::::I agree with what's been said above and will only add that if the duplicate page is located at an implausible search title (e.g. a duplicate article about [[Birmingham]] at [[Biggringham]]), either [[WP:CSD#G2|criterion G2]] (test page) or [[WP:CSD#G6|criterion G6]] (housekeeping) could apply. [[WP:CSD#R2|Criterion R2]] (redirect from implausible typo or misnomer) could also apply once the page is redirected ''if'' there is no useful page history. –'''[[User:Black Falcon|Black Falcon]]''' <sup>([[User talk:Black Falcon|Talk]])</sup> 06:27, 9 March 2009 (UTC)
::::I agree with what's been said above and will only add that if the duplicate page is located at an implausible search title (e.g. a duplicate article about [[Birmingham]] at [[Biggringham]]), either [[WP:CSD#G2|criterion G2]] (test page) or [[WP:CSD#G6|criterion G6]] (housekeeping) could apply. [[WP:CSD#R2|Criterion R2]] (redirect from implausible typo or misnomer) could also apply once the page is redirected ''if'' there is no useful page history. –'''[[User:Black Falcon|Black Falcon]]''' <sup>([[User talk:Black Falcon|Talk]])</sup> 06:27, 9 March 2009 (UTC)
::::R3 actually. --[[Special:Contributions/76.69.168.166|76.69.168.166]] ([[User talk:76.69.168.166|talk]]) 20:16, 10 March 2009 (UTC)
== Clarify A7 ==
== Clarify A7 ==
Revision as of 20:16, 10 March 2009
Read this before proposing new or expanded criteria
Contributors frequently propose new (or expansions of existing) criteria for speedy deletion. Please bear in mind that CSD criteria require careful wording, and in particular, need to be
Objective: Most reasonable people should be able to agree whether a page meets the criterion. Often this requires making the criterion very specific.
Uncontestable: It must be the case that almost all pages that could be deleted using the rule, should be deleted, according to consensus. CSD criteria should cover only situations where there is a strong precedent for deletion. Remember that a rule may be used in a way you don't expect, unless you word it carefully.
Frequent: Speedy deletion is intended primarily as a means of reducing load on other deletion methods such as Wikipedia:Articles for deletion and Wikipedia:Proposed deletion. These processes are more discriminating because they treat articles case-by-case, and involve many points of view; CSD sacrifices these advantages in favor of speed and efficiency. If a situation arises only rarely, it's probably easier, simpler, and fairer to delete it with one of the other methods instead. This also keeps CSD as simple and easy to remember as possible, and avoids instruction creep.
Nonredundant: If the deletion can be accomplished using a reasonable interpretation of an existing rule, just use that. If this application of that rule is contested, consider discussing and/or clarifying it. New rules should be proposed only to cover situations that cannot be speedily deleted otherwise.
If you do have a proposal that you believe passes these guidelines, please feel free to propose it on this discussion page. Be prepared to offer evidence of these points and to refine your criterion if necessary. Consider explaining how it meets these criteria when you propose it. Do not, on the other hand, add it unilaterally to the CSD page.
At it's been pointed out that we need something to counteract abuse of the book feature for hosting vanity, spam, attack pages and general crap. As books are essentially lists of articles with a main subject, I propose the following criterion:
G13 - any book whose contents would be subject to speedy deletion as an article.
I agree. The userspace guidelines (and speedy deletion criteria) handle books in userspace fine. I think MER-C's formulation is the way forward for project-space books - they're a permutation of our article content, so article criteria should apply. The general criteria also apply to books in either namespace, of course. Also, let's keep in mind that many people saving books in either place simply might just not know what they're doing - so we should be willing to userfy content that would be fine in userspace, and provide instructions for saving local copies of books that don't meet userspace guidelines. — Gavia immer (talk)03:00, 1 March 2009 (UTC)[reply]
I think since these could end up in multiple namespaces, we would need to look into creating a new category of CSD's just for Books. These could apply either in the Wikipedia:Books/ space or in User: space; basically anything that is appropriately stuck in Category:Wikipedia:Books; sort of similar how a Template CSD can apply to an inflammatory userbox in userspace.
MER-C, I like your idea of any book whose contents would be subject to speedy deletion as an article. I'd like to add a suggestion for another one: any book that does not contain an article, and whose contents do not assist users in using the encyclopedia or understanding its administration. This way, we avoid the creation of purely userspace books, but allow ones to remain that contain lessons for new users, related policies, and the like. For example, the user page design center is entirely in the Wikipedia userpsace, and my adoption lessons are wholly contained in my userspace; books of entirely these pages would be acceptable under this guideline.
All of the above was before an edit conflict; considering what Gavia said, perhaps any CSD's we do set up should be subject to a time limit as some of the image criteria are, in order to allow users the chance to export their books or otherwise resolve the problem. Hersfold(t/a/c)03:35, 1 March 2009 (UTC)[reply]
I fully agree with the spirit of the proposal. The wordage and expansion should happen, but lets not lose focus: this is about a new type of content, which is already creating trouble, and for which we have no deletion policy. We need something rough and fast out there, now, and then we can polish. I commend MER-C for the initiative. --Cerejota (talk) 06:23, 1 March 2009 (UTC)[reply]
I think that the existing criteria cover what we already have to the extent that they should, and we probably don't have enough experience yet with what people will create to know whether we have a reason to extend the criteria. It's a waste of time to focus upon making up rules for things that haven't happened yet.
For examples: Wikipedia:Books/Accademia Leonardo I suspect is already speedily deletable as a combination of "user test" and "advertising". It's a book that contains nothing but Wikipedia:Introduction. (The extra content on the page isn't rendered into the book, note.) Similarly, Wikipedia:Books/yeah contains nothing but Wikipedia:Upload and is probably speedily deletable as a "user test" as well. Clicking "add wiki page" on the book toolbox next to a random page (or a sandbox) and then hitting save is just as much a user testing the editing function as a user clicking on a random feature of the editing toolbar and hitting save.
As for Wikipedia:Books/new stadiums in Bulgaria: Remember that the extra content isn't rendered into the book itself. The closest parallel to this seems to be templates. We don't speedily delete valid templates because the <noinclude> text in them happens to contain irrelevant rubbish. We remove the irrelevant rubbish. We only delete the templates if they aren't valid as templates, and we don't do that through speedy deletion in most cases. It's probably best to think of books in the same way. That book contains one article that seems to be relevant to its title, and seems to be valid as a book. The spurious text can be edited out, and the bad title and subtitle can be edited too. If the book itself doesn't seem maintainable, or valid as a book, the correct forum is Wikipedia:Miscellany for deletion, just as it is for (say) WikiProjects or portals that don't seem valid. We don't need a speedy deletion criterion to cover it.
I suspect that we'll have to augment the speedy deletion criteria slightly eventually. I suspect that many editors would want Wikipedia:Books/sabina wantoch and anything similar to be deletable on sight, but currently there's nothing to cover it, since the page is in the project namespace. Moreover, Wikipedia:sabina wantoch with exactly the same content wouldn't be speedily deletable, and it is just as easy to create. On the gripping hand, it's currently a set with one element. Let's wait to see whether this actually becomes a problem. Such pages are not a particular problem for the project namespace (as a whole) at the moment. Uncle G (talk) 06:55, 1 March 2009 (UTC)[reply]
I am CSD tagging all I see that can meet the existing criteria, and MfD those who don't. Wikipedia:Books/Accademia Leonardo was deleted after I tagged as spam, so it seems we have some protection until we can have a more formal policy.
Not to be BEANy, but my main worry is that once this feature spreads around the internet, and it becomes known there is community paralisys, we will find ourselves in a situation where the number of bad books is higher than those of the good books. There are all kinds of policies and guidelines that need to be modified to take into account this new content. I really wish this feature had a more wide community discussion so we could prepare (and if it was had, I apologize, but you guys overlooked a bunch of obvious stuff). For example, this thing should have its own Space, "Books:".
I think for example that we need:
Deletion criteria/discussion space - I am not sure MfD is enough. We are addressing CSD here.
MoS - I mean, in userspace let the people pig-out, but I would like Community Maintained books to be pretty and coherent - as well as compliant with core content policy like WP:NPOV
Editing policy - In articlespace, we have separate polices for the article and the talk page. In wikispace, we have the same thing, and also have levels from policy to essay, and include proposals and the wikiprojects. All of these covered by policy.
I have proposed to disable the creation of books in project space from Special:Book, here. And instead, to have a few books created and maintained by the community (instead of just being created by any user without any control). Cenarium (talk) 15:09, 1 March 2009 (UTC)[reply]
Support the general idea of this proposal: I'm really not the kind of person who works much with legalese, even here, so I'm not going to participate in a discussion of how-exactly-do-we-word-this, but without a doubt this is a reasonable proposal. No reason that this new feature should enable people to have things on Wikipedia that would never be permitted on here otherwise. Nyttend (talk) 15:40, 1 March 2009 (UTC)[reply]
Comment Some recently created books (and many of those that will created in near future) may be deleted as test pages (G2). Many of these books are indeed test pages, and I just deleted a couple of them. I think until specific CSD criteria are agreed upon, the (creative) use of G2 can partially solve the problem. Ruslik (talk) 20:08, 1 March 2009 (UTC)[reply]
Yes, we have too many criteria already. If something is in fact a "test page" the creator is unlikely to protest the deletion once they know whether the "test" succeeded or failed. Something like {{prod}} would probably be better. — CharlotteWebb21:43, 1 March 2009 (UTC)[reply]
If we create a namespace for books, we could use specific CSD criteria for them. We could replace G13 by a criteria similar to P1:
We don't need a new speedy criterion for such books. We already delete article drafts that wouldn't survive in mainspace and if the articles within a book are speediable, the links won't be working anyway. Let's first see if it's actually abused to the point we need criteria to battle it. - Mgm|(talk)08:50, 2 March 2009 (UTC)[reply]
I think you miss the point that these are a completely new category of content, for which we might want to keep under circumstances that would be speedy delete or delete under circumstances that might not be speedy delete now. I do tend to agree with you to wait it out, but we just had some MfDs on books which should have been speedy deleted/userfied based on the purpose that books exist for, but required long discussion. I feel that is adding unnecessary bureaucracy. CSD is after all really an extention of WP:IAR, so all I am saying is we tell admins it is okay to delete that book.--Cerejota (talk) 21:21, 7 March 2009 (UTC)[reply]
I've added this (boldly, hopefully this isn't a problem) as it's already codified at WP:OLDIP (also official policy) and in the interests of centralizing speedy deletion policy. If there's a problem with this, please say so here. =) —Locke Cole • t • c20:08, 3 March 2009 (UTC)[reply]
My reading of the OLDIP discussions is that there is not concensus, it was not widely advertised (i.e. here), and it is seriously disputed.
The U4 non-contributor idea relates to old, non-contributors with play subpages, or possibly just with userpages, and is connected to a series of userpage WP:MfD nominations for declined CSD tagged userpages. Unfortuantely, the cases include a high proportion of users who have edited in the last year. See WT:UP.
Given the now exposed routine out-of-process deletions that have been occuring, I propose the following:
No pages are to be deleted unless covered by a current CSD criterion or authorised by an accepted XfD process.
Exceptions to the above are to be reported here (or logged on a specificied page).
I've removed it for now. Though I think making G6 a broad stroke like this is a mistake and it makes more sense to have this as an actual criteria (since it will be an ongoing activity, and people looking for justification at CSD are most likely to look under the User criteria, as I did). I get the need to avoid instruction creep, but I think this is one case where it's a sensible and reasonable addition (if for no other reason than to reduce potential confusion). Then WP:OLDIP could be redirected here, and the verbiage at WP:USERPAGE could be significantly trimmed. —Locke Cole • t • c21:01, 3 March 2009 (UTC)[reply]
I had not seen WT:Criteria for speedy deletion/Archive 32#Deletion of old IP talk pages. A new years eve discussion is not a point to develop consensus, but the thread established that there was not consensus. Moving the discussion and deletion authorisation to WP:UP was forum shopping in a policy backwater. WP:UP should not be used to circumvent CSD restrictions. I strongly disagree that deletion of userpages can be called "housekeeping", even if they are IP userpages. --SmokeyJoe (talk) 21:04, 3 March 2009 (UTC)[reply]
I agree with you that the only things that should be speedy deleted are those things set out explicitly on this page, hence why I'm suggesting moving WP:OLDIP here to consolidate/simplify our policies. I note you've removed it from WP:UP, you may wish to reconsider after reading this discussion on WT:UP (which seems to indicate consensus support for this). —Locke Cole • t • c23:28, 3 March 2009 (UTC)[reply]
Some subtantial things were done on the basis of unclear and fragmented discussion, all during a Christmas to New Year period. It doesn't look good, and it wasn't good. I don't agree that the discussion "seems to indicate consensus support", only that a consensus seemed to be forming. Wikipedia:Silence and consensus doesn't apply to things done in backwaters and while others aren't looking. The mass deletions are not easily undone. Discussion should have been broadler, and actions more tentative. --SmokeyJoe (talk) 00:53, 4 March 2009 (UTC)[reply]
In all seriousness, is there some sort of calendar for these things? Yes, it was during the late part of December. But generally there are more people on the site, not less, during these times. (Quite a few trying to escape family functions, I imagine. ;-) --MZMcBride (talk) 00:58, 4 March 2009 (UTC)[reply]
Normally I prefer week or two week comment periods (RFC) so people have a chance to talk out any concerns they have. OTOH, I don't think this is a major issue given the ultimate criteria (the year of inactivity, lack of incoming links, etc). I think it's very unlikely you'll hit a false positive with those criteria. —Locke Cole • t • c01:21, 4 March 2009 (UTC)[reply]
I agree strongly with Cole's motive. If there is not consensus around this policy, it shouldn't be policy, and it should be removed from WP:OLDIP. Otherwise, it should be listed here, for the sake of clarity, since evidently it is not "routine housekeeping" to many of the people who take issue with it. Other policies should not be permitting speedy deletion not expressly permitted here. Dcoetzee21:41, 3 March 2009 (UTC)[reply]
Agreed. If there's consensus for old IP talk pages (and, personally, I agree with the reasoning behind it) then it should certainly be listed here. I dislike using G6 for specific things that could have their own criteria; and this is a specific reason that could use one. ~ mazcat|c22:49, 3 March 2009 (UTC)[reply]
My reading of this discussion on WT:UP shows me there's some consensus for this. Barring any major objection I'll be adding U4 to this and removing the note from G6. I'd also suggest we tighten G6 up to be limited to specific tasks, not open ended or subject to individual interpretation. As an aside, apparently there was a discussion at WP:AN on this subject as well; I haven't read that archive yet. But again, given the two discussions linked above, there seems to be existing support for this. —Locke Cole • t • c23:26, 3 March 2009 (UTC)[reply]
I strongly oppose WP:OLDIP authorising CSDs at WP:UP. I would prefer that Old IP userpages were not deleted, as per others' opinions, but if they are to be speediable, then OLDIP deletions must be here authorised here. --SmokeyJoe (talk) 00:44, 4 March 2009 (UTC)[reply]
I suggest not deleting old IP talk pages as it makes it harder to find patterns of abuse from an IP. Perhaps blanking after X amount of time of inactivity so that new users to the IP don't get greeted with a bunch of old warnings. Chillum01:26, 4 March 2009 (UTC)[reply]
It occurs to me that while discussing these old IP talk pages, it may also be good to discuss this category, which also isn't documented on this page. Perhaps one new criterion covering both is possible? --MZMcBride (talk) 00:42, 4 March 2009 (UTC)[reply]
The routine deletion of old IP pages is another step towards forcing all editors to register. Am undecided as to whether this is a good or bad thing. I’ll admit that I dislike talking to IPs, or other unpronounceable handles. --SmokeyJoe (talk) 00:57, 4 March 2009 (UTC)[reply]
The criteria that was agreed to back in late December wouldn't seem to have the effect you're thinking it would. The talk page must be inactive for a year. The editor must not have made an edit in the past year. And the talk page itself must not have any incoming links (which, if the IP leaves just a single message on a talk page and signs their comment, the incoming link will be automatically generated). I'm personally iffy on deletion of these pages, but I see the point and don't think many actual editors will fall prey to this. —Locke Cole • t • c01:12, 4 March 2009 (UTC)[reply]
I'd prefer to see all such pages just blanked, but if they are to be deleted, a new CSD criteria covering both is a good idea. --SmokeyJoe (talk) 01:07, 4 March 2009 (UTC)[reply]
WP:CSD#G6 covers this, it is just cleaning out old skeletons from our closet. We have always deleted these pages after enough time has passed. We usually leave them up for months before deleting, so it is really not a "speedy" deletion. It is just a way of denying blocked users a trophy page. Chillum01:14, 4 March 2009 (UTC)[reply]
G6 is much too broad IMHO. Criteria should be narrowly defined with little room for interpretation. I note that G6 currently calls out a number of specific instances (this is good), so either G6 needs to explicitly define the remaining valid instances of deletion under that criteria, or perhaps additional criteria should be created to address them. —Locke Cole • t • c01:32, 4 March 2009 (UTC)[reply]
As an aside, I don't think CAT:TEMP and OLDIP would work well in a single criteria. I think we're looking at likely U4 for OLDIP and U5 for CAT:TEMP; OLDIP just has too many criteria for it to merge well with CAT:TEMP I'd think. —Locke Cole • t • c01:32, 4 March 2009 (UTC)[reply]
Fine, not G6. Also, not a speedy deletion, just an admin best practice. I don't think deleting temporary user pages of indefinitely blocked users after a long period of time is a CSD issue, nor do I think we need a special policy for what we have been doing all along. I only mentioned G6 because this is fairly routine maintenance. Chillum01:35, 4 March 2009 (UTC)[reply]
I don’t think that any userpage cases are necessarily uncontroversial enough for G6. There’ll be false positives in the Old IP pages. There are bad blocks made for “inappropriate usernames”. For a non-experienced editor, finding their userpage deleted is a pretty confrontational rejection. To then find through the deletion log link that it was deleted as “general housekeeping” is incredibly insulting, for the rare false positive.
Temporary and IP pages should be inactive for at least one year. I think that this should be a strict requirement. We should make allowance for part time editors, etc. Denial can be accomplished by blanking. I don’t want to see routine deletion of userpages just because someone gave an indefinite block because he considers the username to have too many characters.
What “we usually do” is a very poor standard. Such standards of practice always degenerate until the problems become big. Better to have flexible documented procedures, and records with correct categorisations. --SmokeyJoe (talk) 01:37, 4 March 2009 (UTC)[reply]
Please keep in mind that these users are both blocked and inactive for 30 days before the page is deleted. Also keep in mind that this is how we have done it for a very long time. It is not that "we usually do", it is that "it is the long standing best practice"(ie what policy is meant to reflect not direct). It is uncontroversial by the very nature that it has been a common practice for so long without issue. The have not been problems, big or otherwise. Lets wait for problems before seeking solutions. Chillum01:45, 4 March 2009 (UTC)[reply]
I certainly do agree it does not belong on CSD. I also don't think it needs any special policy. This is already taken care of, there is a bot that deletes the pages when it is time. It became such a common and accepted practice that it is now automated. Chillum01:48, 4 March 2009 (UTC)[reply]
Uncontroversial does not mean "longstanding best practice," it means "unlikely to be disagreed with by any reasonable user." We are reasonable users; we disagree. G6 does not apply, period. Admins always retain the option to delete at their own discretion, but they must take responsibility for such deletions, and I assert that other policies should never sanction speedy deletion; our policies are labyrinthine enough without spreading CSD over multiple pages in bits and pieces. Dcoetzee01:55, 4 March 2009 (UTC)[reply]
As I have already said a couple times here, yes this is not a CSD G6, it is not even CSD. Putting a page in a category for 30 days before deleting it is not a "speedy" deletion. This is the wrong page for this discussion imo. You can take any such deletion to WP:DRV and see what the community thinks of it. I am sure the deleting admin will take responsibility, though in this case most of the deletions are done by an admin bot so I suppose it would be the bot operator that would gladly take responsibility. Chillum01:58, 4 March 2009 (UTC)[reply]
Many CSDs (e.g. F11) have waiting periods. "Speedy" means "with limited process," not "quickly." This is the right place to discuss them and the right policy to sanction them. Dcoetzee02:02, 4 March 2009 (UTC)[reply]
This is not a limited process thing. It first requires that a user be blocked and the users page becomes inactive for a long period of time. It has limitation on when it can be done, process if you like. This really have been long settled, and I see no new issues arising from these deletions. If a consensus springs up that we should do things differently then fine, but that would have to be a significant consensus. I really wonder why you some people think that indef blocked users deserve to have their content and history posted on Wikipedia? How does that help the encyclopedia? Chillum02:16, 4 March 2009 (UTC)[reply]
I'm not terribly concerned about the policy, just that it be documented in the right place. To clarify, "speedy" means "not via a process involving consensus, such as articles for deletion or proposed deletion." Every criterion imposes conditions; the conditions for F11 are actually similar to those proposed here. I do disagree with OLDIP personally, but I'm not interested in arguing about that - I'm much more concerned about centralized documentation of policy. Dcoetzee02:27, 4 March 2009 (UTC)[reply]
Okay. Fair enough. I don't think it needs codification until there is a problem requiring it, but I am not dead set against the idea. It should reflect our existing best practices. That is not to say that consensus cannot alter it, I would suggest that 60 days inactivity would be better than 30 days. And it should be kept simple. Chillum02:30, 4 March 2009 (UTC)[reply]
Proposed U4
Okay, so we need a criterion to talk about here, so here's what I came up with, off the top of my head.
4. Inactive user talk page. The user has made no edits in the past year, nor has anyone edited their user talk page. The user talk page has no incoming links and no unsubstituted templates.
It is a good start. I think a year is a bit long, we are at 30 days now. I suggest 60-90. I think the spirit of WP:DENY behooves us to deal with these things before too long. Chillum02:49, 4 March 2009 (UTC)[reply]
Linking the unfamiliar terms seems to help people a great deal. "incoming links," for example. Or "no edits." Other than that, sounds pretty good. But it leaves out the never been blocked requirement. Should we remove that? (Others have argued that we should in the past.) Or perhaps say no blocks with the past __ years? --MZMcBride (talk) 02:48, 4 March 2009 (UTC)[reply]
People in this category are indef blocked. There should be a requirement that the user is blocked indef as well as not active. We don't delete the pages of users in good standing because they are inactive, we delete the pages of old indef blocked users in the spirit of WP:DENY. Chillum02:51, 4 March 2009 (UTC)[reply]
Okay, second draft, splitting this into two criteria for clarity:
4. Inactive blocked user talk page. The user is an indefinite blocked user whose user page has been included in Category:Temporary Wikipedian userpages for the last 60 days. They have made no edits in the last 60 days, nor has anyone edited their user talk page in the past 60 days. The user talk page has no incoming links and no unsubstituted templates.
5. Inactive anonymous user talk page. The user is an anonymous IP user who has never been blocked and has made no edits in the past year, nor has anyone edited their user talk page in the past year. The user talk page has no incoming links and no unsubstituted templates.
I am all for #4, good wording which seems to reflect current practices. However I don't think there is a consensus for #5. It was briefly put in a policy by someone but has since been removed. We should not block IP talk pages because the reasons that we have for blocked users are not there. We don't indef block IPs, they sometimes change owners they sometimes do not. Deleting old IP pages makes it hard for admins to see the history of the IP and impossible for non-admins.
Blanking of old IP pages so new users don't get a page full of warnings when they show up is a good idea, but that can just be done by people, it is not really part of CSD as it is just blanking. Chillum03:00, 4 March 2009 (UTC)[reply]
I agree, but some people e.g. MZMcBride are very staunch supporters of an OLDIP criterion and this is where I'd like to have that discussion. Mainly I'm just trying to figure what it should look like if it were enacted, so that we have something to talk about. Dcoetzee03:03, 4 March 2009 (UTC)[reply]
Ok, I will make it official:
Support 4 and Oppose 5(versions posted 02:56, 4 March 2009) for the reasons I have given above. The wording of both are very clear. Chillum03:10, 4 March 2009 (UTC)[reply]
(e/c)Number 4 doesn't really reflect current practice for CAT:TEMP at all. 30 days with no edits has been the de facto standard for as long as I know. AFAIK, there's never been any check for incoming links. I've basically been handling all of these for the past month or so with User:CAT:TEMP deletion bot, the criteria it uses are:
No edits to the page in the past 30 days
The user is indef blocked
There is no sockpuppet-related template on the userpage or the user talk page
I would support that criteria. You can find if there are incoming links by loading the "what links here" page. Though I don't think it is a big deal if a username in an ANI archive somewhere goes from blue to red. Chillum03:23, 4 March 2009 (UTC)[reply]
Okay, next attempt. I also just realised it's redundant to say a blocked user has "made no edits in the last 60 days" since they can only edit their talk page anyway. :-P
4. Inactive blocked user talk page. The user is an indefinite blocked user whose user page has been included in Category:Temporary Wikipedian userpages for the last 30 days. No one has edited their user talk page in the past 30 days, and neither the userpage nor user talk page contains any informational templates other than the "indefinitely blocked" template.
Chillum: Have you read the past discussions regarding OLDIP? If so, what do you object to? (Please, don't make me re-type the old arguments. My fingers, they ache. ;-) --MZMcBride (talk) 03:45, 4 March 2009 (UTC)[reply]
My arguments are above(scattered throughout the U4 heading), where are yours? I have not read them. In short it makes tracking the history of an IP awkward for for admins and impossible for non-admins. I also don't think much is gained by deleting when we could simply blank the pages. I certainly support blanking the pages because they a) could contain personal opinions of trouble makers, and b) could be full of warnings which a new user to the IP should not have to see. Chillum04:22, 4 March 2009 (UTC)[reply]
Tracking the contributions is trivial, indeed. However, tracking if users with a certain modus operandi have been warned in the past is practically impossible (especially for non-admins) if the talkpage has been deleted. I would really be more comfortable with archiving and replacing with a newly designed re-welcome/blocked/archived template ('this account has been blocked indef, if you want to start contributing again, do this, your info has been archived here'). Friendly, and still the tracks are there for everyone to see. --Dirk BeetstraTC09:50, 4 March 2009 (UTC)[reply]
I think we may be talking cross purposes. We are talking about two things, indef blocked users which are by definition not welcome back, there would need to be an agreement from the blocking admin or a consensus elsewhere to unblock before they were welcomed back. The other topic is IP addresses which generally change owners. I am all for blanking or archiving the content of old IP pages, but not deleting for the reasons you have given. Chillum13:57, 4 March 2009 (UTC)[reply]
Can you give an example of a page that has been deleted where it would've been useful to have been able to see the page history? Most of these (the large, large majority from what I've seen) are ancient warnings by people using Twinkle or other automated, templated warnings of some type. Is there any evidence anywhere that deleting these pages is inhibiting vandalism-fighting? (And wouldn't other projects like the German Wikipedia not delete these if there were?) --MZMcBride (talk) 17:30, 4 March 2009 (UTC)[reply]
Erm, perhaps I'm missing something really obvious, but that list is almost entirely user account pages, not anonymous user talk pages. And the few anonymous user talk pages I saw at the bottom you deleted as "user request."
Here's a partial list[1] of my several hundred restorations of recently deleted spammer talk pages. I finally despaired and gave up as the task was so large; I figured I had several tens of thousands of deleted IP pages to review for spam warnings. This was a big loss. Perhaps someone can develop an admin bot to temporarily restore all those deleted IP talk pages, look for a spam warning, then redelete if not spam-related. --A. B.(talk • contribs)19:49, 4 March 2009 (UTC)[reply]
I clicked a random link from the page and saw that the user hasn't edited since 2004 (Special:Contributions/219.82.139.89). I'm still having a hard time seeing the benefit to keeping such pages around forever (and in fact we may be inadvertently calling people spammers when they're simply using an IP of someone else...). However, because I don't work in this area, I added the check and all subsequent pages have been skipped. --MZMcBride (talk) 20:43, 4 March 2009 (UTC)[reply]
No, MZMcBride, it is not that we want to keep to call this editor a spammer, I am all for archiving the stuff to a subpage, and replacing the page with a friendly welcome notice. But DELETING the data removes the track. See my example below with the 15 IPs. Did we sufficiently warn these IPs? I don't see ANY warnings. Still we decided to put the stuff on the meta blacklist. Why, because there are warnings. If the two editors that were warned had the stuff in an archive, but still the page revisions accessible, then all editors could immediately see which editors were contacted at some point. But that data is gone (for non-admins). And so for hundreds (if not thousands) of pages. You don't see the use, but we see the loss, does that also mean nothing? --Dirk BeetstraTC20:50, 4 March 2009 (UTC)[reply]
Restored U4
I've restored the following text (which I originally added at the beginning of this discussion):
4. Old IP talk page. Talk pages of anonymous users may be blanked or deleted as part of routine housekeeping if they meet the following criteria: 1.) Never been blocked; 2.) Not using any unsubstituted templates (e.g., {{SharedIPEDU}}); 3.) No edits within the last year; 4.) No talk page activity within the last year; and 5.) No incoming links to the page.
I've also removed the OLDIP references from G6. I believe there's consensus for this based on prior discussions. I think CAT:TEMP should ultimately be a U5 criteria as I don't see a good way to merge these highly specialized deletions in to a single rule. —Locke Cole • t • c18:43, 4 March 2009 (UTC)[reply]
I am not sure we agree totally on the boundaries, in fact, MZMcBride did add another clause (which gets it closer, but not there yet). --Dirk BeetstraTC10:13, 5 March 2009 (UTC)[reply]
Perhaps it could use some modification and refinement, but that proposal seemed to have consensus around New Years and has been enforced for three months without much objection (see WP:SILENCE, or WP:CON which also spells out silence as consensus). Again, not trying to discourage changes or refinements, just justifying the addition as-is. —Locke Cole • t • c10:26, 5 March 2009 (UTC)[reply]
The volunteers most involved with spam cleanup and blacklisting argued strongly (begged actually) that these pages not be deleted. Spammers tend to sockpuppet a lot, using multiple IPs and/or multiple user accounts, sometimes getting away with spamming for several years. Blocking is ineffective (they just switch accounts) and blacklisting their domains is the only way to stop them. At the same time, many non-Wikimedia sites running wikis using Media-Wiki software use our spam blacklists for their own spam filtering, so we are reluctant to blacklist a domain unless we're sure they know our rules and intend to ignore them indefinitely. For this reason, we need to know how many warnings they've received across their multiple accounts. Deleting talk pages really screws us up. If we're going to delete old talk ages, I'd like to see an exception carved out for accounts that have received spam warnings. --A. B.(talk • contribs)03:53, 4 March 2009 (UTC)[reply]
Yes, that seems perfectly reasonable to me. Looking for certain key words seems to be the best, most reliable method. (\bspam\b|\bpromote\b|\bpromotion\b) is what I use. --MZMcBride (talk) 03:58, 4 March 2009 (UTC)[reply]
Strongly agreed here. I often times fight spam and vandalism and have seen a lot of pages on my watchlist deleted because they were old. Spammers routinely return after a set time and its important to keep the tags on their pages to identify the history of editing from that IP. Regular IP users shouldn't amass spam tags on their pages so the argument that IPs can be edited by more than one person doesn't hold up with spammers. If a person spams and returns a year later to spam, odds are likely its the same person. Themfromspace (talk) 05:24, 4 March 2009 (UTC)[reply]
While not currently active in that area I've dealt with extensive cross wiki spam for quite a period of time. It would seem utterly daft to delete such pages. I agree completely with A. B. --Herbytalk thyme08:01, 4 March 2009 (UTC)[reply]
I am fully against any deletion of old talkpages or unsollicited removal of userpages (except if the userpages are meeting other deletion reasons, as being attack pages, only for advertising or similar), whether it are spammers, vandals or POV pushers. It deletes an active record of what happened in the past. I see that after some people complained about it, the active practice has changed a bit, but deletion goes on, while there was discussion on the content guideline talkpage. I still do not agree that these are deleted. And it NOT only spammers. POV pushers are just as bad, and no, they don't always get blocked, if they perform just 2-3 edits per account, then most IPs will not have had a block until admins get so upset that they start blocking an IP on the first edit, hoping the editor will see that, or if an editor get so upset that he starts tagging them all as socks. But they don't get tagged always or similar, often it is just one or two warnings. I have been handling such editors long enough to know how creative (and ignorant) spammers, POV pushers and vandals can be.
For IPs: yes, they change, but deletion of the page makes it appear as if a user has never been warned in the past about not abusing a link or pushing a POV (i.e. that we tried to inform the editor that his actions are unwanted). Yes, there may be now a completely different user on that IP, and the 'offender' may now use a completely different IP, but still, if the modus operandi of the editor is the same (and believe me, if they don't get it, they keep doing the same, and companies get payed for advertising for other companies, they don't change their way of operating, never, it is what they are payed to do!), then it is very likely that that user used the old IP in the past and attempts have been made to warn them there. Tracking which IPs have been used in the past to perform certian edits is easy (history tabs, etc.), but when the talkpages are deleted, tracking if the editor has been warned when they was using the first IP is quite some work (if there are 50 old IPs for a certain modus operandi, and 10 have been warned and then the pages deleted, then you have to go through all 50 deletion histories to see if the editor has been warned, and an admin has to do that, non-admins can not help with this task. So it appears that editors with this modus operandi have not been warned for their actions, so is abuse without warning enough to blacklist a link? And we do get critisised for blacklisting without warning!). A very bad situation, which can easily be avoided.
A similar situation is for users who make single-purpose accounts. Also there, users are creative, and the criteria can so easily be met, that it is likely (though slighly less likely) again to remove tracks.
I really don't see any problem with a bot going round, archiving old talkpages of IPs, and replacing them with a newly designed {{re-welcomeanon}}. That removes messages which are likely not for the new user of the IP, but could inform the editor of the past situation. That message can be friendly and welcoming, and it may even attract the editor more then not seeing 'you have new messages .. thank you, I get welcomed here when I first enter, this is a nice site!'. (And can we please get a bot undeleting all tracks that have already been lost ..). --Dirk BeetstraTC09:43, 4 March 2009 (UTC)[reply]
I strongly Oppose deletion of IP userpages used for fraudulent, disruptive, deceptive purposes that violate and circumvent enforcement of Wikipedia policies. The use of an "IP" is not only the principle source from which spam origionates, it is the primary method used by spammers extensively is "stealth sockpuppetry" by quickly changing or using Multiple IP's. These records need to be retained as stated by others. Additionaly, these type if IP pages are an exception to deletion for the same reasons put forth on Wikipedia:User page#Removal of warnings;
“
Important exceptions may include declined unblock requests and confirmed sockpuppetry notices (while blocks are still in effect), or for anonymous editors, shared IP header templates. In these cases it may be legitimate in order to keep a user from gaming the system. Such templates are intended not only to communicate with the user in question, but to share important information about blocks and sockpuppetry with other users..
The community has strongly rejected deletion of these types of uerpages/IP's as a legitimate means of keeping a user(s) from gaming the system and to prevent future abuse.--Hu12 (talk) 17:06, 4 March 2009 (UTC)[reply]
Did you read the specifics of this proposal? I think it alleviates most of your concerns (for example, any IP that has been previously blocked wouldn't have its page deleted). And there must be at least a year of inactivity both by the IP and on the user talk page. Does this address your concerns? --MZMcBride (talk) 17:25, 4 March 2009 (UTC)[reply]
Many spam cases are multi-year SEO efforts and frequently use "one-time throw-a-way" IP's. Deletion after a year of inactivity or lack of a block makes it easy to game the system and exploit wikipedia further. The "block" or "one-year" criteria does not affectively help or prevent the most pervasive abuse (next to vandalism) on Wikipedia, spam. I have no issue with normal IP's for all the good reasons they should be deleted for recycling, however abuse accounts such as sockpuppetry and spam IP's should not be deleted.--Hu12 (talk) 18:14, 4 March 2009 (UTC)[reply]
Agreed. Personally, I scan the page for certain key words, specifically "spam", "promote", and "promotion" to ensure that the IP talk pages aren't spam-related. --MZMcBride (talk) 18:17, 4 March 2009 (UTC)[reply]
Yes, you have been diligent with your communication, and have suggested various ways to detect these pages. I personaly thank you for that.--Hu12 (talk) 18:26, 4 March 2009 (UTC)[reply]
I know, MZMcBride, and I thank you for that part (still, many had already been deleted), but no, it does not alleviate my concerns. There are often spammers who are warned with a vandalism remark. Still they are spammers, but the word 'spam', 'external link' or something similar is not on the page, and hence you would delete it. For me there still is no reason to delete them, there is no timelimit to vandalism, spam or pov pushing, and I don't see the problem with archiving, and replacing them with a friendly re-welcome notice. In that case we can see immediately that there has been communication with the editor, we can't see that from a redlink. --Dirk BeetstraTC19:17, 4 March 2009 (UTC)[reply]
Funny, I came accross User talk:Kekkomereq, the user has some warnings from XLinkBot. The user runs into an indef block on the 23 of March, 2008, resulting in the talkpage to be on the 'temporary page list'. On the 2nd of April and 24th of May the user gets two warnings about images, and on the 2nd of August, the talkpage gets deleted. Then after that, the page gets recreated, as two new warnings on the 17th of September and on the 13th of December. I am sure that that is also useless, but since it is not tagged anymore, it does not get deleted. So in the end, possibly useful pages get deleted, and other useless crap does not (at least, yet). Too bad. Similarly, User talk:Theprominence, blocked for another reason, but spamming a link with a possible coi (on a blogspot). Possibly a useful track, as there seems to be a coi (and indeed there is a sock, also blocked etc.). Talkpage got recreated for another reason on the 2nd of october (over 5 months ago), after being deleted on the 2nd of August. I wonder how many cases there are which are less easy to follow. I'll have a look. --Dirk BeetstraTC19:53, 4 March 2009 (UTC)[reply]
Our spam-related indefinite-block templates used to automatically categorize the talk pages as temporary. They now categorize them as indefinitely blocked spammers. I restore a few on my watch list after they were deleted as temporary; I would then go and remove the "temporary" categorization from the indefinite-block notice. Maybe those are the ones you saw. --A. B.(talk • contribs)19:57, 4 March 2009 (UTC)[reply]
15 spam IPs (who added a link), 9 are active here, 17 link additions, bot two have been warned for spam (well, one with the word spam in it, the other one a friendly message ..).... But which? I feel I become a bit pointy here. --Dirk BeetstraTC20:27, 4 March 2009 (UTC)[reply]
I find the above discussion a little difficult to follow. Is it agreed that WP:CSD#U4 "Old IP talk page" has consensus support? --SmokeyJoe (talk) 20:08, 4 March 2009 (UTC)[reply]
I think we're still sorting it out. It's unclear whether there's support to scrap oldIP or to keep it. But then again, maybe I'm just confused, too. --A. B.(talk • contribs)20:16, 4 March 2009 (UTC)[reply]
Didn't notice this section until just now. I added a criterion about pages not being spam-related. Personally, I think this ensures that nothing of value is ever lost with these pages (they're almost entirely automated, templated warnings). Though I'll re-read the conversation above to ensure I'm not missing anything. --MZMcBride (talk) 20:34, 4 March 2009 (UTC)[reply]
See the example above with the IPs. The link is blacklisted on meta, one of the two warnings would still be deleted according to the current rules, and it is a lost-spam track. And I don't think that spam is the only problem that suffers here. --Dirk BeetstraTC20:37, 4 March 2009 (UTC)[reply]
I think there was consensus based on the old discussions at WP:UP and WP:AN. I think there's consensus forming here for it as well, but it's not as far along. I added it based on the old consensus, removal (if people seek it) should be established via agreement. I announced the restoration at #Restored U4, above (where seemingly nobody has responded, heh). —Locke Cole • t • c22:24, 4 March 2009 (UTC)[reply]
Looking at this more broadly
Currently, we have vandalism fighters who patrol the RecentChanges feed and they warn thousands of anonymous users (IPs) each day. The warnings tell the user that they're behavior is inappropriate and explains it in a templated manner. It's almost like a game for some vandalism patrollers, which is fine, because the users need to be warned so that the behavior hopefully stops and the user can become productive.
However, most of these are "easy-on" taggings. They one-time deals. So in that spirit, deleting these pages should be no big deal. Easy-on, easy-off. And that's also why so many precautions are taken when looking at which pages to delete. If the IP ever commented anywhere, the page won't be deleted. If the IP has ever been blocked, it won't be deleted. If the IP has edited in the last year (or if their user page has been edited in the last year), the page won't be deleted. And if there are any helpful notices about {{SharedIP}} or the like, the page won't be deleted. And, in addition to all of this, the page text is scanned to ensure that the warnings are spam-related so that we don't inhibit spam tracking.
With all of these checks, the likelihood of deleting anything useful is nearly non-existent. An IP was warned with Twinkle six years ago for vandalizing Apple? Who cares? Easy-on, easy-off. --MZMcBride (talk) 21:14, 4 March 2009 (UTC)[reply]
Well, some of us do care. It is part of our 'game' here. Twinkle etc. makes it easier to warn, still sometimes 'the wrong warning' is given (something that is spamming is tagged vandalism). Others don't like our pre-fab warnings (it did not take me too much time to come up with the 15 IPs above, one has a friendly remark from a user who did not use the standard template, still, the user did remove the spammed links). There is no quick scan for these things, and if I go through my list of deleted edits to talkpages, there are numerous talkpages which already have been deleted. Of the 3 editors who have been warned for these link additions 2 have no easily trackable warning history anymore.... If you would now give me that as proof for blacklisting, I would say 'pff, OK, many additions, and it is useless, but please first try to warn the editor a bit more, maybe they will respond ..'. Or on the other hand, if now some editor comes, asking me for de-blacklisting, I would say on this proof 'sure, this was a bit quick, the editor has hardly been warned'. It really cost me time to find the warnings again, only one is in my deleted contris. More, current, examples: "89.216.66.248 89.216.66.36 89.216.66.54 89.216.66.213 89.216.66.95 89.216.66.183 89.216.66.19 89.216.66.80 89.216.66.204 89.216.66.224 89.216.64.142", "189.92.76.100 189.92.12.25 201.23.160.73 201.23.160.1 201.23.160.80 189.94.29.19 189.92.5.106 189.92.60.137 189.92.94.127 189.92.112.79 189.92.20.131 189.92.24.114 189.94.61.53 200.169.124.131 189.94.18.243 201.23.160.67 189.94.18.187 189.92.29.233 189.92.30.189 189.92.62.202", "86.150.201.168 86.153.149.73 86.153.150.184 81.156.233.53 86.152.191.89 86.170.178.200 86.170.181.57". Have a look at m:User:COIBot/XWiki, most cases involve many, many IPs on many wikis.
I indeed do not care if someone was six year ago warned for vandalizing Apple. But how do you see from the warning, without a good look at every single edit the editor performed, that this user was not a rogue member of the feared associtation of stroopwafel addicts, who was using a large group of IPs with as a single goal: wipe apple trees from this very planet. Most of the time, it is just simple vandalism, but you don't always see that from the tag. With spamming it is pretty clear (though in cases of mis-warning or non-standard warning it is still unclear), but with POV pushing it does not have to be clear. You are right, MZMcBride, in by far the most cases, it is simple vandalism, but there are several cases out there which look like simple vandalism, but are actually much more than that. Easy-on, easy-off .. and then a lot of work in some cases (did you already find the 2 IPs which were warned for spam?). I again ask, most of you don't see the use, but we really do see the loss. Deleting is easy, undeleting the ones which should not have been deleted practically impossible. Please consider archiving again, it serves both purposes well, and please precede it by blanket undeletion of all those user talk pages which were deleted (except the ones which in themselves were blatant spam). It only increases the number of edits a bit, and the db is a bit bigger (but that is not our concern). --Dirk BeetstraTC21:50, 4 March 2009 (UTC)[reply]
Just looking at the example you mentioned, "If you would now give me that as proof for blacklisting, I would say 'pff, OK, many additions, and it is useless, but please first try to warn the editor a bit more, maybe they will respond ..'." If someone came to you and said, "Hey this IP vandalized in 2007, got warned, and just did it again, please block." I think any admin would say, "You need to warn as we have no way of knowing if the IP belongs to same person any longer and the old warnings don't particularly apply over a year later." And, any contributions to any (non-deleted) page are still visible to any user, even though when dealing with IP addresses, it's always best to assume they've changed hands over the course of time, because that happens frequently with a lot of these IPs. So regardless of whether an IP was warned (for vandalism, spam, POV hoaxes, whatever), I simply don't see the relevance years later. A month later? Sure, I could understand. But if the IP hasn't edited in a year and the page hasn't been touched in a year, the account is very likely dormant or has changed hands. At which point all of the warnings mean very little (except in the case of spam tracking, as I've discovered and accounted for). --MZMcBride (talk) 22:11, 4 March 2009 (UTC)[reply]
Not that it really needs to be pointed out, but those IP talk pages wouldn't be deleted because (a) they're far too recent (they don't meet the 1 year last edit or 1 year last page activity requirements); and (b) they would fail the spam requirement. --MZMcBride (talk) 22:24, 4 March 2009 (UTC)[reply]
If the modus operandi is EXACTLY the same and pretty unique, then I would block an editor who had a couple of warnings in 2005, and started again now. For many spammers and POV pushers, that is often true. The IP has changed, it is still the same company, user. --Dirk BeetstraTC22:19, 4 March 2009 (UTC)[reply]
I think an important question here that MZMcBride has not addressed is why deletion is preferable to blanking in these cases. Dcoetzee22:26, 4 March 2009 (UTC)[reply]
Blank pages on a wiki drive me insane. When I come across a blue link, I expect there to be content on the page. If not, my mouse nearly instinctively reaches for the history tab to see why there isn't. Red links indicate no content, not blue links.
But more than philosophically, there's no need to keep these pages around (a fact confirmed by other wikis deleting them), there's the issue that only bots can blank a page without causing the orange bar of death to appear, and (quoting from previous discussions) "we end up with thousands of blanked talk pages sitting around. Truly, if they're going to be blanked, they may as well be deleted. These things do add up when doing things like database queries. You have the thousands of revisions on the talk pages currently plus thousands of more revisions to blank the page." Further, "Assume we have 1,000 talk pages each with one revision. If you go through and blank them, you're now at 2,000 revisions. If you delete them, the revisions are moved to the archive table (essentially going from 1,000 to 0) and are no longer cluttering up the revision table or my SQL query results. ;-) And the fact that the revisions aren't truly gone is yet another reason why deletion is really no big deal, no? Blank pages really are simply annoying and if we can avoid creating blank pages, I would prefer that option. When you come across them, it's unclear whether a blank page is the result of vandalism, a right to vanish attempt, or whatever. "So just leave a note saying it was blanked by a bot." That seems rather unnecessary. A red link is much quicker and easier to understand than a blue link that leads to a note about blanking." (Apologies for the copy-pasting, but re-keying all of this seems unproductive.) --MZMcBride (talk) 22:34, 4 March 2009 (UTC)[reply]
MZMcBride, I would block if there were warnings to other editors with the exact samemodus operandi. If Grawp is changing IP, then you know that he has been warned, and you block regardless if this new IP is warned or not. Why, because a) you know he has been warned, and b) you know it is the same person, even if he is using another IP. There are very similar cases here, but not on such a huge scale as Grawp. The problem is not our run-off-the-mill vandal, the problem is not Grawp, it is the cases inbetween.
But I see we don't get any further, I have been sleeping over this (and thinking for some time while being awake), and I read the above message "it must be the case that almost all articles that could be deleted using the rule, should be deleted, according to general consensus. CSD criteria should only cover situations where there is a strong precedent for deletion. Don't forget that a rule may be used in a way you don't expect if not carefully worded.". This part of the speedy is contestable. Under the current wording:
4 Old IP talk page. Talk pages of anonymous users may be blanked or deleted if they meet the following criteria: 1.) Never been blocked; 2.) Not using any unsubstituted templates (e.g., {{SharedIPEDU}}); 3.) No edits within the last year; 4.) No talk page activity within the last year; 5.) No incoming links to the page; and 6.) Is not spam-related.
I contest the part 6 here. The following is getting closer:
4 Old IP talk page. Talk pages of anonymous users may be deleted if they meet the following criteria: 1.) Never been blocked; 2.) No edits within the last year; 3.) No talk page activity within the last year; 4.) Not using any unsubstituted templates (e.g., {{SharedIPEDU}}); 5.) No incoming links to the page; and 6.) all edits of the IP are 'simple' vandalism, i.e., does not involve spamming, POV pushing, self-promotion etc. (see notes). If the criteria 4-6 are not met, consider archiving and replacing with an appropriate welcome template.
Notes:
Changing the external link of a high-school to 'redtube.com' is not spamming (even if it warned as such), addition of a personal blog however can be spamming.
Before deleting, check also if there are deleted contributions for the editor which may not be 'simple' vandalism.
The warnings that are actually on the talkpage are not a reflection of what the editor has been warned for, or a reflection of what the editor has been actually doing; some editors use custom warnings, sometimes 'wrong' warnings get placed, and sometimes warnings get removed by the editor.
(can be expanded, reworded). Note now, that here talkpages which are tagged with a spam template can now be deleted, as they constitute not actual true spamming, but simple vandalism. And others should now not get deleted, as they are part of a 'bigger' problem (which may include possible cross-wiki problems, so they are not necessarily visible here!). I am sorry, but blanket deletion is simply deleting pages which are of use to some other editors.
I am against blanking, I think it is for a user (i.e. someone who is just coming here to read a page) very strange to see the orange banner, and then clicking to a blank page. That is very confusing, even more confusing then clicking to a page full of warnings which are not meant for this user (and as I understand, that is the main reason for deletion, it clears the message flag, and there is simple nothing there). Blanking also would make the page being subject to this same rule after another year, where it is then absolutely subject for deletion, so the U4 is a sort of self-fulfilling prophecy.
I therefore also changed the order of the items and added a sub-clause. If the editor is likely to have changed (and now we can even say, after 2 months in stead of 1 year, which is quite long), move all warnings etc. into a subpage, and replace the contents with an adaptation of {{welcomeanon}}, which is 1) not directly focussing on possible contributions but on using the information (so not having a 'thank you for your contributions', but 'nice that you consult our database for more information'), b) discusses shortly and to the point that the IP was used (and maybe abused) before etc., c) friendly tries to suggest the editor to stay, contribute etc., and d) contains a convenience link to the archive. One could even consider a bot-run that does: 1) move the page to a subpage, 2) delete the redirect that is left behind, so the message flag is reset, and 3) makes a minor bot contribution to put the welcome-template and a link to the archive, so that the message flag does not get set.
I believe this takes away MZMcBride's concern about bitey messages on talkpages which are useless for the new user of the IP (who is maybe not even a contributor, yet!), and still leaves us with the track that we sometimes do need. I know, MZMcBride, you don't see the use of these tracks, and we keep telling that for some of us some of these are tracks that we do need, but apparently these concerns don't mean a thing here. For most of you, these pages are useless, and therefore a blanket deletion is applied. The current form of U4 is contestable, and seeing the above banner, therefore should not be applied in this way, even if it is just true for a couple of editors. I am sorry, but this deletion, as well as the deletion of the related case with the 'temporary user pages', is doing a big disservice to some of us, as A. B. said, he lost a lot of data, and undeletion is an almost impossible task. --Dirk BeetstraTC09:50, 5 March 2009 (UTC)[reply]
I oppose the proposed U4, on the grounds that identifying "simple vandalism" is too subjective. There's a big grey area here. If someone posts text saying "JOHN DOE IS A PAINTER IN TEXAS" is that self-promotion, an attempted outing, or simple vandalism? If someone posts "FORD TRUCKS RULE" is that simple vandalism, or spam? You get the idea. Dcoetzee19:45, 5 March 2009 (UTC)[reply]
Heh, Doetzee, that is the whole thing. See the header of this text "Uncontestable: it must be the case that almost all articles that could be deleted using the rule, should be deleted, according to general consensus. CSD criteria should only cover situations where there is a strong precedent for deletion. Don't forget that a rule may be used in a way you don't expect if not carefully worded.". The CURRENT U4 is contestable, as the criteria are wrong. It should not be there at all. --Dirk BeetstraTC09:44, 6 March 2009 (UTC)[reply]
Even stronger, I think that Hu12, A. B., me and some other currently contest the U4 that we have (which is actually still actively used). --Dirk BeetstraTC09:46, 6 March 2009 (UTC)[reply]
I don't think creating IP subpages is a good idea, if a bot is to do this it should just add the {{older}} template or even a permalink to the page at the state of blanking (the way I do it when cleaning IP talk pages of older warnings). –xeno (talk) 14:27, 6 March 2009 (UTC)[reply]
I could live with that as well.
I have done some more research, here. The second case is the most problematic. 3 out of the 4 talkpages in this spam case would still be deleted (in the beginning of next year, if you ignore the now incoming link from this research page). I really would like this to be discussed further. --Dirk BeetstraTC14:31, 6 March 2009 (UTC)[reply]
An IP vandalizes Apple. Nobody blocks, nobody brings it up anywhere and links to the user, nobody does anything except hit a tab in Twinkle and leave a warning. Three years later, without any further activity from the IP or activity on the talk page, why does anyone care? There are six strong safeguards here (including checking page text). Other wikis do this without any issues. What is the big deal? --MZMcBride (talk) 03:28, 7 March 2009 (UTC)[reply]
No, for those, no-one cares. Yes, most can be deleted, the trouble is, which ones not to delete but to blank. The safeguards are not strong, they are too easily to get around. --Dirk BeetstraTC17:58, 8 March 2009 (UTC)[reply]
That's a problem connected to dynamic IPs. This won't solve that substantially at all. And if we did want to solve this by removing the warnings a bot blanking the page after a certain amount of time would work just as well without damaging transparency. JoshuaZ (talk) 18:12, 6 March 2009 (UTC)[reply]
We don't have to do everything with a reason. There doesn't have to be a benefit to it; if people want to delete such pages, they can, beneficial or not. We do not have to provide reasons for doing everything. Majorlytalk18:31, 6 March 2009 (UTC)[reply]
Actually we should be doing things with the reason "build an encyclopedia". We actually do need reasons to delete things. That is what the little box you type your delete reason is for. Chillum18:39, 6 March 2009 (UTC)[reply]
1) Damaged transparency. The project is committed to openess whenever possible. This undermines that. 2) Increased log sizes which make it harder to find specific items in the log 3) Increased server use (deletion is a relatively expensive operation and deleted edits take up more room than undeleted edits). None of those are very strong reasons aside from the first reason, but even the last two mean that we shouldn't be doing this without something resembling a coherent reason other than people simply wanting to delete the pages. JoshuaZ (talk) 18:59, 6 March 2009 (UTC)[reply]
1) Openness for the sake of openness is as bad as any arbitrary rule, worse in some cases. 3) Don't worry about performance. A 5 MB image takes up about as much space as 2000 deleted edits, and we allow images up to 100 MB. Deleting them also reduces the size of the database dumps of all pages. Mr.Z-man18:21, 7 March 2009 (UTC)[reply]
Openess for the sake openess isn't a bad thing as long as we don't give it that much wait. However, having the default situation be open and transparent is completely reasonable. That's the situation here. JoshuaZ (talk) 01:18, 8 March 2009 (UTC)[reply]
I have said before, and I am going to repeat. Really, most of these pages can safely be deleted, as they are related to single vandalism. However, there are cases where there is more going on, it is coordinated, or ongoing, or long term vandalism, it is continued POV-pushing, or is widescale persistent spamming. Those pages get used in other wikipedia processes. Deleting these does not only hamper those investigations (we are not all admins here, and even for admins, it is still quite some work to find from a list of 15 'random' IPs which one were warned for this typical modus operandi of 'vandalism' if the talkpages are deleted, you have to go to all 15 of them, check the deletion history. The process becomes even worse if the pages get recreated, then you have to check behind the page if there are also deleted contributions. The problem is, that under the current settings, the criteria are sloppy and incomplete. And probably they will never be complete enough for a bot to delete them.
A second point is transparency. If we blacklist, and the editor comes with 'I did not know', then with the deleted talkpages we can't say 'you were warned' (with again looking them all up ..). If the histories are there, even a non-admin can do it, simple. If the situation is cross-wiki (and that is what some of us handle) the situation becomes increasingly more difficult.
As for the pages about continued POV-pushing, long-term vandalism, etc. Those pages will likely have links, categories, and templates on them, and many will be blocked. I don't see how you would be able to say "you were warned" if the talk pages weren't deleted without looking them up, unless you just assume that a blue link talk page meant that they were warned, which is about as bad an assumption as assuming a talk page that hasn't been edited in a year still belongs to the same person it did a year ago, or that you can connect it to someone else despite having no categories, templates, or links to track it. I think I said this before somewhere else, but never got a straight answer: Why is putting a template or a category on pages we might want to keep such a horrible idea? We've been doing it for sockpuppets for years. It'll make tracking and writing bots/scripts to aid in tracking easier and it'll allow us to more easily delete the pages that we don't want to keep. Mr.Z-man18:21, 7 March 2009 (UTC)[reply]
Josh: Damaged transparency? All of the IP's contributions are still visible. The only thing that isn't visible are the templated warnings (that everyone seems to agree are often inaccurate). And, of course this only applies to IPs that have never been blocked, haven't edited within a year, aren't link to from anywhere, have no recent edits to their talk page, etc. So what transparency is lost? You can't see that someone warned an IP six years ago for petty vandalism using Twinkle? Who cares? --MZMcBride (talk) 03:24, 7 March 2009 (UTC)[reply]
My main complaint with the criterion as it stands is (6) "all edits of the IP are 'simple' vandalism." Not only is "simple vandalism" highly subjective, but this is likely to never be true - most IPs will have non-vandalism edits in their history somewhere, especially dynamic IPs. We are deleting these pages under the assumption that a good user may use the IP in the future - but won't that very act make it ineligible for deletion under this criterion ever again? This is not easy to fix - if you only consider a subset of their contributions, you introduce subjectivity in the choice of what contributions to include. Dcoetzee04:04, 7 March 2009 (UTC)[reply]
Oi, you are right there, "all edits of the IP are at the worst 'simple' vandalism". The point is filtering out those which are POV-pushing, spamming, etc. If they are not that, they are simple vandalism, if they may be any of the other, don't delete, 'blank' (in whichever form ..). Behold, blanking SHOULD be done, the IP is indeed probably not the same user anymore. All I am arguing is, in case of doubt, make sure that the page history does not get hidden. That is the transparency that I mean. And even, that loss in transparency is often just a minor problem, but in some cases it certainly is not. We do get spammers saying 'we did not get warned' .. same goes for contributors: 'how do you dare blacklisting that, the editor did not get warned, why not try that?', the same is true for POV-pushers and some other forms of persistent vandalism. --Dirk BeetstraTC17:56, 8 March 2009 (UTC)[reply]
I think Beestra has a very valid point that speaks against such a criterion. Blanking the IP's talk page using a bot will get the same result (new user with this IP will not be confused by old warnings) but will not lose the page history to non-admins. I have yet to hear a reason why only deleting will achieve an acceptable result. SoWhy20:03, 8 March 2009 (UTC)[reply]
New talk page categories for spammers
As of today, when new spam, advertising or coi warning templates are placed on spammers' talk pages, the template when substituted will automatically add the page to the appropriate new category below:
Can we include the coi template as well, and there may be more warning templates which point towards a more structural problem with the edits then 'simple vandalism'. And this still does not resolve the problem of editors warning with {{uw-vandalism}}, while the edit is clearly promotional. --Dirk BeetstraTC17:37, 9 March 2009 (UTC)[reply]
Problems with the the proposed U4
If we want to see the contribs of the user, then we have already found him (those tracks are generally pretty clear). But has the user been warned for the actions? That is done by checking timing of warning and contents of warning against the contribs of the editor. If the revids of the talkpage are deleted, it comes down to admins to check these. Know, that if the editor adds exact the same piece of text 2 year later, then it is physically the same person, even though the editor has now a completely different IP and the old IP is now used by someone else. I see with XLinkBot often cases of some IPs spamming, me putting the link on XLinkBot, and then XLinkBot reverts and warns a couple of times. Then 3 months later suddenly an account comes up, autoconfirmed and all, who spams exactly the same articles. What happened? Their IP gets warned, they create an account, do some unrelated or minor edits or wait three months doing nothing, and start spamming. Timing of account creation only minutes after the last IP stopped spamming. No blocks yet, abuse stopped .... As I have given as example earlier, if Grawp comes with a new IP, you block on sight, why? Because you know his modus operandi, and know he was warned. These cases are often less bad, still for some, if they return after a week, a month, or two years ... we sometimes know it is the same physical person, and hence immediate block is often in place.
I know that speedy criteria are never complete, but if I check 4 existing talkpages and see 3 fail in a range of 256 possible talkpages .. and on the other hand, another range of 256 talkpages had some which stayed and could probably be deleted. But the first 75% failure rate, 50% on the other example I mentioned. I really hope that the error rate over tens of thousands of pages (I heard A. B. saying that he was getting tired of scrolling through pages with 5000 deleted pages) is getting way below 1% .. but I think it is going to be difficult to get it below 5% .. that is in the end also thousands of pages.
About incomplete:
Blocks: blocks includes rangeblocks (easy to check locally), and since some time also for blocks applied mediawiki wide (used nowadays while e.g. waiting for meta blacklists to kick in).
Blocks 2: IPs often don't get blocked, they change their IP and go on. Only if we are really fast, that makes sense. And some IPs we simply can't block or only if it really gets bad (those UK IPs which were used widely after the album cover image, government IPs?).
Incoming links 1, if the case is cross-wiki, there are not always local reports, but the incoming links are coming from other wikis (mainly meta).
No activity by the IP, that should include deleted activity.
Incoming links 2, not all actions require a signature (is that not omited for RFCs?).
Incoming links 3, especially IPs don't always know that they should sign. Sinebot is active since August 2007, pages before that were often not signed, are we sure that that criterium can be met? And then still, Sinebot does also make mistakes, I see it miss signatures. And some IPs sign by hand, not leaving a link to the talkpage. And where those IPs should have signed, those are often discussions that matter.
Tagging: Some of the deletions I saw had an IPsock tag on the userpage, not on the talkpage. Indeed, talkpage is not tagged.
Incoming links 4, some of our reports link to the userpage, or to the special:contributions (I do that with the XLinkBot log), or may just state the whole range and not link to individual user(-talk) pages.
Warnings: as mentioned, warnings are not always proper (vandalism warnings for spam, etc.).
Warnings 2: also mentioned, people use still custom warnings, 'Hi, I have reverted your edits, please don't insert your own links. Thanks. (sig)' .. still spam!
Users are allowed to blank warnings, under the idea that they have then read the warning. If the warnings for spam get removed by the IP, the word spam is not anymore on the talkpage, so it may be deleted.
On the other hand:
Some unsubstituted templates are just as confusing as a spam message which is not for the editor. Why keep it if the IP(-range) is hardly used)?
Talkpages with only a nice and friendly welcome message get deleted.
By far not all the 'spam' warnings for users are for spam, it may be plain vandalism where the user replaced e.g. the homepage of their school with a movie on redtube. Those get by XLinkBot tagged for spam (well, for inappropriate external link addition I prefer to think), but are just plain vandalism. Similar for taggings for inclusion of example.com. Difficult to distinguish!
Blocks 2: If the user has changed, then having had a block on the IP is not a reason not to at least blank the page. The user is long gone.
There may be more points of interest here, but this shows in my (humble) opinion, that automated deletion is dangerous. I have too many concerns, some of which already raised over a month ago, which under the current policy were not met, or which were impossible to check automatically (or which the current automated process did not seem to follow). I know the policy was discussed over and over, and was (apparently) carefully thought through, but I am afraid it was not done good enough. (copied from my talk). --Dirk BeetstraTC18:13, 8 March 2009 (UTC)[reply]
"... we sometimes know it is the same physical person, and hence immediate block is often in place." This implies a tracking system of some sort. And to my knowledge there isn't one in place.
The criteria were created the way they were because it provides an easy way to find users / user talk pages where there hasn't been any substantive activity. 365 days without any activity; 365 since the last edit; no templates; etc. I've asked you previously, but are there any examples of returning users coming back after years that also would have their user talk pages deleted? Some of the examples you cited in your /DeletedTalks page would not be deleted due to reports like Wikipedia:WikiProject Spam/LinkReports/public-domain-image.com.
And I don't think you addressed the issue of why warnings matter when they're so often wrong. Contributions are still visible to everyone. --MZMcBride (talk) 20:00, 8 March 2009 (UTC)[reply]
Tracking the users, not perfect, but yes, if the modus operandi is too similar, and the IP is in the same range, then that assumption is quite justified. And even if not so, it is returning abuse, so the pages are of interest.
I'll have a look for returning abusers. I know I blocked an editor not too long ago who 1) spammed as an IP, but never got blocked, 2) stopped spamming, but created an account immediately, and then 3) returned months later. I'll have a look.
The warnings are not wrong, the example with the 4 IPs of the public-domain-image were all four warned because what they were essentially doing: spamming (though vandalism warnings were used). That does not mean they were not warned. Indeed, these editors are covered because they do have incoming links, but as I said, some cases handled on meta do not get such local links, except if they are there for other reasons. --Dirk BeetstraTC20:08, 8 March 2009 (UTC)[reply]
So, let's look at one of your examples: User talk:89.216.66.107. It's a generic {{uw-vandalism1}} warning. It obviously wouldn't be deleted for another year. But what's the purpose of keeping it after a year? It's a generic template and as you say, the IPs often rotate for these spammers. And the contributions would still be visible to everyone. The only thing lost is a generic warning template. Why did you cite this as problematic? --MZMcBride (talk) 20:11, 8 March 2009 (UTC)[reply]
The template is a generic uw-vandalims1, but the actual action was spam. 4 Editors got warned for spam .. 8 editors adding it, I would say, blacklist locally. But now envisage long terms between the edits, or cross-wiki. And by the way, we have long-term POV pushers or vandals as well. Is it so hard to see that we have examples of editors who stay away for a year but keep doing the same edits (I'll come up with an example of one, though I am pretty sure there are enough tracks there not deleted, he is persistent)? --Dirk BeetstraTC20:43, 8 March 2009 (UTC)[reply]
Examples: diff, diff, etc. etc. reverted many many times.
Page protected: diff by me on the second of August, 2007, expires the second of February, 2008, 6 months.
22 Feb, 2008, diff, editor is back, 20 days after the deprotection. In a couple of days we have one year after I reprotected it again (protection is indef now), but we can do the experiment, do you want to see if the same editor returns?
Here there are many, many talkpages tagged in various ways, and I am sure that there are discussions pointing to many of them. I don't think all of them will be deleted, but it is an example of 1 editor, always the same one, who does return after long times of absence. I have absolutely NO doubt that it is always one and the same physical editor. --Dirk BeetstraTC20:57, 8 March 2009 (UTC)[reply]
No, this case is indeed quite big, though some talkpages in this huge range of editors have already been deleted (will have trouble finding one, have to follow all the deleted revisions ...).
The purpose: we don't blacklist if there have not been attempts to warn the editor. The editor has been warned that these edits are unwanted, it is therefore support for us that we can safely blacklist: the editor persists after attempts to communicate. --Dirk BeetstraTC21:18, 8 March 2009 (UTC)[reply]
Active on 15 September 2008, 17 September 2008, 5 October 2007 and 8 March 2009. Same link, 6 months and 15 days after first IP
OK, the first IP did not get warned, but while it is quite clear that these 3 editors are the same physical person (same link to same pages, close range, activity periods that overlap), it shows how persistent editors can be, they will return after months and months of inactivity. Deleting after one year (when warned on 29 February 2008) and having the bad luck of no incoming links (there are none!) would result in the old talk page being gone. I hope this shows to what length people go to spam. 2 of these editors do have incoming Now imagine a SEO being payed for links being added ... --Dirk BeetstraTC21:53, 8 March 2009 (UTC)[reply]
I routinely deal with accounts that have spammed off and on going back to 2006. They often have multiple domains, too. So you get one IP adding one domain for a while, then another account taking the baton and spamming another domain for a few months, etc. We need to know who's gotten warnings and how many. Some of our volunteers are not admins. Likewise, most of the meta admins that deal with the Wikimedia global spam blacklist are not en.wikipedia admins, so they can't see our deleted talk pages either.
Do they have a policy for this, I'd like to see their reasons why they think certain pages are not necessery anymore (I will also ask them myself). --Dirk BeetstraTC09:11, 9 March 2009 (UTC)[reply]
See McMzBride's comment above under my comment "I think an important question here that MZMcBride has not addressed is why deletion is preferable to blanking in these cases." I don't find it terribly convincing myself - I think maintaining transparency and keeping notice history available to non-admins is more important than the cosmetic issues he raises, regardless of their merit. Dcoetzee10:44, 9 March 2009 (UTC)[reply]
Reply to a question from MZMcBride above: The purpose is: the physical user, that is the person who performs this type of 'signature vandalism', was warned on the old IP. That is information which can be of interest later on. And we are still talking about vandals and spammers, there are also occasions where users discuss on their talkpage and do edits based on those discussions. They may not have been involved in other discussions, or those may not have been signed (yes, we NOW have Sinebot, but this deletion also deletes pages from 2005!). How are you going to detect if these users have been involved in other, possibly useful, discussions? You can partially catch them by loading their contribs outside of mainspace and see where they contributed, but then you still have the problem of deleted contris, you also have to scan these, and vandals may also vandalise outside of mainspace, so it again cuts both ways. Still .. There are a lot of scenarios thinkable which are very, very difficult to detect.
MZMcBride, you strongly give the feeling here, that because you (and quite some others) don't see the use of them, that we are really just moaning, and actually don't have a use for them. Can everybody here really say that by far the most of the old talkpages of IPs that fall under the current #U4 conditions are utterly useless, and that by far the most that do not fall under the current set of conditions for #U4 are useful, and that there are no alternatives to this? Seen the above set of arguments I posed, and the different examples that I have shown, what do we think the 'error rate' is? I hear that A. B. already gave up undeleting the ones which are of interest (to him).
However, blanking the page has the same non-bitey effect as deletion of the whole talkpage (I still prefer to 'archive' or replace with a new template: A bot could do that easily, record the current revid, and replace contents with a new template: {{subst:re-welcomeIP|oldrevid}} on it, a blank page may still be confusing). We might even be able to find a way to 'reset' the new messages bit. This all without us even having to discuss a) which to delete and why are they useful and others delete them as well etc. etc., seen that we can (easily!) come up with examples where the info is clearly related to old cases (I'd like A. B. to show a couple as well), and b) we have your deleted contribs, a bot could just go through them now, undelete, and recreate them as said with the {{re-welcomeIP}} and we are all happy (well, except for those who worry about having all these talkpages hanging around). Further archiving could be done by a (non-admin!) bot, with less detection problems. I don't care about the #U4#6 'involved in spamming' anymore, replace them; the time period could be 3 weeks for volatile IPs, 3 months for static IPs (not a year, a lot of users use some IPs in a years period, getting all those annoying messages!!), unsubstituted templates (which may also not be true anymore!), and one could even do this for blocked users (which IPs can also be used by new physical users or be assigned to other ISPs etc., block templates can be very, very bitey for the next user!). The only problem that I see is that you make 100.000 new revids in the database for talkpages (how many did you say you deleted?) but that is not our problem, we'll discuss about that if the developers have problems with it. IMHO, this is clearing out many, many MORE old IP talkpages then the current conditions, is much LESS bitey then leaving the ones that are still left, and is not 'destroying' useful information (whatever who thinks what is useful and useless), resulting in improved transparency! --Dirk BeetstraTC10:49, 9 March 2009 (UTC)[reply]
Name of policy
(unindent) Not to completely stir the pot, but ... should this page be renamed to something a bit more descriptive? We've had the whole "speedy doesn't actually mean quickly" issue for years. --MZMcBride (talk) 02:07, 4 March 2009 (UTC)[reply]
No, definitely not stirring the pot. :P I've always accepted "speedy deletion" to not mean "instant deletion" but rather "deletion done out of normal processes" (XFD, DRV, etc). Hence why OLDIP and CAT:TEMP make sense to be included here, IMHO. —Locke Cole • t • c02:17, 4 March 2009 (UTC)[reply]
Good question, hmm. Looking at the first line of the policy - "Criteria for speedy deletion specify the limited cases where administrators may delete Wikipedia pages or media without discussion" - I suggest "Criteria for deletion without discussion." Dcoetzee02:34, 4 March 2009 (UTC)[reply]
Noting that CSD is an administrative method of deletion (XfD is community deletion), I did a thesaurus search on administrative, and further thesaurus searches on initial hits, selecting for words starting with “s”. This gave:
Criteria for summary deletion (see summary judgment) would be my first preference were it not for the fact that it could lead some people to draw parallels between Wikipedia and a court of law or between deletion and extrajudicial killing. Of course, that being said, "speedy deletion" is also be misleading because for most people 'speedy' = 'quick'. By the way, does anyone know why this page is not at Wikipedia:Speedy deletion (a redirect)? What function does the "Criteria for" serve? –Black Falcon(Talk)06:46, 4 March 2009 (UTC)[reply]
Actually that's a great name - the analogy is perfect, and we don't have to change the initialism. :-) The association of "summary" with summary execution is particular is unfortunate, but I don't think we'd find anything better. Dcoetzee07:07, 4 March 2009 (UTC)[reply]
Oh no, not "summary deletion", people may well think that it is for the deletion of unwanted edit summaries. We should just redivide the deletions between "deletions without discussion" (which includes prod and all CSDs) and "deletions after discussions (which includes all the *fD pages). Fram (talk) 10:09, 4 March 2009 (UTC)[reply]
Here's a relevant thesaurus entry. It lists: "summary, immediate, instant, instantaneous, direct, prompt, speedy, swift, rapid, without delay, sudden, abrupt, hasty, peremptory, without discussion, without formality." I think given the available options and the desire to retain the original initialism, we're either going to stick with what we have now or go to Wikipedia:Summary deletion (Fram's concerns about confusion are legitimate - my uncertainty revolves around whether the word "speedy" or the word "summary" is more misleading.) Dcoetzee21:33, 4 March 2009 (UTC)[reply]
If this thread had been started by Jimbo saying "find an alternative name for WP:CSD", then yes, maybe. How can the work of volunteers be busywork? If no one thought that a task here was worth doing, then it would never be done, simple as that. The fact that people are prepared to consider such a change indicates that those people don't think it is a pointless exercise. If there are valid objections to the proposal, that's one thing. Saying that it should not be done because you personally don't think it is important is another matter entirely. I agree that this is way down the list of Things That Are Broken On Wikipedia. The wonderful thing about being part of this community, however, is that we can work on anything we like from that list, no one can order us to start at the top. Happy‑melon13:09, 5 March 2009 (UTC)[reply]
To avoid getting sidetracked, yes, this is not the highest priority thing to do; the only purpose would be to avoid some confusion caused by the current name and I could let it drop. I think Stifle is well aware that volunteers will work on what interests them. Dcoetzee20:11, 5 March 2009 (UTC)[reply]
Language affects the way that people behave, and the word "speedy" seems to create an impression that the speed of deletion is more important that whether it is correct. I remember recently seeing a disputed speedy deletion (I'm not sure whether it was at WP:DRV or on a talk page) where the admin said in defence that there were lots of pages in Category:Candidates for speedy deletion so it was necessary to go through them quickly. It's good that the word "summary" brings to mind summary execution, as it should help focus admins' minds on what they are doing to people's contributions. I support changing the wording to "summary deletion". Phil Bridger (talk) 20:20, 5 March 2009 (UTC)[reply]
I could support "criteria for summary deletion", although I don't think we should necessarily worry about keeping the acronym consistent, so I also propose "criteria for administrative deletion" or "criteria for undiscussed deletion" (I especially like the last one, if only because WP:CUD sounds like we should take some time to chew it over before deletion. *grin*) Anyway, when I was a noob I was certainly under the impression that "speedy" meant "quick", as opposed to "undiscussed", and it took me quite a while, and a lot of poking through archives of old discussions, before I realized that CSD was invented to reduce the load on the XfD processes, as opposed to getting rid of stuff quickly.--Aervanath (talk) 02:25, 6 March 2009 (UTC)[reply]
Okay, you'd think this could fall under G1 as a hoax, but I wouldn't consider this to be "patent nonsense", anyway. I think we need a new image criteria for hoax images, including unofficial or "fake" versions of cover art, packaging, etc.
Blatant hoax, an image that is blatantly false or is a hoax, unless it is being used to illustrate the fact that it is a hoax. Examples include "fan-made" cover art for photographs and blatantly modified images.
Main issue is subjectivity. It can be difficult to distinguish a well-made hoax from a difficult-to-believe real thing. It can also be difficult to distinguish a poorly-made hoax from a... poorly-made real thing. And well, if we excluded modified images, all our magazine covers would be right out. :-P Dcoetzee03:42, 4 March 2009 (UTC)[reply]
I agree with SoWhy. If the fake cover is so obvious as to be clearly a hoax, G3 already covers it. If it might not be a hoax, it shouldn't be speedied. Stifle (talk) 12:45, 4 March 2009 (UTC)[reply]
User pages eligible under G11?
I sometimes see user pages or user subpages (which often look like articles-under-development) nominated for speedy under G11. I cannot find any past discussion about whether user pages are even eligible for deletion under this criterion. The text of the criterion refers to articles, but it is in the G rather than A section. My inclination is not to delete things out of a user's userspace unless there is some clear harm to Wikipedia. Is there any established consensus on this issue? --R'n'B (call me Russ) 16:50, 4 March 2009 (UTC)[reply]
True. But G11 talks about "Pages that exclusively promote some entity and that would need to be fundamentally rewritten to become encyclopedic. Note that simply having a company or product as its subject does not qualify an article for this criterion." The two underlined phrases raise doubt in my mind. User pages are not supposed to be encyclopedic, so the first phrase does not seem to establish any meaningful criterion for assessing them. And the second, I think, is obvious. --R'n'B (call me Russ) 19:03, 4 March 2009 (UTC)[reply]
I agree - we may want to delete spam user pages, but this criterion as stated makes very little sense for them (a user may reasonably ask, how do I fix it? It doesn't tell them.) I suggest rewording G11 to describe different requirements for articles and other types of pages. Dcoetzee21:13, 4 March 2009 (UTC)[reply]
(Edit conflict) The user pages that should be speedied under G11 are the ones that are blatant ads — either fashioned to look like an encyclopedia article but needing to be fundamentally rewritten to become encyclopedic, or in some cases just pure marketing copy using Wikipedia user space as a web host. The ones disguised as articles can be difficult to distinguish from sandbox works-in-progress, but only as difficult as it is to distinguish G11 and A7 deletions in article space. If you'd seen any of the pure marketing copy ones you wouldn't be so doubtful about why this applies to userspace (and sometimes user talk space) as well as article space. —David Eppstein (talk) 21:17, 4 March 2009 (UTC)[reply]
Okay, why not something like this?
Blatant advertising, any content that exclusively advertises some entity with no non-promotional content worth saving. Note that simply having a company or product as its subject or the introduction of promotional language into an otherwise neutral page does not qualify an article for this criterion.
Pages created exclusively to promote some entity and that would need to be fundamentally rewritten to become encyclopedic. Pages outside the article namespace fall under this criterion if they would be deleted as an article. Note that simply having a company or product as its subject does not qualify a page for this criterion.
”
It takes the idea from P1, i.e. that it's not important which namespace is chosen to promote the entity. It's strict enough as long as admins do not see this as a free pass to self-promoting userpages and suchlike, which G11 never was anyway. An userpage with "I'm John Doe and I am the best user here!" is promotional but it's not a G11. "Buy Viagra now at www.fakeviagrastore.com!" would be deleted at Buy Viagra Now! and at User:Buy Viagra Now. Regards SoWhy11:55, 5 March 2009 (UTC)[reply]
Well, as R'n'B points out, the current wording is a bit strange because for example user pages cannot be rewritten to "become encyclopedic". I think it needs a better wording to make clear that it's not only for articles but I admit that I am not very good at it. SoWhy09:57, 6 March 2009 (UTC)[reply]
Content in any namespace has the potential to become encyclopedic. I could draft an article on my userspace then move it when it is done. The wording is accurate in that we use the likelihood of it ever being encyclopedic to determine if it is advertising regardless of the namespace. Chillum18:44, 6 March 2009 (UTC)[reply]
The concern is that, for example, a user's user page may be deleted because it would "need to be fundamentally rewritten to become encyclopedic"; but it may not need to be fundamentally rewritten just to be an acceptable user page. Dcoetzee23:36, 6 March 2009 (UTC)[reply]
You might adopt some wording from this User Talk template I use when tagging spam User pages ({{spam-warn-userpage}}):
A tag has been placed on your user page, User:EXAMPLE, requesting that it be speedily deleted from Wikipedia. This has been done under the criteria for speedy deletion, because the page seems to be blatant advertising which only promotes a company, product, group, service or person, and which is a violation of our policies regarding acceptable use of user pages: user pages are intended for active editors of Wikipedia to communicate with one another as part of the process of creating encyclopedic content, and should not be mistaken for free webhosting resources. Please read the general criteria for speedy deletion, particularly item 11, as well as the guidelines on spam, the guidelines on user pages, and, especially, our FAQ for businesses.
If you can indicate why the page is not blatant advertising, you may contest the tagging. To do this, please add {{hangon}} on the top of the page in question and leave a note on this page explaining your position. Please do not remove the speedy deletion tag yourself. Feel free to leave a note on my talk page if you have any questions about this.
The various screenshots used in Oddworld: Abe's Oddysee have now been tagged to speedy deletion a total of three times by the same user, the first two having been undone by two different users. In neither the second nor the third instance of tagging did the tagging user leave a summary of why s/he was tagging the page after it had been removed. Is there some rule about this? It just doesn't seem right to edit war of a file's speedy deletion without explanation from the nominator... I tried to leave a decent explanation when I first removed the tags, and again at Wikipedia:Featured article review/Oddworld: Abe's Oddysee, but there's been no response.
Reasonably disputed CSDs should go to AfD (for example, if the person removing it provides a good reason why it does not satisfy the criterion or otherwise should not be deleted). Unreasonably disputed CSDs should be re-tagged. I realize that's rather subjective, but that's generally how we do it. Dcoetzee22:54, 4 March 2009 (UTC)[reply]
Dcoetzee has it exactly right. If the person removing the tag has given any kind of halfway reasonable rationale, it's no longer uncontroversial. AfD would be the correct way to go at that point. Now if we could just get more admins to look at the history, and not speedily delete an article 10 minutes after someone else declines the same tag... *sigh* --Fabrictramp | talk to me23:00, 4 March 2009 (UTC)[reply]
(ec)That's the thing; when I'd removed the CSD tags, I said that it should be taken to WP:FFD (since there were at least two users who opposed the speedy), but the tags were simply readded without further explanation. -–Drilnoth (T • C) 23:13, 4 March 2009 (UTC)[reply]
About all you can do at that spot is talk to the individual editors involved. I find a friendly note gets about a 50% response rate of "*facepalm*, you're right, I should have checked the history" and about 50% getting ignored. So that's about 50% better than if I say nothing. :)--Fabrictramp | talk to me23:18, 4 March 2009 (UTC)[reply]
Note that the images were not tagged for speedy deletion, they were tagged as disputed fair use rationales. If the tag had persisted for seven days, they would then have become candidates for speedy deletion. Stifle (talk) 11:37, 5 March 2009 (UTC)[reply]
For example, if we had (hypothetically) a stub article created at ManUtd FC, or an article that completely duplicates Manchester United F.C., we don't technically have a criteria to cover this clear duplication. This would therefore require an AFD debate in order to delete it through the correct process. Therefore - and I'm just throwing this out there - perhaps a new criteria for "Articles with consist of content that duplicates that already available in existing articles". The problem I can immediately see is that it could interfere with expanding coverage of a topic, but if a topic is already covered on Wikipedia and an article does not offer anything substantive worth merging, a criteria for such scenarios could be an idea. Esteffect (talk) 00:32, 7 March 2009 (UTC)[reply]
In that situation, the new stub should generally be just made into a redirect to the article. If someone's ended up creating it, it's probably a likely search term. Algebraist00:35, 7 March 2009 (UTC)[reply]
Agree or use {{merge}} if there is content that can be moved there. As Algebraist points out, there is no reason to delete it because it's a likely redirect candidate then. SoWhy18:28, 7 March 2009 (UTC)[reply]
I agree with what's been said above and will only add that if the duplicate page is located at an implausible search title (e.g. a duplicate article about Birmingham at Biggringham), either criterion G2 (test page) or criterion G6 (housekeeping) could apply. Criterion R2 (redirect from implausible typo or misnomer) could also apply once the page is redirected if there is no useful page history. –Black Falcon(Talk)06:27, 9 March 2009 (UTC)[reply]
Well, my proposal to speedy delete Thing Thing Arena 3 got rejected because of someone's interpretation that online games do not count as web content per A7 (which says it doesn't apply to "software"). But, if anything, it should distinctly acknowledge this, so I propose a little tweak to A7:
An article about a real person, an organization (e.g. band, club, company, etc., except schools), or web content (e.g. a web site, blog, image, video, browser game, etc, except software) that does not indicate why its subject is important or significant. This is distinct from verifiability and reliability of sources, and is a lower standard than notability.
It's pretty obvious that there are online games that are not web content (like pretty much any modern retail PC game). I'm hesitant even in the case of web-based online games to sanction this expansion. Dcoetzee14:22, 9 March 2009 (UTC)[reply]
Certainly "browser game" is a better phrase to use, "online game" is a very general one. I'd agree with you that, given a reasonable interpretation of the current A7 standard, a small Flash-based browser game should be deletable under A7 (as "other web content") but a conventional computer game which is played online should not. Certainly specifying "browser game" is better if you want to change the wording here without consensus for an overall change of the scope of A7. ~ mazcat|c14:29, 9 March 2009 (UTC)[reply]
I don't have any opinion on the wording change, but I wanted to clarify a misconception on ViperSnake151's part. I was the admin who rejected the A7. The article as it was at that time (I haven't looked at it for a couple of days) did not say if the game was a browser game, other online game, or a game you play entirely on your own computer. I assumed the latter, which would not be eligible for A7 even under this proposed wording. When ViperSnake151 complained about it on my talk page, I did a quick gsearch, saw s/he was correct that it was an online game, and apologized on my talk page for the error.--Fabrictramp | talk to me14:44, 9 March 2009 (UTC)[reply]
Agree, no change necessary. A7's wording covers flash and browser games already and this was a unique case where Fabrictramp correctly declined the speedy based on the information in the article because it was not clear that it fitted A7. There is no need to change the wording as noone debates that those games are within A7's scope. SoWhy07:30, 10 March 2009 (UTC)[reply]
