Wikipedia talk:Criteria for speedy deletion

Curious about A2

Just why are non English articles that exist on other Wikimedia projects speediable? --Ron Ritzman (talk) 05:41, 11 January 2009 (UTC)

If an article is in German, then it belongs on the German wikiproject, not the English wikiproject. And if it exists there, then we don't need it here to get the information from to translate into English.---Balloonman ^PoppaBalloonCSD Survey Results 06:04, 11 January 2009 (UTC)

I am not really happy with this rule. If it is here it should be either translated, or--if not appropriate by our rules--deleted. There are enough people here that simply removing it because it hasnt been translated yet makes no sense--yes it's on the other project, but who would know to look. We should keep it for the same reason we keep stubs. DGG (talk) 04:23, 13 January 2009 (UTC)

Many times an article that gets tagged A2 ends up being translated by its creator (either partially or entirely) even before an admin can review the speedy tag. I, for one, never use this tag, even though I recently proposed that it be expanded to cover articles speedied by other projects, although I can see its use: a would-be translator would be able to get the original text from the other project anyway. I say A2 should stay, but an admin who deletes an A2-tagged article should consider making a request at WP:TRANSLATION (though if the article is spam and hasn't been identified as such, that would only make it more visible than warranted). -- Blanchardb -^{Me•MyEars•MyMouth}- timed 15:24, 13 January 2009 (UTC)

Perhaps one also could drop a little note at WP:PNT, for a (hopefully) speedy evaluation of the merits of the article. Lectonar (talk) 15:32, 13 January 2009 (UTC)

Proposed template "db-speedy2" to parallel prod2

User:Davidwr/Templates/db-speedy2 is meant to parallel {{Prod2}}. It's meant to give alternative reasons for deletion in the face of a {{hangon}} or a marginal criteria like a weak A7 or A9.

Purpose: Add other speedy deletion criteria under which this article may be deleted. This is a hint to the deleting admin that even if he disagrees with the first suggested criteria for deletion, it should be deleted under another criteria instead.

Note: If you do not think the article should be speedy deleted, either remove it entirely or, if you are a major contributor, use the hangon tag.

Parameters:

• disagree - optional, if set, changes text to indicate disagreement.
• reason= reason - required, must be a valid criteria for speedy deletion.

Usage:

To add additional criteria, for example, to counter a {{hangon}} tag or to bolster a claim that is marginal:

• {{db-speedy2|reason=[[WP:G12|G12]] - blatant copyright violation.}} produces:

To disagree with the stated criteria but to offer a substitute criteria:

• {{db-speedy2|disagree=yes|reason=Not an attack, but [[WP:A7|A7]]-no claim of notability.}} produces:

Would this be helpful to anyone but me? davidwr/(talk)/(contribs)/(e-mail) 23:14, 11 January 2009 (UTC)

I'd just say if you think someone else also applies, add that tag as well. There is no harm when there are two CSD tags. And if you think the article is mistagged, just replace the tag with what you think correct. The reviewing admin must apply the correct tag anyway, not the one that the article has been tagged with. Regards SoWhy 23:15, 11 January 2009 (UTC)

I agree with SoWhy; the more the merrier with regards CSD tags. I never really understood why the same principle didn't apply to PRODs too. Happy‑melon 23:20, 11 January 2009 (UTC)

Ah well, in my understanding {{prod2}} serves as a way of expressing a delete-!vote, i.e. saying "I too think that this article should be deleted because of the reasons specified". SoWhy 09:45, 12 January 2009 (UTC)

This is one of the few cases where "rolling your own" tag with {{delete}} is useful. I used something like this once...

{{delete|the article is blatant advertising (CSD G11) and the first paragraph was copied from http://www.example.com (CSD G12)}}

--Ron Ritzman (talk) 23:35, 11 January 2009 (UTC)

This doesn't do anything that can't be done already. To add another speedy deletion reason then simply add another template, and if you disagree with the reason given but think another criterion applies then replace the template. The reason for having the prod2 tag is that the prod tag starts a timer (which is why it should be substituted), so a different tag is needed for additional reasons to avoid confusion about when the article can be deleted. That doesn't apply to speedy deletions because they can be implemented immediately anyway. Phil Bridger (talk) 00:06, 12 January 2009 (UTC)

... and for comments I just place a {{comment}} box directly beneath the CSD box, although it doesn't use the correct color coding. A {{db comment|Bla}} template as a shortcut for {{mbox|type=speedy|text=Bla}} could be useful there, I noticed at least once that my comment was overlooked, probably due to the wrong color.

Amalthea 18:09, 12 January 2009 (UTC)

It was suggested in another thread one could use the {{db}} tag along with a reason and that works fine here as well for multi criteria that may fit. I also echo what Ron, Phil and Amalthea say but add on that as the deleting admin can use their own judgment to either delete under the given CSD or delete under another one anyway I don't think a new option tag is needed. Soundvisions1 (talk) 19:10, 12 January 2009 (UTC)

if you think that a tag doesn't fit, but that another tag would, be bold and change it. When we admins come around to check we'll look at the whole history (or at least we're supposed to do that, and most of them time we do). I choose the deletion reason that I think applies the best, and often give several if several apply--it can save arguments later, e.g. "G11 promotional, and also no claim to notability" DGG (talk) 04:20, 13 January 2009 (UTC)

Per DGG. Dlohcierekim 04:23, 13 January 2009 (UTC)

A proposal for matching descriptions to templates

As a newbie to the administrative side of WP, I see many experienced members and admins are overworked. I'm sure they'd like some assistance. One thing that I find strongly discourages me (and likely many others) from lending a hand is the challenge of associating these helpful descriptions for reasons to applying speedy deletion to the templates themselves. The lack of a direct association with the templates themselves is frustrating and seriously inhibits usability, requiring user to search for and match up the appropriate template.

Among experienced Wikipedians a shorthand lexicon associated with various templates has emerged, for example, A2 and G11. Veterans toss them around like so much pizza dough. I haven't a clue. Why not provide the appropriate template syntax within the context of the description, or link directly from the descriptions to the template, and perhaps number the descriptions in like manner? This would immensely improve more new comers' ability to make sense of this complicated syntax and structuring. -- btphelps ^{(talk) (contribs)} 00:43, 12 January 2009 (UTC)

You can always look up the template named db-##, replacing ## with A7, G11, or whatever. Stifle (talk) 09:39, 12 January 2009 (UTC)

Yes, if you know what the reference number of the criterion is, you can link to it using WP:CSD#Xn and template it using {{db-Xn}}. It's mostly a question of familiarity with the numbers. I agree that this is not something that comes overnight, which is why we're always trying to consolidate and simplify the system so there are fewer tags to remember. Happy‑melon 11:07, 12 January 2009 (UTC)

The recent changes in them in which the drop down list varies from day to day, has not helped people keep straight what to do. It might be a good idea to discuss here first before implementing one's latest ideas. When changes are agreed on, they should be made everywhere, consistently. DGG (talk) 04:14, 13 January 2009 (UTC)

The only one that's been changing from day to day is PROD, although I agree with you that that one has been changing too often for comfort. Now that we've got the sysop javascript sorted out again, things should be steady, and most importantly, since they come from one place (the templates themselves) they'll stay consistent. If we can just persuade Twinkle to adopt the same summaries, then they'll be universal (at last). Happy‑melon 10:57, 13 January 2009 (UTC)

The unified A7 description that has emerged in the dropdown is a step backwards because the basis provided is cryptic. We were much more transparent when we had four separate descriptions for A7 (1 companies and corporations; 2 people, groups; 3 web content; 4 bands, singers, musicians, or musical ensembles), and when each of them actually tracked the criteria ("that does not assert the importance or significance of the subject" as opposed to the oblique "no indication that the aricle may meet guidelines for inclusion). Or course, all of the dropdown reasons provide a link to the specific section of the criteria used. It's natural to click on that and be taken directly to the relevant section of WP:CSD so I don't see there really being a problem with associating the shorthand numbering schema to the criteria.--Fuhghettaboutit (talk) 11:59, 13 January 2009 (UTC)

If the page is tagged with a CSD tag, the autofilled reason matches the tag: I just deleted Richard Cranor, which was tagged with {{db-bio}} - look at the deletion summary. Happy‑melon 12:11, 13 January 2009 (UTC)

Maybe it's a result of my preferences or monobook scripts but in most cases I get the generic "does not meet guidelines for inclusion" or nothing and go to my offline cheat sheet. So the autofill doesn't work at least for me, and the dropdown menu doesn't contain the options to make A7 tailored.--Fuhghettaboutit (talk) 12:31, 13 January 2009 (UTC)

I'm afraid you are all proving my point. You are all busy people, I'm sure you'd like extra hands to assist on your various projects. I've been editing on and off for four years, more intensively for 3-4 months, and I have no idea what "drop-down" you are referring to, nor can I recall from memory what "PROD" is. Forcing someone to use an "offline cheat sheet" is exactly the kind of challenge the current system fosters. It raises the bar for new members to help out to a sufficiently high level as to discourage people. Is there any reason we can't link from the page describing the types of clean up reasons to the associated templates? -- btphelps ^{(talk) (contribs)} 19:40, 13 January 2009 (UTC)

You won't be able to view the dropdown menu, because it's part of &action=delete, which is only available to sysops. The text that makes it up, however, is available at MediaWiki:Deletereason-dropdown.

We do have redirects set up for most (maybe all) of the templates. {{db-nonsense}} redirects to {{db-g1}}, for example. Some others, such as {{db-person}}, {{db-web}}, and {{db-club}}, merely have more precise names themselves. The system of numbers are letters, while somewhat difficult to memorize unless one uses the CSD often, are fairly straightforward and provide a simple way of referring to each individual criterion. If you think of any template titles that are fairly intuitive and aren't already in use, feel free to create them as redirects to the proper criterion. I hope that answers your question. Cheers. lifebaka++ 20:00, 13 January 2009 (UTC)

Btphelps, where exactly is it confusing? On the tagging end, all of the criteria have intuitive names, copyvio? db-copyvio; attack page? db-attack; spam? db-spam; web, group, corporation, person fails to indicate importance? db-web, db-group, db-corp, db-bio. It's only when someone already knows the numbering scheme that they use db-[letter/number]. On the deletion end, almost all admins use either a tailored or software provided deletion reason which links to the specific section of the criteria they provide as a basis. With the link provided, there shouldn't be any confusion. As for warning users that's intuitive also; nn-warn; spam-warn; and a few others. The stuff about the dropdown menu was a tangent—it was an admin end thing. So can you provide a specific example of a common use of the naming scheme that is olikely to confuse those unfamiliar?--Fuhghettaboutit (talk) 01:48, 14 January 2009 (UTC)

Merge U1 and G7

Since we seem to be on a roll of combining and consolidating criteria, I never did understand why these were separate entities in the first place. U1 excludes User talk: pages, but G7 makes no mention of it, so currently pages in User talk: can be deleted under G7 anyway. U1 doesn't say anything that couldn't be done in a sentence in G7, and I usually get the two muddled up anyway. Thoughts? Happy‑melon 23:19, 12 January 2009 (UTC)

The author of a user page may not be the same as the user whose name it is under. More than once I've seen someone botch a move to user space and it wound up in the "wrong" user place. Granted, these could be G6'd as technical deletions but it's good to tell people "if your name is foo and you see something under User:foo/, you can have it zapped [almost] no questions asked." By the way, U1 does not apply to User_talk:, although db-author may apply to a user_talk page edited by only 1 person. davidwr/(talk)/(contribs)/(e-mail) 02:57, 13 January 2009 (UTC)

Also, G7 cannot be applied to User talk: as well, as the talkpage-user is seldom the same as the creator (who most likely was someone leaving them a message/welcome/warning). I am with David on that, U1 serves it's purpose when someone else created something in your userspace as G7 cannot be applied if you haven't created it. SoWhy 12:12, 13 January 2009 (UTC)

Agree, no need to go overboard on mergers like this. Stifle (talk) 12:38, 13 January 2009 (UTC)

No transparency

Once again I need to remind people that the speedy deletion community (which is indeed a community, as I was surprised to learn last year) is less than respectable because of a lack of transparency, and something needs to be done about it.

Every time I bring this up people complain to me that I'm just saying that mistakes sometimes happen and it's unreasonable to expect infallibility. However: it was quite by accident, rather than through a systematic process that does not exist and is badly needed, that I discovered the irresponsible deletion on December 20th of Poisson hidden markov model. From the talk page:

Logs for this page

I moved this page from Poisson hidden markov model (with a lower-case m) to Poisson hidden Markov model (with a capital M).

The logs do not move along with the edit history. That means the logs for this page fail to indicate that user:VirtualSteve deleted the page without discussion on the grounds that it was "patent nonsense, meaningless, or incomprehensible".

Obviously this was a case of "I am ignorant of this topic, therefore I will call an article in this subject area 'nonsense'". Searching "Poisson hidden Markov model" on Google would have revealed in one or two seconds that this is NOT "patent nonsense".

Deleting with no discussion just because one has no knowledge of the topic area is irresponsible to say the least. Michael Hardy (talk) 15:59, 13 January 2009 (UTC)

We don't have control over how the logs are or aren't moved here at WT:CSD. I agree that ideally logs should be moved just like edits when page moves are made, but you'll have to submit a bugzilla request for it (or find one that was already submitted). Cheers. lifebaka++ 16:57, 13 January 2009 (UTC)

So what? "we create, they delete" - this is the essence of wikipedia. Either outpace the sysops by throwing more firewood, or lose. NVO (talk) 19:43, 13 January 2009 (UTC)

end of excerpt

As I said, I found this only by accident. I was not looking through any list of speedy deletions of articles in which I could specify search terms that appear in the article or its title or its edit summaries or its category tags. No such thing exists, as far as members of this community have been willing to tell me. I was looking at the list of contributions of a new user, who had created the article.

The two-second Google search, and also a search of relevant terms in Wikipedia (Markov model, Poisson process, hidden Markov model) should have been done. Last time I suggested adding to the Wikipedia:Criteria for speedy deletion page a statement specifying some concrete things like that, that should be done to ascertain whether an article should be deleted, I was told that "you can't legislate common sense". But the problem is that such things really are needed. Calling this article "patent nonsense, meaningless, or incomprehensible" is really offensive. It says that ignorance is a qualification to pass judgment. If there is one illiterate somewhere who has never heard of a topic, it must be nonsense and it must get deleted. How is that consistent with the fact that those who have never heard of a topic come to Wikipedia to find out such things?

That earlier occasion when I was told that "you can't legislate common sense" was when someone proposed SPEEDY deletion of an article, on the grounds that it was "blatant advertising", when

• The article was created when Wikipedia was unknown and no one would have thought of using it for advertising, and the organization that the article was about was universally known and respected.
• THOUSANDS of internal links in Wikipedia linked to that article.
• THOUSANDS of external links in Wikipedia linked to the website of the organization that the article was about.
• Those links were put there in many cases by well-respected professors with expertise in the topics they were writing on.
• A highly prosperous WikiProject prescribed the use of standard templates for external links to the organization's web site.

I proposed that Wikipedia:Criteria for speedy deletion should say that one should consider those things before speedily deleting. I was told: No. You can't do that. You can't legislate common sense. But if the sort of irresponsible deletions cited above are dismissed as merely cases of inevitable fallibility, then we really NEED such prescriptions.

We also need transparency. There's no way to know how many similar cases escaped accidental discovery. We need systematic, rather than accidental, ways to know what is happening. Michael Hardy (talk) 16:26, 13 January 2009 (UTC)

• here is your permanent log. People make mistakes, man. Protonk (talk) 16:33, 13 January 2009 (UTC)

Unfortunately, there is no way to monitor the performance of admins in this area. "People make mistakes, man" is very unhelpful. DuncanHill (talk) 16:36, 13 January 2009 (UTC)

Well them's the breaks. Either propose a technical solution to the problem (namely that CSD is a category, not a persistent log and that only through a detailed review of the deletion log and deleted contributions of an admin can we judge efficacy and accuracy of their deletions) or stop harping on it. We have been over this before at WT:CSD and the result was agreement that VS screwed up in deleting that page. What's the next step? Desysop him? Hang him from the neck until dead?

So you tell me. How much more helpful is the complaint "there is no way to monitor admin performance in this area"? I'll say the same thing I did last time: if you care about the quality of speedy deletions, spend time watching the CSD page and decline speedies yourself (you can, so long as you aren't the author), spend time watching RfA and vote against people who make rash CSD tags, and spend time on NPP/RCP and improve new pages in mathematics (or whatever your area of interest is). Those are labor intensive, human solutions but they are the only ones we have. Coming back to WT:CSD to complain about this again isn't a solution to anything. Protonk (talk) 16:48, 13 January 2009 (UTC)

The problem with declining speedies yourself is that there are some really piss poor admins out there who will delete anything with a CSD tag on it, and there are enough people who will continue to re-add the tag until their article is deleted. And then they feel justified that they were correct when said admin deleted said article. I recently was talking to somebody who re-added the tag 3 times, had 3 different people tell him that it wasn't a speedy candidate. While we were trying to explain it to him, a fourth admin deleted the same article and the NPP'er basically said, "I knew that it would be deleted if I kept readding it."---Balloonman ^PoppaBalloonCSD Survey Results 21:41, 13 January 2009 (UTC)

Then those admins should be trout slapped, either for not checking the page history or for actively encouraging admin shopping (or a minor form of wheel warring even). That's not a problem of declining an SD nomination. --Amalthea 21:59, 13 January 2009 (UTC)

(ecx3) Let's see if we can cut out the hundred kilobytes or so that usually follow before we actually get to the real question: What do you propose is done about this?? You can stand on this talk page and complain until you're blue in the face, and indeed you have done so, but unless you propose a constructive response, that's all you're doing, and people will eventually start ignoring you. The one concrete idea you suggest above is not a viable response: you can't legislate common sense. No one has or (in all likelihood) will dispute the fact that the article you mention did not satisfy WP:CSD#G1; why do you think that admins are more likely to read the parts of WP:CSD that are not part of the actual criteria, than they are the criteria themselves? Allowing non-admins to view deleted material has been explicitly vetoed by the Foundation; [[ is a comprehensive, accessible (to those with the admin bit) and completely transparent way of recording every deletion that occurs here, and why the admin deleted it. What do you want to do? If you don't have anything constructive to suggest that we do about your alleged issue, please, for the love of god, let's not have another bitch fight over it. Happy‑melon 17:01, 13 January 2009 (UTC)

Part of what I proposed to do is crystal-clear: edit the policy to include the things I proposed above. Concerning the other part I will say more later. Michael Hardy (talk) 19:31, 13 January 2009 (UTC)

All right, continuing with the above: Your assertion that "you can't legislate common sense" is just dogma. That sort of "legislation of common sense" happens every time a page in the "Wikipedia" space suggests good ways of doing things. So we should edit the policy to suggest the following:

• Before deleting something on the grounds that it's "patent nonsense", search via Google and Wikipedia and the like for relevant terms to see whether it's something that might have appeared to be gibberish because you're not familiar with the subject. That can happen when the person who wrote the page is a newbie who didn't do enough context-setting and the like.
• Before deleting speedily deleting anything, check to see what links to it, and hesitate of a large number of pages link to it, since that can indicate why the topic is notable or otherwise shed light on the nature of the page.

The suggestion to look at Special:Log/delete is ridiculous. It's like suggesting that someone search for a specific topic on Wikipedia by looking at a list of all Wikipedia articles, instead of entering search terms. I just looked at the first three pages of 500 deletions each, and all 1500 of them were in less than 24 hours! We need a way to search by keywords, titles, categories, etc. Only then can persons with expertise in specific areas be able to see what's going on in cases where such expertise is needed. Michael Hardy (talk) 21:11, 13 January 2009 (UTC)

Your second suggestion is already implemented on the delete form, where it is much more visible than on a policy page. Your first suggestion is only a good idea if the article appears to be about a technical concept but does not make this clear through wikilinks etc - perhaps we could add "descriptions of technical concepts" or something similar to the list of types of article that do not qualify for G1. Recently someone proposed that the deletion log should be altered to allow the user to filter by namespace, and the developers rejected it because of performance issues, so some of what you are suggesting may not be technically feasible. Searching for categories in deleted articles will do very little because most speedy deleted pages (including the one you discuss above) are not in any categories. Hut 8.5 21:36, 13 January 2009 (UTC)

DAMN EC! Hut covered most of what I want to say, except for that if something is, indeed, actually patent nonsense or giberish, a Google or other search wouldn't be necessary to verify. For instance, it would be pointless to do a search for an article which has the text "blue monkey green blue green monkey red monkey monkey key eat skull" or some similar string. Any case where the deleting admin wants to perform a search to check simply shouldn't be speedily deleted, regardless of the results of said search. Cheers. lifebaka++ 21:42, 13 January 2009 (UTC)

The fact that we delete thousands of pages every day (over four thousand this afternoon, for instance) is indicative of how important deletion, and CSD in particular, is in keeping this wiki from being silted up with sludge. I don't think you really realise how massive this operation is. Correct me if I'm wrong, I don't mean to speak for you here, but I think you see CSD as a minor process involving a 'clique' of perhaps a few dozen admins that does as much harm by deleting good articles as it does good by deleting bad ones. CSD is a process that involves hundreds of admins, the majority of whom don't even watch this page. They delete thousands of articles every day, a process which, if stopped or even significantly slowed, would destroy XfD under the additional load. Now, given that you get what you pay for, I think that that cadre of unpaid volunteers does remarkably well, both in accepting the complete lack of recognition when they 'get it right', and in accepting the sometimes-deserved-but-often-entirely-unjustified flak when they 'get it wrong'. To cut another hundred kilobytes out of this discussion, I'll jump to another question that we always ask when you start these threads: why don't you get involved yourself? Happy‑melon 23:31, 13 January 2009 (UTC)

Just thought of something related to the sortable logs idea. Something could be whipped up on the toolserver that could search a users deletion log for items with a certain phrase in the title, such as namespace, (assuming it doesn't already exist), which would at least help make the process more transparent. Or an external script or some such could work on the same using database dumps, though it would be slightly out of date all the time. Would either of these ideas be tenable? lifebaka++ 03:46, 14 January 2009 (UTC)

I and others have from time to time checked the log, either for all admins, or for specific users. and there are even a few of my fellow admins whose deletion logs I spot check on an irregular basis. In fact, when asking the mop, I said specifically that i wanted to be able to follow this up by examining the articles. The error rate depends on what one calls errors. I & others find there are about 5% - 10% of speedies that do not conceivably deserve to be speedied at all on any criteria, about half to a third of which would probably make it as articles. The reasons are divided between carelessness and deliberate refusal to follow policy. Sure, it would help to have better sort criteria. It wouldn't be easy to do this, because one can give any non-standard deletion reason. The only reason I decided not to follow this up in detail, to be perfectly frank, is the desire to avoid making enemies--and the knowledge that I too make mistakes, as can be seen from the occasional justified complaints on my user page. As for searching before deletion, i do it when I think it advisable to check myself. So I do not accept Lifebakas view that if I need to do a search , it isn't a valid speedy. I prefer to play it safe, and double check, just in case the article, while not indicating notability, might be about something notable, or whether a G11 or copyvio is worth rewriting. DGG (talk) 22:04, 13 January 2009 (UTC)

Michael Hardy, you've complained many a time here when one or two of the articles you've created has been deletedarticles you are interested in has been deleted. However, you've left out the fact that the deletion has been reversed, the person who made a mistake has realised it, and people have improved. If you want to propose a concrete solution to a specified problem, please do. So far, all I can see is "tell people to be careful when speedying articles", which we already do, and "make a better way of searching through deleted articles", which would have to go to bugzilla. I'm really not trying to be dismissive of a reasonable grievance, but you are the only person who is regularly here complaining about a very small percentage of speedy deletions being wrong, and sometimes it's time to drop the stick and back slowly away from the horse carcass. Stifle (talk) 16:42, 14 January 2009 (UTC)

I have NEVER complained here when an article I created has been deleted. In fact, I don't know of any case in which an article I created has been deleted. Michael Hardy (talk) 02:29, 15 January 2009 (UTC)

Apologies. I have corrected the statement accordingly. Stifle (talk) 09:03, 15 January 2009 (UTC)

I think there is a legitimate point buried in here, which is that it would be great to keep a log of all speedy deletions which were subsequently reversed; this would help to inform policy discussion and be a way to teach new admins about common mistakes. This can probably be queried for pretty easily automatically or semi-automatically. We could have a subpage here and link it, and group them by criterion. Dcoetzee 10:03, 15 January 2009 (UTC)

Where is the alleged list?

Someone has told me:

CSD is an area often back-logged and I was working through a large list of such items.

Can someone tell me where this list is? I've looked over the CSD page and I don't find any obvious links to it. Michael Hardy (talk) 23:23, 13 January 2009 (UTC)

There's only the dynamic list at CAT:CSD#Pages in category. --Amalthea 23:31, 13 January 2009 (UTC)

Ad hominem

I was surprised to find this as a criterion for speedy deletion:

Banned user. Pages created by banned users in violation of their ban, with no substantial edits by others.

That is a clear case of the ad hominem fallacy. If Adolf Hitler points out that 2 × 3 = 6, the fact that Hitler is evil does not diminish the cogency of his observation. If an article is in all other respects a valuable contribution to Wikipedia, then readers are to be deprived of good information on grounds altogether irrelevant to the merits of the article.

This criterion should be deleted. Michael Hardy (talk) 00:20, 14 January 2009 (UTC)

Any editor, including an admin, is free to undo and re-do the edit, preserving a record in the edit history that the edit is worth keeping. The reason banned editor's contributions are deleted is without that, a ban has no meaning. Since we can't fine or arrest people for trespass like people do in the real world when they are banned, we silence them. davidwr/(talk)/(contribs)/(e-mail) 00:34, 14 January 2009 (UTC)

That doesn't make sense. It does not answer the objection I raised above as to why this rule is fallacious, and the ban is still just as effective in preventing the banned use from vandalizing pages or doing whatever other harm is to be prevented by the ban. Michael Hardy (talk) 02:55, 14 January 2009 (UTC)

The criterion is longstanding and in line with the current wording and intent of WP:BAN. I'd personally love to see the way we handle bans changed myself, but the chances of getting a change here before a change there is fairly slim. Cheers. lifebaka++ 03:48, 14 January 2009 (UTC)

It's necessary to keep our cult status with the IRS. Wait, what? --NE2 04:03, 14 January 2009 (UTC)

Imagine this scenario: a user who was banned for tendentiously pushing original research begins creating articles describing more of their original research. There's no CSD under which we could delete this, so it has to go through AfD, perpetuating the very same disruption and needless conflict that their ban was intended to eliminate. Since banning is already our last resort, without this criterion there's no way we can really prevent a user from disrupting Wikipedia. Dcoetzee 06:34, 14 January 2009 (UTC)

Also, as with all articles, no administrator HAS TO delete such creations. If a previously banned user makes a good edit as a sockpuppet, we can keep it if keeping the edit is more positive than enforcing the ban. As Dcoetzee points out, the criterion serves its main purpose in cases where the banned user comes back to violate their ban by disrupting again. Regards SoWhy 17:47, 14 January 2009 (UTC)

Quite. It is fallacious and somewhat disingenuous to assume that anything that could be deleted under a CSD, must be deleted. From WP:BAN: "By banning a user, the community has decided that their edits are prima facie unwanted and may be reverted without any further reason. This does not mean that obviously helpful edits (such as fixing typos or undoing vandalism) must be reverted just because they were made by a banned user, but the presumption in ambiguous cases should be to revert". Substitute "edit" with "page" and "revert" with "delete" and you have the admins' version. We do not ban users because we believe they are fascist world-dominating megalomaniacs, we ban them because we do not trust them to claim that 2x3=6 without having an ulterior motive for doing so. Banned users have proven themselves incapable of making constructive edits, to the extent that we make a "formal revocation of editing priviledges". What does that mean, exactly, other than "your edits may be reverted on sight"? Happy‑melon 19:37, 14 January 2009 (UTC)

Not quite true User:Everyme was banned not because of vandalism or issues along those lines, but rather because he was curt with people and didn't give a damn about civility. His edits were generally in good faith, but his attitude torpedoed him.---Balloonman ^PoppaBalloonCSD Survey Results 19:51, 14 January 2009 (UTC)

Yes, which is where Happy-melon's first sentence and SoWhy's comment would apply. Mr.Z-man 23:44, 14 January 2009 (UTC)

CSD'er of the Week

Some of you may have noticed that over the past month I've been awarding people the CSD'er of the Week award. This award is given for excellent work at CSD. The Award Criteria can be found here, but basically it is given for adhering to the policy and not relying upon "IAR" to delete articles because you can. So far, we've had four winners. I'm doing this with the hope of inspiring people to improve their practices. Right now, the award is only going to admins (as they are the one's who can do the most damage via deletion) it will be opened up to non-admins in a few weeks.---Balloonman ^PoppaBalloonCSD Survey Results 14:51, 14 January 2009 (UTC)

• Cool. Issue the star, than CSD the awardee. NVO (talk) 19:41, 17 January 2009 (UTC)

Why is WP:CSD#A1 restricted to very short articles?

I occasionally look through the speedy deletion category looking to rescuing articles that deserve more consideration, but I sometimes come up against articles where, in respecting policy, I have to remove WP:CSD#A1 tags from articles that have no context but are not "very short" as required by the criterion. I don't see any reason why articles of any length that fail to provide sufficient context to identify the subject of the article shouldn't be speedily deleted. Can anyone justify the "very short" get-out? I've also noticed that the inclusion of this wording means that many new page patrollers concentrate on the "very short" aspect and tag under this criterion even if context is there - I mean articles such as "AAA is a village in BBB" or "CCC is a book by DDD". Removing the "very short" would make it clearer that context is the important part of this criterion, not article length. Phil Bridger (talk) 21:25, 14 January 2009 (UTC)

I agree. It should be the lack of context that is important, not the length. FlyingToaster 23:44, 14 January 2009 (UTC)

Don't quite agree with the first part of the logic (an article of decent length should be long enough for you to figure out what it's about with or without overt context), but you're spot on for the second bit. Removing the text should avoid confusion. If there's no objection (which I doubt there will be much of) we can make the change in a few days or so. Cheers. lifebaka++ 02:30, 15 January 2009 (UTC)

If the article is more than a sentence or two, then it would have enough that you should be able to identify the subject. If you can't. then it is possibly wp:nonsense But make sure you know what WP:NONSENSE is before noming something as such. G1 is the category that gets the most abuse.---Balloonman ^PoppaBalloonCSD Survey Results 02:32, 15 January 2009 (UTC)

I think Phil Bridger needs to give some examples of the kind of articles he means, just to show if its possible for a decent length article to have no context. Epbr123 (talk) 17:25, 15 January 2009 (UTC)

Either way, I support a change to the criteria. If an article has no context, it should be deleted for having no context, regardless of its length. How is "very short" defined, anyway? Epbr123 (talk) 20:05, 15 January 2009 (UTC)

I don't have any examples to hand, but I have seen example of quite lengthy articles written about subjects that exist in some fictional universe or belief system whose subjects can't be identified because the context isn't provided, even though the language used is quite coherent, so they wouldn't qualify as patent nonsense. On thinking further I accept there are probably not enough of these cases to make speedy deletion necessary, but I still think that the wording of the criterion should be changed so that taggers concentrate on the lack of context rather than article length. It seems that many of our new page patrollers don't have the attention span to read past the first few words of a criterion, so the important part should be presented first. How about

Articles lacking sufficient context to identify the subject of the article. Example: "He is a funny man with a red car. He makes people laugh." Context is different from content, treated in A3, below. This will usually apply only to very short articles.

Phil Bridger (talk) 19:54, 16 January 2009 (UTC)

• Here's an example that I just came across of an article that isn't very short but provides no context. Phil Bridger (talk) 01:03, 18 January 2009 (UTC)

Well, this article doesn't fit A1... just because you are not a Muslim and don't understand it doesn't mean that it doesn't tell you exactly what it is in reference for. It is instructions on how to Performing Hajj & Umrah---religious ceremonies. The article is, however, a how-to manual and thus should probably be AfD'd, but right now I don't have time to do so. But, this is the an example of why A1 doesn't work on long articles.---Balloonman ^PoppaBalloonCSD Survey Results 01:21, 18 January 2009 (UTC)

But it doesn't tell us what "Ameen Aziz" is, which is supposedly the subject of the article. Phil Bridger (talk) 01:30, 18 January 2009 (UTC)

• Here's a better example of what I mean. Phil Bridger (talk) 22:03, 19 January 2009 (UTC)

WP:CSD#A7 Needs to be more Useful

There's been some past discussion on it, but this has not yet moved forward... so I'm bringing this up again with the hope of leading to actual policy change. Why does WP:CSD#A7 only relate to "real people, organizations (e.g. bands, clubs, companies, etc., except schools), or web content? This seems a completely arbitrary collection of things which may not be notable, and our efficiency at getting rid of poor content is hampered as a result. For instance, why can "web content" fall under this category, but not "magazine content?" Why can I mark a person or company under this category, but not a pair of pants, a hamster, a song just made up, a slang word used in a particular school? Instead, let's make A7 apply to any entity that is obviously not notable. Seems silly that we can tag a non notable person as A7 and watch it be speedily deleted, but an article about a new dance step needs to be WP:PRODed or WP:AFDed. Thoughts? FlyingToaster 23:41, 14 January 2009 (UTC)

Relevant past conversations: [1] [2]

Short answer: Because there was no consensus for it.

Long answer: Because those were the only instances that we agreed to that admins should be able to judge whether importance was asserted. Please note that this does not mean notability but something less, whether something is notable or not is never decided, only that there is no indication that it might be. And as usual leaving an article for PROD does not harm us at all, so there is no reason to increase the scope of A7 just because we have to PROD now. Leaving a bad article for five days does not really hurt us but it allows others to expand on it, so why change it? SoWhy 23:50, 14 January 2009 (UTC)

Indeed, you are right that this is not an exact call for notability. But I don't see why non-admins can be expected to be good judges of whether there is indication for whether people are important, but not a pair of pants. My main complaint with WP:PROD for articles which are obviously not notable is that in five days everyone forgets, the original author deletes the tag, and the article either stays or stays way too long. FlyingToaster 00:16, 15 January 2009 (UTC)

To be fair, A7 is not "this subject isn't notable," it's "this article doesn't point out why the subject is notable." That said, I've never considered the exact mentioned items to be a limiting factor when deleting; if someone posts an article consisting of "My dog Ralph is a golden retriever," then yes, I'm going to delete it as an A7, exact verbiage be damned. EVula // talk // ☯ // 23:52, 14 January 2009 (UTC)

Indeed, EVula. And I do find that the majority of admins do the same. But every now and then I get a comment like "cannot delete under A7: subject is a hamster and not a person" that makes me sigh. It's a place I'm happy to WP:IAR: I just wish the rules were a bit better so 1. we don't have to always IAR and 2. the very few people that do not IAR here slow the process of deleting bad content. FlyingToaster 00:16, 15 January 2009 (UTC)

Wrong. A7 has nothing to do with notability. --NE2 02:21, 15 January 2009 (UTC)

The article about the hamster is probably deletable as vandalism or A3. But the reason why we allow admins to do people, websites, groups, and companies is because these are the ones that are likely spam/vanity candidates. Personally, I would tighten the criteria for A7, as it is often abused. A7 does not require that the subject meets our notability requirements, in fact the threshold is explicity set lower than notability, all it has to do is assert importance/significance.---Balloonman ^PoppaBalloonCSD Survey Results 02:30, 15 January 2009 (UTC)

The reason is that those three items - real people, organizations, and web content, get a lot of "novelty articles" and by requiring people to at least bother to assert significance, it gets some of them to read the notability criteria and not bother writing articles about non-notable subjects. In the cases where that doesn't work, it provides a speedy deletion which in turn deters other novelty writers. The reason schools are exempt is because so many schools, including almost every government-run school offering high school diplomas and just about every accredited post-secondary school offering degrees, are "inherently notable." Well, not really inherently notable but you would have to be willfully blind to not be able to find enough of an indication of notability to pass AFD. See Wikipedia talk:WikiProject Schools/Criteria for Speedy Deletion A7 for more discussion. The reason other types of articles are not A7 is because they aren't abused as much. As for non-notable pets, non-notable places like "My house is a 2-story house in Springfield where me and my family live" etc. that do not assert notability, usually there is another criteria they fail, but if there is not, PROD works nicely.

By the way, when in doubt, don't speedy. For A7 and A9, this also means when in doubt of the actual notability of the article, don't speedy. If Johnathan A. Doe doesn't assert notability, but you think he could be notable, don't speedy. Either improve the article or tag it {{notability}} and perhaps PROD it. davidwr/(talk)/(contribs)/(e-mail) 02:40, 15 January 2009 (UTC)

davidwr - So it doesn't bother you that articles like The Tale of Gregory Lovell fall under none of the criteria, and thus will stay up until their prod expires? Gotta say it seems bizarre that if the writer had mentioned their story was on a website or blog, it could be deleted as "web content." As it stands... we have to wait, and I do think Wikipedia is ever so slightly worse the longer articles like this are left standing. FlyingToaster 07:34, 15 January 2009 (UTC)

Not at all. By waiting for the prod to expire, you give the author a chance to salvage the article. This means the author doesn't feel bitten. They had a chance to save the article. Also, while in the current stance, it deserves to be deleted, but does it do the project any harm? No. Now what if there was more to the story than we know? What if there is more to the story that actually makes it notable or interesting? The problem with a lot of articles is that it isn't as cut and dry as some people think... especially, during the first few hours of an articles existence---when most articles are tagged by NPP for CSD. I would rather have that article on WP for 5 days, than to delete a similar article 2 minutes after creation only to discover that the article would have been worth keeping, but we've now lost an editor. Keeping that article prodded for 5 days, does not harm the project. It's prodded so outsiders will know it doesn't meet our standards, thus it doesn't damage WP's reputation. Thus it does no harm. If it does no harm, why not give the author the benefit of the doubt, and afford them every opportunity to salvage it? And if, as you suggest, "Wikipedia is ever so slightly worse" for having this article, I think the cost of deleting and losing a potential contributor is worse.---Balloonman ^PoppaBalloonCSD Survey Results 08:13, 15 January 2009 (UTC)

Nicely put. As I, too, said above, keeping it for 5 days does not do any harm and allows editors time to rescue it. It only harms us if it violates BLP/copyright or spams but in those cases we can G10/G11/G12 it. There is no need to ignore the criteria under IAR just because they don't fit if the alternatives to speedy deletion do not actually harm anyone. I think it is very disturbing that some people think that we need to get rid of bad articles on sight...but cannot explain why the project is harmed if they aren't. Regards SoWhy 09:01, 15 January 2009 (UTC)

No, it doesn't bother me at all. If we had a bunch of volunteers who were good at AfD who stepped up and said "I'm not much of an article writer but I'll participate in AFD" then we wouldn't need A7/A9 at all, we could PROD and AFD all of these. Personally, I'd like to see Speedy Deletion split into "gotta zap it" stuff like copyvio and attack, and a slightly slower "it won't kill anyone if it lasts 24 hours" 1-day-prod which would be a dated tag like dated prod. This would give editors at least 1 wake/sleep cycle to address the problem.

I agree with the idea of a new PROD - one that's a little slower than a speedy but still doesn't allow dumb articles to remain up there for long. However, I still don't see any reason why certain articles are excluded from A7. There's no reason why an article about something that doesn't assert its notability - whether or not it's a person or group - shouldn't be speedied. Speedy deletion is there for a reason, and we shouldn't keep bad articles around for the sake of not biting the newcomers. How is speedying an article about a corporation any different, bite-wise, than speedying an article about a word someone made up one day? Graymornings_(talk) 00:13, 17 January 2009 (UTC)

The restrictions on A7 are there for a reason, as explained; I think it's okay sometimes for an admin to - at their own risk - invoke the snowball clause and get rid of a clearly useless article, but attempting to codify this process is notoriously tricky. Another reason that A7 is restricted to only certain topics is because those cover a good majority of new articles that are missing assertions of significance; the remaining ones are not so numerous as to overwhelm other methods of deletion. Dcoetzee 09:58, 15 January 2009 (UTC)

Let's say we have an article We have a band in our garage and a couple of daughter articles Our band once burned its music on a CD. The former gets {{db-band}}ed. The latter gets {{prod}}ed, deprodded, listed on AfD, "debated" for five days, then deleted. WTF?

If the album was a notable Grammy winner, major chart hit, etc, the band would be notable too. A notable band can have a non-notable album but the reverse is a bit hard to imagine. Prove me wrong: list some encyclopedically notable albums where the band does not pass notability criteria. I dare you :-). The rule "speedy the band, AfD the CD-ROMs" is kind of backwards to begin with. 88.112.34.160 (talk) 15:41, 18 January 2009 (UTC)

Actually, A9 could be used to remove the albums, assuming they have no claim to importance beyond being albums by said band. Cheers. lifebaka++ 15:46, 18 January 2009 (UTC)

I had an idea about A7

Quite a bit of A7'd articles seem to be autobiographies. Could we just do a "db-autobio", which is essentially db-person (A7) with the phrase "and which seems to be created by the subject"?. Oh yes, I am not proposing a new Asomething, but rather, an extension of A7. Narutolovehinata5 ^tccsdnew 10:05, 15 January 2009 (UTC)

As WP:COI mentions, an autobiography can be a conflict of interest, but is not forbidden; considering that A7 already covers the cases with no assertion of significance, I don't see any clear reason for a redundant extension. Dcoetzee 10:52, 15 January 2009 (UTC)

Agree. Autobios can be written by notable subjects after all; A7 applies to all articles regardless of who authored them but applies only when there is no claim that the subject may meet the guidelines for inclusion. SoWhy 18:56, 15 January 2009 (UTC)

Overturned speedy deletions

To help provide some useful feedback on why some speedy deletions are overturned, for the purpose of discussion and training, I've started a new subpage called Wikipedia:Criteria_for_speedy_deletion/Overturned_speedy_deletions. This page lists articles whose speedy deletions were overturned in DRV, organized by criterion, along with their delete/restore date and summary, link to DRV discussion, and brief summary of why it was overturned. I've only done maybe a couple weeks of DRVs, let me know what you think. Dcoetzee 12:26, 15 January 2009 (UTC)

Interesting. It could be useful to get an idea of our most problematic criteria, if trends emerge. --Moonriddengirl ^(talk) 13:00, 15 January 2009 (UTC)

Moon, if you haven't looked at the study linked at the end of my name you might want to... it also has some other "surveys" that I've done on the subject.---Balloonman ^PoppaBalloonCSD Survey Results 01:37, 16 January 2009 (UTC)

I actually tried to participate in that one, but I think I must have done something wrong, since it didn't look like my comments made it. :) --Moonriddengirl ^(talk) 23:42, 19 January 2009 (UTC)

Why only those that were overturned in a DRV? Why not include those that were undone by another administrator because the deletion was unjustified? Michael Hardy (talk) 03:22, 16 January 2009 (UTC)

I've started including some of those too - the DRV ones are ideal because they've had input from a number of people who agree it was in error, and they're easier to find. I've been trawling through deletion logs though and I now have several not listed at DRV. Dcoetzee 07:27, 16 January 2009 (UTC)

what is scary is the thought that these are just the ones that are overturned... there are a large number of items that are speedily deleted that shouldn't be, that are also unworthy of being recreated.---Balloonman ^PoppaBalloonCSD Survey Results 07:29, 16 January 2009 (UTC)

Keep in mind that not all of these represent incorrect deletions; in some cases they were restored because someone promised to improve it or new facts came to light. They really deserve careful study. Dcoetzee 07:44, 16 January 2009 (UTC)

Which tells me that some of them were probably deleted WHILE the author was working on said article, which is where a lot of people get turned off of WP.---Balloonman ^PoppaBalloonCSD Survey Results 23:56, 16 January 2009 (UTC)

A7 autofill/deletion dropdown: opaque and shouldn't refer to "notability"

On the dropdown end, I understand there's a scrolling issue when the description is made longer (and something been added recently making it scroll wide again) but the deletion summary "CSD A7: No indication that the article may meet guidelines for inclusion" is fairly opaque. I think providing a more transparent and targeted deletion summmary takes prominence over a scrolling issue. Also, I just bet there's a workaround to make the dropdown box wrap in some way of otherwise be fixed so it doesn't have to be the same length as its longest included summary. Some coordination with WP:VP/T might get somewhere. Let's make it four summaries (as it used to be) with some real criterion tracking language, such as:

• [[WP:CSD#A7|A7]]: Article about a real person or persons that does not assert the importance or significance of the subject
• [[WP:CSD#A7|A7]]: Article about a company or corporation that does not assert the importance or significance of the subject
• [[WP:CSD#A7|A7]]: Article about web content that does not assert the importance or significance of the subject
• [[WP:CSD#A7|A7]]:Article about a band, singer, musician, or musical ensemble that does not assert the importance or significance of the subject (or maybe this should be folded into the real person description, now that that has been done in the criterion)

With regard to the "other reason" autofill (where are these located in order to edit them?), when it works, which for me is rarely for reasons unknown, it is more tailored but it doesn't track A7, instead stating for bios "[[WP:CSD#A7|A7]]: Article about a real person, which does not assert notability" (emphasis added) and similar descriptions using "notability" for other A7 subjects. It is an oft repeated canard that A7 and notability are totally unrelated, but nevertheless, we should not be using that wording in the autofill. These should likewise use "importance or significance" language, tracking the actual criterion.--Fuhghettaboutit (talk) 12:42, 15 January 2009 (UTC)

I implemented a fix to MediaWiki:Sysop.css to prevent the box overflow, but it was reverted because it broke some browsers (see the history). Perhaps a useful line of enquiry would be why it "breaks in webkit browsers" and whether there is an alternative. I agree that the traditional "does not assert the importance or notability of the subject" wording is preferable, but I don't think it is necessary to have four separate options for the four types of subjects. The category that the article falls into is usually obvious from the title, and is certainly clear from the deleted page content; the only advantage to spelling out the subcategories is to reinforce the point that only elegible subjects can be deleted under A7, which although a recurring problem is not one that I think will be remedied by simple technical measures - it is as always a matter of education and correction of wayward admins. I would support the wording "article about an elegible subject that does not assert the importance or significance of the subject".

The autofilled "other reason" field is set on the CSD templates themselves; the |summary= parameter content is appended to a wikilink to the criterion and posted into a hidden span given the "delete-reason" id. That span content is picked up by javascript in MediaWiki:Sysop.js and the link in the 'delete' tab is modified to pass that content on to the 'other reason' field when the link is clicked. The same content is appended to the 'delete' link on the template. So simply navigating directly to http://en.wikipedia.org/w/index.php?title=Foo&action=delete, or getting there by any fashion other than clicking the 'delete' tab link, will not autofill the summary. How do you access the delete confirmation screen? To answer your question, though, in general to edit the autofill reason that appears when a particular CSD template is on a page, edit the |summary= parameter on that template. However, I think we should be trying to avoid continually changing these summaries, and particularly trying to keep them consistent, such that the deletion summary for a particular situation is similar or identical no matter how you reach it. Happy‑melon 17:42, 15 January 2009 (UTC)

Fixed both. I see that it would be difficult to tailor the autofill message because all the specific a7 templates draw from the summary in db-a7. I would prefer more narrowly tailored but can live with eligible subject just fine. Thanks for the info on where the autofill is. Now let's see how long it takes for the dropdown to be reverted based on the scroll.--Fuhghettaboutit (talk) 23:59, 15 January 2009 (UTC)

Proposed change to I10 (Useless media files) to respect copyleft

As of today, I10 is worded as follows:

Useless media files. Files uploaded that are neither image, sound, nor video files (e.g. .doc, .pdf, or .xls files) that are not used in any article and have no foreseeable encyclopedic use.

But a .xls file that OpenOffice.org Calc can open may contain the original data for a graph in PNG or SVG format, and some copyleft licenses used for images here and on Commons require that the original data be preserved in an editable form. So I hereby propose an amendment to clearly exempt a file's original data from this criterion:

Useless media files. Files uploaded that are neither image, sound, nor video files (e.g. .doc, .pdf, or .xls files) that are not used in any article, are not the transparent copy or source code of a file used in an article, and have no foreseeable encyclopedic use.

Any objection? --Damian Yerrick (talk | stalk) 14:38, 15 January 2009 (UTC)

I'd rather see a more general assertion such as "encyclopedic use, and are not required for license compatibility", if anything at all. I'm not aware of this being a widespread problem - is it even an evident problem, or is it a theoretical exercise? A note on the file description page explaining the situation and why the file needs to be kept is probably more useful (and likely to be more successful) than a rather technical addition to the criterion here. In a notification template you have the ability to explain the situation in plain english (including how its existence thereby has "encyclopedic use", as required) and also add tracking categories etc. Probably a better solution than playing with the criterion to proscribe individual exceptions. Admins are supposed to have common sense, after all. Happy‑melon 17:49, 15 January 2009 (UTC)

Not to start up anything but there was a strong indicator from a few admins and editors that terms such as "encyclopedic use" were frowned upon as it would relate to a situation such as this. In this case, as currently worded, to be considered "useless" the file must not be an image, sound or video file and it must be unused. The other qualifier is the questionable one because it allows one person alone to determine the file has no "encyclopedic use". Yes, details can be given in the actual notification, but when the file is tagged it should also be very clear what files should be included and what file should not be. That's why I would like to see the wording of have no foreseeable encyclopedic use removed and am for for wording along the lines of what Damian Yerrick has suggested in order to start to zero in on what a files contents must contain in order to be not considered part of "Useless media files". Soundvisions1 (talk) 20:07, 15 January 2009 (UTC)

I don't see this as necessary. In the described situation, the file clearly does have an encyclopedic use. On top of that, I don't think .xls files can be uploaded here. Stifle (talk) 20:18, 15 January 2009 (UTC)

No, they can't, and there are a grand total of 8 left. I can't see the point of making a change in policy that could only possibly impact 8 pages. Hut 8.5 21:13, 15 January 2009 (UTC)

I agree with this change, and my concern here is not so much license compatibility as editability. If someone for whatever reason chooses to build media content using proprietary software, it's really convenient for someone else who has that software to be able to load it and make small changes. The alternative of making changes directly to the media is frequently nearly impossible. And it's not realistic to mandate that all media is created with free software; people use what they know how to use, and some features are only available in proprietary software. Another good example of this is images generated by a program, where we always want to have the source code available in case we want to generate it differently. However, the restriction on extensions of uploads makes this sort of thing problematic. Dcoetzee 21:42, 15 January 2009 (UTC)

I agree with the intent of the change, but how many such pages are there which this would cover? If the number is fairly small, {{go away}} would be a simpler solution than changing the criterion here. Additionally, I prefer Happy-melon's more general wording, as there are licenses other than the GFDL and GPL we might have to worry about (most likely in the case of fair-use images). Cheers. lifebaka++ 15:50, 18 January 2009 (UTC)

Change on removing CSD tag and hangon

This is big change. Did I miss the discussion on this? It seems like a poor choice that only an admin can remove the CSD tag if incorrectly placed. HANGON should be for the creator. If anything, it is already to easy to speedy delete an article that does not really meet CSD criteria. Dlohcierekim 01:26, 16 January 2009 (UTC)

I think everyone missed the discussion on that - or rather, there was no such discussion at all. I have reverted the change. DuncanHill (talk) 01:30, 16 January 2009 (UTC)

Concur with revert. There really needs to be broad discussion before making such a change. Dlohcierekim 01:31, 16 January 2009 (UTC)

Absolutely. I don't see that many incorrect removals of CSD tags to warrant that change. If it's removed in good faith, it's typically not a non-controversial deletion. And if it shouldn't wait (G10, G12) then there's always AN/I, or a swift AfD. --Amalthea 01:36, 16 January 2009 (UTC)

Not to mention that such a change to this policy only would create a policy inconsistency, since Wikipedia:DP#Speedy deletion says "Anyone except a page's creator may contest the speedy deletion of a page by removing the deletion notice from the page." --Moonriddengirl ^(talk) 01:38, 16 January 2009 (UTC)

Since I joined Wikipedia (in 2005) editors were not supposed to remove speedy tags. As far as I can tell, someone just added a notice here changing that. That change had no consensus I could find. Naturally if the CSD is obviously wrong anyone can remove it. Otherwise we shouldn't be removing speedy tags from a page just because we don't think it should be deleted. It should only be removed if it doesn't meet the CSD. Prodego ^talk 01:39, 16 January 2009 (UTC)

It's been this way since before I was an admin. Cheers, Dlohcierekim 01:44, 16 January 2009 (UTC)

... at least since July 30, 2007. --Amalthea 01:46, 16 January 2009 (UTC)

There's been some other discussion about this lately. My understanding is that anyone except the page creator can remove a CSD tag, but only if they contest that it is valid, and they are generally expected to explain their reasoning; the bar is higher than it would be for removing a PROD tag, but is not a prohibition. Dcoetzee 01:45, 16 January 2009 (UTC)

<<ec>>Naturally if the CSD is obviously wrong anyone can remove it. Exactly. If any user besides the creator removes the tag, deletion is too controversial for a speedy. Dlohcierekim 01:47, 16 January 2009 (UTC)

(ec)Admins are grossly overworked (or so some of them keep telling us). The removal of incorrect speedy tags by non-admins reduces the workload on admins, as well as reducing the chances of an admin incorrectly speedying an article. DuncanHill (talk) 01:49, 16 January 2009 (UTC)

but but but... I thought the lesson of CSD was to keep re-adding the tag until SOMEBODY actually deletes the article...---Balloonman ^PoppaBalloonCSD Survey Results 01:50, 16 January 2009 (UTC)

Well, that does seem to work :( DuncanHill (talk) 01:51, 16 January 2009 (UTC)

The key thing is that only someone who is both knowledgeable about the CSD (which any admin clearing C:CSD must be), and uninvolved with the page (which any reviewing admin is also virtually ensured to be). It should not be open to the sort of removal prod tags are open to. Unlike proposed deletion, the CSD are either met, or aren't. Only someone competent to judge that may remove the tag. If you tag an article as U1 deletion, anyone can judge that is wrong. But if you tag a page as A7, or A9, or G10, etc, it isn't going to be so obvious. The idea is that you put it on CSD once someone who is highly experienced looks at it once, and then it is either deleted, or that reviewer sends it to AfD or prod. Prodego ^talk 01:52, 16 January 2009 (UTC)

I've removed plenty of blatantly incorrect speedy tags, and will continue to do so. I don't need some trumped-up popularity contest like RfAd to tell me that I can understand the criteria. DuncanHill (talk) 01:55, 16 January 2009 (UTC)

And "and then it is either deleted, or that reviewer sends it to AfD or prod" - so no possibility in your view that the reviewer decides that it isn't a candidate for deletion at all? DuncanHill (talk) 01:56, 16 January 2009 (UTC)

No, whoever reviews that CSD is obligated to send it somewhere. It could be prod, or AfD, but it has to go somewhere, assuming it is even mildly reasonable (e.g. not U1 on an article). Prodego ^talk 01:58, 16 January 2009 (UTC)

No, that's nonsense. Are you saying that you send articles to AfD when you do not believe they should be deleted just because someone put an incorrect speedy tag on it? DuncanHill (talk) 02:03, 16 January 2009 (UTC)

Any editor acting in good faith should be able to remove a speedy. Likewise, any editor who in good faith believes a speedy was removed in bad faith or by someone who was clueless should be able to restore it, but the restorer has a burden to explain why the removal was improper. Restoration is especially allowed in cases like blatant attack or proven copyvio. Issues of "good faith" or "bad faith" shouldn't be that common, and can be handled on a case-by-case basis not by rules within CSD. In any case, the deleting administrator should consider the history of tag additions and removals before deciding to delete the file. davidwr/(talk)/(contribs)/(e-mail) 02:02, 16 January 2009 (UTC)

Agree with DuncanHill. ((are you an admin, if not maybe I should nom you.)) whoever reviews that CSD is obligated to send it somewhere. Absolutely not. If the tag is wrong, it's wrong. You don't have to be an admin to see that. I the original tagger thinks it should be deleted, it's up to them to PROD or AFD. Dlohcierekim 02:11, 16 January 2009 (UTC)

Not an admin and would be first to oppose myself if nominated. But kind of you to suggest it. DuncanHill (talk) 02:15, 16 January 2009 (UTC)

←With respect to Prodego's concern that someone just added a notice here changing that, language allowing anyone to remove the tag was added here, in April 2007. It was changed to reflect deletion policy, to which it had been added here in March that same year, as a part of the language overhaul discussed here, evidently publicized in various forums as well as its own talk page. -Moonriddengirl ^(talk) 02:14, 16 January 2009 (UTC)

If someone raises a valid concern about an article, but it doesn't meet CSD, you had better not just remove the deletion tag. Prodego ^talk 02:24, 16 January 2009 (UTC)

Then the editor with the concern can go to AfD, the person removing the tag is under no such obligation. DuncanHill (talk) 02:28, 16 January 2009 (UTC)

That isn't exactly a very helpful attitude. Prodego ^talk 02:53, 16 January 2009 (UTC)

Prodego, what you're suggesting is patently unreasonable, though. If a db- tag is inappropriate, a reviewer should remove it. Sometimes I'll PROD or AFD it, but large numbers of pages tagged for speedy deletion just aren't viable candidates for any form of deletion. No one should ever take an article to AFD if they don't believe it should be deleted. WilyD 03:10, 16 January 2009 (UTC)

Do you explain why the tag is removed? When I de-speedy, I try to explain it either in the edit summary, the article talk page, or the talk page of the editor who speedied it in the first place. This is easy in obvious cases, less so in cases like A7. With those I typically wind up slapping a PROD or AFD on it anyways if I think it should be deleted but there was an assertion of importance. davidwr/(talk)/(contribs)/(e-mail) 04:10, 16 January 2009 (UTC)

Yeah, in the edit summary typically. Very occasionally I'll chastise the tagger if they're absolutely out to lunch. But usually just "Poland is notable" when declining an A7 or whatnot. WilyD 04:32, 16 January 2009 (UTC)

I usually put it in the edit summary. I notify the tagger, but not as often as I should. I may add a talk page note if it's something extra-ordinary. Dlohcierekim 04:13, 16 January 2009 (UTC)

No if the page doesn't merit a prod/afd, in the eyes of the person who removed the tag, then they are not under any obligation to do anything. It is the repsonsibility of the person who tagged the article to monitor them. If they disagree with the opinion of the person who removed the tag, then they can tag the article themselves.---Balloonman ^PoppaBalloonCSD Survey Results 04:33, 16 January 2009 (UTC)

Which is what I said. If the tag has any merit at all, it should be proded or sent to AfD by the person who removed the tag. Prodego ^talk 08:02, 16 January 2009 (UTC)

Maybe your choice of words is throwing people off. If I decline a speedy request where an AFD would be reasonable, I'll make a note of it in the decline statement (typically the edit summary). But unless I believe it should be deleted, I must not nominate it. In a "Well, this might or might not be deleted at AFD, I am personally inclined towards keeping" case, I shouldn't nominate it - "Nominator isn't asking for deletion" is realistically a "speedy close" rationale at AFD. WilyD 12:02, 16 January 2009 (UTC)

Where is this activity people are taling about?

Quoting from above:

Someone has told me:

CSD is an area often back-logged and I was working through a large list of such items.

Can someone tell me where this list is? I've looked over the CSD page and I don't find any obvious links to it. Michael Hardy (talk) 23:23, 13 January 2009 (UTC)

There's only the dynamic list at CAT:CSD#Pages in category. --Amalthea 23:31, 13 January 2009 (UTC)

But also above you see "Happy-melon" telling me that thousands of pages are speedily deleted every day. The "dynamic list" above does not back that up. I am also told by "VirtualSteve": "CSD is an area often back-logged and I was working through a large list of such items." It doesn't look that way from the page at CAT:CSD#Pages in category.

So what am I missing?

This came up after "VirtualSteve" speedily deleted Poisson hidden markov model on the grounds that it lacked enough context to identify the subject, even though a few seconds on Google or on Wikipedia would have identified the subject. After I was told of the page at CAT:CSD#Pages in category, I looked there and found the page titled Saint-Adèle tagged as giving insufficient context to identify the subject. That was nonsense—it was very badly written but I identified the subject in a few seconds and after deliberating decided to add a "mergeto" tag to it, pointing to Sainte-Adèle, Quebec.

Is the page at CAT:CSD#Pages in category really so horribly backlogged that it is unreasonable to expect an admin to take a few seconds to do that? To say that the page titled Saint-Adèle lacks sufficient context to identify the subject is unreasonable, just as with Poisson hidden markov model. Michael Hardy (talk) 03:45, 16 January 2009 (UTC)

CSD is only backlogged if you have rediculous standards of what constitutes "backlogged". Usually when articles sit in C:CSD for a long time, it's because they're too marginal to either accept or decline the speedy, and nobody wants to do either. WilyD 03:50, 16 January 2009 (UTC)

<ec>At the risk of sounding like a broken record, CSD's mostly come in two categories-- absolute rubbish, and likely rubbish. In the former instance, I generally delete as fast as my neurons process all the way through. In the second, I generally search for RS with V to try to save the article. And good save on the Sainte-Adèle. I think I looked at that and passed by as not speediable, but I did not follow through for the save. And like WilyD said . . . . Dlohcierekim 03:58, 16 January 2009 (UTC)

I would imagine that the references above are to Special:NewPages, which has a massive backlog. I'd say a good half of new pages are speedy deleted, with about 10% incorrectly tagged, usually fixed by a "new page patroller" or admin. (EhJJ)^TALK 04:39, 16 January 2009 (UTC)

As before, Michael Hardy, people make mistakes. Most of them are caught and reversed, and really, harping on about them isn't going to accomplish anything. Stifle (talk) 09:53, 16 January 2009 (UTC)

Special:Log/delete is a complete record of every page that is deleted, indeed that has ever been deleted, on en.wiki. Every log summary that includes a link to WP:CSD is a speedy deletion. As you can see, there are thousands of them. CAT:CSD contains a list of pages that are tagged for speedy deletion at the present time, but which have not yet been either deleted or kept - that is, pages where a CSD decision is 'pending'. Happy‑melon 10:32, 16 January 2009 (UTC)

• Err, yes. CSD is often busy but never actually noticeably backlogged. WilyD 12:03, 16 January 2009 (UTC)

"Stifle" wrote:

"really, harping on about them isn't going to accomplish anything."

That is false. Harping on them did acomplish something in March 2008. For about six weeks in February and March we saw repeated speedy deletions every day of articles that were deleted ostensibly because they were gibberish, when the real reason was that the were on mathematics. That ended abruptly when I wrote an emphatic complaint about it here on this page. Michael Hardy (talk) 21:00, 16 January 2009 (UTC)

The numbers are, I believe, half guess-timation and half combing through database dumps of the deletion logs. If you'd really like some hard numbers, I suggest more of the latter. Cheers. lifebaka++ 21:09, 16 January 2009 (UTC)

Michael Hardy, back in August, during one of your many previous rants on this subject, I asked you for a list of these articles. You didn't provide one, so I looked up your original complaint. Copying my conclusion from then: "I read the entire section twice and all I see is a complaint about two articles being tagging (admittedly inappropriately) with speedy deletion tags. Both speedies were correctly declined. Only one of these two articles was tagged shortly after creation. In a separate complaint in May, an article was speedily deleted, and subsequently restored."

For someone who is interested in mathematics, I fail to see how two articles makes "repeated speedy deletions every day" over a six week period. (Yes, I did major in math, so I'm fairly certain that 2 is not equal to 6 times 7). And even if it was, there's absolutely no evidence of cause and effect between your complaint and a decrease in speedy deletion of mathematics articles. I learned about the dangers of assuming cause and effect in my statistics classes -- I would have thought you did, too.

If you had taken the advice of many, many editors here and spent more time working on CSD instead of rehashing the same old complaints, you would have seen that "patent nonsense" is a widely misapplied speedy tag across all kinds of articles, not just mathematics articles. You'd also have learned that the people who are misapplying this tag are not the people who are reading this page. The only conclusion I can draw is that you enjoy complaining and pretending to be the injured party more than you like doing anything constructive.

Those who know me will know that calling Michael Hardy out like this is totally out of character for me. But I've just had enough of this constant groundless whining. Yes, there are a tremendous number of things wrong with CSD. But we can't fix imaginary problems, only real ones.--Fabrictramp | talk to me 21:41, 16 January 2009 (UTC)

I can't speak for Michael, but one of the reasons I don't keep a list of bad deletions is that I believe I would be blocked for compiling evidence of admin incompetence. DuncanHill (talk) 22:00, 16 January 2009 (UTC)

Being blocked for that reason and no other would be travesty. But I would certainly hope that if you claimed time and time again that there had been a rash of improper deletions which had only stopped because of your complaints, that you would be able to point out at least one when asked about it. Or stop bringing it up until you could come up with one. And the thing is, Michael Hardy is an admin -- he could restore those deletions. If they are truly bad deletions, would he be that worried about being de-sysoped, but not be at all worried about bringing the subject up time and time again? It just doesn't pass the smell test. --Fabrictramp | talk to me 22:44, 16 January 2009 (UTC)

OK, well there was a rash of speedy deletions of articles about Fellows of the Royal Society. There were speedy deletions of recipients of the George Cross (which got an admin not just threatening me but lying about my deleted contributions). All were on AN or ANI. DuncanHill (talk) 22:50, 16 January 2009 (UTC)

Are these Michael's math article deletions, or is this on a different subject? I'll gladly take a look at them if you want, but you'll need to be a bit more specific (article names, AN dates, something. I'm not going to slog through pages and pages of archives to find these for you.)--Fabrictramp | talk to me 23:08, 16 January 2009 (UTC)

You have my permission to keep a record of my bad speedy deletions if you like. WilyD 23:49, 16 January 2009 (UTC)

I kept no records of the events of February and March. Maybe I can find some of it. "FabricTramp", I have two questions for you: (1) what makes you think I haven't done those restorations, and (2) when did I ever express any worry about being de-sysoped? Michael Hardy (talk) 23:36, 16 January 2009 (UTC)

I can't speak for Frabrictramp on the second question, but for the first this would tend to suggest that you weren't performing restorations related to your complaint in March. Cheers. lifebaka++ 00:58, 17 January 2009 (UTC)

(ec)Michael Hardy, I don't have any reason to think you did or did not restore the articles in question. Mainly because other than your original complaint, where the articles were not deleted at all, I've never seen you mention which articles they are. You might be reading more into my reply to DuncanHill than is really there.

I also never said you expressed any worry about being de-sysoped. DuncanHill posited that a fear of keeping a list may have prevented you from doing so. Besides the fact that I've never read anywhere that you had such a fear, I was also trying to explain that such a fear isn't a logical explanation. Still, I await some concrete examples of a problem, and am even more anxious to see some evidence that the problem "ended abruptly when I wrote an emphatic complaint about it here on this page".--Fabrictramp | talk to me 01:03, 17 January 2009 (UTC)

No I didn't, I said that a fear of being blocked has stopped me from keeping such a list. DuncanHill (talk) 01:04, 17 January 2009 (UTC)

My apologies for assuming that you were offering your own fear as an explanation for Michael Hardy's actions. Since you inserted it into this thread, I thought it was about the same discussion.--Fabrictramp | talk to me 01:07, 17 January 2009 (UTC)

The words "I can't speak for Michael" were intended as some kind of clue, obviously far too cryptic. DuncanHill (talk) 01:09, 17 January 2009 (UTC)

My experience has been that when someone starts out with "I can't speak for...", they continue on to do just that. Again, my sincere apologies. Just goes to show that humans still rely so much on non-verbal cues for meaning, and writing is easy to misinterpret.--Fabrictramp | talk to me 01:33, 17 January 2009 (UTC)

While I am not worried about being desysopped for criticizing other admins, I am worried about losing their good will, and I am also concerned about not appearing overly combative. I therefore do not go out of my way to look for bad CSDs, though I usually do comment on particularly unjust ones i encounter. If I were to devote myself to finding them, I am sure I could do so based on limited probes. I estimate then number that I personally think are actually wrong are about 5 to 10 percent of the total so, and the number where there is no possibility of rescue at all, possibly twice that. Looking back over them, I could probably find some reason to contest some of my own: I will delete many a G11 where i could, if I had infinite time, rewrite the article from scratch. But I agree totally with the comment above, that the real problem is inadequate patrolling. I've has a little more insomnia than usual recently, and I have been doing some new pages work, sometimes from recent, sometimes from older, and there is a great deal of material getting by that needs either deletion or drastic fixing. We need a better way of doing things than patrolling at random. DGG (talk) 02:20, 17 January 2009 (UTC)

Although NPP is a bit off subject here, I can say for certain from my work at WP:DEP that the change to allow pages to be marked as patrolled made a huge difference in reducing the compost that slips through the cracks. Still not perfect, but the difference was dramatic.--Fabrictramp | talk to me 15:03, 17 January 2009 (UTC)

Indeed. Random checking is not working as well as I'd like. Dlohcierekim 02:37, 17 January 2009 (UTC)

In my research for Wikipedia:Criteria_for_speedy_deletion/Overturned_speedy_deletions it appears that most bad deletions are either located by the users who created the article that was deleted, by users who encounter a salted page and wish to recreate the article, or by admins who stumbled across a page they expected to exist and encountered a suspicious deletion record. Random checking is taxing since most (I'd say at least 90%) of speedy deletions are valid or at least defensible. Dcoetzee 07:24, 17 January 2009 (UTC)

There should be a less drastic way to do this

SOPHIA (European Foundation for the Advancement of Doing Philosophy with Children) was tagged for speedy deletion on the grounds that "It was created by a banned user in violation of his or her ban, with no substantial edits by others. See CSD G5." It has been pointed out that such deletion is also required by WP:BAN.

So at a time when its edit history looked like this:

• (diff) 02:48, 14 January 2009 . . Michael Hardy (Talk | contribs | block) (401 bytes)
• (diff) 00:20, 14 January 2009 . . Michael Hardy (Talk | contribs | block) (415 bytes)
• (diff) 21:34, 13 January 2009 . . Sticky Parkin (Talk | contribs | block) (409 bytes) (Requesting speedy deletion (CSD G5). (TW))
• (diff) 21:03, 26 September 2007 . . CmdrObot (Talk | contribs | block) (395 bytes) (sp: a open→an open)
• (diff) 12:53, 15 September 2007 . . SmackBot (Talk | contribs | block) (394 bytes) (Unrefernced to refimprove where appprpriate and/or general fixes.)
• (diff) 19:33, 6 September 2007 . . SmackBot (Talk | contribs | block) (396 bytes) (Date/fix the maintenance tags or gen fixes)
• (diff) 14:49, 4 September 2007 . . Melsaran (Talk | contribs | block) (379 bytes) (Reverted 1 edit by Epic consultants to last revision by Melsaran; Copyvio from http://sophia.eu.org/About%20SOPHIA/about_sophia.htm.)
• (diff) 13:39, 4 September 2007 . . Epic consultants (Talk | contribs | block) (20,169 bytes) (Description of SOPHIA and the constitution of SOPHIA)
• (diff) 15:29, 3 September 2007 . . Melsaran (Talk | contribs | block) (379 bytes) (wikify)
• (diff) 13:38, 3 September 2007 . . IPSOS (Talk | contribs | block) (370 bytes) (create stub)

the edits by others being quite minor, I deleted it and then recreated it as a new article, saying in the initial edit summary

(Recreation of article without the banned user who created this in the edit summary. Minor contributions were made by CmdrObot, SmackBot, Melsaran, and me.)

That way the banned user gets no credit, and that seems to bring it into compliance with the policies.

But obviously the exact nature and times of the edits by others are no longer available to non-admins.

The best solution would seem to be to simply delete the banned user's identity from the edit history, so that it would say "BANNED USER" instead of "IPSOS", making the latter information available only to admins.

I propose that we ask developers to create the technology needed for this.

Opinions? Michael Hardy (talk) 19:32, 17 January 2009 (UTC)

• Where are the credits to the original author? There's not a lot of contribution in this example, but isn't damnatio memoriae a direct violation of GFDL? NVO (talk) 19:38, 17 January 2009 (UTC)
• Yes, it's problematic with respect to the GFDL, and it should be discussed at WT:BAN since the criterion is only a consequence of the policy there, and your proposal applies both to page creations and other contributions. --Amalthea 19:42, 17 January 2009 (UTC)

• I understand the concern about credit to the original author. The policy says the page may be speedily deleted if created by a banned user. That would allow deletion followed by recreation by another user. That's what happened in this case. The policy appears to be intended to deprive banned users of credit for their contributions. What I'm saying is we need a better way to do that, so that other contributions can be fully an accurately reported while nonetheless complying with that policy.

Another alternative would be to change the policy. But I think this would not be the right page for that; it would have to be done at the talk page for WP:BAN. Michael Hardy (talk) 20:20, 17 January 2009 (UTC)

• Surely the intent of the policy is to ban a person from editing, not to breach copyright by allowing their edits to stand without attribution. Phil Bridger (talk) 20:25, 17 January 2009 (UTC)

• (after edit conflict) Retaining or recreating verbatim content but removing the attribution in the edit history is clearly a breach of GFDL. I can't see anything in WP:BAN that says that speedy deletion is required. Under WP:BAN#Enforcement by reverting edits it says "this does not mean that obviously helpful edits (such as fixing typos or undoing vandalism) must be reverted just because they were made by a banned user" and "if other editors have unwittingly made good-faith contributions to the page or its talk page, it is courteous to inform them that the page was created by a banned user, and then decide on a case-by-case basis what to do". Phil Bridger (talk) 20:20, 17 January 2009 (UTC)

Michael's example is a blatant misunderstanding of the policy. It is never okay to oversight edits where the content from that edit is being retained; only reverted edits can be oversighted under the GFDL. Under G5, you could delete this entire article, since the edits by others are not "substantial", but not only the banned user's revisions. If the article is restored, all revisions should be restored, including those of the banned user. Moreover, WP:BAN does not "require" deletion of anything; it merely authorizes it. Dcoetzee 23:25, 17 January 2009 (UTC)

I've restored the edit history of the SOPHIA article because of these copyright concerns. Michael Hardy (talk) 23:55, 17 January 2009 (UTC)

Clarify

The policy says an article by a banned user MAY be deleted speedily, NOT that it MUST be deleted speedily.

This raises two points:

• Someone requests speedy deletion on the grounds that its an article by a banned user with no contributions from others. Should they be told that that's not a good enough reason? After all, it says "MAY", not "MUST", so there should be a reason beyond just that, before it SHOULD be deleted. With the SOPHIA article, no further reason was given. When I declined to delete speedily, the person who requested it asked my why, and repeated that it's by a banned user. It's as if I had the burden to show it should not be deleted, even though he had never given a reason to delete it, except that the rules say I MAY delete it. That the rules say I MAY delete it is not a reason why I SHOULD delete it.
• After the admin declines to delete speedily, is that act of declining enough of a "substantial contribution by others" to the article that other admins may no longer come along and delete without AfD? I'm inclined to say it ought to be so, and I think the rule should clarify that point.

So the rules said that I MAY delete speedily, and the person who pointed that out gave no reason to delete except that—no reason why it SHOULD be deleted. That's enough reason to decline to delete if it's a good article that should exist. Michael Hardy (talk) 01:12, 18 January 2009 (UTC)

May basically puts it at the mercy of the administrator, and the speedy-tag becomes a request to judge the article not the validity of the claim, which in this cases is self-evident. The administrator should ask "is it better for the encyclopedia that I delete this or keep it?" One solution is to rewrite it from scratch and delete the banned editor's edit, effectively WP:DENYing him recognition while keeping the information that was in the article. Other solutions are to let the article stand as is or to delete it. davidwr/(talk)/(contribs)/(e-mail) 04:29, 18 January 2009 (UTC)

Keep in mind that WP:DENY is about denying recognition for vandalism. Generally speaking, if someone has declined a speedy, that is not a "substantial contribution," but it does provide evidence that a speedy deletion on those grounds would be controversial and should not be pursued further. Occasionally admins will wipe out an article that has a declined speedy because they either didn't review the history or disagreed with the reviewing admin, but I disagree with this practice. Dcoetzee 06:00, 18 January 2009 (UTC)

I had never heard of this particular reason for speedy deletion until I came across this particular article with a "speedy" tag, which I deleted from the article, and then I got a message on my talk page asking why I had done that despite the page's being a creation of a banned user, phrased in such terms as to suggest that what I had done was surprising, unreasonable, and unacceptable. I found this all quite alarming, to the point that it led to my starting this present thread on this page. So should I take the comments above as implying I'm not the one who was unreasonable? Michael Hardy (talk) 06:10, 18 January 2009 (UTC)

No, you were fine. An admin always has discretion about whether to delete or decline, regardless of whether or not the criterion is technically satisfied. From the policy: "Deletion is not required if a page meets these criteria. Before nominating an article for speedy deletion, consider whether it could be improved, reduced to a stub, merged or redirected elsewhere or be handled with some other action short of deletion. If this is possible, speedy deletion is probably inappropriate." Dcoetzee 06:44, 18 January 2009 (UTC)

I generally agree: If a page is helpful for the encyclopaedic content then it shouldn't be delete just because it was created by a banned user. WP:DENY helps to discourage vandalism, with banned users it depends on why they were banned. That's why the criterion stresses that it's only for page creations in violation of their ban. So from taking a glance at the arbcom case and the article, I'd say you were quite correct to deny deletion: he apparently was banned in parts for spamming, but I don't see that that was the motive behind this article. --Amalthea 12:56, 18 January 2009 (UTC)

Toward transparency

We now have Wikipedia:Criteria for speedy deletion/Overturned speedy deletions.

We do not yet have any log of pages tagged for speedy deletion and their outcomes. We have only the page at CAT:CSD#Pages in category listing those CURRENTLY proposed for speedy deletion. There is the deletion log (not searchable!) but there's no record of those proposed for speedy and declined.

I've also sent Brion Vibber this email:

begin copy of email

Hello. I wonder if this searchability issue can be dealt with:

The log at

http://en.wikipedia.org/wiki/Special:Log/delete

is not searchable.

I.e. I cannot enter some search terms to identify which articles containing certain

(1) words or phrases or

(2) category tags,

either

(1) in the body of the article, or

(2) in the title, or

(3) in the edit summaries, or

(4) in the list of users who have edited it, or

(5) in the links _to_ the new article, or

have been deleted during a time period that I specify.

Only such searchability can make it possible for those with expertise in a particular subject to see what's going on with deletions that they would recognize as having meaningful content where a non-expert would not. What happens repeatedly is a badly written new article on a topic worthy of an article is speedily deleted by an admin who says it's incomprehensible gibberish.

If I point out in such a case that five seconds of Google searching, or often of searching within Wikipedia, would identify the subject, I am lectured about how unreasonable it is to ask anyone to do that when there's a huge backlog of new articles.

Speedy deletion proposals like this one:

http://en.wikipedia.org/w/index.php?title=Saint-Ad%C3%A8le&oldid=263979110

or this one:

http://en.wikipedia.org/w/index.php?title=Poisson_hidden_Markov_model&oldid=259124759

just seem irresponsible.

For about six weeks in February and March 2008 several articles were deleted every day on the grounds that they were new articles on mathematics. That ended abruptly when I wrote a strident complaint at Wikipedia talk:Criteria for speedy deletion, but that is not a systematic or permanent solution.

Is it possible that that log page could be made searchable in a way that satisfies points (1) through (5) above?

Thanks. -- Mike Hardy

end copy of email

Michael Hardy (talk) 21:40, 19 January 2009 (UTC)

Frankly, if the speedy is proposed and declined, I believe nothing has gone awry; the very reason deletion is not given to all users is that regular users are not expected to be familiar enough with policy and/or mature enough to exercise due caution in deleting articles. For the same reason I omit articles from Overturned speedy deletions that were restored almost immediately by the admin that deleted them (I think we can forgive an admin for changing their mind after further thought). Searching deletion logs would be great, but full-text search functionality like this is notorious for performance issues, since there's no full-text index on the summary fields - Brion may be likely to object on those grounds. Dcoetzee 22:03, 19 January 2009 (UTC)

I agree that nothing has gone awry if all that has happened is that a speedy has been proposed and declined.

I also thing nothing has gone awry if a new article has been created. So should we erase all record of that too?

There far too much that's opaque about the whole process; I want to get a picture of what goes on. Michael Hardy (talk) 23:23, 19 January 2009 (UTC)

Actually, that's exactly what happens: new page creation is logged in the recentchanges table, which is systematically purged after 30 days. The ability to search for new articles is lost, you have to look through articles individually to see if they were created 31 days ago. Since there is no enduring log of new page creations at all, the deletion process is actually more transparent. Happy‑melon 23:37, 19 January 2009 (UTC)

MfD for Overturned speedy deletions

If interested, please consider commenting at Wikipedia:Miscellany for deletion/Wikipedia:Criteria for speedy deletion/Overturned speedy deletions. Dcoetzee 22:30, 19 January 2009 (UTC)