Aristocrat said: I apologize if this is the wrong place to ask this, but I need to know why it's considered such a big deal and frowned upon to "artificially" increase one's upload and tag-edit count by going back to a post after uploading and adding any missed tags.
Legitimate tag gardening is fine. The problem, I think, comes when people artificially increase their productivity (frivolous note edits, pointless tag edits, etc.) in order to either increase their e-peen, attempt to attract attention for higher status, or both. Using underhanded methods to achieve higher privileges is what's frowned upon.
Whenever I put up a new upload or notes, I always proofread the tags and notes after I'm done to see if there are any errors, which are promptly corrected. Not that I always catch everything, but it helps. As far as I can see, there's nothing wrong with that.
glasnost said: When I see a new comic posted by an artist I like, then find out another translator has beaten me there and started working on it, a little voice in the back of my head goes "oh darn", and knowing that the voice is silly doesn't make it go away.
Guilty as charged. The little voice torments me too. Ultimately though, I get over it and move on. Not having much time at all in the past year or so to do any translating anyway has helped to keep that annoying little bugger quiet.
Heh, I don't upload nearly enough any more to care about "The Game". Though, I admit, I was getting pretty close to one million tag edits when Albert turned tag edit counts off in profiles, and I'd rather like to know what I hit that silly milestone. Just a number in the end though.
Anyway, this is a very hard thing to regulate. Any hard-coded rules will run into situations where they just don't work or make sense. It's also hard to draw a line at how many tags should be present in the initial upload, how many you can add later, and how long of a wait is "excessive" before adding more.
It's almost never a big enough issue to warrant serious mod intervention.
jxh2154 said: Anyway, this is a very hard thing to regulate. Any hard-coded rules will run into situations where they just don't work or make sense. It's also hard to draw a line at how many tags should be present in the initial upload, how many you can add later, and how long of a wait is "excessive" before adding more.
Should people give records for a perceived lack of tagging then? Otherwise people will have their own arbitrary guidelines for what is considered lacking.
I know extreme cases like Mr_GT's deserve a boot to the ass, but Coconut's record seems absurd.
Neutral record is a better first step. But people leaving records also need to be prudent about when they give them. Focus on real offenders, not just the dude who added 3 gentags but left off some tag you personally think was important.
I'm against records about this issue, but I still wouldn't encourage situations like here.
recklessfirex said: Legitimate tag gardening is fine. The problem, I think, comes when people artificially increase their productivity (frivolous note edits, pointless tag edits, etc.) in order to either increase their e-peen, attempt to attract attention for higher status, or both. Using underhanded methods to achieve higher privileges is what's frowned upon.
I don't know where Aristocrat would have heard that but this is a non-issue, as people doing this, if any, would eventually be found out. Pretty much all I've seen was a user getting invited for note edits that weren't actual translations but sfx, typos, resizing and character names, and some janitor or mod said about it that it still was a useful job. Then another one who looked like a gardener before it turned out that he was only spending multiple tag edits on all of his uploads (that earned him a promotion). And that's about all.
The actual issue is really the other way round, that is, plenty of good people out there taking forever to get noticed. Partly because you can't request your own invite, and partly because few people are willing to look out for them. So even if some member was smart enough to know that a thousand dummy note resizings could potentially get him an invite, he's stuck waiting for about half a year while crossing fingers that people checking for invites won't ever dig through his history. I'm not really worried.
Fred1515 said: Honestly, I don't see what all the fuss is about. Unless you only care about that certain handful of famous artists, there will always be plenty of stuff to upload, provided you're patient enough to look for them.
It's incredibly much more time consuming to try finding good leftovers in the depths of pixiv (blogs, etc.), than monitoring the surface. Posting missed images tells more about your standards since you have much lower chances to find no-brainers, but you don't want to take this route if you're aiming for contrib because it's just too damn long.
Fred1515 said: That's true only for the most well-known and high-profile artists, and there aren't that many of them.
Can't agree here either. On pixiv, anything truly good doesn't last more than 10 minutes, unless you're lucky enough to be the only uploader browsing pixiv at this particular hour (rare and unpredictable). It gets much better on other sources but even there I'm surprised to see how frequently other uploaders check them.
A search on gentags:0 still turns up 590 pages of posts. So I'd say that it's a problem, quite aside from concerns about people 'snatching' uploads of good images. Some images are slipping through the cracks with no tagging or painfully minimal tagging.
how about not approving posts with deliberate bare minimum tags? or those posts will not appear in pending for approval unless the uploader adds sufficient existing general tags (those that appear already in the database with about 100+ entries)?
i could see 1-3 tagged posts a month old. maybe we could prevent the list from growing if we don't approve them in the first place. they already have 3 days to return and re-tagged their uploads but they seemed to not bother at all.
Xabid said: A search on gentags:0 still turns up 590 pages of posts.
This is the real problem here, and this is the thing people should be punished for if they continue to contribute to it.
ghostrigger said: How about not approving posts with deliberate bare minimum tags?
This was suggested / done in the past, but it's a bad solution. This method ends up causing otherwise good posts that would be useful or of interest to other users to become auto-deleted. It's not fair to the userbase to kill good images based solely on their Danbooru meta-data.
Shinjidude said: This is the real problem here, and this is the thing people should be punished for if they continue to contribute to it.
It's actually substantially less of a problem than it looks like, a database issue popped up somewhere, I cannot say when or how or what posts, and a bunch of posts have incorrectly recorded tagcounts.
It's actually substantially less of a problem than it looks like, a database issue popped up somewhere, I cannot say when or how or what posts, and a bunch of posts have incorrectly recorded tagcounts.
I've been idly working on that list off and on for a long time, and I know many other people have been, too (it used to be around thousand pages long IIRC.) I'm reasonably certain that the vast majority of images there genuinely have no general tags -- at least, I've never come across an image there that seemed to have the wrong count. Granted, I usually start with the oldest ones and work my way forward, so they might be at the other end.
Shinjidude said:
This was suggested / done in the past, but it's a bad solution. This method ends up causing otherwise good posts that would be useful or of interest to other users to become auto-deleted. It's not fair to the userbase to kill good images based solely on their Danbooru meta-data.
Hrm, how about if it the submission form simply gave an error if you put in absolutely minimal tags, without even leaving the page (so you don't need to re-enter things, just enter a few tags?)
Of course, this might result in a bunch of images getting extremely generic tags slapped on them for breasts, hair color, etc. But that'd be better than nothing at all.
Xabid said: Hrm, how about if it the submission form simply gave an error if you put in absolutely minimal tags, without even leaving the page (so you don't need to re-enter things, just enter a few tags?)
There's really no point because the limit would need to be like 2-3 so you didn't get locked out from posting abstract or scenery pieces.
ShadowbladeEdge said: There's really no point because the limit would need to be like 2-3 so you didn't get locked out from posting abstract or scenery pieces.
I suspect that just about any image could surpass that in theory, though I agree that it could be a huge pain at times.
At the very least, you'd have scenery or abstract, plus no_humans and/or still_life, (or solo, though if it has a character even abstractly they almost certainly contain many more tags), plus a tag for the overall color and color-scheme in many cases (including not just the color tags, but things like monochrome, colorful, etc.) Scenery will almost without exception get one of the nature, landscape, cityscape, or sky tags. Most abstract images would also get a tag for the medium or art style, or something like minimalist or the like.
Those tags are all kinda underused, and are all useful to have when looking for a certain type of image.
Just glancing over the abstract and landscape categories with gentags:1 -- the landscape one is empty, and all the abstract ones seem to clearly have multiple tags that could be added to them (almost all are monochrome in a single color, which is two tags already.)
Forcing people to put at least one gentag on every image they upload isn't going to be a huge pain -- every image could have at least one. More than two or three could become a pain, yeah. But, I mean -- an image with virtually no tags might as well not exist. Nobody's going to find it.
And even one or two gentags can make a big difference, since people can use it to find the image and apply other tags, or can rapidly go through and apply likely-implied tags based on it.
Sorry for reviving this topic I would like to ask at which point should a lack of tagging, when an uploaded picture is easy to tag, should be considered worth having a record, whether negative or neutral?
Nials said: Sorry for reviving this topic I would like to ask at which point should a lack of tagging, when an uploaded picture is easy to tag, should be considered worth having a record, whether negative or neutral?
For me I guess I would say:
1. If they upload a few images at a time but never bother tagging (day later) and they have be informed about tagging.
2. Uploading a ton (10+) of images in a row to up their post count and snipe images constantly.
Comics would be the exception here I think since they can come in large chunks and it can be quicker to tag them all at once since the tags overlap a lot.
I don't think point two is nearly as much of an issue anymore with Mr.GT gone.
I'm asking this, because i'm looking at someone's tagging and it seems that out of 98 posts, that person has tagged at most the characters and/or the author, and translation_request and that was about it. for a user that's been registered since last year, you'd think he'd know tagging is what you should be doing after uploading images.
Going back to the suggestions above, I think it would be a good idea to only accept posts with at least one copyright, one character, one artist and 2 gen tags.
Aristocrat said: You'll have one hell of a hard time uploading posts with original and artist_request then.
I forgot to add that danbooru should accept original as copyright & character tag and artist_request as artist tag. Same procedure with similar request tags like character_request.
