Flagophile October 24, 2007 8:19 AM Subscribe
Does it matter what I flag a post/comment? Does "offensive" have a higher priority than "other?"
Just seems like other than "double" they are all about the same level of priority. I'm never sure what to flag an axegrindy post.
Oh, and for my pony request: Can we add "axegrindy" to the list of flags?
Just seems like other than "double" they are all about the same level of priority. I'm never sure what to flag an axegrindy post.
Oh, and for my pony request: Can we add "axegrindy" to the list of flags?
"A lot of flags, regardless of what they say" has a higher priority than "one flag from one dude, no matter what it said." That's about the size of it.
posted by Wolfdog at 8:25 AM on October 24, 2007
posted by Wolfdog at 8:25 AM on October 24, 2007
Can we add "axegrindy" to the list of flags?
No. The last thing the flagging system needs is any of that type of vague-but-satisfying flag. People has suggested "troll" as a flag before as well, and variation on "shit" or "stupid", and none of them would help anything because it'd be far too easy to flag something without having to identify a core, at least semi-objective issue with the comment/post.
If a comment breaks the guidelines, or is arguably offensive, or is noise, flag it as such; if it's not quite any of those but you don't like it anyway, take a breath and move on, essentially.
posted by cortex (staff) at 8:25 AM on October 24, 2007
No. The last thing the flagging system needs is any of that type of vague-but-satisfying flag. People has suggested "troll" as a flag before as well, and variation on "shit" or "stupid", and none of them would help anything because it'd be far too easy to flag something without having to identify a core, at least semi-objective issue with the comment/post.
If a comment breaks the guidelines, or is arguably offensive, or is noise, flag it as such; if it's not quite any of those but you don't like it anyway, take a breath and move on, essentially.
posted by cortex (staff) at 8:25 AM on October 24, 2007
And the different flags have some de facto priority in what we look at first, but not so much what we do or don't consider; so "it breaks the guidelines" flags, esp. in AskMe, kind of get first dibs, along with straightforward neutral problems like "double" or "display error". But we see every flag, and a post with a dozen flags on it, as Wolfdog says, will get attention sooner than a post with one.
posted by cortex (staff) at 8:28 AM on October 24, 2007
posted by cortex (staff) at 8:28 AM on October 24, 2007
I flagged this post as "flaggy". What? You don't have "flaggy" in your list of options? Hold on, I'll whip up a GreaseMonkeytm script for ya.
posted by blue_beetle at 8:31 AM on October 24, 2007
posted by blue_beetle at 8:31 AM on October 24, 2007
All the admins have their own little math in terms of what gets paid attention to first, later, etc. I've spelled mine out a few times in here but basically ones that imply "something is on fire!" or "something broken that can be fixed quickly!" usually get my attention first but we see all of them.
The real thing is the combination of flags. If one post gets 10 "offensive" flags [i.e. everyone sort of agrees that something is offensive, which almost never happens] that sort of says something.
posted by jessamyn (staff) at 8:35 AM on October 24, 2007
The real thing is the combination of flags. If one post gets 10 "offensive" flags [i.e. everyone sort of agrees that something is offensive, which almost never happens] that sort of says something.
posted by jessamyn (staff) at 8:35 AM on October 24, 2007
It would probably help users, when flagging a something, there was a sound of a heavy piece of wood hitting flesh, followed by a cry of pain and little pop up message saying "You showed that mofo!", with a thumbs up sign in neon green. I'm only half joking.
posted by Brandon Blatcher at 8:40 AM on October 24, 2007 [5 favorites]
posted by Brandon Blatcher at 8:40 AM on October 24, 2007 [5 favorites]
Brandon, make that thumbs up sign blink, and I'm sold!
posted by NationalKato at 8:43 AM on October 24, 2007
posted by NationalKato at 8:43 AM on October 24, 2007
No. The last thing the flagging system needs is any of that type of vague-but-satisfying flag. People has suggested "troll" as a flag before as well, and variation on "shit" or "stupid", and none of them would help anything because it'd be far too easy to flag something without having to identify a core, at least semi-objective issue with the comment/post.
Woah, wait, what? Right now, people can already flag anything without having to identify a core issue, nothing can stop them from flagging it as "offensive" or whatever, so I'm not sure 1 or 2 more specific flags would suddenly allow that. On the other thing, posts have gotten deleted for being axegrindy or trollish before, so it is possible for people (namely the mods) to recognize those types of posts. What it sounds like you're saying here, is that there are categories for deletion that can be recognized by mods, but that are not recognizable by the users in general. I'm not trying to accuse you of something here, but that is what your post here reads like to me.
posted by Snyder at 8:45 AM on October 24, 2007
Woah, wait, what? Right now, people can already flag anything without having to identify a core issue, nothing can stop them from flagging it as "offensive" or whatever, so I'm not sure 1 or 2 more specific flags would suddenly allow that. On the other thing, posts have gotten deleted for being axegrindy or trollish before, so it is possible for people (namely the mods) to recognize those types of posts. What it sounds like you're saying here, is that there are categories for deletion that can be recognized by mods, but that are not recognizable by the users in general. I'm not trying to accuse you of something here, but that is what your post here reads like to me.
posted by Snyder at 8:45 AM on October 24, 2007
Please add a "LOL" flag so I can guide the admins to that which is LOL.
posted by mullacc at 8:56 AM on October 24, 2007 [5 favorites]
posted by mullacc at 8:56 AM on October 24, 2007 [5 favorites]
I think the point is, Snyder, that certain categories for deletion are by their nature liable to generate too many false positives if the userbase at large is allowed to cry out about them.
posted by Wolfdog at 8:56 AM on October 24, 2007
posted by Wolfdog at 8:56 AM on October 24, 2007
"People has suggested 'troll' as a flag before as well, and variation on 'shit' or 'stupid'".
Seldom has people asked "is our cortex learning"?
A "shit" flag would be awesome, though. No variation needed. Just "shit".
posted by mr_crash_davis at 9:03 AM on October 24, 2007 [1 favorite]
Seldom has people asked "is our cortex learning"?
A "shit" flag would be awesome, though. No variation needed. Just "shit".
posted by mr_crash_davis at 9:03 AM on October 24, 2007 [1 favorite]
If one post gets 10 "offensive" flags [i.e. everyone sort of agrees that something is offensive, which almost never happens] that sort of says something.
I can't be the only one that really really wanted to flag Jessamyn's post offensive, right?
posted by inigo2 at 9:11 AM on October 24, 2007 [2 favorites]
I can't be the only one that really really wanted to flag Jessamyn's post offensive, right?
posted by inigo2 at 9:11 AM on October 24, 2007 [2 favorites]
Woah, wait, what? Right now, people can already flag anything without having to identify a core issue, nothing can stop them from flagging it as "offensive" or whatever, so I'm not sure 1 or 2 more specific flags would suddenly allow that. On the other thing, posts have gotten deleted for being axegrindy or trollish before, so it is possible for people (namely the mods) to recognize those types of posts. What it sounds like you're saying here, is that there are categories for deletion that can be recognized by mods, but that are not recognizable by the users in general. I'm not trying to accuse you of something here, but that is what your post here reads like to me.
The thing is that "offensive" is value-neutral in a way that e.g. "axegrindy" or "troll" is not. Of course someone can be offended by a comment because they think it's axegrinding or trollery, and they can flag it as offensive if it crosses their personal threshold of being worth dropping a flag for.
The problem with "axegrindy" or "troll" as flags is that they are loaded terms that significantly lower the barrier to entry. If Alice dislikes Bob's stance on an issue and sees Bob make yet another comment on said issue, that "axegrindy" flag stands as a sort of tacit approval of Alice's annoyance, and, huzzah: we get treated to a bunch of flags that indicate personal dislikes without any presumed need for there to be anything else actually wrong with the comment beyond the fact that there's a flag to match Alice's sentiment.
Likewise "troll", handy for anything that Charlie thinks is disingenous or baiting or whatever else in Darren's comments. The value-judgment built into the flag itself defeats some of the purpose of having the neutral, bring-your-own-rationale flags we currently use.
It has nothing to do with the admins and the users having different abilities to distinguish categories, and everything to do with trying to avoid creating easy-target pre-approved peeve flags.
I personally think of flagging as a kind of trust exchange. I trust that someone flagging something would be able to write up a paragraph describing why, exactly, they're flagging this thing as noise, or offensive, or guidelines breaking, or even fantastic. I trust that there is a fair-minded rationale motivating the flag, whether or not I would agree with it personally. On the other end, I want users to trust that when they flag something, I'll assume such good intent and do my best to try and understand where they might be coming from when I check out the flagged item.
But writing (and reading) a paragraph for every flag that hits the system would be a tremendous amount of effort, and get quickly redundant besides, and so we don't do that; we just go with a trust exchange and make do as best as possible, and every once in a while fire email in either direction to try and understand a puzzler better.
So why do things get deleted as trollish or axegrindy? Because in the hotseat we have the strange priviledge of making that kind of ultimate administrative judgment call—and of taking as necessary the heat for it, given that even when we decide to nix something like that it can be a contentious issue. And because when we do delete something, we're pressed by circumstance and tradition and mostly just an aim of transparency to offer some brief summary of what the issue is, which leaves us using descriptions that may capture the situation (that a post seems very axegrindy in context, or that something has been kind of collectively concluded to be trolling, or that a post was just awfully weak) that wouldn't actually make for very good flag options for the reasons I've blathered about above.
posted by cortex (staff) at 9:21 AM on October 24, 2007
The thing is that "offensive" is value-neutral in a way that e.g. "axegrindy" or "troll" is not. Of course someone can be offended by a comment because they think it's axegrinding or trollery, and they can flag it as offensive if it crosses their personal threshold of being worth dropping a flag for.
The problem with "axegrindy" or "troll" as flags is that they are loaded terms that significantly lower the barrier to entry. If Alice dislikes Bob's stance on an issue and sees Bob make yet another comment on said issue, that "axegrindy" flag stands as a sort of tacit approval of Alice's annoyance, and, huzzah: we get treated to a bunch of flags that indicate personal dislikes without any presumed need for there to be anything else actually wrong with the comment beyond the fact that there's a flag to match Alice's sentiment.
Likewise "troll", handy for anything that Charlie thinks is disingenous or baiting or whatever else in Darren's comments. The value-judgment built into the flag itself defeats some of the purpose of having the neutral, bring-your-own-rationale flags we currently use.
It has nothing to do with the admins and the users having different abilities to distinguish categories, and everything to do with trying to avoid creating easy-target pre-approved peeve flags.
I personally think of flagging as a kind of trust exchange. I trust that someone flagging something would be able to write up a paragraph describing why, exactly, they're flagging this thing as noise, or offensive, or guidelines breaking, or even fantastic. I trust that there is a fair-minded rationale motivating the flag, whether or not I would agree with it personally. On the other end, I want users to trust that when they flag something, I'll assume such good intent and do my best to try and understand where they might be coming from when I check out the flagged item.
But writing (and reading) a paragraph for every flag that hits the system would be a tremendous amount of effort, and get quickly redundant besides, and so we don't do that; we just go with a trust exchange and make do as best as possible, and every once in a while fire email in either direction to try and understand a puzzler better.
So why do things get deleted as trollish or axegrindy? Because in the hotseat we have the strange priviledge of making that kind of ultimate administrative judgment call—and of taking as necessary the heat for it, given that even when we decide to nix something like that it can be a contentious issue. And because when we do delete something, we're pressed by circumstance and tradition and mostly just an aim of transparency to offer some brief summary of what the issue is, which leaves us using descriptions that may capture the situation (that a post seems very axegrindy in context, or that something has been kind of collectively concluded to be trolling, or that a post was just awfully weak) that wouldn't actually make for very good flag options for the reasons I've blathered about above.
posted by cortex (staff) at 9:21 AM on October 24, 2007
Flags work like this: they're democratic. Democracy is driven by marketing. So: if you see something you don't like, post something keening about it in MeTa, which will drive traffic to it, and therefore increase its flag count. When the flag count goes up, the admins delete.
posted by koeselitz at 9:31 AM on October 24, 2007 [1 favorite]
posted by koeselitz at 9:31 AM on October 24, 2007 [1 favorite]
Does it matter what I flag a post/comment? Does "offensive" have a higher priority than "other?"
Well... not when you flag it.
But in all seriousness, this thread has taught me one very important thing: always flag things you don't like as "offensive." It doesn't matter if it is or isn't, what matters is getting the mods to pay attention to me.
posted by shmegegge at 9:31 AM on October 24, 2007
Well... not when you flag it.
But in all seriousness, this thread has taught me one very important thing: always flag things you don't like as "offensive." It doesn't matter if it is or isn't, what matters is getting the mods to pay attention to me.
posted by shmegegge at 9:31 AM on October 24, 2007
Also, cortex's comment above:
If Alice dislikes Bob's stance on an issue and sees Bob make yet another comment on said issue, that "axegrindy" flag stands as a sort of tacit approval of Alice's annoyance, and, huzzah: we get treated to a bunch of flags that indicate personal dislikes without any presumed need for there to be anything else actually wrong with the comment beyond the fact that there's a flag to match Alice's sentiment.
was the first time I was ever really strongly tempted to get a sockpuppet account. I would have named it Alice just so I could come in to this thread this one time and say "You know what, cortex? Fuck you." and never post again.
posted by shmegegge at 9:34 AM on October 24, 2007 [3 favorites]
If Alice dislikes Bob's stance on an issue and sees Bob make yet another comment on said issue, that "axegrindy" flag stands as a sort of tacit approval of Alice's annoyance, and, huzzah: we get treated to a bunch of flags that indicate personal dislikes without any presumed need for there to be anything else actually wrong with the comment beyond the fact that there's a flag to match Alice's sentiment.
was the first time I was ever really strongly tempted to get a sockpuppet account. I would have named it Alice just so I could come in to this thread this one time and say "You know what, cortex? Fuck you." and never post again.
posted by shmegegge at 9:34 AM on October 24, 2007 [3 favorites]
But in all seriousness, this thread has taught me one very important thing: always flag things you don't like as "offensive." It doesn't matter if it is or isn't, what matters is getting the mods to pay attention to me.
Sounds more to me that while they'd notice your name more, but they'd start to take you take you less seriously. Might as well change your nick to "Crying Wolf" at that point.
posted by inigo2 at 9:45 AM on October 24, 2007 [1 favorite]
Sounds more to me that while they'd notice your name more, but they'd start to take you take you less seriously. Might as well change your nick to "Crying Wolf" at that point.
posted by inigo2 at 9:45 AM on October 24, 2007 [1 favorite]
always flag things you don't like as "offensive."
My example was meant to say "when a lot of people agree that a post is bad in one certain way, we'll pay close attention to it" and offensive was the example I used. Actually, for AskMe "breaks the guidelines" will usually get quicker attention but again once we realize any one person is abusing the flagging system [see what cortex said re: trust, above] we have to correct for that on our end.
If you see something offensive and you flag it as HTML error, we may not even be looking for content, for example, just trying to figure out what needs fixing.
posted by jessamyn (staff) at 9:49 AM on October 24, 2007
My example was meant to say "when a lot of people agree that a post is bad in one certain way, we'll pay close attention to it" and offensive was the example I used. Actually, for AskMe "breaks the guidelines" will usually get quicker attention but again once we realize any one person is abusing the flagging system [see what cortex said re: trust, above] we have to correct for that on our end.
If you see something offensive and you flag it as HTML error, we may not even be looking for content, for example, just trying to figure out what needs fixing.
posted by jessamyn (staff) at 9:49 AM on October 24, 2007
I think blue_beetle was on the right track. There should be flags for *everything*. They wouldn't actually show up anywhere, but the flagger would get that moment of satisfaction. Nothing could possibly go wrong...
posted by lorimt at 10:05 AM on October 24, 2007
posted by lorimt at 10:05 AM on October 24, 2007
I think "offensive" is a bit too much of a catch-all for "I don't like this."
Maybe it should just be "I don't like this." It's honest.
posted by dw at 10:43 AM on October 24, 2007
Maybe it should just be "I don't like this." It's honest.
posted by dw at 10:43 AM on October 24, 2007
It would probably help users, when flagging a something, there was a sound of a heavy piece of wood hitting flesh, followed by a cry of pain and little pop up message saying "You showed that mofo!", with a thumbs up sign in neon green. I'm only half joking.
Now that's a GreaseMonkey script I can get behind.
posted by dw at 10:45 AM on October 24, 2007
Now that's a GreaseMonkey script I can get behind.
posted by dw at 10:45 AM on October 24, 2007
When I flag something, I find it helps to say aloud "bullshit," or "retarded," or "wrong," thus imbuing my flag with a super-textual impact beyond what it would normally have had.
But things I flag never get deleted anyway, so it's really just to make myself feel better.
posted by klangklangston at 10:52 AM on October 24, 2007
But things I flag never get deleted anyway, so it's really just to make myself feel better.
posted by klangklangston at 10:52 AM on October 24, 2007
When you flag a post a little box should come up that says "We're sorry that you find this post $reason_for_flag. Your feelings on this matter have been noted in the record. Thank you and good day."
and then an animated gif of a foppish gentleman should lift his top hat towards you and then walk away.
posted by drezdn at 10:52 AM on October 24, 2007 [1 favorite]
and then an animated gif of a foppish gentleman should lift his top hat towards you and then walk away.
posted by drezdn at 10:52 AM on October 24, 2007 [1 favorite]
lorimt: "I think blue_beetle was on the right track. There should be flags for *everything*. They wouldn't actually show up anywhere, but the flagger would get that moment of satisfaction. Nothing could possibly go wrong..."
On that point, which reminds me somewhat of the idea of creating alternate-reality metafilters where banned people would think they were commenting but where no one else would see their comments: I wonder if admins might be able to put together some algorithms to maximize the validity of flags. For example, if you cross-referenced the metrics for how many times a particular user has used a particular flag, or flagged in general, and then revalued the flag based on that info. So that, for example, a post would get more flag-value if it was flagged as "offensive" by people who'd never flagged anything as "offensive" than if it was flagged by the ten people who flag a post as "offensive" every day.
I don't know how much that would help. I don't even know if the flagging system needs helping, or if maybe the admins are already doing this. But I do wonder, along the lines of cortex' example of Alice and Bob, whether there are people who habitually flag based on personal opinion rather than an unbiased appraisal of what's being flagged. If that's a constant problem, then putting the flags on a curve based on user-specific frequency seems like it would help. It doesn't even seem that mathematically trivial, although I don't know a damned thing about coding, and I can foresee that user-specific stats on flags are probably either hard or impossible to obtain.
posted by koeselitz at 11:06 AM on October 24, 2007
On that point, which reminds me somewhat of the idea of creating alternate-reality metafilters where banned people would think they were commenting but where no one else would see their comments: I wonder if admins might be able to put together some algorithms to maximize the validity of flags. For example, if you cross-referenced the metrics for how many times a particular user has used a particular flag, or flagged in general, and then revalued the flag based on that info. So that, for example, a post would get more flag-value if it was flagged as "offensive" by people who'd never flagged anything as "offensive" than if it was flagged by the ten people who flag a post as "offensive" every day.
I don't know how much that would help. I don't even know if the flagging system needs helping, or if maybe the admins are already doing this. But I do wonder, along the lines of cortex' example of Alice and Bob, whether there are people who habitually flag based on personal opinion rather than an unbiased appraisal of what's being flagged. If that's a constant problem, then putting the flags on a curve based on user-specific frequency seems like it would help. It doesn't even seem that mathematically trivial, although I don't know a damned thing about coding, and I can foresee that user-specific stats on flags are probably either hard or impossible to obtain.
posted by koeselitz at 11:06 AM on October 24, 2007
I'd kinda like a Pepsi Blue flag. Pepsi Blue is easy enough to spot when it's the subject of an FPP, but less so when it's someone who strolls into an ongoing thread and stealth spams (like the guy who dropped in on a movie thread I was involved in a few weeks ago and, under the pretense of contributing, worked in a plug for his favorite DVD subscription service -- which was, he assured us, vastly superior to its competitors! -- and then helpfully provided a sign-up link...this was his second comment, by the way). I guess this fits under the general category of "breaks the guidelines," but I think it's sufficiently pernicious to warrant an extreme prejudice type heading of its own.
posted by kittens for breakfast at 11:13 AM on October 24, 2007
posted by kittens for breakfast at 11:13 AM on October 24, 2007
koeselitz -- your system shouldn't involve frequency, it should involve accuracy. If your flagged posts consistantly do require admin attention, they're more often paid attention to.
posted by garlic at 11:20 AM on October 24, 2007
posted by garlic at 11:20 AM on October 24, 2007
For example, if you cross-referenced the metrics for how many times a particular user has used a particular flag, or flagged in general, and then revalued the flag based on that info.
I find the question in abstract interesting, and I've actually done a little bit of flag-behavior analysis (very top-level at this point) that I might write up if I ever dig in far enough to find anything really interesting.
But I doubt we'd ever make that sort of thing part of the actual flagging system. We occasionally notice slightly weird flagging behavior, but when it comes right down to it someone is either abusing/misusing the flagging system or they aren't. Any one user lacks the power to disrupt the system, and the only likely result of some obsessive flagging campaign on their part is us learning to shrug a bit at that particular user's flags re: their pet cause. That binary Hrm?-or-not granularity is all we functionally need at this point to actually do our jobs.
posted by cortex (staff) at 11:21 AM on October 24, 2007
I find the question in abstract interesting, and I've actually done a little bit of flag-behavior analysis (very top-level at this point) that I might write up if I ever dig in far enough to find anything really interesting.
But I doubt we'd ever make that sort of thing part of the actual flagging system. We occasionally notice slightly weird flagging behavior, but when it comes right down to it someone is either abusing/misusing the flagging system or they aren't. Any one user lacks the power to disrupt the system, and the only likely result of some obsessive flagging campaign on their part is us learning to shrug a bit at that particular user's flags re: their pet cause. That binary Hrm?-or-not granularity is all we functionally need at this point to actually do our jobs.
posted by cortex (staff) at 11:21 AM on October 24, 2007
I'd kinda like a Pepsi Blue flag. Pepsi Blue is easy enough to spot when it's the subject of an FPP, but less so when it's someone who strolls into an ongoing thread and stealth spams
Flag it as "breaks the guidelines", or, hell, something else negative if you prefer. Probably a few other people will too—this happens now and then when someone pulls that sort of stuff. If you flag it, check back a day later, and it's still there, send us an email detailing what you see as problematic.
posted by cortex (staff) at 11:25 AM on October 24, 2007
Flag it as "breaks the guidelines", or, hell, something else negative if you prefer. Probably a few other people will too—this happens now and then when someone pulls that sort of stuff. If you flag it, check back a day later, and it's still there, send us an email detailing what you see as problematic.
posted by cortex (staff) at 11:25 AM on October 24, 2007
Also, I assume you guys are still working on putting up that "delete this post" button for me.
posted by koeselitz at 12:02 PM on October 24, 2007
posted by koeselitz at 12:02 PM on October 24, 2007
Cortex wrote: "If a comment ... is noise, flag it as such"
I don't see "noise" as an option anymore (I happened to notice this yesterday when I went to flag that crappy list of pseudo facts--dammit, I just lost the game of flag-it-and-move-on). Bug or feature?
posted by Horace Rumpole at 12:28 PM on October 24, 2007
I don't see "noise" as an option anymore (I happened to notice this yesterday when I went to flag that crappy list of pseudo facts--dammit, I just lost the game of flag-it-and-move-on). Bug or feature?
posted by Horace Rumpole at 12:28 PM on October 24, 2007
We trimmed some of the the flag options out of the post flag list—posts and flags now have seperate lists, huzzah!—including "noise". We'd found that "noise" didn't really tell us much about a post flagged as such, and it wasn't as meaningful as, say, said flag on an askme one-liner.
posted by cortex (staff) at 12:37 PM on October 24, 2007
posted by cortex (staff) at 12:37 PM on October 24, 2007
posts and flags now have seperate lists, huzzah!
I gather you mean posts and comments, because I do see noise and derail as options for flagging a comment.
So is a post I'm flagging just on the basis of generalized lameness "breaking the guidelines", or just "other"? Or should I simply STFU noob?
posted by Horace Rumpole at 12:44 PM on October 24, 2007
I gather you mean posts and comments, because I do see noise and derail as options for flagging a comment.
So is a post I'm flagging just on the basis of generalized lameness "breaking the guidelines", or just "other"? Or should I simply STFU noob?
posted by Horace Rumpole at 12:44 PM on October 24, 2007
I gather you mean posts and comments
Ha! Yes. Awful typo. Though a thorough search of the metatalk archives would probably turn up someone asking for the ability to flag flags. Heh.
So is a post I'm flagging just on the basis of generalized lameness "breaking the guidelines", or just "other"? Or should I simply STFU noob?
I'd be inclined toward "guidelines", yeah. Insofar as a post is lame enough to deserve deletion, it's definitely breaking the prevailing "something interesting" guideline.
posted by cortex (staff) at 1:03 PM on October 24, 2007
Ha! Yes. Awful typo. Though a thorough search of the metatalk archives would probably turn up someone asking for the ability to flag flags. Heh.
So is a post I'm flagging just on the basis of generalized lameness "breaking the guidelines", or just "other"? Or should I simply STFU noob?
I'd be inclined toward "guidelines", yeah. Insofar as a post is lame enough to deserve deletion, it's definitely breaking the prevailing "something interesting" guideline.
posted by cortex (staff) at 1:03 PM on October 24, 2007
All eight flag reasons really mean the same thing: "mod, please take a look at this." Given that interpretation, I think one would find moderators very responsive.
posted by klarck at 1:35 PM on October 24, 2007
posted by klarck at 1:35 PM on October 24, 2007
“I think ‘offensive’ is a bit too much of a catch-all for ‘I don't like this.’”
Yeah, I was surprised to see cortex defend that as a more objective judgment. But it's not. People tend to claim that something is generally offensive if it is offensive to them. And being offended is an emotional reaction, very subjective.
“Axe-grindy”, though prone to the bias cortex described, is nevertheless much more of a rational judgment than it is an emotional judgment. I predict that it's inclusion would actually be more reliably impartial than “offensive” is.
“Offensive” should be changed to something that isn't so equivalent to “I am offended”. Something that is more necessarily a judgment about external, not internal, things. (Yes, strictly speaking, “offensive” should connote that something is offensive to more than just oneself. But I don't think people reliably use it that way.) “Needlessly Provocative” would be good, except that it doesn't really include deeply offensive things. Maybe “offensive“ should be “offensive to others”, which would hopefully be interpreted by flaggers as “offensive to more people that just me”.
Something else that would be more objective and also quite helpful to the admins would be the flag “Escalating Angry Argument”. Those happen a lot and it would be great if the admins knew specifically when such things are flagged and could do something early on.
posted by Ethereal Bligh at 7:42 PM on October 24, 2007
Yeah, I was surprised to see cortex defend that as a more objective judgment. But it's not. People tend to claim that something is generally offensive if it is offensive to them. And being offended is an emotional reaction, very subjective.
“Axe-grindy”, though prone to the bias cortex described, is nevertheless much more of a rational judgment than it is an emotional judgment. I predict that it's inclusion would actually be more reliably impartial than “offensive” is.
“Offensive” should be changed to something that isn't so equivalent to “I am offended”. Something that is more necessarily a judgment about external, not internal, things. (Yes, strictly speaking, “offensive” should connote that something is offensive to more than just oneself. But I don't think people reliably use it that way.) “Needlessly Provocative” would be good, except that it doesn't really include deeply offensive things. Maybe “offensive“ should be “offensive to others”, which would hopefully be interpreted by flaggers as “offensive to more people that just me”.
Something else that would be more objective and also quite helpful to the admins would be the flag “Escalating Angry Argument”. Those happen a lot and it would be great if the admins knew specifically when such things are flagged and could do something early on.
posted by Ethereal Bligh at 7:42 PM on October 24, 2007
“Axe-grindy”, though prone to the bias cortex described, is nevertheless much more of a rational judgment than it is an emotional judgment. I predict that it's inclusion would actually be more reliably impartial than “offensive” is.
When used, aptly, for the very narrow ground it covers. When used inaptly, it is worse than useless; when used aptly, it will refer to so small a subset of problematic comments that its inclusion in the list seems odd.
So by including it we'd be gambling on the spread between vendetta-driven abuse and careful, rational neglect. Where's the win?
Offensive works not because things are generally objectively offensive-or-not, but because there is no pre-stocked ideological baggage in the term "offensive". The ten dozen different ideas of what is and is not offensive drive individual users' application of the flag, and we well know that, but they have to at least decide that something is, in fact, transgressive enough to merit flagging on the grounds of its offensiveness.
The whole idea is that it's not a free ride, and that your rationale is not handed to you on a plate to remove the burden of actually deciding why you're flagging it.
posted by cortex (staff) at 8:18 PM on October 24, 2007
When used, aptly, for the very narrow ground it covers. When used inaptly, it is worse than useless; when used aptly, it will refer to so small a subset of problematic comments that its inclusion in the list seems odd.
So by including it we'd be gambling on the spread between vendetta-driven abuse and careful, rational neglect. Where's the win?
Offensive works not because things are generally objectively offensive-or-not, but because there is no pre-stocked ideological baggage in the term "offensive". The ten dozen different ideas of what is and is not offensive drive individual users' application of the flag, and we well know that, but they have to at least decide that something is, in fact, transgressive enough to merit flagging on the grounds of its offensiveness.
The whole idea is that it's not a free ride, and that your rationale is not handed to you on a plate to remove the burden of actually deciding why you're flagging it.
posted by cortex (staff) at 8:18 PM on October 24, 2007
cortex, do you long for the days when all you had to do around here was make wonderful music?
posted by Effigy2000 at 12:07 AM on October 25, 2007
posted by Effigy2000 at 12:07 AM on October 25, 2007
it's been raised and shot down before, so I'm not getting my hopes up, but I keep finding myself wishing 'noise' were an option for posts as well as comments.
posted by Arturus at 1:15 AM on October 25, 2007
posted by Arturus at 1:15 AM on October 25, 2007
Bah! Embracing the creative potential of the human soul is what arguing policy in Metatalk all about.
posted by cortex (staff) at 8:00 AM on October 25, 2007 [1 favorite]
posted by cortex (staff) at 8:00 AM on October 25, 2007 [1 favorite]
dw writes "I think 'offensive' is a bit too much of a catch-all for 'I don't like this.'
"Maybe it should just be 'I don't like this.' It's honest."
Well I don't like the mystery meat youtube posts but I don't find them offencive.
posted by Mitheral at 1:51 PM on October 25, 2007
"Maybe it should just be 'I don't like this.' It's honest."
Well I don't like the mystery meat youtube posts but I don't find them offencive.
posted by Mitheral at 1:51 PM on October 25, 2007
« Older You do read digg and reddit, don't you? | Why is it only axe grinding if certain people say... Newer »
You are not logged in, either login or create an account to post comments
posted by NationalKato at 8:25 AM on October 24, 2007