 PM on October 16, 2016 [3 favorites]


to provide an individual user story, I have been participating in a desultory manner in the current election threads; previously as far as I recall I mostly avoided them.

loading the threads on my phone or iPad has been problematic, as is commenting in them after 1500 comments or so. The time commitment to skim up 3-400 comments from the current comment-box also has reduced the amount of time I spend on the site in other threads. So while to an extent my time spent on the site engaged in user activities with an interactive component (favorites, writing a comment, checking for comment overlap) has increased, my participation outside of the election threads is very definitely down, in part because the election threads are on their own more time consuming to interact with due to the density of user input and consequent new content to review.

In short, my user experience supports the hypothesis that the 2016 election threads have accelerated a decline in user participation while simultaneously increasing my personal user participation and time commitment to the site.
posted by mwhybark at 7:27 PM on October 16, 2016


Rhaomi: "What I would like to see is a breakdown of the participation in the megathreads. Are they drawing in a microcosm of the site's userbase, or are they dominated by a handful of highly active users? And how does that compare to the average thread? It would be more worrisome if the mammoth threads sucking up all the air were really the product of a small core of commenters."

Here's a chart of the percentage of comments written by the top 10, 100, and 1000 commentators in 2015 and 2016. As you can see, in Jan. - Sept. 2015 across the Blue the top 100 commentators accounted for about 25% of the total comments, and the top 1000 commentators about 75%. So we might take that as a baseline for a non-election year.

These numbers hold true, more or less, for the posts under 1000 comments this year. The posts > 1000 comments are a different story though. There the top 100 commentators wrote more than 50% of the total comments, and the top 10 commentators wrote more than 13% (!).

I don't know if you would say that means the election threads are dominated by a handful of highly active users -- but certainly it's a smaller number of people participating more intensively than in the rest of the site.

--

More broadly speaking:

* The number of people posting at least one comment each month has continued to decline this year from its peak in Oct. 2011. ~3400 people posted at least one comment last month, a number last seen in Sept. 2008. I grant you that the rate of decline here has substantially slowed over the last few years compared to 2012, but it is still worrisome that even with the increase in comments this past year there has not been an increase in number of people commenting -- it's a smaller number of people writing more, not a greater number of people participating.

* Multiple people have suggested that participation may be declining, but readership may be flat or increasing. The infodumpster doesn't contain any information on readership, so it's hard to tell from there, but the Alexa data seems to suggest that traffic has generally declining over the last year, with a particularly sharp drop from June to the present.

I'm not saying that Metafilter is going to die tomorrow. I am saying that I don't think the election threads are good for the long-term health of the site.
posted by crazy with stars at 10:12 PM on October 16, 2016 [3 favorites]


crazy with stars: "Here's a chart of the percentage of comments written by the top 10, 100, and 1000 commentators in 2015 and 2016. As you can see, in Jan. - Sept. 2015 across the Blue the top 100 commentators accounted for about 25% of the total comments, and the top 1000 commentators about 75%. So we might take that as a baseline for a non-election year.

These numbers hold true, more or less, for the posts under 1000 comments this year. The posts > 1000 comments are a different story though. There the top 100 commentators wrote more than 50% of the total comments, and the top 10 commentators wrote more than 13% (!).
"

For completeness' sake, what does the breakdown for participation in 1000+ comment threads look like for that Jan-Sep 2015 baseline you cited? I.e., is the greater participation of fewer people in megaposts unique to this year, or has it always been thus? I'd guess the latter; we are like most social sites subject to the 90/9/1 rule, and megathreads strike me as more enticing to power users than more casual readers.
posted by Rhaomi at 12:43 AM on October 17, 2016


the vast majority of heavily favorite comments are now on the election threads (like right now, there are literally zero comments on the Popular Favorites page that are not election thread comments).

I would love to be able to filter the comments column on the Popular page to remove specific tags or posts. I have taken to scanning the comments for more current days because the rest are from 3-4 days ago.
posted by soelo at 8:40 AM on October 17, 2016 [1 favorite]


I must say, as a foreigner who isn't following your election, it seems to be going really well on MetaFilter: continued flow of interesting articles and discussions, some heated, some less so. Well done Mod Team!

If the election is a disaster on MetaFilter, it's brilliantly contained. Thanks for all the hard work.

Like a swan, right? Gliding along on the front page, frantic mod paddling in the murky election-thread depths.
posted by alasdair at 5:11 AM on October 18, 2016 [8 favorites]


There the top 100 commentators wrote more than 50% of the total comments, and the top 10 commentators wrote more than 13% (!)

I read some of the threads but don't participate. Sometimes I go into election threads and Ctrl+F user names. I thought was my own confirmation bias. But yeah, even with the frequent new posts, the resident goldfish are growing to the size of the tank. (I know that is a myth.)
posted by kimberussell at 8:05 AM on October 18, 2016


Rhaomi: "For completeness' sake, what does the breakdown for participation in 1000+ comment threads look like for that Jan-Sep 2015 baseline you cited? I.e., is the greater participation of fewer people in megaposts unique to this year, or has it always been thus? "

I've added breakdowns for 2015, 2011, and 2008 to give a sense of what it's looked like over the (relatively) recent history of the site. You're absolutely right that fewer people have always participated in megaposts, at approximately similar ratios. I don't think that really changes my point though -- even though megaposts have always attracted fewer people participating more actively, that doesn't change the fact that this year's megaposts are still dominated by a relatively small number of people compared to the rest of the site.

And more broadly the obvious difference this year is that so much of the site's activity takes place in such threads: 132k comments on megaposts so far this year (33% of all comments), compared to only 15k comments last year (3% of all comments). If megaposts are inherently going to be dominated by a small number of people, I don't think it's in the site's long-term interests to encourage the growth of such posts, especially if it comes at the cost of decreasing activity across the rest of the site.

I've also created a chart of the number of people who wrote at least one comment in megaposts per month over the last year. On average only about 20% of the site's active userbase posted at least one comment in a megapost over the last year. Participation steadily grew from April to July but has declined over the last few months.
posted by crazy with stars at 11:01 AM on October 18, 2016 [2 favorites]


I don’t think anyone is actively encouraging these mega-posts; rather, both the mods and the community seem to be doing what they can to both keep US election posts from taking over the front page and to keep the discussions substantive and on-topic. Although I do wish there were a wider range of posters, and that the heaviest posters would take it a little easier, the fast-moving nature of the mega-posts can make it hard to get a comment in while it is still relevant unless you are somehow able to keep caught up in a thread wherein comments are being made faster than most of us can read them.

I think, however, that people who are not interested in these threads may not understand that to many of us, they are more than a distraction from other posts on the front page. They are a lifeline to logical discourse that is very hard to find in this election year, when the candidates themselves are using Twitter as a primary communication conduit—a medium that is basically just a huge comment section with no moderation, littered with all of the hate speech and ugliness one has come to expect in such an environment. As noted above, we have seen quite a few readers creating accounts just for the opportunity to offer thanks for maintaining a discussion that allows them to keep up with the important events occurring every day in this shitshow of an election without constantly feeling physically ill.

While I understand people’s concerns over the existence of these mega-threads, I think it would be pretty hurtful to those of us who are good-faith users of the site to make any move toward doing away with them. It is our love of this community—a community we have worked for years to create and foster—that makes these threads so vital to us in trying times such as these. The fact that they are drawing users away from other posts is unfortunate, but temporary. A lot of us who are reading these threads are frightened, disgusted, and often at odds with the people around us to such an extreme that families and friendships are being torn apart, and even mild statements of opinion in most other venues pretty much always result in spittle-flecked attacks from actual fascists.

I know that Metafilter never claims to be a “safe space,” but for many, it’s the safest space they have right now, and that means a lot in a situation wherein sexual assault is a recurring talking point and threats of political retaliation are being made on live television.

My point is just that this situation is not normal. These threads have turned out to be really, really important to many of us, and I think that in the long run the existence of Election 2016 mega-threads will do more good for the community as a whole than harm.
posted by obloquy at 12:55 PM on October 18, 2016 [2 favorites]


I'd be interested in seeing the results of a poll indicating how many people have changed their mind/opinion/voting plan as a result of those mega-discussions.
posted by Greg_Ace at 2:20 PM on October 18, 2016


I changed my mind on Sanders during the primary, in part because of the Metafilter election threads. I suppose it was Sanders himself and his campaign that changed my mind, but since Metafilter was my main source of news about it, I guess I can blame you people, too.
posted by ryanrs at 5:45 AM on October 19, 2016 [1 favorite]


To expand on that a bit, early in the primaries, I joked/not_joking that I couldn't decide if I wanted to vote for Sanders or Trump. Both of them seemed likely to increase the chances of a progressive administration.

But by the California primary, Trump had evolved from a ridiculous joke candidate to a scary joke candidate, and Sanders was looking more and more like a bitter spoiler. Noting the change, I hung onto my absentee ballot until the day of the election, then voted for Clinton.

As for the general election, in the past I might have voted Green, especially in such a lopsided race. But I don't like the direction of Stein's campaign, so that's out. And after setting aside that lol protest vote, that leaves Clinton, which is cool since she's the person I actually want to be president.


Were it not for Metafilter, it's more likely I would have voted for Sanders or Trump in the primary, and Stein in the general. None of which would have mattered, of course, since I live in California.

democracy, heh
posted by ryanrs at 6:17 AM on October 19, 2016


There the top 100 commentators wrote more than 50% of the total comments, and the top 10 commentators wrote more than 13%

Hamilton wrote THE OTHER FIFTY-ONE!
posted by mbrubeck at 12:47 PM on October 19, 2016 [4 favorites]


I wanted to comment the other night in an election thread---which I've been keeping up with on my phone---and it crashed mobile safari: first I got a metafilter "something went wrong" error message, and then I got Safari's "sorry, fatal error" (or something). I can read ok, but commenting 5 days into an election thread was not happening. (But I don't comment much anyway, so no loss I guess?)
posted by leahwrenn at 7:42 PM on October 19, 2016


I can't understand how anyone can keep up with these election posts. I use a GM script to filter by favorites, which I have never done before and hate doing, but I still can't keep up if I want to have any chance of fulfilling my real-world responsibilities.

I am very interested in seeing a stat-rich post-mortem of Metafilter election activity.
posted by double block and bleed at 10:47 AM on October 22, 2016


cortex: We're muddling through, and muddling is never going to be terribly satisfactory. I really, really look forward to mid-November.

On the one hand, these election threads have been literal sanity preservers. I agree that they are sucking a lot of air out of the room - I've noticed astronomy and science news threads scroll by and not had the energy to go back and participate in them. But the stakes are really high, especially for those of us in the US, and especially for those of us who aren't native-born / white / men. If I wasn't getting my politics fix in the election threads here, I'd have to resort to some of my older favorite sites, or - yikes - talk to people, but they're usually several cycles behind the MeFi discussion.

On the other hand, if I had to guess, I'd say that these threads are good for MeFi because at least some new commenters are likely to stick around. Smart[*], informed[**] commentary with high signal-to-noise ratio is definitely the best of the web.

And on the third hand, I feel like Cortex is tempting fate with his statement, because after the election, we will have the post-election litigation thread, the Inauguration thread, the first hundred days agenda thread, the impeachment thread, and then it's on to the mid-terms...

[*] Sometimes
[**] Sometimes
posted by RedOrGreen at 1:42 PM on October 23, 2016


« Older New category in AskMe?   |   Can we stop attacking... Newer »

You are not logged in, either login or create an account to post comments