Wikipedia talk:Criteria for speedy deletion/Archive 17

From Wikipedia, the free encyclopedia
Jump to: navigation, search

A one week long trial run

I'd like to propose something that may be controversial, but whatever. I'd like to see a one-week trial run of A7 speedies being transformed into WP:PROD. An examination of PROD in other areas has shown that it is working, and many prodded articles that would be valid A7s otherwise are holding out for their full 5 days. A week long trial to see if the amount of contested prods for A7 going to AfD outweighs the benefit of having AfD less overloaded may be helpful in seeing if a change in policy is needed. Thoughts? --badlydrawnjeff talk 18:41, 23 January 2007 (UTC)

I'm unclear on what problem this is attempting to solve, or how it solves it. Postdlf 18:45, 23 January 2007 (UTC)
You realize that, after the standard five-day period we would have 20,000 - 30,000 articles in Prod? Also, based on standard practice in CSD, we'd probably have 1000 a day contested. What do we do with those? And do you really believe we should leave all the attack pages ("so-and-so is the gayest student at xxx high school") sitting out there? Jeff, have you watched the new stuff that comes in over a typical hour? Fan-1967 18:50, 23 January 2007 (UTC)
Don't be hyperbolic, Fan. Attack pages are a completely different criterion and would not be subject to this proposal (which I am, admittedly, unclear on the intent of). Jeff are you trying to show that A7s can be prodded and rescued with less contentiousness? -- nae'blis 19:15, 23 January 2007 (UTC)
I have seen what comes through in an hour, and I don't think this would cause a problem. I don't know if 20-30k articles is really realistic, either. --badlydrawnjeff talk 19:18, 23 January 2007 (UTC)
I suspect Fan mistook your suggestion as converting all Speedies to Prod for a week, not just A7s. I would add the caveat that anything tagged as A7 but falling under another explicit criterion would be exempt from the experiment, but realistically, I don't know that you'll have much success convincing admins to follow the experiment, much less the myriad tagging users. -- nae'blis 19:40, 23 January 2007 (UTC)
The problem is that too many good articles are being speedied, and that PROD is better to handle them. --badlydrawnjeff talk 19:18, 23 January 2007 (UTC)
The vast majority of what's being speedied (even limited to the A7's) is being created by high school or college students, and is about themselves, their friends, their teachers, or their garage bands. If you let it pile up for even a week you'd have a gigantic mess. There's an additional issue, which is that many of these kids keep creating articles until they get the message that they will be deleted. If you leave the articles out there, they'll just keep making more. Fan-1967 19:45, 23 January 2007 (UTC)
What makes you think that too many good articles are being speedied? Postdlf 21:32, 23 January 2007 (UTC)
I recently checked the December deletion statistics, and we were deleting about 3,500 pages of all types a day via all processes. Assuming a reasonable share of talk pages, images, user pages, etc... we probably have at most 2,000 articles a day. About 200 go to PROD/AFD. So maybe 1,800 articles a day via all criteria for speedy deletion. If even half of those are A7 (no idea), we'd see maybe 5,000 in a five day run. So the scale is challenging, but not overwhelming, if we can make the rest of the experiment work.
The trouble would be keeping track of all the ones to which the expirement was applied, and evaluating the experiment at the end. This idea has some value, but the proposal isn't fully baked yet. How will we keep track of the articles that were run through it? How will we evaluate the results at the end. We might send the experiment idea to the folks at WP:SCISSORS and ask them to design an experiment that we could actually learn something from, and cleanup from afterwards in case the (pseudo-?)prod tags are removed without any improvement to the articles.
If we are going to run an expirement, I'd wait a few days more so that the SEO community really knows that we've implemented nofollow (see the community portal). That in and of itself should help with the true spam levels and shrink the experiment size. GRBerry 21:52, 23 January 2007 (UTC)
I'm with Postdlf on this one: what exactly are we supposed to learn from that 7-day experiment? When you do newpage patrol, you realize how much junk is created. GRBerry guesses around 5000 a day and that means that you'll end up with a lot of contested prods that go end up in AfD giving them another 5 days worth of pointless debate. As long as pages are not salted, there's really not a tremendous amount of harm being done when an article is deleted under A7. If a mistake has been made, then the article will be recreated eventually and I seriously doubt that anyone seeing their last creation deleted under A7 will simply turn away from Wikipedia in disgust. We're adults (well, some of us :-) ), we have a talk page message that says: "contact me for further questions". The only point I see in the experiment is that when all is said and done people who disagree with A7 will say "see, these x articles survived AfD" and the supporters will say "see, these 5000-x did not" and we'll be back at square one. Pascal.Tesson 22:46, 23 January 2007 (UTC)
Minor quible, I estimate 5K for a 5 day run, not 5K in one day. We don't delete 5K pages in all spaces for all reasons in a single day. GRBerry 23:41, 23 January 2007 (UTC)
So the revised process might go something like this: user adds junk article, article is labeled A7, A7 is converted to Prod, user removes prod (improperly), user is notified of error and prod is added back, user (whose main interest is to keep article) pulls prod again, user is blocked for a day, prod is reposted, user pulls it again, user is blocked for a week, user re-registers as new user and pulls prod, sock puppet case is filed, sock puppet is identified, user is blocked as a sock puppet, prod (finally) expires (assuming no futher sock puppets), article is deleted. Certainly is a different approach. -- John Broughton | (♫♫) 00:01, 24 January 2007 (UTC)
Actually, that sequence frequently occurs in the 30 minutes before an article is speedied. For five days, expect at least a dozen socks and dozens of reverts on the tags. Fan-1967 00:06, 24 January 2007 (UTC)
We can learn whether prodding them increases the load on AfD. If it doesn't, then prod is a better alternative because it allows more eyes to see the deletions. --badlydrawnjeff talk 00:03, 24 January 2007 (UTC)
Anyone with experience at WP and sufficient stubbornness can prolong procedure for a long time, even now. If everyone did so WP would have collaspsed long ago. That wont be the usual case. Usual will be speedy, contested, to prod, most contested prod are contested the first day & go to AfD, AfD 5, and almost never do they go to DelRev. This is about 5 or 6 days for the fully contested ones. Probably a good number, once the first ed. has a chance to fix while at prod, will not actually be nominated for AfD. Most of the ones I de-prod don't get nominated for AfD. Some do, and about half make it and half don't. If I find it is still unfixed at AfD I don't necesarily vote for it myself. DGG 03:00, 24 January 2007 (UTC)
The thing is if even 10% of these prods are contested that would double the load on AfD. That's alot to ask unless there are a several truly improper A7's going on every day. I'd really like to see some evidence that articles are being deleted despite a consensus (rather than the opinion of a few editors) that they should be kept. A7 reflects the community consensus that non-notable subjects are not appropriate for the encyclopedia and it should not be suspended (which is what this proposal would do) unless there is strong evidence that it is being actively misused. Eluchil404 06:31, 24 January 2007 (UTC)

The major problem we have with new articles is abandonment. Newbies write something and forget about it. If speediable the article gets tagged for speedy, if not it gets tagged with "cleanup", "sources", etc., is subsequently forgotten, and we have some piece of crap in article space. So users don't really come back that often and ask what happened to their article. I don't know how many users ask the original deleter, but the volume we get at DRV is a fraction what we speedy on a daily basis. I'm mostly in favor of Jeff's proposal restoring A7's and G11's on request under two conditions:

  1. A7's,s G11's and Prods are treated the same way, namely they can be undeleted and listed by any admin who sees enough merit to give them a run at AfD. Articles tagged as vandalism, attack pages, copyvios have to go through the normal DRV process.
  2. The passus that undeletion means the original deleter was out of process is struck from WP:UNDEL. What undeletion means in this case is simply that enough divergence in opinion exists on the article to seek community input.

And while I'm at it I can pimp (and ask for input on) my attempt to streamline DRV. ~ trialsanderrors 04:28, 24 January 2007 (UTC) ||| Corrected to clarify that I don't support turning A7/G11's into prods, but I support restoring A7/G11's and Prods on request iff at least one admin sees enough merit in the article to list it at AfD. ~ trialsanderrors 23:46, 24 January 2007 (UTC)

I'm sorry, but if this is based on Jeff's assertion that "too many good articles are speedied" then I would like to see some evidence of that. Can people bring to the attention of the discussion any (let alone a high rate) of "good articles" that have been speedied under A7. By which, I mean articles that actually should not be deleted, not merely ones that should have been deleted via a different process that CSD. Mangojuicetalk 13:51, 24 January 2007 (UTC)
Well, offhand I can think of Jimi Tenor; I originally tagged the article A7 (there was no assertion of notability, and even now the assertion isn't very clear), but it was despeedied by someone who knew of the subject of the article (which I was unaware of); it's conceivable that the article could have been deleted if nobody who knew of the subject had come across it. The page seems to have been edited since, although I'm not sure if it's in a good state at the moment. I would prefer it if A7 was changed to give a time limit between addition of the tag and deletion (maybe 15 minutes, and applicable to new articles only?), so that situations of this sort can be corrected by the original article creator. --ais523 13:57, 24 January 2007 (UTC)
If you're adding such a caveat, we're not going to get anywhere - articles that shouldn't be speedied under A7 shouldn't be speedied under A7. --badlydrawnjeff talk 14:00, 24 January 2007 (UTC)
Writing a Wikipedia article about someone with no assertion of notability doesn't automatically make the person non-notable. (It does lead to the article being pretty useless, though.) This is possibly why A7 is as controversial as it is. The waiting period would give time for the original article-writer to state notability, thus making the article keepable and avoiding biting the author. --ais523 14:05, 24 January 2007 (UTC)
Jeff -- I know you hate it, but keep in mind, WP:IAR. Process is not the be-all-and-end-all. If the articles you're concerned with should still be deleted, then they should still be deleted. The whole point of CSD is to delete things that should be deleted. If you think it's broken, you'll have to convince people that articles are being deleted that actually should not be deleted. Otherwise it's just a question of process. As for Jimi Tenor, bad example because (1) it wasn't actually deleted, and (2) it wasn't properly tagged in the first place -- the version you tagged, Ais, listed a long discography, over ten albums. If I'd been closing that one, I would have easily denied the speedy, and not even changed it to a Prod (which I always do when I think the article should be deleted anyway). Mangojuicetalk 14:23, 24 January 2007 (UTC)
IAR shouldn't be invoked to push a speedy through when it obviously isn't one. I'm sure I can bring up a list of improper speedies - tough because I don't have access to the text - but it'll take me a while. One admin in particular has about a 10% fail rate by my calculations, and I'm sure that admin is in the middle of the pack. Is that tolerable? The point of CSD is to delete unambiguous articles, not just "things that should be deleted." --badlydrawnjeff talk 14:35, 24 January 2007 (UTC)
I was pretty new when I tagged that, and it wasn't obvious that a discography had anything to do with notability to me then. (Even nowadays, I tend to mentally blank 'data dumps' of that sort when reading an article, which is why nowadays I stay away from that sort of A7 (not trusting myself to detect an assertion of notability properly).) A7 is a problematic criterion; clearly it's needed, but it's not sufficiently unambiguous to make a good CSD in my opinion. It would be nice if the CSDs could be worded so that it was impossible for anyone reasonably intelligent to make mistakes in applying them, but I understand why that's unrealistic. --ais523 14:30, 24 January 2007 (UTC)
Several people have asked for positive evidence about the volume of articles that are being improperly deleted before starting the test. While I have my own reservations about the test, I don't think that's an appropriate prerequisite, at least as it's being applied above. Speedy-deletions are unreviewed prior to the deletion and only rarely reviewed after the deletion. That's the nature of a speedy - it escapes the oversight and discussion of an XfD discussion. There is no statistical source for the data being requested.
That leaves us with anecdotal evidence. In the discussions above, however, anecdotes are being rejected as evidence because either 1) the error was found and overturned (which means the process worked and excludes the datapoint) or 2) has not been confirmed by community decision or 3) was escalated to the community via AFD - which means the process worked and the datapoint is again excluded. I have not been able to come up with even a theoretical scenario that would provide clear evidence satisfying all of the objections above.
Before sending Jeff on a wild-goose chase for data that will address your concerns, I challenge the naysayers (including myself) to clearly describe the analysis that we would need in order to be convinced and to provide some idea for how that data can realistically be obtained. Rossami (talk) 15:42, 24 January 2007 (UTC)
(conflict..) I think the burden should go the other way around. Jeff is proposing a rather radical trial here. I know he's had a long-standing problem with A7 deletion and is one of the staunchest inclusionists you can come across. But okay, since Jeff isn't an admin and can't look through deleted history, this is sort of hard. So I'll do the following. I'll go into the deletion log, pull out the last 100 deletions under A7, and I'll use my judgement as to whether any of them would have (1) survived prod, and (2) survived AfD. I'll compile the results here; it may take me some time. Mangojuicetalk 15:49, 24 January 2007 (UTC)
Thanks Mangojuice. I was going to offer to go through a day's worth of deletions at some point and look for false positives - that should be about 3500 according to GRBerry's stats, right? Starting with a presumption of guilt is too much, but a review is healthy and would provide more data than we usually get in this argument/discussion. -- nae'blis 15:55, 24 January 2007 (UTC)
Up to 20 samples so far. Only one that even has a chance of going through AfD, but in that case, the article didn't meet A7... and in that case, I don't think it has much of a chance. Mangojuicetalk 17:50, 24 January 2007 (UTC)
Again, "survived prod" or "survived AfD" is a horrible way to figure out the context. If that's why you're speedying articles - because you personally don't think they're going to survive a separate process - this is exactly why an overhaul is likely necessary. The criteria are strict for a reason. --badlydrawnjeff talk 17:59, 24 January 2007 (UTC)
Yes, but there's a difference between whether or not the rule is faulty or just whether certain deletions are faulty. Obviously the latter happens, but I just don't see what the big fuss is if (1) no one brings up a deletion review, (2) no one recreates the article after it's deleted, (3) the article would never have a chance at AfD, and (4) the only problem is that someone stretched a speedy criterion too far. Jeff, this might be a good time to remember about Wikipedia is not a bureaucracy. The best approach here is to educate admins, not to get rid of the rule. Mangojuicetalk 22:10, 24 January 2007 (UTC)
It's both, really. The rule is faulty because it allows too much interpretation (it's not objective and too ambiguous) and because admins don't know how to properly use it. I'd challenge more of these if I could see the information but, well, that's not happening. That's big fuss #1. Big fuss #2, maybe it would be recreated, but that's another can of worms, I've been scolded for doing that before, even with obvious notability. Big fuss #3 is irrelevant - "not a chance at AfD" isn't a speedy criterion for a good reason. It's interesting - you only hear WP:IAR/WP:NOT a bureaucracy when there's no other leg to stand on. Maybe I should IAR and just remove A7 from the policy altogether, eh? That would be a bit much, but there's an expectation with speedy deletions that clearly isn't being met, and it, not too surprisingly, is exactly what was predicted when A7 was originally proposed. --badlydrawnjeff talk 00:30, 25 January 2007 (UTC)
(Edit conflict...) Jeff, you can't just go around and make silly accusations like "One admin in particular has about a 10% fail rate by my calculations, and I'm sure that admin is in the middle of the pack." What does "fail rate" even me in this context? If you believe that some admins are overusing A7 then take it up with them. Judging from my own experience at newpage patrol, most articles speedied under A7 are at least read twice because they are tagged first by non-admins (or careful admins) and deleted a few hours later. That prevents some of the problems. Pascal.Tesson 15:47, 24 January 2007 (UTC)
In response to Rossami, unless I'm somehow confused about admin access to deleted pages, it should not be too hard to carefully review a random sample of recent A7 deletions and to do enough research to figure out whether or not these articles would have ever stood a chance to meet the various relevant guidelines and policies. It should also be fairly straightforward to compare deletion logs prior to the expansion of A7 and the current ones to understand how many extra A7 deletions are performed on those grounds. Finally, statistics on the activity at AfD are fairly easy to compile. I'm not saying that data tells the whole story but it's a good start. Pascal.Tesson 15:52, 24 January 2007 (UTC)
A "fail rate" is "articles speedied that did not meet any of the criteria listed." I can use one example of an administrator easily, with some work I can come up with a few more. I would be more okay with the CSD as a whole if there was no unilateral deletion, but instead one tags and one deletes, but that doesn't address the broader issue. Mangojuice's comments above worry me - how many other admins who work on speedy deletion use the "will not survive PROD/AfD" measurement as opposed to the criteria listed? --badlydrawnjeff talk 17:59, 24 January 2007 (UTC)
Just to be clear, I don't go around speedy deleting articles just because I think they'd never make it through an AfD. However, you're saying that A7 is so broken, we should replace it with prod, and when authors object, we have to take those things to AfD, which is a really large burden on the project. Obviously the current system isn't perfect, I'm just saying, in the end, how often does an article actually get deleted via CSD that really shouldn't be deleted, period? If it basically never happens, then the proper thing is to educate those who are misusing speedy deletion and advocate for the correct process, but in the end, we keep the criteria we have because they basically work as long as they're applied correctly. Mangojuicetalk 18:26, 24 January 2007 (UTC)
I again reject your framing. I asked it down there, I'll ask it up here, too - is a process that only works properly 60-70% of the time really beneficial? Let's say those 21 that were ambiguous by your account go straight to AfD - if 10 of them are salvaged, was that worth it? Maybe switching them to PROD isn't the answer, maybe changing them to PROD with an important caveat regarding the PROD process is the answer, I'm not entirely sure. But I can say this much - speedy deletions are supposed to be objective and uncontestable, and it's clearer now than it was when I proposed this that A7, as used, is not. I may be the most inclusionist regular editor on the project, but I'm rarely wrong when it comes to understanding our processes and policies. This is a Big Deal. --badlydrawnjeff talk 00:06, 25 January 2007 (UTC)

Mangojuice's analysis

Okay, I'm done with my analysis at User:Mangojuice/a7. I only went up to 50 articles, but I have other things to do, anyone may feel free to continue my analysis. In the 50 I looked at, 3 had a greater than 0 chance of surviving AfD; of those, only one contained any actual work worth keeping (SofTech, Inc). However, an awful lot of them would have become AfDs if they weren't speedily deleted. There are a few examples of misapplications (IMO) of the criteria, but with those 3 exceptions, I think the chance that Wikipedia lost something valuable is basically nil. Keep in mind there were a LOT more A7/G11s that I skipped over because certain admins made a lot of deletions. And I'll point out that the ones that I skipped over the most (Steel359 and Jni) had solid records. So don't take this as 6% of speedy A7/G11 might have problems, the real percentage is probably much lower than that. Mangojuicetalk 22:05, 24 January 2007 (UTC)

The one you cited as possibly worth keeping, SofTech, has been repeatedly deleted as spam for the last year ([1]). -- Fan-1967 22:17, 24 January 2007 (UTC)
Going only off your analysis and nothing else, 29/50 were rated "5." This means that 21 of the A7s listed were not completely unambiguous. That is a major problem. I had the poor hit rate at 10%, your poor hit rate could be as high as 42%, and this is without you sampling some of the worst admins when it comes to A7 deletions as well. Again - it has nothing to do with a survival rate at AfD/PROD. --badlydrawnjeff talk 22:30, 24 January 2007 (UTC)
First of all, the survival rate at Prod is potentially 100%. Those are the Prod rules, anyone can cancel it. Imagine that your estimate is correct, and 42% of A7's have some glimmer of a shadow of a hope of a possibility of notability. Imagine also that any substantial percentage of those authors would contest the Prod. Guess what: We cant handle that many in AFD. We barely handle what we get now. Of course, the alternative would be just to keep everything. Why not just make this a vanity site about really talented sixth-form footballers? Fan-1967 23:55, 24 January 2007 (UTC)
There are other alternatives, of course, but it seems like you're more interested in extremes. Do you consider a process that only works 60-70% of the time a net benefit? --badlydrawnjeff talk 00:02, 25 January 2007 (UTC)
Without commenting yet on the rest of the discussion, I'm going to disagree with two of Fan-1967's points immediately above. First, it is not my experience that "100%" of Prods are simply undone. That's theoretically possible but doesn't really happen. (If it did, Prod would be pretty pointless.) Second, I disagree with the assertion that "we can't handle that many in AFD." AFD is a very robust process that successfully handles a lot more than most people give it credit for. Would this add to the load at AFD? Yes. Is it an unbearable load? We don't have enough data yet to know. Is the potential gain worth the additional load? Again, not enough data yet but Mangojuice's analysis is getting us closer. Rossami (talk) 00:12, 25 January 2007 (UTC)
Whoa there. I didn't say 100% of Prod's are undone. I just said it is "potentially" (maybe I should have said "theoretically") 100%. I just asked what the load would be if any substantial percentage were contested, and I think that is a very real possibility. Fan-1967 00:17, 25 January 2007 (UTC)
I disagree with Jeff's interpretation that 42% have some hope. No, 3 of them had some hope. The 4s were ones I think everybody would agree on, but there is merely a minimal element of judgement involved. As an example, the first article was on a band: the claim of notability amounted to that some of the tracks on their unreleased album are publically available and have "been praised." That's what a 4 is to me. And 3s are ones that I think clearly fit in the criterion, but require real judgement to decide. An example: one was a prof -- that's all, just a prof, no other explanation. Not a claim of importance, but you have to use judgement. Mangojuicetalk 00:23, 25 January 2007 (UTC)
If there's any judgement involved, why are we speedying it? No, really, this is important. Speedy deletions are supposedly unambiguous, so there shouldn't have to be any judgement. --badlydrawnjeff talk 00:30, 25 January 2007 (UTC)
Because we have to. On a daily basis, we're being inundated with dreck, and bands who claim someone praised one of their songs on myspace, or that it must be good because dozens of people have downloaded it, are a component of that dreck. Should we go to AFD for every band with a myspace page? Fan-1967 00:33, 25 January 2007 (UTC)
No, we don't have to. Should every band that asserts notability go to AfD - yes, without question. --badlydrawnjeff talk 00:53, 25 January 2007 (UTC)
You really need to spend more time on RCP. Frequently, the myspace page, and the 173 downloads from it, are the assertion of notability. So, based on that, AFD or A7? Fan-1967 01:00, 25 January 2007 (UTC)
That's an A7. 173 downloads isn't an assertion of notability. I spent a lot of time on RCP at one time (I love how that's supposed to be an argument - "Oh, you'd support more speedies if you saw what gets submitted." - Yes, we do know, and we're still bitching). Ironically enough, I got sick of it when too much was being deleted that shouldn't have been. That was over a year ago, and it's frustrating to have evidence that it hasn't been improving like I assumed it had. --badlydrawnjeff talk 01:09, 25 January 2007 (UTC)
You're using judgement there. Good judgement. Just like admins do, for most of their admin actions, including speedy deletion. Mangojuicetalk 01:52, 25 January 2007 (UTC)
There's no judgement to be found there. No one's making a judgement as to whether 170 downloads is an assertion of notability, nor is it meant to be. If people are making judgements regarding deleting, they're not doing their jobs, quite frankly. If there's even a question, it's not supposed to be speedied. --badlydrawnjeff talk 02:41, 25 January 2007 (UTC)
Ok -- how about an article on a doctor that, apart from clearly ordinary doctor stuff, adds "Whilst at University he won the Francis Dudley Memorial prize in Anatomy, and also the Southmead Shield case presentation award." How about the article on the band that has two albums that are free and self-released, but says this is "by choice." Mangojuicetalk 03:16, 25 January 2007 (UTC)
No to the first, without even thinking about it. Yes to the second without even thinking about it. Neither of these are even close to judgement calls. --badlydrawnjeff talk 04:43, 25 January 2007 (UTC)
Rated those both as 3s. The "by choice" band was a harder call than I made it look, because the article made it read like it was a philosophy rather than that they don't have a record deal, and I could picture a legit band doing that... but that meant their records weren't really a claim. It struck me as a little unfair that a similarly successful band that sells their records could very well meet WP:MUSIC while this band qualifies for speedy deletion. The first, to me, is an easier delete: the awards are obviously nice but also totally irrelevant to anything, it's like a CV, and I can't imagine someone being notable for having won an award in med school. If anything, their inclusion makes the article worse. Just because you say "without even thinking about it" doesn't mean you aren't exercising judgement, just that you're very sure of your judgements. Mangojuicetalk 11:14, 25 January 2007 (UTC)
Then I'm not even sure about your ratings anymore, either. Or judgement. --badlydrawnjeff talk 12:21, 25 January 2007 (UTC)
Nothing is unambiguous that depends on human interpretation, and good rule-making acknowledges this and allows for it & takes precautions. The conditions are an attempt, & would help if used right. There is a trace of it: it takes two people for a speedy or prod in general. DelRev is there, but it is very clumsy and therefore little used. One step would be to require recent & original editors to be notified, instead of just suggest. (I'd like to say automagically, but it is not automatic identifying the right editor(s).)
It might be possible to devise rules that would handle at least some of the caseload. (ie pages with fewer than 10 words) Would take a little testing.
I think it perfect reasonable for anyone to nominate for AfD, whether or not after a prod or a speedy, for other reasons than to prevent deletion: in order than a broader number of comments might be heard, as is often specified in AfD nomination. (sometimes when it seems to be a question where the usual practice is unclear, sometimes if it looks to be a question of bias or hoax., or just if one sees something that looks like junk, but in a field where one cant tell oneself.) Whatever the defects of AfD procedure, it does allow for a variety of views. I find it very quick to scroll through the daily AfD list looking for the few about which I have something useful to say. . DGG 00:46, 25 January 2007 (UTC)

We try to minimize the chances of good stuff being deleted, but the chances in these situations are so low that it helps far more than it hurts. We can't be brought to a screeching halt because a tiny little percentage of articles might not be speedy'able. Wiki is wiki, and speed is our ally as well as our enemy, as long as things are in an acceptable balance. Where we draw the line might be up for debate, but it is vital that we do draw such a line.

As far as an experiment goes, I think doing what Mangojuice is doing would be a far better idea (post-review) than the original proposed test, if such a test is still desired at this point. So far, what Mangojuice has shown seems to show we have a more than acceptable balance of "good with the bad" for this process. -- Ned Scott 04:58, 25 January 2007 (UTC)

I'm okay with my idea being put in the dustbin for the time being, but I'm not sure what kind of "acceptable balance" we're observing here. Being generous, 25-35% improper speedies does not seem acceptable to me in the least. --badlydrawnjeff talk 04:59, 25 January 2007 (UTC)
Well, maybe if that were true, but it's not. Which wiki are you watching, because it's not this one. -- Ned Scott 05:03, 25 January 2007 (UTC)
I don't see what's untrue about it. His random sample of 50 is very clear that many articles are being speedied that do not fit the criteria. --badlydrawnjeff talk 05:04, 25 January 2007 (UTC)
No, 2 out of 50 don't fit the criteria. 3 more fit the criteria only dubiously (and one of those, it turned out, was a copyvio, but as that information was probably unknown to the closer, I called it a 2/5). The rest definitely fit the criteria. That's what's untrue. What I see as useful, though, is that this kind of review may help point out which admins are interpreting the criteria broadly vs. more narrowly, and for those who interpret broadly, we can attempt to convince them it would be better to be more narrow. I'd like to suggest, though, that if anyone wants to repeat the experiment, they pick a time period (say 4 hours, though a day would be better) and evaluate all deletions in that period, regardless of how many any given admin closes. I had a feeling, from doing the experiment myself, that I had disregarded a lot of very solid decisions, and fixing a time period would help clarify that. Mangojuicetalk 11:14, 25 January 2007 (UTC)
You keep saying "2 out of 50," but that's entirely misleading. Perhaps another admin who can see the text can go over them and add his/her own ratings just to get a second opinion. i mean, how broadly were you interpreting it? --badlydrawnjeff talk 12:21, 25 January 2007 (UTC)
I've asked Nae'blis, who offered to help earlier, to do an independent set of ratings. As to how broadly I was interpreting it, I'll say I probably would have still speedied all the 3s, 4s, and 5s, although perhaps one or two of the 3s I might have converted to Prod to be charitable. The 2s, I would generally not speedy, myself, but would prod or AfD, but since they were clear deletes, I can't fault the decision too much. Generally speaking, I'm pretty strict, too. Take band articles, for instance -- if they say anything that makes me think they are remotely successful, beyond, say, having a few local gigs, I shift to Prod, AfD, or simply deny the speedy. I have a high standard especially for A7 corporation or website deletions, because I think 'claim of significance' has to be interpreted quite broadly there. The kind of corp article I would A7 are for businesses I know there are hundreds or thousands of (such as software companies or web design companies, or law firms) with no claim of being important, where even things like being relatively old or large, or international can be claims of importance. But if a company is doing something sufficiently unique, I don't speedy. G11s I'm tougher on: detailed spam is still spam; I'll deny one if it's easy to remove the adspeak but otherwise I don't consider such contributions to be "salvageable". A7s on groups or clubs other than bands are not as well-defined: again, anything unique about them will save it (though I don't count doing-x-at-y-college as "unique"), including being old or big. Mangojuicetalk 15:45, 25 January 2007 (UTC)
Interesting analysis, it suggests to me that CSD appears to be working. Odd that my deletions got 2 of the 8 footnote mentions ;), upon review I think my decisions were correct. As long as we don't bite the authors who are trying to create valid articles, the occasional mistake can be corrected without much of a problem. But given the volume involved, we seem to be doing a good job. NoSeptember 12:17, 25 January 2007 (UTC)
How do you feel it's working with this evidence? --badlydrawnjeff talk 12:21, 25 January 2007 (UTC)
Any good article deleted from a newbie risks losing him; any good article omitted from WP diminishes its status. Summary procedure is needed in cases of actual harm, but it is detrimental everywhere else. There will be errors, but that does not mean we should despair of minimizing them. I would much rather have 25% the amount in AfD than do the work of sorting every few hours through speedies. DGG 01:32, 26 January 2007 (UTC)

This debate makes me confident that I was right when I suggested above that if we went through with the experiment Jeff was proposing, we would end up with supporters of A7 saying "see, I told you it works" and opponents saying "see, I told you it doesn't". If anything, the stats provided show what we all believed: A7 mostly ends up deleting content that we all want to see deleted but is occasionally used to delete things that should have gone the AfD route. And let's not kid ourselves, the stats aren't showing that it fails in 40% of the cases, no matter how much one tries to bend the data to prove a point. DGG says that every good article deleted risks losing a newbie and diminishes Wikipedia's status. Fair enough although it would take a pretty thin skinned newbie to be so deeply offended when user talk page templates are used appropriately to motivate the decision. On the other hand, every article going to AfD with absolutely no chance to survive wastes Wikipedia's most valuable resource: time of dedicated voluntary editors. And every spammish crufty non-notable stub diminishes the credibility of Wikipedia. This debate is all about balancing the need for efficiency and the need for prudence. The rough consensus, as far as I can tell, is that A7 is fairly close to the right balance, especially if admins are encouraged to be careful in its application. Pascal.Tesson 06:33, 26 January 2007 (UTC)

Single Day Data

I've pulled the data for 24 January (all day EST). To the extent I've parsed it (haven't even gotten to reasons yet), it is at User:GRBerry/Deletion Log Stats. Activity is up from the December average, with 4,239 log entries of which onyl 50 were restores. Top spaces were (Main) 2,093 entries on 1,865 pages; Image 1,028 entries; and Talk 468 entries. The top 11 admins did 50% of the activity. Parsing reasons is going to take some serious thinking before I code, and it is pushing midnight, so definitely not tonight and probably not till next week. I obviously can upload a file of it all, or a specified fraction. Let me know special requests, but this is on my home machine so do not expect to see anything except late in the US evening. GRBerry 04:38, 26 January 2007 (UTC)

No rush, but is there a status report on this? --badlydrawnjeff talk 14:55, 29 January 2007 (UTC)
Not yet. Probably not tonight or tomorrow night. But I did just think of it again. GRBerry 00:16, 6 February 2007 (UTC)
I did an analysis of 1/3 of a day at User:Mangojuice/a7.2. I suppose I should admit I'm not going to finish it at this point. It is exactly 1/3 day, though, 1600 to 2359 UTC on 1/16. Mangojuicetalk 01:04, 6 February 2007 (UTC)

Updated with reason tables for all spaces. Some pages had multiple reasons given. So the tables have two variants - one reflecting each unique combination and not overcounting, and one counting each page under each reason, so overcounting in total. Top reasons (overcounting method): A7 678 entries, G8 330 entries, I4 268 entries, PROD 241 entries, G11 213 entries, and G1 207 entries.

I could now field requests for space/reason or user/space/reason analysis, as needed. I'm certainly not going to do all the cross-combinations, unless there is a specific request. GRBerry 03:33, 9 February 2007 (UTC)

BJBot & Possibly large number of images in ORFU categories

Wikipedia:Bots/Requests for approval/BJBot is being designed to go through the 100,000+ orphaned images and mark all the orphaned fair use images with the ORFUD template. Estimates for the number of images that are orphan fair use are 19,000 (via analysis of November 30 dump) and 40,000 (my calculation based on my experience of going through orphan images manually). The bot will likely operate at 1 or 2 image edits per minute, thus, the bot could be placing up to 2880 images PER DAY in the Orphaned Fair Use categories. This will create 6 to 13 days (estimated) of categories with almost 3000 images in them (more days if the image runs at a slower rate, but each category would have less images). I believe this to be okay, since at least the images would be marked, and if the categories aren't cleaned out for a month or two afterwards, that would be acceptable. A possible alternative is to create categories especially for the Bot would would still apply under CSD I5, but not cause backlog in the ORFU normal categories. The advantage of this would be clearing out this "hidden backlog" of images, with the days of categories after the Bot run being likely very minimal (now they currently go from 90-300 (the higher numbers are likely days I manually tagged images)) because the orphaned images would likely mostly be taken care of. As part of the bot approval, it was suggested mentioning it here to ensure that no one would have a problem with the possibly (likely) large number of images in the categories. Is the alternate method of creating the special categories for the Bot preferred? Are admins that work in this area up to the monumental task of deleting of at least ~19000 images in the next month? Are there other questions or concerns that anyone has? Any comments or questions are appreciated. Thank you. --MECUtalk 02:05, 24 January 2007 (UTC)

Go for it. I hope one day we have something built into the software that deletes orphaned fair-use images automatically after 48 hours (for images uploaded after 13 July 2006). It might make the task of deleting larger, but that's only because the task really is that big. -- Ned Scott 02:14, 24 January 2007 (UTC)
Hmm, that thought makes me wonder if anyone has made such a feature request. -- Ned Scott 02:16, 24 January 2007 (UTC)
I like the idea of specialized categories myself but that's just an opinion without much basis in experience. Also, I take it there's some concern with bots deleting images? Orphaned fair use images seem appropriate to simply delete without much oversight. (this coming from a supporter of fair use on Wikipedia. If it's not being used then it probably shouldn't be there at all.)--Jeff 02:20, 24 January 2007 (UTC)
A bot has never been given sysop to date, so I don't think it will start soon. BJTalk 02:52, 24 January 2007 (UTC)
In this case there might be problems with people fooling the bot into thinking a page was ORFU when it wasn't, causing the deletion of a useful image, which is why such a deletionbot is unlikely to be approved. --ais523 13:59, 24 January 2007 (UTC)
That's easy to code around ... a hypothetical deletion bot (which this bot is NOT - this is purely hypothetical) would need to check to see that the orphaned template was added over a week ago and that the image is really in a fair use category. Besides, since images can be restored now, this isn't as big of an issue as it would have been six months ago. Heck, a deletion bot could even be made to accept requests to restore images that it deleted so that if it did delete anything in error, it could be restored with no admin intervention. Sadly, though, if we wouldn't approve using a bot to protect the daily featured article from penis images, I doubt we're going to approve one that would actually delete stuff. --BigDT 00:13, 25 January 2007 (UTC)
This is also all hypothetical but BJBot already checks what you mentioned and a check to make sure the image hasn't been edited in 5 or so days would be good to have. Really I have a better chance of getting sysopped right now than a bot with sysop ever happening anytime soon. Plus if BJBot ever had a chance to be sysopped I would have to hand over the code to somebody else because they will never sysop a bot ran by a non-sysopped user. BJTalk 00:22, 25 January 2007 (UTC)
User:BJBot has made it's first run using the dump from late November 2006. It looked at ~14,000 images and marked ~10,000 for deletion. The ORFU categories for Jan 27, 28, 29 and 30 will have elevated numbers, especially 28 and 29. However, I have gone through orphaned images and there are many that BJBot "missed" for one reason or another, I estimate that BJBot only got 40% of the ORFU images, which means there are still 15,000 still out there. --MECUtalk 00:52, 1 February 2007 (UTC)

Grace period on newly created articles

I suspect it's not uncommon for new users to create an article with essentially no content, and then spend some time adding content in the next revision -- only to discover that the article got speedied in the mean time when they click "save page". In the spirit of WP:BITE, I propose adding, say, a two hour grace period during which a new article is immune to A1, A3, A7. —Ashley Y 00:10, 29 January 2007 (UTC)

That's what we have the {{hangon}} tag for. Articles don't actually get deleted that fast, so nearly all the time the author would have time to add the hangon tag before deletion. The way you propose basically isn't workable, though -- people need to look at new articles right away, and checking for creation time is an extra step admins shouldn't have to take all the time (though, it would be nice, and I sometimes will delay a deletion when I have a suspicion that the article is too new. Mangojuicetalk 01:07, 29 January 2007 (UTC)
I saw a recent case where an article got a speedy deletion tag slapped on it 26 seconds (apparantly) after it was created. I removed the speedy tag (the article had references from the moment it was created), and the article was listed at AfD and eventally kept. Carcharoth 02:50, 29 January 2007 (UTC)
I've had an article have an A7 speedy tag incorrectly slaped on it less than 15 minutes after I saved it, and an admin deleted it 3 minutes later. I went directly to the admin instead of deletion review; the article is still there. Argyriou (talk) 03:00, 29 January 2007 (UTC)
Mangojuice, It takes exactly the same amount of time to delete pages 2 hours after creation as it does to delete them 2 minutes after creation. It is not difficult to click on the history tab , look at the time, and then close the tab or window and go on to the next one if the time is very short. I just now times it, and it takes between about 5 seconds each once you get started. Try it, you wont find it so bad. Anyway, it should be not necessary. for admins to check it manually, because it would be trivial programatically to add the time to the displayed list, or to display them in one hour segments. For those accustomed to the current procedure, I point out that unless it is used carefully and sensitively, we will be alienating new users--and also those not so new.
Having all admins being as careful as you is one solution, of course, but I don't think its very likely. DGG 06:18, 29 January 2007 (UTC)
Couldn't this be addressed with the addition of some ParserFunctions to the relevant templates ... such that the article wouldn't actually be added to Cat:CSD until X hours after deletion? -- Visviva 06:41, 29 January 2007 (UTC)

Making users more aware of the fact that they can use their user space to create and tinker with an article until it is ready for the main space would help as well - X201 10:31, 29 January 2007 (UTC)

  • Limited support. I oppose this for CSD rules relating to vandalism, patent nonsense, attack pages and other clear cut objective criteria, because these are unlikely to become useful and are easier to dispose of as quickly as possible from Recent Changes patrol instead of reviewing new pages. I support it for more subjective CSD rules such as A7 and advertising, which are notorious for false positives. Deco 19:26, 29 January 2007 (UTC)
  • Strong Oppose Sometimes an article just doesn't deserve to exist. While on RC patrol earlier today, I deleted Pirates Of The Caribbean Fans Group and Jalice within a minute of their creation (the former was blatant advertising for some fan club, while the other was a vanity article that so succinctly defines G1 that it should be made into a monument.. if it wasn't utter crap). Both had been tagged for speedy deletion in less than a minute after their creation, and their authors given sufficient notice on their respective talk pages.
    Sorry, but WP:BITE shouldn't take priority over maintaining the integrity of the encyclopedia. Tough shit for the newbie if their first contribution is of dubious nature; we don't have a "wait two hours before reverting bad first edits" rule, so I see no reason that we should have one for new articles. EVula // talk // // 20:16, 29 January 2007 (UTC)
    • I'll just point out that the delay suggested was intended to apply to all new articles, not just first articles by total newbies. And that there should be a way to both maintain the integrity, and keep good all articles capable of improvement. WP people, by and large, are ingenious.DGG 20:30, 29 January 2007 (UTC)
  • Partial agreement and partial disagreement. Criterion A1 appears to be somewhat subject to debate, but I don't know how often it actually gets used. Criterion A3 should not be subject to delay - if all you can do is put up a couple of links, instead of rewriting some text from one of those links, the article can get nuked. Criterion A7 is the most error-prone and subjective of all the A criteria; perhaps we should eliminate A7, and make all such articles,which don't qualify under other criteria, go to AfD instead. Argyriou (talk) 20:21, 29 January 2007 (UTC)
    • The problem isn't A7. The problem is that A7 is being misapplied. A7 is intended to provide for the speedy deletion of high school students and personal websites. It wasn't intended to be a substitute for AFD in every case where the subject may not be notable. --BigDT 20:26, 29 January 2007 (UTC)
      • The solution to A7 being misapplied is either to get better people to apply it, or to clarify or drop the rule entirely. We're not going to get better administrators, so we need to fix the rules. Argyriou (talk) 20:33, 29 January 2007 (UTC)
        • We may not "get better administrators", but if A7 were to be reworked slightly, I think it would be entirely appropriate for a bot to spam the talk pages for all administrators to inform them of the revised guidelines (as part of an effort to make sure the rule is fully understood). EVula // talk // // 15:50, 2 February 2007 (UTC)
  • Oppose. Two hours is too long, by far. The typical A7 article reads like this: "Seeoh Aye is the greatest 10th grader ever! He attends Enn Eff Tee High School and gets straight As. If you want to talk to him, you can contact him at <personal information>." This is not a borderline case, and if it were taken to AfD the consensus would certainly be "speedy A7." This article does not deserve two hours, or even two minutes (although I agree we shouldn't bite, we also have to show that we won't tolerate nonsense). Criterion A7 is not meant to be subjective -- it's usually very clear what is, or is not, an assertion of notability. In cases that it isn't, many admins will PROD or AfD the article anyway. Perhaps the ideal thing to do with articles not asserting notability is to "speedy userfy" instead of speedy delete -- most of them are COI anyway, and would be acceptable in user space. --N Shar 06:14, 7 February 2007 (UTC)
  • Oppose. A grace period means that many speedyable articles will never be caught. We need to make it clearer that you must provide a decent article when you first submit it, though. -Amark moo! 15:31, 7 February 2007 (UTC)
  • support grace period There are some articles under A7s which can well go immediately, and the way to indicate them is to use less confusable criteria, such as "little or no content" --which describes the one or two liners mentioned above. I don't think userification for these pages makes any sense at all, for we will just be accumulating more nonsense in user space--anyway, what's the point of userifying an article from an ip address?
    It's the A7s with content that are the problem. Among the articles listed for A7 have been articles about a member of the National Academy of Sciences, and a full professor at Caltech. (both kept after AfD). Now, these should not have been nominated in the first place, at least not for Speedy. But such articles are nominated and should be kept. Perhaps the solution to improper use of speedy is to replace A7 with a much more specifically worded replacement, but we've been debating the wording for some time. In the meantime we need to have a way of screening articles. The suggestions for even more drastic procedure assume perfect or nearly perfect admins--and as a previous comment mentioned, that applies to many, but not all admins. DGG 16:18, 7 February 2007 (UTC)
    • DGG, I understand your point but in your first paragraph above you are improperly stretching the definition of other CSD criteria in your attempt to address it. There is no CSD criterion which allows speedy deletion for "little or no content". Criterion A1 allows speedy deletion for little or no context. N Shar's example above has context - it's a biography about a student at a particular high school. Criterion A3 allows speedy deletion for "no content whatsoever", that is, a page with only links, tags, etc. The criteria are very narrowly written. None of the criteria except A7 allow for the speedy-deletion of the example N Shar described. Rossami (talk) 21:19, 7 February 2007 (UTC)
    • Let me quickly point out that we won't have the problem of being unable to userfy an article because it came from an IP, since IPs can no longer create articles. --N Shar 00:33, 8 February 2007 (UTC)
  • Oppose. I think the integrity of the encyclopedia is served best by deleting obviously inappropriate articles as quickly as possible. I believe people are less likely to create such articles if they are unable to show them off to friends, email links around, etc. I have seen unwise usage of A7 (and G1), but (1) those nominations are a tiny percentage of all speedies; and (2) practically all of those articles would have been deleted anyway, via PROD or AfD. A two-hour (or other) "grace" period just allows Wikipedia to be a laughingstock or object of ridicule for two hours per article that should have been deleted on sight. And leaving those articles up encourages repetition of similar behaviour. --MCB 22:30, 7 February 2007 (UTC)

G4 seems to have a wording problem...


...Specifically, overbreadth. This has been nagging at my hindbrain for about 13 months; after digging around all this time I can't seem to find any way around the conflict, so here it is: G4 seems to imply that any reinstitution of a deleted article (other than by creation of an all-new article that happens to have the same name) should be SD'd. The problem: Many articles are deleted because they are unsourced or questionably sourced (i.e. WP:V, WP:N, WP:RS, WP:INDY, WP:AUTO, WP:VANITY, etc.; all of them and more have relevant passages). Meanwhile, WP:N, WP:V, etc., indicate that (if other faults are not present) adequate sourcing is enough to save an article. So, G4 appears to be in unintentional conflict with these other policies/guidelines when it comes to reinstitution of articles that were deleted on such grounds, but later restored verbatim but for the addition of (multiple, independent, reliable — PNG) source citations that vindicate them. I think the simple addition of a sentence or even a phrase with regard to this could easily rectify the deficiency in G4. The way G4 is presently worded, I strongly suspect that anyone who has created and had deleted an unsourced article would a) have a great deal of trepidation about restoring the article even after extensive source citation researching, and b) would actually be discouraged from even bothering to attempt such sourcing. G4 presently reads like a "once deleted, always deleted no matter what" policy, even though it surely is not actually intended to be one. — SMcCandlish [talk] [contrib] 08:53, 31 January 2007 (UTC)

G4 generally applies to articles deleted via WP:AFD (indeed, PRODs if recreated need some other process, and anything speediable under some other criterion may not be G4'ed, although generally the same criterion applies again to recreations). So if an article is G4'ed, there already exists a discussion about what's wrong with it. If you can fix those concerns so that it's no longer substantially similar to the article AFD deemed inappropriate, you're in the clear. >Radiant< 12:25, 31 January 2007 (UTC)
I guess my point is that this is not at all clear from the text of G4. If I've been scratching my head (as to policy reasoning, not actual practice, which is readily observable) about this for a year or more, surely others are also. NB: I wasn't attempting to draw a distinction between Prod'ed, SD'd or AfD'd articles in the above; just saying that G4 appears to imply that once an article has been deleted once (for any reason, by any process) it will be SD'd, no matter what, if it comes back unless it is an all-new article. After a while around here I did of course realize this is not actually true, but a) it isn't spelled out at G4 (or anywhere else I'm aware of) why and under what circumstances this is so, and b) it shouldn't mislead anyone into such a misconception in the first place. I'm just saying it needs a clarity twiddle. You'll probably find it uncharacteristic of me, heh, but I'm not in any way disputing the actual intent/purpose/legitimacy/tone/etc. of the section, just this very narrow inspecificity of its wording. :-) Or, in case I'm not getting the exact point across: It's not clear in G4 that if an article is AfDd for verifiability/sourcing reasons (under WP:V, WP:N, etc.), that G4 won't come into force if the article is restored with new citations that fix those issues but is otherwise word-for-word. Hope the concern is clearer now. I think if some phrase were added, like "absent improvements that address the reason(s) the article was originally deleted", that would totally solve the problem. — SMcCandlish [talk] [contrib] 12:52, 31 January 2007 (UTC)
Interesting point. I've reworded G4 to read "Substantially identical means that the new article fails to address the reason for which it was deleted. " Would that help? Please copyed. The beauty of it is that it obviates the need for the "except CSD" exception. >Radiant< 13:34, 31 January 2007 (UTC)
Perfecto! — SMcCandlish [talk] [contrib] 20:32, 31 January 2007 (UTC)
For what it's worth, I've always interpretted G4 more narrowly... that G4 only applies when it's clear that someone used a fair bit of text from the deleted version (eg. saving their own copy before deletion, copy-n-pasting from Google cache, ...). "Fails to address the reason" requires the deleting admin to make a determination of whether the XfD/etc arguments mostly apply to the new article... which admins may be capable of doing, but it's still somewhat of a subjective decision. An alternative for when an article has all-new text but still falls under the previous deletion discussion is to just prod it, noting the previous deletion. --Interiot 18:25, 31 January 2007 (UTC)
I'm seeing G4 and G8 be interpreted very broadly myself. Mathiastck 14:46, 1 February 2007 (UTC)
Noted, but the word "interpreted" is almost a red flag itself. After quite a while I figured all this stuff out on my own, but the immediately-previously italicized part is/was the issue — it shouldn't be a long-term excercise in wikigeeking, but immediately plain. The "[if the] article has all-new text but still falls under the previous deletion discussion" part is noteworthy but pretty much the opposite of the case I was raising (if I interpreted you correctly). It certainly should be addressed, it but didn't get to what I was after, as it were. "(R)equires the deleting admin to make a determination of whether the XfD/etc arguments mostly apply to the new article" is what I was getting at (or, rather, more broadly, that the ""rejuvenated" suspect article should be examined (and policy/guidelines should cause it to be examined ⇐ the real point!), by said admin, or SOME admin, or the community in AfD or whatever, just as long as the examination, rather than reflexive and unexamined deletion, is in policy/guidelines somewhere. Radiant's edit ack'd above pretty much just nailed it. Topic/issue resolved as far as I'm concerned.
[Argh, THREE edit conflicts back-to-back. Mangojuice, Blis, and Badlydrawnjeff, feel free to refactor the indent levels and whatnot if your comments aren't where they should be. Let's see if I can actually submit this now...]
SMcCandlish [talk] [contrib] 20:32, 31 January 2007 (UTC)
Good catch, SMcCandlish. It wasn't all that long ago that we had to clarify that G4 did not, in fact, apply to speedied/prodded articles. There's a fine line between circumventing Deletion Review and being bold, and I think this gets us closer to where we should be. I'll also mention this at DRV's talk page to get a broader input. -- nae'blis 20:16, 31 January 2007 (UTC)
Glad my nagging brain-itch has led to useful results then. :-) — SMcCandlish [talk] [contrib] 20:32, 31 January 2007 (UTC)

G4 for redirects:not resolved

Can we deal with the problem that G4 always applies to deleted redirects; once there has been a "consensus" at RfD, even if it's two editors, the redirect can never be recreated? Septentrionalis PMAnderson 02:49, 2 February 2007 (UTC)

I would love it if we could, but the only solution for now seems to be to take it to DRV or be bold and hope that consensus has changed. -- nae'blis 16:17, 2 February 2007 (UTC)

"Attempt" to address

I've changed the wording to "Fails to attempt to address the reason for deletion" to make it clear that the question isn't whether or not all concerns have been met, but whether an attempt has been made. If there has been an attempt, it should at least be evaluated at AfD. Mangojuicetalk 19:51, 31 January 2007 (UTC)

That seems more logical. I support this language. --badlydrawnjeff talk 20:13, 31 January 2007 (UTC)
  • No, if it has already been discussed on AFD there's no need to discuss it again unless there is an actual improvement, as opposed to an attempted improvement. Any recreation that's not fully identical can be argued to be an attempted improvement. >Radiant< 12:18, 1 February 2007 (UTC)
    • Rereading the draft, I don't like that "substantially identical" is defined to be something very different from "identical." The current wording implies that any new article that probably meets the same old reason should be speedy deleted, which is a significant broadening of the criterion. (And I don't agree with it -- in the interests of fairness and not biting the newbies, I think if someone writes an article that wouldn't be speedy deleted in the first place, we can at least do them the courtesy of having a debate, or at least using PROD.) I would like to remove the sentence we're discussing and change the first sentence to: "A substantially identical copy, by any title, of a page that was deleted via Articles for deletion or another XfD process, and which clearly does not address the reason(s) for deletion." This solves my concern, and I think it addresses Radiant's as well. I like having the "clearly" qualifier, though, because admins should not be making the call if it bears discussion. Mangojuicetalk 15:14, 1 February 2007 (UTC)
"clearly" is much better than "attempts". -- nae'blis 15:17, 1 February 2007 (UTC)
Yeah, we can't use "attempts to", or it doesn't work. If I attempt to address a sourcing problem by adding a bunch of forum links, then why does it need another AfD discussion, where it will have the same problem? -Amark moo! 21:05, 3 February 2007 (UTC)

Looking back

So, is this working? --badlydrawnjeff talk 01:50, 7 February 2007 (UTC)

New 'not policy' CSD for templates

I'd like to add a 'T2' (or T3 / T4 if we are 'reserving' historical numbers) criteria to allow speedy deletion of templates which purport to enforce a policy or guideline where no such exists. Having essays in Wikipedia or User space which don't have the weight of policy is fine, but when you start plastering templates on articles and talk pages telling people what to do, there needs to be consensus acceptance of it. I'm thinking of things like:

  1. Templates saying that 'this article contains offensive material'... despite consensus that we don't use such disclaimers.
  2. Templates saying that users are not allowed to remove warnings from their talk page... despite there being no policy to that effect.
  3. Templates saying that NPOV violations (as perceived by any user) are vandalism... despite policy clearly indicating that they aren't.

And so forth. Templates like these give a false impression of Wikipedia's policies and practices as they seem 'official' while actually being incorrect. Any cleanup/warning/banner template which can't be linked back to a specific approved policy or guideline ought to be CSD material. --CBD 13:03, 1 February 2007 (UTC)

  • Reasonable, but note that the first two of those would also fall under the "recreation" clause (G4) because they've been discussed and deleted on TFD several times. >Radiant< 13:10, 1 February 2007 (UTC)
    Right. Actually, all three are examples of things I have seen previously. I'm just positing that there are others like this out there and/or will be more in the future... and I think it ought to be possible to zap them without the big debate. A template with 'approval by a solid minority' might be kept as 'being found useful by some portion of the community'... but no way do we adopt new policies unless they have solid approval from most users. There are vastly different standards for the two things... so templates which express a policy/guideline view should require that policy/guideline to actually exist and be approved by the community (or enacted by WikiMedia). --CBD 13:18, 1 February 2007 (UTC)
      • You're right about that one. I recall this issue about the "removing talk page warnings" issue, where we had the problem that you need a consensus in favor to create that as policy, and yet a consensus in oppositino to get rid of the templates. In the end they were speedied anyway. So yes, anything that sounds official but really isn't should be removed, reworded or deleted. >Radiant< 13:30, 1 February 2007 (UTC)
  • While I certainly hold such templates in contempt, I'm going to claim that the ones that don't fall under G4 are sufficiently rare that they fail to meet "arise frequently" and so can be handled through TfD. Deco 14:17, 1 February 2007 (UTC)
    • Perhaps (if there are more than three examples of this) we should create a Wikipedia:List of undesirable templates that explains the kind of templates that have been repeatedly discussed-and-deleted already. In fact, WP:NDT already does that for one case. >Radiant< 14:27, 1 February 2007 (UTC)
  • I would agree that the examples CBD describes above are all deletable but I am very uncomfortable with the idea of making them speedy-deletable. These are all issues which deserve discussion and debate. Some templates do have basis in policy and not all admins know about every conceivable policy page we have. Blank the template during the discussion period if you're worried about it being misused. but don't speedy them. Rossami (talk) 14:44, 1 February 2007 (UTC)
    • 100% agree with Rossami here. If the templates appear to change policy, change the templates, and nominate them for TfD. Mangojuicetalk 15:01, 1 February 2007 (UTC)
Agree with Rossami. Reword them into compliance if the wording is a problem, nominate them for TFD or utilize G4 if it continues. There's no bright-line for defining what tries to enforce/not enforce policy that requires a separate speedy category, in my view. -- nae'blis 16:12, 2 February 2007 (UTC)

Proposal of new criteria for speedy template deletion

I'm a bit noobish in Wikipedia so bear with me if I don't explain well, or if I explain too much ^_^ I noticed a minor problem in a page related to Monkey Island (series) and its navbar. While trying to fix it, I noticed that the template of the navbar had two names, one was Template:Monkey Island (already deleted), and the other was Template:Monkey Island series. I noticed that the template was moved to the new name to match with naming conventions, and the old name was left there as a redirect.

However, during my clean-up, all the articles which used the template were changed to use the new, more correct name. I checked the "what links here" page to ensure that no pages used the template redirection. Since it wasn't an article redirection, but a template redirection, and it wasn't used anymore, the template was rendered useless. Any new pages which want to use the navbar will use the new name, because all pages use it, and any template search for it will come up with the new name rather than the old. It's not like articles where you have redirection pages due to probable different spellings or alias of the article name, templates only have (in the majority of cases) one name, and I think that's how it should be.

Since the page was useless, I wanted to tag it for deletion, but it didn't really match any speedy deletion criteria. I tagged it with a generic template, gave my reason, and came here to explain all this thing. I'm guessing that every template move will leave behind a template redirection which should be cleaned, and once not used, it should be deleted. While it is not critical, this process will help to maintain Wikipedia a little bit more clean, isn't it?

I also wonder if there are any bots which will look at template redirection pages and delete them if they aren't used anymore. And btw, how do I access a list of template redirections?

IMHO the exact text of the new criteria, and where should it be put, should be done by a more experienced wikipedian, but I'd nominate templates, and not redirection, as the right place for it. This is because this problem only comes with templates, not with redirections in general. I hope I made sense with all this blurb! --Pi (π) 14:50, 1 February 2007 (UTC)

This kind of thing comes up very frequently. In my view, there would be no harm keeping Template:Monkey Island as a redirect, since it's a plausible mistake someone could make in an article. There is a process for this: WP:RFD, which is much like WP:AFD but for redirects -- this kind of thing bears discussing at least, it's not so clear-cut as to be worth adding a speedy deletion criterion. Mangojuicetalk 15:05, 1 February 2007 (UTC)
Well, I don't think that RFD really covers the issue I've mentioned. There must be a reason for redirects. For article redirects, it's a plausible mistake. This is not a reason for templates, only very used templates have alternative names, I guess many are kept for historical reasons. When you use a redirect, you already know the correct name, either from a search, or from looking at the source of somewhere where the template is used. Template redirects are used for moved templates, I'm guessing *until* all the places where it's used have the new name. But once the redirect is not used anywhere, then there is not a reason for the redirect to be kept, unless it's kept for historical reasons, as I've mentioned. IMHO, these cases fall into speedy deletion, as there's not much to discuss about an unused redirection (as long as no page uses the redirection, of course). I tagged the redirection I put as an example, and it was promptly deleted. I'm just proposing an specific criteria to be used with this class of unused template redirects. --Pi (π) 16:03, 1 February 2007 (UTC)
I would recommend against deletion in this case. Redirects do much more than just to support the search engine. Redirects provide history and tracability to our actions. What if, for example, someone reverts one of the pages using the template to a version which had the old template? If substituted, you have a reference to a template which no longer exists (and no easy way to find the replacement template). If transcluded, you end up with a dead link. RfD is well suited to working out these issues since you are discussing the deletion of the redirect, no longer discussing the deletion of the first template. Rossami (talk) 16:16, 1 February 2007 (UTC)
That's a good point, didn't notice the problem with reverts. As I said, I'm a bit new in Wikipedia and maybe I'm lacking that "global scope". Thanks for paying attention, tho. --Pi (π) 19:00, 1 February 2007 (UTC)

Speedable image question

Not-so-hypothetical situation: an article is speedily deleted (twice, no less!) under G11. Its images, however, are still around (and going through IfD). Is there an actual criterion for speedily deleting them as well? I would consider it to, possibly, be G6 (Housekeeping), but it isn't specifically mentioned.

And, if there isn't a criteria, shouldn't there be one?

For those that are curious, the article is The Dodos, and the images are Image:Dodos.jpg and Image:Dodoguitar.jpg. EVula // talk // // 05:01, 2 February 2007 (UTC)

I've tagged images left over from speedy articles as housekeeping in the past, and that seems to have worked. Although, if I recall correctly, most of those images were obviously of no value and were either something silly or plain vandalism. -- Ned Scott 05:13, 2 February 2007 (UTC)
Hah, I was just looking for you in #ev3 to get your opinion (and, eerily enough, you popped in just as I was writing that...). I'll wait until a few others weigh in, but if G6 is being used for such deletions, we should probably expand it's "official" scope. EVula // talk // // 05:14, 2 February 2007 (UTC)
I agree with deleting images uploaded only for speediable articles, but it should probably be a new criterion, rather than a G6 subcriterion. The wording would need to be thought about carefully. --ais523 13:17, 2 February 2007 (UTC)
I don't see a reason to bother. Images like this end up qualifying for CSD I4 or I5 before too long. Mangojuicetalk 14:15, 2 February 2007 (UTC)
My understanding is that, for I4, I5, and I6, it needs to be tagged in a particular way for seven days. I'd rather live up to "speedy" when dealing with these. :) EVula // talk // // 15:45, 2 February 2007 (UTC)
Yet another reason to call the process "streamlined deletion" or something along those lines. -- nae'blis 16:16, 2 February 2007 (UTC)
That's a serious point (and one that's been confusing issues at WT:CSDUA, among other places); 'speedy' can be a confusing name for many of the slower CSD processes. I like your 'streamlined' suggestion; any further thoughts? --ais523 18:08, 2 February 2007 (UTC)
I also thought "straightforward" or "simplified" could be accurate, if we wanted to keep the longtime CSD acronym. -- nae'blis 18:26, 2 February 2007 (UTC)
Have you noticed how the 'C' changes its meaning from page to page? It's 'Criteria' here, but 'Candidates' in CAT:CSD. --ais523 18:40, 2 February 2007 (UTC)


I've noticed {{db-afd}} is used when An admin has closed an Articles for deletion debate as a "delete" but not deleted the article. However, the template has been changed by Ais523 to generalise across all deletion debates including AfD. I'm proposing to move the template to {{db-xfd}} and add the following description according to the change: An admin has closed a deletion debate as a "delete" but not deleted the article. Thoughts? Michaelas10 (Talk) 14:49, 2 February 2007 (UTC)

'but not deleted the page', surely, or my changes to make it applicable to other deletion debates will have been in vain. --ais523 16:12, 2 February 2007 (UTC)
I was about to say s/article/page, but other than that it looks like a uncontroversial change. I like the way that template determines what XfD it's talking about by the namespace it's substituted transcluded in... -- nae'blis 16:15, 2 February 2007 (UTC)
It works even without subst, as generally speaking db-templates aren't substed. --ais523 17:16, 2 February 2007 (UTC)
Yarrr. I meant transcluded; I've had a coffee break and need to be retrained been out of the loop for a couple of days...
I went ahead and changed it. Michaelas10 (Talk) 17:29, 2 February 2007 (UTC)

Change to I8

The current rule for "speedy" deletions of images moved to Commons was adopted in July 2006 after the proposer got two supports - the only comments received. One of the supports said that this would be better than the 5-10 day wait of going through IFD. However the rule requires a one week wait - hardly shorter than the 5 days mentioned. The tags for the process - Template:NowCommonsThis makes no mention of a one-week wait but, in fact, says images may be deleted "immediately after satisfying these conditions (CSD I8). The words "these conditions" are linked to Category:Images with the same name on Wikimedia Commons which contains instructions for admins which also contain no mention of the one week wait. I would suggest that the one week wait be removed as it serves little purpose and had little support when instituted. Admins already need to check the images, revisions, licensing and links before deletion - the addition of a one week wait seems needless. Rmhermen 17:02, 2 February 2007 (UTC)

Agreed, there's no reason for a delay if all instances have been corrected. EVula // talk // // 17:10, 2 February 2007 (UTC)
I too would like to remove the provision The image has been marked with Template:NowCommons for at least one week. if all other provisions have been met. Any objections? --Golden Wattle talk 21:53, 1 March 2007 (UTC)
Disagree, the one week wait provides one the opportunity to object to the deletion. And I routinely object since moving images to commons prevents one from using the Watchlist to identify vandalism and talk page comments. Since many of the images I have created are high profile, the descriptions are collectively vandalized several times per week (9 times in the last week, for example). Dragons flight 22:06, 1 March 2007 (UTC)
Could we have a balancing provision, if the nomination for nowcommons was made by the original uploader then no delay necessary? That would then allow uploaders to object in cases similar to User:Dragons flight whre others have moved the image against his better judgment or wishes, but where the uploader herself / himself has moved the image deletion can power ahead. It would mean an admin would not clear out the category earlier than 1 week but a direct contact between admin and user could be legitmately actioned by the admin in accordance with the criterion.--Golden Wattle talk 20:36, 3 March 2007 (UTC)
  • Given no response in over a week, I propose to change the criterion to reflect uploader's wishes--Golden Wattle talk 19:52, 11 March 2007 (UTC)
    The change you made to I8, i.e. appending "except where the uploader has moved the image and tagged it", is confusingly vague. This addition implies images moved and tagged by the original uploader are not eligible at all for an I8 speedy delete, as opposed to not needing to wait a week. I think "...for at least one week. Waiting one week is not necessary if it was the uploader who moved the image and marked it." would be clearer. —RP88 20:28, 11 March 2007 (UTC)
thanks for clarifying, I will amend as suggested.--Golden Wattle talk 21:07, 11 March 2007 (UTC)

G8 needs revision

Now that protected deleted pages are being protected as actually deleted, there's going to be talkpages for redlinks on them. These shouldn't be speedied, though, because they're the specified place to gather information to request recreation. Night Gyr (talk/Oy) 22:46, 3 February 2007 (UTC)

I strongly agree that G8 needs revision. As I recall a couple months ago when I looked at it, it was weaker. Did something happen justifying strengthening it? I feel much to much of a noob here to edit it myself, but I would like to open a larger discussion about the various reasons to delete talk pages. Mathiastck 11:24, 4 February 2007 (UTC)
I added a note on protected deleted pages, which should be uncontroversial as it is no change from existing practice. Any other revisions should probably be discussed here first. Christopher Parham (talk) 05:30, 5 February 2007 (UTC)
I would rather see (temporarily) orphaned talk pages allowed as the place to gather information on very-soon-to-come articles, but the template that supported such a concept got shot down at TFD. I'm not sure I see the point in allowing an exception when the page is protected, honestly. If anything, those pages (Talk:GNAA, Talk:Encyclopedia Dramatica) have been more likely to cause problems than Talk:Chiba-Ken. -- nae'blis 16:27, 6 February 2007 (UTC)

User warnings

Right now, most of the speedy deletion templates say, "Please consider placing ... on the User Talk page of the author." When I patrol CAT:CSD, I generally find that the warning template has been applied only around half of the time.

There are obvious reasons for the warning. If the article is merely missing some pro forma requirement, for example, the person is notable, but the article itself does not assert the significance, or the topic is important, but there isn't enough context and it fails WP:HOLE, a warning allows the contributor to fix the problem and we get another needed article. If a contribution was made in good faith by someone who just wasn't aware of our policies, the warning makes them aware. I have had multiple occasions where I have applied the warning and the contributor has thanked me for the explanation and said that they did not realize that Wikipedia was not interested in bios of averge high school students. In this case, using the warning may help us to keep a potential contributor and it serves to answer their question of, "what the heck happened to my article". The third case is a bad faith contribution and the warning allows an admin working WP:AIV to see that the person needs to be blocked.

Would anyone see a problem with changing "please consider placing" to "IMPORTANT: please place"? That may get the message across a bit better that these warning templates really do matter. --BigDT 13:35, 5 February 2007 (UTC)

I agree. I believe you should always place the tag on the user's talk page that way he/she knows his article is up for deletion.↔NMajdantalk 14:17, 5 February 2007 (UTC)
I disagree. Yes, people should place warnings, but I would rather have an editor who doesn't place warnings continue to do so and mark pages for deletion than not, because someone made an issue of the user warnings. I think we already have the appropriate level of emphasis on warnings. Mangojuicetalk 16:00, 5 February 2007 (UTC)
Concur with Mango. Interested parties are encouraged to use watchlists, that's what they're there for. Forcing people to use warnings is overly bureaucratic, and using terms like "IMPORTANT" in all caps give the impression that it's mandatory. >Radiant< 16:06, 5 February 2007 (UTC)
Unfortunately in the case of speedy deletions, watchlists are about useless, unless you happen to check it between the addition of the tag and the actual deletion. But that's probably better addressed by technical means. Christopher Parham (talk) 22:53, 5 February 2007 (UTC)
Furthermore, some admins will speedy delete a page immedately, without letting it be on the list at all. This may be justified in exceptional cases, and I can see the convenience, but it means that no other person gets a chance to review. I have followed up 1 or 2 and in each case it was certainly justified, but it doesn't mean all of them always are. DGG 01:48, 6 February 2007 (UTC)
"Justified in exceptional cases"? The speedy deletion tags are just there for convenience, to allow non-admins to flag content that they think meets the criteria. They are in no way a necessary prerequisite for speedy deletion. --bainer (talk) 22:58, 6 February 2007 (UTC)
I'd much rather see a technical solution whereby deletion log entries show up on your watchlist, for many of the above reasons. Creator abandonment is too high for me to expect that user warnings being mandatory will have much effect. I agree that in some of the borderline cases (A7, G11) they could be used more often, when paradoxically it seems I see them more often with straightforward vandalism/nonsense/attack pages. But that might just be my selection bias based on which CSD candidates I tend to evaluate. -- nae'blis 16:23, 6 February 2007 (UTC)

removing tags

The template says if you are the author of the article do not remove the template, which certainly would normally be interpreted to say that anyone else can. (Hangon is provided for the author as an alternative). In fact, i do remove such tags two or three times a week, when I think them incorrectly used for cases that do not fit clearly in the rules.
I have just been challenged on this by another editor, and I would like comments. It would appear to me inequitable that anyone can place, but only an admin remove. It makes sense that the author himself as an interested party shouldn't do so.
Some of the discussions above about inappropriate uses of speedy make the need for this particularly appropriate. DGG 21:36, 6 February 2007 (UTC)

While I certainly can understand and appreciate DGG's perspective, my concern is that allowing a non-admin to remove the tag could be an invitation to Wheel War. If I understand policy correctly, wheel warring is considered an extremely bad thing. Regardless of my beliefs, what is the policy on removing the tags? Are they sacrosanct or not? --SilverhandTalk 21:41, 6 February 2007 (UTC)
There are lots of invalid taggings going on, so I see no problem with someone who isn't an admin removing tags that are clearly wrong. However, I would ask that only people who know what they're doing do this (or the article may just be retagged), and furthermore that anyone removing such tags take care not to let pages fall through the cracks -- in other words, if the page should be deleted, just not speedily, enter it into another deletion process. Non-admins should certainly not de-propose pages for speedy deletion if those deletion requests were valid. Doing this a lot would be disruptive. Mangojuicetalk 22:46, 6 February 2007 (UTC)
Anyone may remove such a tag if they disagree with the tagger's assessment of the article. If that happens, it's time for articles for deletion (or whatever other "X for deletion" is appropriate, or another system like Wikipedia:Copyright problems if you think the problem is that it's a copyvio, and so on).
Mangojuice makes a good observation, that people removing tags indiscriminately can be just as annoying for everyone as people adding them indiscriminately (although remember that admins can delete something under the criteria at any time, the tags are just for convenience). To this I would say that people removing tags should seriously consider sending the page straight to the appropriate "X for deletion", rather than simply removing the tags. That's likely to cause less friction. --bainer (talk) 22:55, 6 February 2007 (UTC)
Before I was an admin, a user questioned my removal of a CSD tag, saying that they thought only admins were allowed to do that. So the idea is out there, but I'm fairly sure it's in the minority, as this was one case out of hundreds that I removed. Anyone may remove the tag just as anyone may place the tag, but it should be done for valid reasons in either direction. -- nae'blis 00:12, 7 February 2007 (UTC)
I think the author and anyone at all should be able to remove the tag if it is unjustified. The suggestion that the author should use {{hangon}} instead implies that the author agrees that the article requires correction and they're going about fixing it. They may think it's a good article just the way it is. The person who added the tag may have simply been wrong, or even malicious. I suggest that no restrictions be placed on what users may remove a tag. Deco 00:46, 7 February 2007 (UTC)
The suggestion to send to prod or AfD is a good one and I will follow it more consistently -- & I will make sure to notify the author, because usually the author has not been warned in the speedy. DGG 02:00, 7 February 2007 (UTC)
I agree. This seems the reasonable path to resolving a situation of this nature. --SilverhandTalk 16:05, 7 February 2007 (UTC)
Then, shouldn't the text be changed? Insteaf of if you are the author of the article do not remove the template, put in small letter in the bottom of the template something in the lines of If you think this article doesn't meet the speedy deletion criteria, please put the article in the appropiate X for deletion category. Maybe something about this should be specifically explained in the CSD page. Pi (π) 09:37, 8 February 2007 (UTC)
There should be no requirement to send a page to AfD/PROD, there are plenty of scenarios where that would not be the best thing. I haven't noticed a big problem with this, it's more that if a user was denying speedies and not doing anything else, I might suggest they consider redirecting the process rather than ending it. In any case, those who care about an article's deletion can always just watchlist the article. Mangojuicetalk 18:48, 8 February 2007 (UTC)

Adding template on the project page, next to each criteria?

Right now, it looks like one has to sort of guess, once a speedy reason has been selected, what the template is for that reason. Would anyone mind if I added (or someone else did - feel free) the template (nowikied) next to each criteria, for that criteria? -- John Broughton (♫♫) 14:56, 10 February 2007 (UTC)

Definitely don't use nowiki. Use {{tl}} if you do anything. But there is a separate table near the bottom, at Wikipedia:Criteria for speedy deletion #Deletion templates because for several of the criteria there are multiple relevant templates. GRBerry 04:49, 11 February 2007 (UTC)
Okay, I didn't see the table, and since I didn't, I'm going to assume (hopefully reasonably) that others have missed it as well. So I added a sentence that mentions it; if that's acceptable to others, then I think the matter is closed. -- John Broughton (♫♫) 02:17, 13 February 2007 (UTC)


I am seeing articles being speedied with explanations along the lines of complaint received on m:OTRS. A recent example would be McLouth High School I've read some of the OTRS material, and don't see anything in there which justified speedy deletion outside of the wp:csd guidelines. What's going on? Is there some new policy which hasn't been added to wp:csd yet? Or are these speedy deleting admins simply acting outside policy? -- RoySmith (talk) 15:14, 10 February 2007 (UTC)

It is more like a pseudo-WP:OFFICE action. Also, while the deleting admin might give a non-descriptive deletion summary for various reasons, these pages are usually A7 material or Blatant attacks or Advertising or Copyright infringement. In any event, an article is not deleted merely because of a "complaint" but because there is a complaint in combination with serious other problems with the article. —Centrxtalk • 16:19, 10 February 2007 (UTC)

I asked the deleter of the article cited above if it was an office action, and was told it was not (um, looking back, I realize I had asked him that quesiton about a previous deletion of his). In several instances I looked at, I could not find any WP:CSD that applied, and in fact one of the articles, after being un-deleted, survived AfD. While I fully understand the need to protect Wikipedia against legal issues, I am concerned that OTRS is creating a sub-caste of admins who are above policy, and that out-of-process deletion is being used in favor of less harsh remedies such as fixing the vandalism and sprotecting the page. If we want admins to be able to speedy articles because of complaints to OTRS, let's at least ammend WP:CSD to say that. -- RoySmith (talk) 16:52, 10 February 2007 (UTC)

Hm. First of all, this page is never going to cover every single possible reason something might get deleted. We, and by we I mean the community, are going to, every so often, encounter some situation where deletion is the best option, but which hasn't been added to this page. That's one of the reasons why we have pages like m:Avoid instruction creep and Wikipedia:Ignore all rules; not because we like chaos, but out of a recognition that big experiments, like giving away a free encyclopedia, demand some flexibility. So don't depend on any policy page, including this one, covering every aspect of every potential situation. Falling in love with process on a wiki will break your heart. As to this specific concern, the Foundation Communications Committee chooses people to help out at OTRS based on a combination of good judgement, courtesy, commitment to the project, and a measure of trust from the local projects. I wouldn't call OTRS volunteers a "separate class of admin", or call OTRS actions "pseudo-WP:OFFICE"; both analogies are a little off. Instead, I'd like to ask editors concerned with an OTRS action to feel free to ask questions about it, and accept that they may not get a satisfying response sometimes. If you want to work on an article about that school, go ask User_talk:David.Monniaux about it, and trust his judgement. All of that said, if the only way you can bring yourself to accept that there are things that we, meaning local editors, don't get to have all the information about or get to decide as if we lived in a world without consequences, is to think of the OTRS team as another class of admin/pseudo-OFFICE, then think of them that way. Jkelly 23:08, 10 February 2007 (UTC)


Does G5 apply for pages which have been edited by other editors after the banned user has created them? --- RockMFR 05:55, 12 February 2007 (UTC)

Apply the "cut your nose off" test: are they useful articles that are worth keeping? If they are, wouldn't deleting them just be cutting off our noses to spite our faces? If the pages have been improved by other editors, then that's a good indication that they are worth keeping. On the other hand, if the pages are not worth keeping, the fact that other editors may have edited the page is not a bar to deletion. --bainer (talk) 06:16, 12 February 2007 (UTC)

Biscuit AFD's

Anyone know where these were?? --sunstar nettalk 11:17, 12 February 2007 (UTC)

I can't recall the exact time frame, but they were done by User:Improv sometime last fall. Most of the discussion is over at the deletion review, I think. --badlydrawnjeff talk 12:29, 12 February 2007 (UTC)

Administrator Review

Does anybody know if there is a template that lets other administrators know that an article for speedy deletion is currently being reviewed by another administrator? There have been several time where I go do a quick google search or try to start discourse with the creating editor and by the time i do, the article has already been deleted? -- Chrislk02 (Chris Kreider) 20:56, 12 February 2007 (UTC)

Well, you could probably adapt the {{hangon}} tag for that. Maybe make a {{hangon-admin}} that says specifically an admin is reviewing it. EVula // talk // // 21:03, 12 February 2007 (UTC)
That is what I was thinking about (just wanted to make sure it did not exist first). Would there be a more apporporiate talk page to bring this up on or do you think it would be ok to just go ahead and create it? I would be interested in a little but of community input on this-- Chrislk02 (Chris Kreider) 21:06, 12 February 2007 (UTC)
WP:AN, perhaps. *shrug* EVula // talk // // 21:08, 12 February 2007 (UTC)
I started a thread here at WP:AN. -- Chrislk02 (Chris Kreider) 21:12, 12 February 2007 (UTC)
To be honest, if a CSD criterion requires an adminstrator so much time to decide that a hangon-admin tag is needed, something is wrong. CSD criteria should be clear and objective and shouldn't take more than a minute to evaluate. It's Criteria for Speedy Deletion. Mangojuicetalk 03:28, 13 February 2007 (UTC)
Agreed. If you have to think about it that long, the CSD tag was very likely wrong in the first place. Just take the tag off while you think about whether one of the other criteria might apply. Rossami (talk) 12:35, 13 February 2007 (UTC)
In fact, {{prod}} might be the best, because that gives 5 days, and it allows for a verbose explanation of what the remaining issues might be (the reason that you're not sure speedy deletion is correct). I have read that giving reasons on a speedy template is frowned upon (They want one of the categories marked, and that's it). EdJohnston 20:38, 13 February 2007 (UTC)
It may truly need to be deleted immediatly. I.E. possible slanderous/personal attacks about people, copy vios or an article on "the best football player ever" who just happens to be a nn high school player. If every one of those articles that a prelim google serach helps with got prodded instead, prod would become massivley backlogged with things that could have been deleted. -- Chrislk02 (Chris Kreider) 20:42, 13 February 2007 (UTC)
I'm not sure I understand the problem here. An admin acting on a CSD nomination can see the deleted revisions even after action has been taken by someone else (although it does take a few more clicks, annoyingly). If you disagree with the speedy/find evidence that invalidates the criterion cited in the deletion log, undelete and/or go talk to the deleting admin. Non-admins can take it to Deletion Review... am I missing anything here? I frequently find pages deleted out from under me after I start researching them, which I take as a good sign that some other trusted user is on the job and I can stop my Googling. I'm worried about {{hangon-admin}} being left abandoned for too long as an admin goes to lunch, falls asleep, etc... -- nae'blis 20:31, 16 February 2007 (UTC)


The criteria currently applies to "redirects to the Talk:, User: or User_talk: space from the main article space", but I'm proposing it should also apply to Wikipedia namespace. I've previously seen several redirects from this type being removed from the mainspace without any deletion debate (e.g. requests for adminship and requested moved). As a part of the change, I also suggest getting rid of {{R to wikipedia namespace}}. Michaelas10 (Talk) 16:27, 13 February 2007 (UTC)

No it shouldn't. At least, not until you get Wikiproject deleted. -Amarkov moo! 16:28, 13 February 2007 (UTC)
It's actually very strange it was kept following this overwhelming consensus. WP:RDR itself states "cross-space redirect out of article space, such as one pointing into the User or Wikipedia namespace" as a deletion criteria. Michaelas10 (Talk) 16:57, 13 February 2007 (UTC)
You know, I really, really hate it when people selectively quote our policies and ignore the parts they don't like. Keep reading on that same section and you will see the section which says "However, avoid deleting such redirects if:" and goes on to list the many exceptions to the bulleted list you quoted from. Many of the redirects to the Wikipedia-space satisfy one or more of the criteria in the exceptions list.
There is not even consensus yet that cross-namespace redirects are inherently bad. (The last time it was proposed, the proposal failed.) There is certainly no consensus that they are so inherently bad that we must speedy-delete them all. Rossami (talk) 17:58, 13 February 2007 (UTC)

There is no consensus to delete all redirects to the Wikipedia namespace. Speedy criteria should only be proposed for cases where the appropriate non-speedy process always produces the result of the proposed speedy criterion. Note that there is no technical difference between a shortcut like WP:CSD and the redirect Wikiproject. Kusma (討論) 18:03, 13 February 2007 (UTC)

I understand that, although the chance that a person will check Wikiproject trying to find a proper mainspace article is a lot higher than the personal checking WP:CSD for the same purpose. I expect there should be minimally a proper mention of this in a policy. It's very strange that Articles for deletion was removed but requests for comment is up for nearly three years. Michaelas10 (Talk) 18:56, 13 February 2007 (UTC)
Speedy deletion and TFD frequently amount to the same thing, because there are some closers at TFD that will speedy-close CNRs hours after they have been opened. The zealotry surrounding CNRs is sort of baffling to me, since we're an in-progress work that can clean up after itself if/when it ever becomes important to remove such "backstage" kludges. -- nae'blis 20:33, 16 February 2007 (UTC)

A7 is getting out of hand

Criteria A7 is being used to speedy delete numerous things that at the very least should be prod'd instead. Things that make marginal claims of notability are still being deleted for having none. I've seen numerous instances of articles whose subjects survive an AfD after earlier being deleted by criteria A7. I'm not an "inclusionist" just worried A7 is being misused. i kan reed 14:09, 15 February 2007 (UTC)

Evidence? Chris cheese whine 14:25, 15 February 2007 (UTC)
  • [edit conflict] This has been discussed to death already. Last time this was raised someone came along with actual evidence (as opposed to here, where there is none) and most reasonable people agreed that A7 is working well enough. Hundreds of articles get deleted under A7 every day. so you need to show that a good percentage of those A7s are wrong. Unfortunately so far all people have been able to provide is a few borderline cases here and there. -- Steel 14:28, 15 February 2007 (UTC)
    • Right now, the middle ground appears to be one in four are wrong. --badlydrawnjeff talk 14:40, 15 February 2007 (UTC)
      • That's odd, I seem to remember the last time someone did a reasonable investigation into it they put it at less than one in forty. Chris cheese whine 14:42, 15 February 2007 (UTC)
        • That's why I said middle ground. He claimed a couple in forty, my analysis was a stricter reading - anything not unambiguous. A larger-scale investigation has not been completed, last I checked. My rating in particular had a 40%+ fail rate - that is, any article that was speedied as an A7 but was not rated a "5", unambiguous. The middle ground is roughly 20-25%. --badlydrawnjeff talk 14:46, 15 February 2007 (UTC)
  • Well, let me add my perspective, this is not exactly scientific, but at least it's not pure conjecture. I did CSD last night and deleted 50 pages, and removed CSD tags from 7. 5 that I removed were A7. At any rate, this seems to suggest about 1 in 7 pages are tagged incorrectly. And that ratio seems about right to me. Although, this is all highly subjective, because we all have different standards for what A7 is, despite attempts to make it not so debatable. Take Jessica Rose Paetsch, one of the articles I took the tag off last night, 13 year old bronze medalist at some ice skating competition... some people might say that's not really much of a claim of importance. But nevertheless I took the tag off. By the way, "less than" 1 in 40 is an absurd number to me... that would suggest you hardly ever encounter incorrectly tagged articles while doing CSD... that's not the case. --W.marsh 15:11, 15 February 2007 (UTC)
    • Right, but you took the tag off that one, right? In that particular case it works perfectly - there was an assertion of notability, ergo it was not deletable under A7, as of right now it remains not deleted. I would still suggest that most of the bad taggings are caught, leaving very few bad deletions. Chris cheese whine 15:23, 15 February 2007 (UTC)
      • This is, by and large, correct. I did some further analysis at User:Mangojuice/a7.2 in which I analyzed all deletions with either "A7" or "G11" in the deletion reason over a contiguous 8 hour block of time. The data is a bit different, partly because (as I suspeceted last time) there were a few admins who did most of the deletion, and they were mostly making correct calls. Out of 161 deletions that could only fit under A7/G11, only 20 got a quality rating of 3/5 or below from me... and of those 6 were 3/5s. Out of the 14 most questionable deletions, only 2 should not have been deleted as they were: one should have become a redirect (and now is), and the other was barely spam, and salvageable. Even being very strict, 7/8 deleted under A7 are just fine. Realistically, about 8.5% of them should have not been speedied, but rather sent to another process. Mangojuicetalk 15:49, 15 February 2007 (UTC)
        • Then again, according to you last time - "Ok -- how about an article on a doctor that, apart from clearly ordinary doctor stuff, adds "Whilst at University he won the Francis Dudley Memorial prize in Anatomy, and also the Southmead Shield case presentation award." How about the article on the band that has two albums that are free and self-released, but says this is "by choice." - were both threes. I'd again love someone who can see deleted histories to rate these as well. I mean, we even have in that list a film (which is not speedyable under A7) deleted anyway. Even if it's only 8.5% (which seems low, but one of the worst of the lot are gone now anyway), if that 8.5% are egregiously poor... --badlydrawnjeff talk 15:57, 15 February 2007 (UTC)
          • Would you be referring to the amateur, unreleased film? Sounds like a perfect candidate for "unremarkable" to me. "zomg we made a film!" is not a valid assertion of notability. Chris cheese whine 16:16, 15 February 2007 (UTC)
            • I've copied and pasted the text here. -- Steel 16:23, 15 February 2007 (UTC)
              • This is why a second person needs to do this analysis. Or maybe that would solve a lot of the problem - A7s and G11s can no longer be deleted on sight, a second admin has to review the tag before the deletion goes through? --badlydrawnjeff talk 16:27, 15 February 2007 (UTC)
                • Too much WP:CREEP. I have to say, looking at the "film", I don't see an assertion of notability in there at all. It seems to be a bunch of kids recreating Jackass and nothing more. Chris cheese whine 16:32, 15 February 2007 (UTC)
                  • God forbid we make an improvement on a controversial criteria because someone might think it's "instruction creep." Can we please come up with a better reason as to why that's a poor idea? --badlydrawnjeff talk 16:35, 15 February 2007 (UTC)
                    • So, not wanting to double the amount of time and effort required to process a candidate article, given the large number of deletions and the small number of people doing them, is apparently not good enough? Chris cheese whine 17:42, 15 February 2007 (UTC)

A lot of this is indicative of a larger problem. Several thousand articles are created every day, roughly half of them need to be deleted. And we have, what, 20 people accounting for 50% of that work? These are all volunteers... often putting in several hours a day to clear CSD backlogs. And that's what they think of themselves as doing, clearing a backlog. They aren't just out to delete articles for the hell of it. There is a tremendous ammount of work being "thrust" upon the handful of people willing to do it. I think we need to remember that. --W.marsh 16:43, 15 February 2007 (UTC)

G12 and OTRS

If an article is deleted using only G12 as the justification, but another criteria would also fit, it can happen that a copyright permission will be sent to OTRS, ending up with the article restored, only to be deleted again. It's a waste of time for the OTRS people, and for the article authors (who go through the work to send the copyright permission through OTRS only to be told later that "well, we don't want it anyway, it sucks"). G12 should only be used when there are no other reasons for deletion. --cesarb 16:21, 15 February 2007 (UTC)

In other words: If it's spam, delete it as spam, because nine times out of ten it's the copyright holder who's writing the article anyway? -- Steel 16:25, 15 February 2007 (UTC)
That's interesting, but I think maybe just changing some guidelines/template messages around might help, too. "Text dumps" are one of the many scourges of newpage patrol... the natural reaction to seeing 5k of unformatted, uncategorized stuff with a totally unacceptable tone is "argh! delete this crap!". Only those aren't really justifications for deletion, but being a copyvio is. I guess a lot of it could be deleted as blatant advertising though. Still, people trying to release their advertising copy under the GFDL so it can be used on Wikipedia should probably see a message at some point telling them that's not a good idea. But can't OTRS tell these people their article didn't have much chance anyway, if they go to OTRS? --W.marsh 16:30, 15 February 2007 (UTC)
Given that there are perhaps hundreds of e-mails per day on OTRS, and each of these requires some investigation, it would be best if the other party recognized without needing to e-mail OTRS that the article is not appropriate for Wikipedia even if the text were licensed under the GFDL, or that the person responding on OTRS is able to handle the e-mail more quickly without needing to investigate any secondary or alternative reasons about the article. The worst situation would be where an article is deleted as a copyvio, the author sends an e-mail asking why, OTRS sends a response that it must be licensed under the GFDL, the author contemplates the matter or discusses it with the company's management, etc. and responds back, then the Confirmation tag is added to the article. Then, the article is deleted because it was crap. That is a lot of wasted time. —Centrxtalk • 16:44, 15 February 2007 (UTC)
I don't think we're doing enough to get the message across that just dumping text into Wikipedia is not a good idea. A few software/warning message tweaks could save us all a lot of time. Take people trying to upload deadend articles... we should have had warnings on those since day one. But we still don't have them. --W.marsh 16:49, 15 February 2007 (UTC)

An adjustment to A7/G11

I think this is important enough to get more feedback on as opposed to it getting lost in the shuffle. On A7/G11, we should prohibit admins from deleting these articles "on sight." Instead, A7s and G11s simply must be tagged by one user and deleted by another. That way, many of the false speedies can be averted because of the extra eyes seeing them. Thoughts? --badlydrawnjeff talk 16:55, 15 February 2007 (UTC)

  • CSD backlogs aren't bad enough already? Some admins already do this. I personally just delete an article if I think it meets a CSD, whether it's tagged or not. At any rate, I am rather sure the most prolific deleters (NawlinWiki, Steel, et al.) get most of those deletions by doing CSD, i.e. they already are deleting articles tagged by others. --W.marsh 17:01, 15 February 2007 (UTC)
  • Usually, the "on-sight" deletions can be prevented ahead of time by using the "Show Preview" button or by building up a copy of the article in a personal sandbox - and if necessary, they can be reverted or undone at WP:DRV. As always, an item needs at least one external and independant reference to survive a speedy-delete - if it's recognizable, it's unlikely that it will even be considered a candidate. --Sigma 7 17:13, 15 February 2007 (UTC)
    • That's not even true regarding the sources, not to mention that the expectation that new contributors know the ropes regarding WP:V and WP:RS is a bit off. --badlydrawnjeff talk 17:16, 15 February 2007 (UTC)
      • I agree with the assessment of W.marsh in this thread and the one above . A few admins devote a lot of their spare time on speedy deletion. They should be praised for keeping the quality standards of Wikipedia to a certain level. Instruction creep will result in discouraging them. I, for instance, have speedily deleted many articles in the past two years. Perhaps a few could have been wrongly tagged without me noticing it. Well so be it, but I wouldn’t have deleted them under A7 if I weren’t convinced that they didn’t belong in Wikipedia. On the other hand, I have salvaged quite a few of thas speedy-tagged articles and transformed them in articles of start-quality of even better, e.g. Anterior cervical discectomy and fusion was originally just one sentence (a short definition) and tagged “db-context”. All right, this wasn’t A7, but what I want to stress is : it is better to salvage and expand the article than just removing the speedy tag and leaving that article in the sorry state it was. JoJan 17:24, 15 February 2007 (UTC)
For reference, the backlog on prod is pushing two days, images on Commons is five days, and fair use issues are a week behind. As of this moment, there are somewhere of the order of 10,000 items that are tagged for speedy. I'm betting the Augean stables are looking good right now. Chris cheese whine 18:02, 15 February 2007 (UTC)
There used to be a clause in WP:SPEEDY that stated admins should not both flag and delete an article because a single set of eyes just isn't enough. When did that get taken out? i kan reed 18:29, 15 February 2007 (UTC)
Maybe you're thinking of the sentence at the bottom of WP:PROD#Procedure_for_admins. That's a different process.--Kchase T 18:39, 15 February 2007 (UTC)
There has never been such a clause in WP:SPEEDY. That possibility was extensively discussed when the speedy-deletion process was first proposed but the community finally decided that it would be more overhead than it was worth. The essense of speedy-deletion and the single factor distinguishing the process from all the other deletion processes is that it can be carried out by one person - that the issues are so clear that it is safe to trust one set of eyes. That's why the speedy-deletion criteria are narrowly written and supposed to be very narrowly interpreted. Rossami (talk) 19:46, 15 February 2007 (UTC)
The main problem I see with this procedure is that (in my view) the tagging is often so wildly bad that this isn't a good way of assuring that the speedies are valid. I probably remove a third of the tags. I've gotten lots of complaints about articles I've deleted and been taken to DRV a few times. By contrast, when I did new page patrol directly, the consequence for improperly tagging an article was usually zilch, as an admin would just remove the tag.--Kchase T 18:39, 15 February 2007 (UTC)
If the specifications for speedy were set so precisely and so accurately that one set of eyes was truly enough, this would be a reasonable argument. But some of the categories are used by some people so generally that this cannot be assumed--as Kchase says, and with much greater experience than my own. To me, the main problem is no assertion of notability, which has been used for full professors and members of the national Academies. But if that criterion were eliminated, some other would be carelessly used. Perhaps true enforcement is needed, such as blocking editors in general from using speedy if they make 2 or more clearly invalid nominations. But the only way to prevent carelessness from admin is to revert to the rule requiring two people.
As sort of compromise is to have a few fixed quantitative rules not requiring two people, e.g. all articles of only one sentence can be speedied, etc. and beyond that takes two people. I know, we're adding complexity, & in effect making a 4th deletion procedure. But any error-correction method take work. I don't think WP can afford to make too many errors in discouraging newcomers, even ones who still have a lot to learn. DGG 20:09, 15 February 2007 (UTC)
Actually, this doesn't take a rule. Some publicity might do it. DGG 20:09, 15 February 2007 (UTC)
  • I don't like the idea of even bigger deletion backlogs, but perhaps giving users warnings/friendly notices when their speedies are declined will help? It would certainly encourage more prodding and less CSD-ing. It doesn't solve the problem of admins simply deleting articles inappropriately, of course. As DGG states, enough publicity might convince admins not to recklessly delete an article that might be borderline notable. Let's not forget that speedy deletion is only intended to get the most inappropriate articles. --N Shar 04:30, 16 February 2007 (UTC)

I try to provide descriptive edit summaries when I remove a CSD... I figure anyone who knows enough to add a CSD will be able to check the history to see why it was removed. But I think people add invalid CSDs knowingly though, because they know some admins will speedy delete stuff for being a "dicdef", "hoax", "failing WP:BIO" and other things that obviously aren't CSDs. Unfortunately I think this is a situation where people are going to keep making invalid taggings if admins keep making invalid deletions based on those tags. I am the first to admit that sometimes an article doesn't meet the letter of CSD but should still be speedy deleted (blatant spam was long an example of this), but a lot of what I'm talking about as "invalid" taggings are ones that really should go to PROD/AfD.

I think we've had problems in the past with admins doing newpage patrol and playing whack-a-mole. While I defend their right to do that job how they want, I do observe some admins taking a much calmer approach now, tagging articles instead of deleting on sight, and only speedy deleting articles that are already tagged, as they do NP patrol. There are other factors involved, but the admins who use this approach do tend to avoid being regular guests at DRV/ANI/etc. I'd oppose making this a requirement (an unenforceable requirement, btw) but it should perhaps be suggested somewhere or other. --W.marsh 15:56, 16 February 2007 (UTC)

A concern about the "new" 48 hour limit on replacable fair use stuff

The criteria as I understand it is that replacable fair use images are speedyable 48 hours after the uploader have been notified. However I notice a lot of images in the 48 hour part of the que where the uploaders have in fact not been notified, notably images uploaded with for example the {{AutoReplaceable fair use people}} selection from the upload page.

I get that the logic is that if they where paying attention during the upload process they would have noticed the faint light gray on white "invalid fair use" heading above those selections (we might want to look into styling that better), and hopefully see the tag after the upload. Still, would it not be better to put those in the 7 day que and instead set up a bot or something to go though the 7 day cat, notifying people and "updating" the tag to the 48 hour style as it goes. I mean I'm all for deleting this stuff, but unless we can get people to understand why the flood will never subside. --Sherool (talk) 11:59, 16 February 2007 (UTC)

Will any CSD cover this?

A user has recently created several redirects concerning Hillary Rodham Clinton. Most are unlikely to be used, although not a typo, and one, President Hillary Clinton, is trying to predict the future. Do you think these should go through WP:RfD or is there a way to justify them with WP:CSD? I hate clogging up XfD with more trivia when not needed. --StuffOfInterest 01:34, 17 February 2007 (UTC)

I tagged the President one with {{db-nonsense}} per not a crystal ball. The others I'm not sure about. (Netscott) 01:38, 17 February 2007 (UTC)
Not fully sure I would consider this "nonsense", but I'll take it. :) More interesting, the user created the same article twice today, first as an actual article and second as a redirect. Guess it may be getting time for a WP:ANB/I posting as his edits are approaching disruptive. --StuffOfInterest 01:45, 17 February 2007 (UTC)
I would suggest if you haven't already you try to counsel to user in the ways of Wikiepdia. ;-) (Netscott) 01:51, 17 February 2007 (UTC)
As I've warned him before about his edits habits, I fear he would not take well to recommendations from me. It really needs a disinterested party. (wink wink, nudge nudge) --StuffOfInterest 01:58, 17 February 2007 (UTC)
User name? (Netscott) 02:02, 17 February 2007 (UTC)
K157. If you, or someone, can drop him a helpful note it will be much appreciated, by me at least. Thanks. --StuffOfInterest 02:09, 17 February 2007 (UTC)
WP:NOT#CRYSTAL BALL is not a speedy-deletion criterion. I would certainly not call it patent nonsense in the narrow way we use that term here. Given the pattern of edits, you could make a case that this was deliberate vandalism but it's a thin case. In the future, send them to RfD for community evaluation. Rossami (talk) 15:03, 17 February 2007 (UTC)

Which template to use for schools?

Which template should be used for schools? db-bio? db-group? Do we need a new template? Regards, Ben Aveling 10:10, 17 February 2007 (UTC)

That depends on which criteria you think the particular articles fall under. --bainer (talk) 12:29, 17 February 2007 (UTC)
I think Ben is referring to a school article that meets A7. Most of the pre-packed A7 templates refer to specific classes. If in doubt, tag it with {{db}} with a suitable description (preferably including "A7" in there somewhere). Articles of the form "X Y is a Y X" come under A3 also. Chris cheese whine 12:43, 17 February 2007 (UTC)
PS, "is a school" is not per se an assertion of notability. Chris cheese whine 12:44, 17 February 2007 (UTC)
Schools are not bands and they're not clubs, and they're certainly not real people or groups of real people. --bainer (talk) 14:06, 17 February 2007 (UTC)
I'm sorry, I wasn't aware of the new rule that says only articles about people and bands have to assert notability for inclusion. Chris cheese whine 14:17, 17 February 2007 (UTC)
Only articles about people and bands that don't assert notability are eligible for speedy deletion. God knows why it's only those things, I think the whole criteria is stupid honestly, but as long as it exists I'll try to make sure that people stick to it and not think that anything they dislike can be deleted under it. --bainer (talk) 14:24, 17 February 2007 (UTC)
The deletion discussions over schools have been so very contentious that I do not think that it is reasonable to use any CSD criterion on them (except maybe vandalism or attack page if it's clearly intentionally derogatory and there is no good version to revert back to). When schools are posted to AFD, they are routinely kept. Personally, I disagree with the reasoning used but we have to acknowledge the will of the community. Attempting to circumvent the community's stated position on schools through CSD would be a clear abuse of the process. Rossami (talk) 15:01, 17 February 2007 (UTC)
I don't believe that is "the community's stated position". It just happened to be the stated position of a group of editors who were numerous and vocal enough to disrupt the process to such an extent. I see no legitimate reason for schools to be exempted from the usual criteria (G1, A3, A6, etc.). I suppose it would be more accurate to say the community's stated position on the matter is "enough fighting already!". Chris cheese whine 15:26, 17 February 2007 (UTC)
Schools are most certainly not exempted from all speedy criteria, but since many people believe schools are inherently notable, they can't be an A7. -Amarkov moo! 15:57, 17 February 2007 (UTC)

Proposal of new guidelines for user category CSDs

Category:Wikipedians who have survived an attack by a crazed astronaut is an obvious joke usercat that doesn't help Wikipedia improve as an encyclopedia. I understand that CSDs must be very narrow so that the appropriateness of such deletions should be obvious to everyone. I thus would like to ask if we can agree on a new rule for usercat CSDs for cats such as the aforementioned one. Xiner (talk, email) 16:14, 17 February 2007 (UTC)

I generally use CSD stupid user category. As much as I would really love a proper CSD criterion to get rid of this rubbish I can't see it happening. -- Steel 16:26, 17 February 2007 (UTC)
But we at UCFD are more cautious than you, Steel. :) Xiner (talk, email) 16:43, 17 February 2007 (UTC)
This is a clear-cut case, of course. But what about things that not everyone would agree are obvious, or even joke categories? Criteria should be as objective as we can possibly make them, and "obvious joke cat" is not. -Amarkov moo! 18:03, 17 February 2007 (UTC)
Hmm, I guess G1 (patent nonsense) suffices. Xiner (talk, email) 18:11, 17 February 2007 (UTC)
So you know for a fact they haven't survived a crazed astronaut attack? --badlydrawnjeff talk 19:06, 17 February 2007 (UTC)
No, I don't. But I'll bet at least one of the three users is lying. Xiner (talk, email) 19:13, 17 February 2007 (UTC)
Not only that, but identify one place where that information might be useful other than in an autobiography, or as an anecdote for a talk show ("Not a lot of people know this, Michael, but I was once attacked by a crazed astronaut ..."). I'm happy to simply classify this one as G0 (i.e. IAR speedies that don't fit a specific criterion but clearly don't belong). :) Chris cheese whine 17:40, 21 February 2007 (UTC)
It might be useful if you are ever working on the article Criticism of crazed astronauts and need some help from an editor experienced in that area. --BigDT 12:54, 6 March 2007 (UTC)

I5 7 day orphans

Hi today, I had to deal with the fact that around 70 images got "semi-automated" deleted by admins due to I5. These images became orphaned because someone made a boo boo when redirecting a template to a new template, resulting in a lot of images being unlinked. Although this was noticed by some editors quite quickly (a lot of articles were fixed in time), no one saw a possible deletion along the way and I guess most editors didn't even realize the template changed. Only when some images started to disappear, the editors felt trouble. However, it took in total over TWO weeks before the problem resulted in messages on related WikiProjects, that made more experienced editors investigate this problem. Over 700 pages had to be checked to find out if images were unlinked and consequently deleted. A lot of wasted work you can say. I'm not really sure if this problem can be avoided, but I thought it would be a good idea to alert people here that the current process (a deletion process with little user review) can be quite problematic in some cases. TheDJ (talkcontribsWikiProject Television) 21:47, 20 February 2007 (UTC)

I'm very confused by this. The only images that can be speedy deleted for being orphans are unfree images, which cannot be used in template space. Can you explain how redirecting a template resulted in I5 deletions? Jkelly 21:56, 20 February 2007 (UTC)
They were arguments to templates. Old template used Image = , new template required Image = Image:imagename.jpg TheDJ (talkcontribsWikiProject Television) 21:59, 20 February 2007 (UTC)
I see. That's a big "boo boo". I'm not sure that there's a reasonable way to guard against this on the deletion side. Images orphaned by bot or by hand can be tracked to a particular article. Images orphaned like this can't be. Jkelly 22:21, 20 February 2007 (UTC)
Its a design fault in mediawiki and one that is virtually impossible to work arround. The problem is we only track history of data not the metadata that we generate from that data. While in principle history of metadata can be regenerated from history of data in practice this is impractical due to the huge volumes of data involved. Plugwash

Split G3 into R4

I suggest removing the "including redirects created during cleanup of page move vandalism" from G3 and move it to a new criteria in the R category. This new criteria would then be:

Disparage or subversive redirects or redirects created during cleanup of page move vandalism.

AzaToth 18:15, 21 February 2007 (UTC)

No it wouldn't be. That introduces "disparaging" and "subverting" as reasons for speedy deletion. -Amarkov moo! 20:04, 21 February 2007 (UTC)
They're G10's anyway. No need to make a more specific redirect criterion. (Besides, there used to be a specific redirect criterion for page-move-vandalism cleanup; splitting it back out of G3'd be a clear step backwards.) —Cryptic 20:16, 21 February 2007 (UTC)

The most obvious cases of redirects created just as insults will fall under G10. Where it's less clear, RFD is more appropriate. I see no need to add a new criteria here, sorry. --W.marsh 14:17, 22 February 2007 (UTC)

G7 clarification

I think G7 needs to be clarified for redirects resulting from page moves. It should state that this criterion does not apply if the page was originally created as the result of a page move and the person who did the move was not the original creator of the article. --- RockMFR 21:42, 24 February 2007 (UTC)

This took me a minute to wrap my head around, but let me give an example to make it clearer.
  1. EditorA creates Article 1 and perhaps writes some content in it.
  2. EditorB, without any prior discussion, moves Article 1 to Article 2.
  3. EditorB requests that Article 1, now a redirect, be deleted. Because they are the sole contributor in the history, this looks a lot like a valid G7.
  4. EditorA does not want it deleted, believing the original title to be valid. They recreate it. It's re-deleted by G4. This is stupid.
The crux of the argument is that since Article 1 is just a redirect, and EditorA chose its title by creating the original article, EditorA has "contributed substantially" to it - hence their permission is required to delete it. I can believe this can be summed up by the statement: "Redirects created by a page move shall be considered to have been substantially contributed to by the person who created the article being moved." I'm not going to attempt to ponder the case of multiple subsequent moves at the moment, but I think this still works.
Note that although this initially appears to contradict WP:OWN, it in fact doesn't, since the only contribution of consequence here is the title - any subsequent contributions to Article 1 were moved to Article 2. Deco 02:06, 25 February 2007 (UTC)
The best way to avoid that situation is really to just apply G4 properly and not use it for speedies. -Amarkov moo! 03:59, 25 February 2007 (UTC)
G4 is not the problem I was seeing with this. Rather, there are problems with the initial db-author deletion if the deleting administrator does not carefully look at the situation. Attached talk pages could get deleted (db-talk), links to the redirect might get removed, etc. It doesn't seem like G4 was created with the intention of allowing moved page redirects to get deleted at the whim of the page mover. --- RockMFR 04:28, 25 February 2007 (UTC)
I guess that would be true. Actually, I would support making G7 not apply to redirects at all; if it doesn't fall under another criterion, I can't see any reason why we would want it deleted. -Amarkov moo! 04:36, 25 February 2007 (UTC)
The scenario above should never qualify for a G7 speedy-deletion. The author making the pagemove is not the only contributor to the page's history (though admittedly it could look like that if you didn't properly read the edit history). Any admin who pulled the trigger on such a request should have his/her hand politely slapped for failing to thoroughly check the edit history and the speedy immediately overturned.
The redeletion is explicitly not allowed under case G4 since G4 may only apply if there has been a prior deletion discussion. Prior speedies are insufficient to qualify a page for case G4.
All that said, I would not want G7 removed from all redirects. If I create an article, realize that I did so at the wrong title (for example, a typo) and move the page myself before any other editors or readers have had a chance to find the page, I should be able to tag my mistake as {{db-author}}. Likewise, if I create a redirect to an existing page but then learn about the manual of style considerations for redirects, I should be allowed to correct my own mistakes. Rossami (talk) 04:48, 25 February 2007 (UTC)
You are correct that G4 doesn't apply - my mistake. The request of the original poster was not a change in the normative meaning, though, only to clarify the history is, as you put it, properly read, which seems fair enough. Deco 10:21, 25 February 2007 (UTC)
I'm with Rossami here; G7 should not really be used in such a circumstance. And it shouldn't be hard to check if this has happened: a G7 deletion always means you have to check the edit history, and if it consists only of a page move, you'll notice it right away. Mangojuicetalk 02:17, 26 February 2007 (UTC)

How Broad is G6?

I notice G6 mentions several specific examples of when it can be invoked...but does G6 apply to all "non-controversial housecleaning deletions"...such as at MedCom the deletion of improperly filed mediation requests? Or deletion of a moved mediation request's subsequent redirect? ^demon[omg plz] 21:53, 26 February 2007 (UTC)

Why wouldn't it? —Centrxtalk • 23:17, 26 February 2007 (UTC)
Speaking of such, do you think that G6 could be extended to requests from WikiProjects to delete pages within its project space or portal, if one exists (assuming that there is consensus within the WikiProject about it)? GracenotesT § 00:13, 27 February 2007 (UTC)
I'm pretty sure it already is, fairly regularly. -Amarkov moo! 00:28, 27 February 2007 (UTC)

CSD for albums?

Is there a CSD that would cover an album of a band page that was deleted?


Squids_and_Chips 01:25, 27 February 2007 (UTC)

  • These articles probably shouldn't be speedied. They should have been bundled with the original AfD, but at this point that's not possible, so the best thing to do is to WP:PROD them. It would be rather borderline to tag them A7 -- this would only be valid if the article for the band had been deleted for notability reasons, which would involve an admin's checking the AfD/deletion log, which would take time. Prodding is really most efficient here. If the prod is contested an AfD will probably result in a consensus of speedy delete, but there's no need to waste the time of the AfDers unless that happens. --N Shar (talk contribs) 01:47, 27 February 2007 (UTC)
Concur with N Shar. I do a lot of work in the WP:ALBUM project and have come across orphaned album articles which I have always applied ProD to. There just isn't a criteria under WP:CSD which covers them, unless they fall foul of one of the included criteria (e.g. short, no context). I have found ProD in these cases to be more than sufficient, and they are rarely contested. Bubba hotep 09:38, 27 February 2007 (UTC)

G3 template

Am I missing it, or does one not exist? CovenantD 09:50, 27 February 2007 (UTC)

Template:Db-vandalism or Template:Db-g3. :) Bubba hotep 09:54, 27 February 2007 (UTC)

Empty maintenance categories that will never be populated again

I propose a new CSD for maintenance categories that will never be populated again. For example, the backlog in Category:Articles to be merged since April 2006 has now been cleared. There is really no reason to discuss this deletion on CfD but no mechanism for speedy deletion at present. This doesn't actually require a new criterion; it can be added to the wording of C1. --Selket Talk 22:17, 27 February 2007 (UTC)

Nevermind, I just realized G6 covers this. --Selket Talk 22:19, 27 February 2007 (UTC)

User pages of non-existent users

It says to check Special:Contributions, but sometimes users have no edits. Should I update it to Special:Listusers? —The preceding unsigned comment was added by -Slash- (talkcontribs) 02:50, 28 February 2007 (UTC).

No need to, the Special:Contributions page for a nonexistent user is sutbly different from one for an existing user. Another trick is to check for the existence of the Special:Contributions link on the toolbox; it only appears if the user exists. --cesarb 03:17, 28 February 2007 (UTC)
Is the only difference the presence/absence of the block log? --ais523 11:23, 28 February 2007 (UTC)
  • At any rate, a user with zero contribs can be considered non-existent. >Radiant< 11:30, 28 February 2007 (UTC)
    • There are legitimate cases for having user and user-talk pages for zero-contrib users (AOL IPs, open proxies, doppelgangers, {{nn-warn}} (because the page that the user created would be deleted, they could have 0 visible contribs), temporarily for usernameblocks), so I don't think deleting user- and talk-pages of zero-contrib users is uncontroversial enough to be a CSD. --ais523 11:41, 28 February 2007 (UTC)
      • And that is why people are supposed to think before deleting anything, and not blindly hitting the delete button for anything that technically meets the criteria. Nobody's going to delete the userpage of a zero-contrib doppelganger. >Radiant< 11:56, 28 February 2007 (UTC)
        • I can think of lots more situations where a zero-contrib userpage should be kept than either situations where a zero-contrib userpage should be deleted, or a no-user userpage should be deleted, so I'd disagree with you that a zero-contrib user can be considered nonexistent. I don't think that there's a rush to clear out incorrect zero-contrib userpages, and so MfD could be used for those (come to think of it, I can think of one that should be deleted at the moment, so I'll go and MfD it now). --ais523 12:20, 28 February 2007 (UTC)
The criterion is meant to allow for deletion of user pages that legitimately belong to no user, but are created by other users for whatever reason. As for user pages of users who have done nothing but create their own user pages, or who exist but have made no contributions, WP:PROD can be used to get rid of them. Mangojuicetalk 13:48, 28 February 2007 (UTC)


What's the tag for A7? {Slash-|-Talk} 04:35, 28 February 2007 (UTC)

{{db-a7}}. All criteria at least have a redirect like that. -Amarkov moo! 04:54, 28 February 2007 (UTC)
Also, {{db-bio}} and {{db-band}} and {{db-corp}} and {{db-group}}.  :) Mangojuicetalk 13:45, 28 February 2007 (UTC)

Instruction creep

This section is overly bureaucratically worded and as such, instruction creep: "For the purposes of this criterion, a redirect created by a page move is considered to have been substantially contributed to by those who contributed to the article that was moved." Anything like "for the purposes of X, treat Y as Z" is worded like a code of law, which is what Wikipedia is NOT. Mango claims this was discussed; I would like to know where, and suggest its removal because we should be careful not to make this page overly complex. >Radiant< 13:49, 28 February 2007 (UTC)

Actually I didn't notice that removal, sorry about that one. But it was specifically discussed: see #G7 clarification, just a few sections back. Personal opinion, yes, we could keep that out, I think it's pretty obvious it should work that way; I'll reinstate the change. The one I noticed was the change to G4, which was discussed at #G4 seems to have a wording problem..., pretty thoroughly. Mangojuicetalk 13:55, 28 February 2007 (UTC)
I would have thought that it was self-evident, too. Unfortunately, recent entries listed on the Redirects for Discussion page convinced me that there was enough confusion that it needs clarification. Would you be more comfortable with simpler language like "Moving a page does not make you the "sole author" of the redirect." Rossami (talk) 03:58, 2 March 2007 (UTC)
Maybe simpler would be to say that G7 doesn't apply to redirects at all? Does it happen very often that a G7 deletion gets put on a redirect that wasn't created via a page move? Mangojuicetalk 16:24, 2 March 2007 (UTC)
It does occasionally happen that G7 is appropriate to a redirect. I don't know how you'd get exact numbers but there are several examples on RFD where a discussion was closed early as an appropriate use of G7 (or where other editors explained that the nominator could have used G7 in the first place). See the G7 clarification discussion several sections above for the scenarios where it can be appropriate. Rossami (talk) 01:27, 3 March 2007 (UTC)
All the more reason to keep the simpler wording. If we have to get into what kind of page-move-left redirects can fit A7 to that degree, it'll just be confusing. I'd rather let the stray problem cases that get objected to go to WP:DRV than clutter up the rules over this. Mangojuicetalk 14:55, 4 March 2007 (UTC)

Issues with CSD G1

I've been having some discussion with an editor over whether Appankoil is CSD G1. I'm pretty sure it isn't by a matter of precedent of how G1 is interpreted. If I'm correct making the narrow nature of G1 more explicit might help. JoshuaZ 05:51, 3 March 2007 (UTC)

I don't know how much more narrow than "patent nonsense" you can get. That's obviously not patent nonsense. -Amarkov moo! 06:11, 3 March 2007 (UTC)
Appankoil is just 1 sentence. and part of G1 states: "This article or other page provides no meaningful content or history,"; therefore it qualifies for G1. --Parker007 06:13, 3 March 2007 (UTC)
Um... no, it doesn't say anything like that. -Amarkov moo! 06:17, 3 March 2007 (UTC)
Well then, the template needs to be changed, because that's not what the criteria says. I'll go do that. -Amarkov moo! 06:38, 3 March 2007 (UTC)

The article wasn't particularly well written, but it clearly wasn't "nonsense". – Steel 13:38, 3 March 2007 (UTC)

Regarding CSD:R2

Shouldn't this criterion also apply to Redirects FROM userspace TO article space, and not simply the other way around? Logical2uReview me! 14:22, 4 March 2007 (UTC)

Um, why should it? – Steel 14:41, 4 March 2007 (UTC)
I don't see a reason to worry too much about it. Surely redirects from userspace to article space that are unwanted will typically be unwanted by that user, and can be tagged with {{db-u1}}. Mangojuicetalk 14:51, 4 March 2007 (UTC)
And redirects from userspace to article space that are wanted are harmless. Redirects out of the mainspace are harmful largely because many, many of our mirrors and other re-users copy only the main namespace; nobody mirrors just user pages. —Cryptic 14:55, 4 March 2007 (UTC)
I brought this up because of User talk:Danielwharton, who moved his userpage to article space, leaving a redirect on his userpage, and there is no real speedy tag for that, but there is for the other way around (IE:Userfication (Infinitive of Userfied?))Logical2uReview me! 17:23, 4 March 2007 (UTC)
I think a redirect from an actual userpage should be prohibited. Nobody should click on a user's signature in a discussion and be led to an article in mainspace on doing that. However, redirects from user sub-pages should not be a problem. I think I have a few of those myself; they happen if you prepare a draft article in userspace and then make it go live. No problem about those, I'd say. Fut.Perf. 17:51, 4 March 2007 (UTC)
How about something like, a redirect from a high visibility userpage to article space (IE: Non subpages, or a subpage in a signature?)Logical2uReview me! 17:55, 4 March 2007 (UTC)
Something like that, but I guess I'd avoid trying to spell it out in all detail. Leave it to common sense and avoid instruction creep. Fut.Perf. 19:28, 4 March 2007 (UTC)

Concerning I6

"Missing fair-use rationale. Any image or media tagged only with a generic fair use template, with no fair use rationale, may be deleted seven days after it is tagged. Images and other media uploaded before May 4, 2006 should not be deleted immediately; instead, the uploader should be notified that a fair-use rationale is needed. Images or other media uploaded after May 4, 2006 can be tagged with {{subst:nrd}}, and the uploader notified with {{subst:missing rationale|Image:image name}}. Such images can be found in the dated subcategories of Category:Images with no fair use rationale." I think this needs some clarification. The term generic fair use template seems to be giving people the impression that this only applies to {{Non-free fair use in}} and {{fairuse}}, even though fair use criterion 9 cleary states that all unfree images require fair use rationales. I've take "generic fair use template with no fair use rationale" to mean any template that includes the words "Note to the uploader: Please provide a detailed fair use rationale" which is almost all of them, if not all of them. I understand not wanting to speedily delete incomplete or misworded rationales, but not speedying all fair use images after a week with no rationale at all doesn't make sense. {{albumcover}} images should not have to go through IFD simply because no one writes a fairuse rationales if we're already speedy deleting other images for the same reason. Jay32183 01:27, 5 March 2007 (UTC)

Agreed. I think the intent of "generic" was a template that doesn't have a fair use rationale in it, but that does need clarification. -Amarkov moo! 02:03, 5 March 2007 (UTC)

While I totally agree that unfree images need fair use rationales, I do not think that lack of one should result in speedy deletion unless {{fairuse}} or {{Non-free fair use in}} was used. For images tagged with other license tags, the uploader should be notified and the image should be put through WP:IFD. The other license tags are descriptive enough to allow the uploader a little more time to add a fair use rationale. —Remember the dot (t) 03:53, 5 March 2007 (UTC)

I think the fact that the tags say "To the uploader: please add a detailed fair use rationale for each use, as described on Wikipedia:Image description page, as well as the source of the work and copyright information." means that there is no excuse for not having a fair use rationale. Using {{subst:nrd}} already requires notifying the uploader and a week of waiting, so I don't actually see what you mean by "more time". Jay32183 04:01, 5 March 2007 (UTC)
  • I think the confusion here is arising mostly because the {{no rationale}} and {{missing rationale}} list {{fairuse}} and {{Non-free fair use in}} as specific examples, because it is quite clear from all the rules that fair use rationales are not limited to those templates and all of the templates say "add a detailed fair use rationale" and CSD regulars do delete images tagged as having no rationale no matter what copyright tag they actually use. Jay32183 19:42, 6 March 2007 (UTC)

The policy originally applied only to images tagged with {{fairuse}}, with the important condition that images uploaded before the policy went into effect were exempt from it (archive 10). This policy went into effect on May 4, 2006. The policy was changed on September 20, 2006 to apply to all fair use tags (archive 13). At the very least, we ought to:

  • clarify the criteria to make it crystal clear that this applies to all iamges
  • exempt images from this policy that were uploaded between April May 4 and September 20, 2006 and tagged with a more specific fair use tag
  • emphasize on the fair use tags that an image without a fair use rationale will be deleted in 7 days.

In my humble opinion, I think the policy should have been left as applying to only {{fairuse}} and {{Non-free fair use in}}. But no matter what it applies to, it ought to be clarified. —Remember the dot (t) 20:10, 6 March 2007 (UTC)

    • I like the clarifying and the emphasizing, but not the exemption. I assume you meant May 4, but that isn't my problem. May 4 2007 is less than two monthes away. On that day I will be proposing that "It has been a year, which should be sufficient time for older images to be fixed. We should remove the date limitation from the criterion, otherwise editors have no encouragement to add the fair use rationales unless we overload IFD." The reason we have speedy deletion is to prevent *FD overloads with things that blantantly violate policy, which unfree images without fair use rationales do. If anything we should stick to the May 4 2006 date rather than pushing it back to 2007-09-20. Jay32183 20:19, 6 March 2007 (UTC)
  • There's a prior discussion at /Archive 16#What is a "generic" fair use tag? (CSD I6). I started a category Category:Wikipedia:Generic fair use tags meant to hold the categories that can be considered "generic" in the sense that just the tag with no further explanation would qualify those images for deletion. I had linked to the category from the policy page but it was reverted.. I think maybe it's time to go back to that. Mangojuicetalk 22:39, 6 March 2007 (UTC)
    • Your list of "generic" fair use tags is very inadequate, espicially since I saw this: [2]. Generic is supposed to refer to all the templates that say "please provide a detailed fair use rationale" which also includes {{tv-screenshot}}, {{albumcover}}, and {{logo}}. What I'm trying to get cleared up here is that I6 is supposed to include all fair use templates because they all require detailed rationales and none of them contain detailed rationales. Because of the notice there is no excuse for this. Jay32183 23:18, 6 March 2007 (UTC)

I do not think it is a good idea to delete half the images on Wikipedia because a poorly explained policy has been "on the books" but not actively enforced. —Remember the dot (t) 23:01, 6 March 2007 (UTC)

We will have to go through the older images at some point, and I'd rather it be sooner than later. The TelevisionWikiProject has received a lot of flak for using to many unfree images improperly and it would be great if we had a better method of policing ourselves, rather than having to overload IFD. I am willing to debate the specifics here, but I think you'll agree that there is a problem with older images without fair use rationales. Jay32183 23:18, 6 March 2007 (UTC)
It would be much better to just add fair use rationales to all the old images than to delete them outright. I doubt that very many of them actually violate the fair use policy (aside from lacking a detailed rationale). Those that do violate the policy and thus need to be deleted can go through WP:IFD.
On top of that, the lack of clarity and lack of enforcement of this policy do not justify mass deleting images until it is crystal clear to every Wikipedia editor that fair use rationales are not only recommended, they are required. —Remember the dot (t) 00:03, 7 March 2007 (UTC)
  • There is no doubt that the criterion should apply to all images, regardless of what tags are slapped on them. The original wording, which only mentioned {{fairuse}} and {{Non-free fair use in}} was partly an oversight (since all fair use images have always required a rationale, for each use) and partly due to the fact that many of the specialised templates weren't even created back then. The intention when I introduced the word "generic" was to clarify that it extends to all templates that contain only boilerplate text. In hindsight it seems that the wording is unclear. I've clarified the wording again. --bainer (talk) 23:25, 6 March 2007 (UTC)
    Since you've changed the wording here to make it clear that all fair use tags without rationale are eligible for speedy deletion, shouldn't it be changed elsewhere? The language here now differs from the language at WP:IFD, {{no rationale}}, and {{missing rationale}}. —RP88 05:20, 7 March 2007 (UTC)
    Done. If there are any other places where the wording is inconsistent then feel free to update them also. --bainer (talk) 13:45, 7 March 2007 (UTC)
    • Whoa, whoa, whoa. This goes WAY too far. First of all, I think that in some cases, boilerplate text really says everything needed for a fair use rationale: in such cases, it's wrong to delete the image. For instance, Image:MidnightOil MidnightOil.jpg, marked with {[tl|album cover}}, is being used exactly as the boilerplate text describes. Second, in many cases where more information may be needed, such as Image:Lore on Descent.jpg, the boilerplate text should be enough to be considered reasonable progress towards an acceptable fair use claim: in such cases, deleting the image is probably a bad idea, instead, someone should finish writing the rationale. Third, even in cases where something decidedly non-obvious is going on, such as Image:BugsBunnyShow.jpg, is it really appropriate to speedily delete the image? Can we be a little more eventualist about this? I6 should definitely apply to any image marked only with a fair use tag that gives only generic reasoning, like {{fairuse}}. But if we start deleting things for being marked only with {[tl|tv-screenshot}} or the like, it's a huge number of deletions which probably aren't really necessary. I know speedy deletion saves time, say, over IFD, but making it too easy actually encourages a larger workload. I don't mind that when the deletions are necessary but I really think most of these are not necessary. Mangojuicetalk 17:43, 7 March 2007 (UTC)
      • We cannot be eventualist with fair use images because of potential legal issues. Also, without speedy deletion there is no incentive for uploaders to make sure they do everything correctly, since they can just wait for some one else to come along. Remember, on Wikipedia, burden of proof falls on those trying to add or keep, not those trying to remove. I feel this isn't strong enough because it is grandfathering images and media that are in obvious violation of FUC#10. Jay32183 19:42, 7 March 2007 (UTC)
        • There's a difference between situations where images are used in a way that isnt fair use and where they are used in a way that isn't Wikipedia:Fair use. I buy the argument that we can't wait when a copyright violation is going on (in other words, when the real concept of fair use is not being followed). I don't buy the argument that when an image fundamentally is being used in a fair way that we can't wait for appropriate documentation to bring the image in line with our internal policy, Wikipedia:Fair use. Mangojuicetalk 15:07, 8 March 2007 (UTC)
          • It is the responsibility of uploaders to provide all necessary information about images or media that they upload, including source information and including a fair use rationale, if it is claimed to be fair use. And because image undeletion has been enabled since June last year, we can afford to follow this through knowing that images can be undeleted immediately when the uploader later provides the rationale. Other users are free to update image pages when they come across them, but we shouldn't force this work onto them simply because uploaders continually fail to comply with the basic requirements. --bainer (talk) 23:29, 8 March 2007 (UTC)

We should not tolerate people ignoring the massive amounts of warnings and instructions on the upload page. Just because a lot of people break the rule does not mean the rule should go. Just be thankful we don't push the 48 hour limit, and go out of our ways to give uploaders a week to comply. This problem is getting bigger and not smaller, by a massive amount. We have no choice but to be more strict about the issue. Users don't have the right to just use any copyrighted image as they see fit without following our policy on them. It's better than no fair use images at all, isn't it? -- Ned Scott 05:13, 9 March 2007 (UTC)

I think this is a major expansion of I6. If we are doing this, I think it really shouldn't apply to old images at all: those should go through IfD. Expectations on uploaders have changed. I wouldn't mind deleting old images that made no attempt at a fair use rationale, for instance, using just {{Non-free fair use in}} or not saying anything. But old images that have a boilerplate template that says most of what needs to be said, when I6 didn't cover this type of deletion before? Mangojuicetalk 13:58, 9 March 2007 (UTC)