Misplaced Pages

Revision as of 02:26, 19 September 2005

Archives: 1 2 3

Template:Wpnamespacemove

Proposal

Per the above, and per several discussions on talk pages, it seems like a good idea to formalize the following:

• (1) If an admin's closing of a deletion debate is disputed, the matter should be taken here, regardless of whether the disputed decision was to keep or to delete.
• If the outcome was one of the following: keep, merge or redirect, and you think it should be one of the others, just be bold and {{sofixit}}, that doesn't require more debate. It's okay to add a note to the VFD debate that the article is merged now. It's also okay to omit that.
• (2) This process page should deal with the technical matter of whether a closing was done properly (as it mostly does now) and should be renamed to reflect that. Proposed name is "Deletion review".
• (3) If something is being discussed on this page, especially if it turns out somewhat controversial, then this process should not be bypassed by starting a similar discussion on other process pages, such as VFD.
• (4) If a deletion debate has shown consensus to delete something, then no admin should unilaterally undelete it. That's what this process is for.
• (5) If something is speedily deleted, or deleted as an apparent mistake or flaw in process, and an admin wishes to undelete it, he should either notify the person who deleted it as a matter of courtesy, or discuss it here.
• (6) Obvious exception - if a new article is written on any subject, then a history undeletion is no big deal and should not require discussion here.

Please don't just yell 'instruction creep', what I'm saying here seems like common sense (and of course is open to discussion) and has been the subject of some recent controversies. No harm in writing it down. Radiant_>|< 09:56, August 29, 2005 (UTC)

I would add the following points in some form or another

• (7) If the closing admin determined that a deletion debate has not shown consensus to delete something, no admin should unilaterally delete it.
• (8) The closing admin may overturn his or her own decision if s/he discovers that he blundered while closing the debate. For example if a vandal altered all the "delete" votes to "keep" votes and the closer failed to notice that, the closer may reverse the decision. No other administrator should do this.
• (9) If a non-admin's closing of a deletion debate is disputed, the debate may be reopened or overturned without a discussion.
• (10) If a debate was closed as a "keep" before the five day lag-time was up, the debate may be reopened without a discussion.

I would also note the first point presented does not rule out the option of putting up a second VFD debate instead, and that this is a better option if a better reason to delete is presented. For example, if someone nominated an article on the Norwegian El 19 locomotive for deletion with the reason "We don't need articles about every locomotive class" and is shouted down by twenty voters saying "yes we do", the place to point out that the El 19 doesn't exist is a second VFD debate.

Another question: Currently, we usually need a simple majority to overturn a disputed deletion, but what would be needed to overturn a disputed "keep" result? About 50%, about 75% or what? I don't think that we should leave it at 50% because that would mean that a simple majority could get an article deleted, while a rough consenus is what should be required. Sjakkalle (Check!) 10:15, 29 August 2005 (UTC)

• Good points all. More common sense, but it never hurts to write that down. Regarding your last point...
  • If an article was closed as keep, then
    • 1. if you believe the admin made a mistake (e.g. missed a sockpuppet) and it should actually be deleted, then either discuss it with that admin or take it here.
      2. if you have found a new reason for deleting the article (e.g. the locomotive you mentioned doesn't actually exist), then take it to VFD again.
• Regarding your question. We should under all circumstances avoid this process being about content. Votes like "delete - not notable" here should be summarily ignored. The question is whether the earlier made decision was in fact supported by consensus.
• Which also brings up the question - under what circumstances should an article processed here be thrown back onto VFD? Currently they generally go back as "procedural" but that seems needlessly bureaucratic. If the decision here is that a previous VFD discussion was valid, there's no point in discussing it again. Radiant_>|< 10:28, August 29, 2005 (UTC)

• I agree that votes like "Overturn the debate and delete, subject is not notable." should(?) be ignored. Only trouble is that the ones who know the process will know that such a vote will be ignored, and instead write "Overturn the debate and delete. Administrator ignored a clear majority.", even if that is not necessarily the real reason. Sjakkalle (Check!) 10:56, 29 August 2005 (UTC)
  • That is true, and it's gaming the system, and unfortunately I can't offhandedly think of a non-instruction-creepish way to fix it that won't allow some other way of gaming it. I'm afraid that occasionally we will have an unfixable mess about deletion (qv GNAA) - regardless of how we word it. Radiant_>|< 11:39, August 29, 2005 (UTC)

I support the concept. VfU can be productively turned into a more formalized appeal process that works for contested decisions in both directions. We might still want to tweak some of the wording. For ease of discussion while we work on the proposal, I'm adding hard-coded numbers to the bullets.

• The wording of bullet 4 could be misinterpreted or deliberately abused since it doesn't specify who got to make the closing decision. I recommend rewording it to parallel the wording of bullet 7. We also need to include the existing caveat that an "out-of-process" deletion may be summarily reverted and does not require discussion.
• Bullet 5 is a courtesy and a good practice but also smells instruction-creepish. Would the policy be significantly worse without that particular bullet? Could we enshrine the practice in some other way?
• Move bullet 7 up under bullet 4.
• In bullet 9, add "..the debate may be reopened by an admin without ..."
• In bullet 10, I would remove the "as keep" clause. Any prematurely closed decision should be able to be reopened whether keep or delete.

Great draft. Thank you. Rossami (talk) 13:59, 29 August 2005 (UTC)

• Regarding your comment on point 10, the reason I said "as keep" is to take into account that some debates are validly closed as a "speedy delete". Sjakkalle (Check!) 14:09, 29 August 2005 (UTC)
  • 4 Reworded. 5, well, possibly - but a certain person has recently been claiming that speedy-deletes can be unilaterally undone. I'd say we can combine points 9 and 10, and 10 should be reworded from "as keep" to "other than by CSD" Radiant_>|< 14:22, August 29, 2005 (UTC)

• I like all the above with the following alterations:
  • (7) If the closing admin determined that a deletion debate has not shown consensus to delete something, no admin should unilaterally delete it but may bring it to Deletion Review for review.
  • (8) The closing admin may overturn his or her own decision if s/he discovers that he blundered while closing the debate. For example if a vandal altered all the "delete" votes to "keep" votes and the closer failed to notice that, the closer may reverse the decision. No other administrator should do this but may bring it to Deletion Review for review.
• Without this option an admin can unilaterally delete an article with all "keep" votes. - Tεxτurε 14:19, 29 August 2005 (UTC)

• Does any of these make it clear that Deletion Review would also be to review kept articles despite all "delete" votes? It should be made plain. - Tεxτurε 14:24, 29 August 2005 (UTC)
  • Yes, the first point ("(1) If an admin's closing of a deletion debate is disputed, the matter should be taken here, regardless of whether the disputed decision was to keep or to delete.") Radiant_>|< 14:27, August 29, 2005 (UTC)

I think probably all the points above are instruction creep, with all the attendant downsides. It will likely turn out to be impossible to abide by them at all times. Kim Bruning 14:45, 29 August 2005 (UTC)

Have you been following the recent controversies on VFU and VFD? It is entirely unclear by current policies whether a closing of a VFD may be disputed in the first place, let alone how to do it. Some people even think VFU may be circumvented by anyone who doesn't like the consensus formed there. Give me one rule that is possible to abide by at all times. Radiant_>|< 14:53, August 29, 2005 (UTC)

This is probably because the fundamental weaknesses of the current deletion process are starting to become more and more apparent as the load on it increses. This has been predicted over and over and over for over a year now. The solution is probably not to keep adding more rules. It'll work for a little while, but every rule you add also adds a possibility of abuse. With this approach, at some point the system will fail catastrophically.

I'm not saying I oppose at this point in time, but realize for yourself that this is a temporary fix. DO make the fix, but DO also work on a long term fix at the same time. Kim Bruning 15:09, 29 August 2005 (UTC)

• meatball:CommunityMayNotScale, something like that? You are absolutely right. Maybe XD will help though, but it only seems to work for uncontroversial things. Radiant_>|< 15:14, August 29, 2005 (UTC)

(Was editing my comment but got edit confliced...)

I am worried about the word unilateral creeping in everywhre. It should be defined or dropped. It appears to be synonymous with WP:BOLD? In that case, check and/or modify WP:BOLD first.

Finally, if people continue to instruction creepize and unilateralize, we're going to have to set up a separate wikipedia:special circumstances with an explicit licence to ignore all these new rules. Kim Bruning 14:58, 29 August 2005 (UTC)

Fewer words!

Good proposal. Too many sentences. Not instruction creep if written right, as it's just a change of scope. I share Kim Bruning's concern that each new rule is a new battleground for the future. How about:

1. Deletion review is the process to be used by all editors, including administrators, who wish to challenge the outcome of any deletion debate or a speedy deletion unless:
   • They are able to resolve the issue in discussion with the administrator (or other editor) in question;
   • The action in question was so clearly out-of-process that it would be reasonable to expect a near-unanimous vote here. They should, as a courtesy, inform the editor whose action they are reversing.
   • In the most exceptional cases, posting a message to WP:AN/I may be more appropriate instead. Rapid correctional action can then be taken if the ensuing discussion makes clear it should be.
   • An administrator (or other editor) is correcting a mistake of their own, or has agreed to amend their decision after the kind of discussion mentioned above.
2. Significant new information has come to light since a deletion and the information in the deleted article would be useful to write a new article. If you can simply write a new article, be WP:BOLD and do so!

• If a deletion debate was closed with a non-delete result early and an editor thinks there is non-bureaucratic value in re-opening it, they may do so.
• This process should not be used simply because you disagree with a deletion debate's reasoning — only if you think the debate was interpreted incorrectly by the closer. This page is about process not content.

That says it all, I think.-Splash 17:33, 29 August 2005 (UTC)

I like the simplicity but feel it needs something regarding restricting admins. This approach does not change the current problems between admins and while this does say that it applies to "all" editors it does not indicate that admins should use this method instead of, dare I say it, unilateral boldness against the consensus and process. - Tεxτurε 18:23, 29 August 2005 (UTC)

I wanted to avoid the "u" word, although I am familiar with the fact that some recent troubles stem from it. It does use the words "all" and "any", however, so there's not much wriggle room. I added an explicit mention of admins. I think the wriggles will come from the second "unless" — that may need phrasing more tightly. -Splash 21:39, 29 August 2005 (UTC)

• Good rewrite.
  1. Do we need a clause to state the obviosity that "converting a keep to a merge or back does not require any vote"?

     Plainly; given the recent complaints that articles were kept and not merged. Septentrionalis 18:51, 5 September 2005 (UTC)

  2. Maybe add to the first line "instead of other processes"? Might also be redundant though.
  3. I believe Sjakkalle recommended that an admin should be able to automatically overturn a closing by a non-admin, and that a non-involved person should be able to automatically overturn a closing by a person involved in the vote. Necessary or creep?

     The first is creep. Since a non-admin is only authorized to close consensus keep/merge/redirect, any other pretended closure should be overturned anyway. The second is necessary; and closure by an involved person should be expressly deprecated. Septentrionalis 18:51, 5 September 2005 (UTC)

  4. I think we should drop the second "unless". It's quite possible that an admin will expect a unanimous vote here and be dead wrong.
• Overall, it's very good, I'm just bringing up some details, which may be very creepish so I won't mention them again unless anyone concurs. Radiant_>|< 10:05, August 30, 2005 (UTC)
• Does the "clearly out-of-process" clause actually add anything to the proposal, other than the potential for abuse? What problem is it trying to solve? Hopefully "clearly out-of-process" happens so rarely that there's no harm in letting the community help make that decision on VfU. We need to be able to trust our admins, but more importantly, the admins needs to be able to trust the consensus process. If there is an admin who continually makes unreasonable out of process deletions, well, the solution is to use our dispute resolution process to reach a satisfactory conclusion, not to cut corners elsewhere. Nandesuka 11:34, 30 August 2005 (UTC)
  • Well, a lot of undeletions in the Deletion Log use the words "revert - out of process". I've not done a statistical review to see how many of them were justified. I know that my own anecdotal evidence suggests that it's not a trivial number. The most common example I can think of are "speedy deletes" which clearly fail to meet any of the speedy-delete cases. However, my anecdotes are old. I haven't patrolled that page since the speedy case list was last expanded. Rossami (talk) 14:02, 30 August 2005 (UTC)
    • Sorry, but no. You cannot enter a summary for undeleting anything (unfortunately). However, a quick count gives about one undeletion per 100 deletions in the log. Seems like there aren't that many improper speedies. Radiant_>|< 14:14, August 30, 2005 (UTC)

      (Interesting. You're right that it couldn't have been in the Deletion Log. But I'm sure I've seen that explanation made for more than a few undeletions. Now I'm wondering where I did see that... Time to do some more digging. Rossami (talk))

      • Or it could just mean your average admin is too crotchety and set in their ways to admit they were wrong. Agriculture 14:17, 30 August 2005 (UTC)
        • While I'm sure that some admins are, I know several admins that are more than eager to prove other admins wrong. Speedies are generally undone by someone else. Radiant_>|< 14:24, August 30, 2005 (UTC)
    • Also, let's say that there were more improper speedies than there actually are. Isn't that an argument that the admins who are improperly speedying need to be brought into line with the community consensus? It seems to me that discussing those speedies on VfU (and proceeding to RfC if that doesn't "help", although hopefully that won't be necessary) is a "better" outcome, because that creates dialogue where the admin can get feedback from the community explaining why she or he was wrong, rather than it just becoming an edit war or personality conflict between two strong-willed admins.Nandesuka 14:19, 30 August 2005 (UTC)
      • Yes. We could then have a page with precedent about what is and is not a good example of a speedy. Radiant_>|< 14:24, August 30, 2005 (UTC)

Ok, there is evident concern that this is still too stretchy. I think admins need to have a little bit of stretch room, however. One situation I was trying to avoid with the "clearly out-of-process" clause was the Garrett thing: some of his earliest deletions were way out of line, and to make those all go through VfU (as well as AN/I and his talk page) would just have been a bit silly. NOT:a bureaucracy blah blah. To Radiant!'s points:

1. I'm not sure we do. It's something that any user can do at anytime, whether we write it in here or not. A new user could do it in blissful ignorance of the VfD closed 1 minute ago.
2. Yes, ok.
3. Misplaced Pages:Deletion process already says the first bit of that and they can't delete articles anyway so the collateral damage is small. To allow the second...well...it would mean that VfU-ing a VfD could be circumvented by almost all admins — just as it currently can be.
4. Ok. Let's try something else. It would also allow for the wide-eyed innocence approach: "dang!, I got it wrong again, huh?".

I made some changes to the original draft, along these lines (use the diff). The thing about AN/I — would people just prefer no such clause at all?

-Splash 14:45, 30 August 2005 (UTC)

The AN/I thing reads better to me, because it makes clear that the guiding principle is community consensus rather than "I, the Lone Gun Admin, am all that stands between the people of Misplaced Pages and Anarchy." Nandesuka 15:16, 30 August 2005 (UTC)

Given a current discussion on the main page, I think we might think about reviving this discussion. It would help greatly to have clarity one way or another. <prods watchlists>.

Naming

If this page is to be renamed, I find Deletion review cumbersome; and easy to confuse with "Deletion Reform".

I much prefer the straightforward Undeletion. Septentrionalis 18:58, 5 September 2005 (UTC)

How about Deletion reconsideration or Deletions up for reconsideration? This seems to be a more apt description of what is happening on the page. NoSeptember 15:57, 14 September 2005 (UTC)

Don't feed the...

I've removed some material from the main page. If someone thinks it should go back, do that of course.
brenneman 23:29, 8 September 2005 (UTC)

CSD A7 clarification proposal

CSD A7 (non-notable bios) has come up for discussion a good deal here. At Misplaced Pages talk:Criteria for speedy deletion#Interpretation of WP:CSD A7 (non-notable bios) is a proposal to clarify the interpretation of this critrion, and particualrly the meaning of "Claim of notability". Please read it and comment if you are at all interested. DES 00:19, 12 September 2005 (UTC)

The scope of VfU

A couple of sections up, in #Proposal and then #Fewer words! there was a formative proposal about extending the scope of VfU to review the outcome of all deletion processes whether resulting in deletion or some other action. Given the current discussion on VfU, we should revive that discussion and complete it.

Originally, a rather convoluted, creepy offer was on the table which was simplified and received some support. The then-current draft of the new scope of VfU was:

This sounds ok to me, but it would since I wrote it. If we can generate some discussion and some consensus here, then perhaps admins fighting over AfD closures might come to an end. -Splash 21:37, 13 September 2005 (UTC)

• Support - Simple has my vote. KISS - Tεxτurε 22:00, 13 September 2005 (UTC)
• Extreme lesbian support! --Phroziac 22:19, 13 September 2005 (UTC)
• Oppose scope creep. Disputed cases of A7 should be taken to Afd, as per Misplaced Pages:Criteria for speedy deletion "If the assertion is disputed or controversial, it should be taken to VFD instead". What is the VfU supposed to do in this case, evaluate if there is a claim or not? So what... let Afd decide if the topic is really worth an article. Kappa 22:20, 13 September 2005 (UTC)
  • I hadn't noticed the A7 heading just above; this isn't related to A7 particularly at all. Specifically, the discussion surrounding Harry Potter trolling made me try this again. It's only intended to prevent the recent wheel warring by bringing within-scope all outcomes of deletion processes rather than just deletes. -Splash 22:32, 13 September 2005 (UTC)
    • If it's not about speedy deletions, leave that out of the proposal. 23:11, 13 September 2005 (UTC)
      • It is about speedies, just not A7 specifically. We should try to stem the recent fights between disagreeing admins by giving them very strong recommendations that they ask the community to make a decision in a deliberative process. -Splash 23:43, 13 September 2005 (UTC)
        • Afd is a deliberative process where those who believe the article has merit have the possibilty of improving it to answer the objects of those who feel it should be deleted. Kappa 00:12, 14 September 2005 (UTC)
          • The "debate it where we can all see it" argument is a good one. My concern with it is that it requires a summary overruling of one admin's decision by another: and will likely result in a further overruling by either the first admin, or a friend thereof. Debates on AfD are not presently completely sacrosanct for 5 days if it's a blatant speedy. By insisting that the undelete come to DR/VfU the war is nipped in the bud before it begins. The provision for "most extreme" circumstances allows an expedited way to fix major mess ups but only once the community has nodded — again there is not the summary reversal required to take it straight to AfD. Although I'm still thinking about this point. -Splash 00:48, 14 September 2005 (UTC)
• I would support this. Please make it clear that VfU is to be renamed to DR as part of this change. I also suggest announcing this proposal fairly widely so that consensus will be clear. Note this is not so much for disputed cases of A7 as for process disputes on AFD or xFD when the decision is soemthing other than delete, in my view. DES 22:22, 13 September 2005 (UTC)
  • Yes, although it obviously wouldn't preclude a delete outcome being brought here too if that were out of process (e.g. the one on the main page at the moment). -Splash 22:32, 13 September 2005 (UTC)
  • I think David raises a very important point that should be clarified. The proposal specifically says "the outcome of any deletion debate or a speedy deletion..." So shouldn't an A7 that someone disputes be brought here? That is in fact the current practice as I understand it, except in those instances when there is an independent undeletion by a sysop or higher, no?—encephalonὲγκέφαλον  23:02, 13 September 2005 (UTC)
    • Short answer is: yes. Long answer is compulsorily if you're not an admin and you can't find an admin to do the job and very optionally if you are an admin. The reason for including speedies is to level that playing field and put an end to a string of recent fights over speedy-unspeedy-respeedy-unspeedy-afd-ani-vfu-talk-yadda. Anyone wants to undelete an article that is not a "most extreme" mistake should bring it here. Imo, anyway. -Splash 23:43, 13 September 2005 (UTC)
      • Absolutely. I think it should also be noted that there's absolutely nothing here that stops anyone from re-writing a new, substantially better article— it merely keeps (almost always) extremely poor articles off WP while a valid community decision is being made via DR.—encephalonὲγκέφαλον  00:27, 14 September 2005 (UTC)
• Weak Support sounds like scope creep and a lot of extra work... yet still a good idea Splash :) Redwolf24 (talk) 22:23, 13 September 2005 (UTC)
  • Thanks. Well, there are very few challenges as it stands both on keeps and deletes so it shouldn't make much extra work. -Splash 22:32, 13 September 2005 (UTC)
• 2 problems I see two problems with the wording above: (1) It doesn't make clear that VfU (or whatever it's to be called) is about resolving issues with the process of obtaining & interpreting consensus, it isn't a place to form new consensus. In other words, if a AfD debate produces clear consensus to delete an article, that should not be challenged on VfU. RESOLVED. (2) The final sentence practically begs editors to re-create deleted articles, which is a waste of time because re-creations generally get deleted too unless substantial new notability is present. I don't see any problem with users interested in re-creating articles to raise the issue on VfU first... it would provide a measure of consensus before the editor creates the new article, as well as protecting it from immediate re-deletion. Andrew Lenahan - Starblind 22:26, 13 September 2005 (UTC)
  • How would a newbie editor know their soon-to-be article had been previously deleted? (Apart from the new link to the history implying it had been.) They wouldn't know the existence of VfU, let alone that they should use it in preference to the edit and save buttons. Perhaps we should rephrase the final sentence to some other form? -Splash 22:32, 13 September 2005 (UTC)
    • Well, yeah, the VfU wording isn't going to make a lot of difference for users who aren't aware of VfU, but the wording as-is practically commands them to re-create deleted articles ("be WP:BOLD and do so!") when in reality that's something that should only be done in certain cases. Since we're on the topic, I do think that the "no article by that name exists yet" page should mention whether a page by that name was deleted (and link to its AfD), which would possibly cut down on re-creations, but that's a whole different topic. I'd say that "be WP:BOLD and do so!" line definitely needs a rewrite, or even a removal. Andrew Lenahan - Starblind 22:59, 13 September 2005 (UTC)
      • Now that I fixed my grammar, I realize that it is actually redundant dressing to the words. How about it is simply removed? We don't need the exhortation, since they already have edit, save and the rest. -Splash 00:05, 14 September 2005 (UTC)
        • Sounds good to me. Andrew Lenahan - Starblind 15:16, 14 September 2005 (UTC)
  • I do apologise, I missed a sentence out about process-not-content in my copy paste from up above. I have added it now. -Splash 22:36, 13 September 2005 (UTC)
    • Much better, thank you. Andrew Lenahan - Starblind 23:01, 13 September 2005 (UTC)

• Excellent work by all the editors involved (Splash, Sjakkalle, Rossami, Radiant et al).

1. Splash, would it make sense to replace "editors, including administrators" with "Users"? There is currently a plausible loophole for someone to claim "Well, I'm a Bureaucrat, so this doesn't apply to me."

   Hmm, well, bureaucrats are admins are editors both. There were suggestions at the time that VfU was only for non-admins to request undeletions and that admins could just do it. In fact, this was the cause of the original discussion. Explicitly placing admins within scope is important in that respect. -Splash 23:32, 13 September 2005 (UTC)

2. This is minor, but what are the exceptional circumstances referred to in 1.2? Can a simple clarifying remark be written about this? I can forsee editors who are so inclined (ab)using this provision to take almost everything to AN/I.

   After a recent RfA the new admin made a number of considerably out-of-process speedies; a number of them were summarily undeleted during a discussion on AN/I (and their talk page...). We should not hem our admins in completely. I would hope that the extreme phrasing and the result being rapid correctional action means those on AN/I would normally just send the complainant to deletion review. I will try to think of a clarifying sentence, but I'm struggling at the moment. -Splash 23:32, 13 September 2005 (UTC)

3. Sentence 2 (Significant new information has come to light...) seems to be hanging. Is it meant to mean: "Deletion review is also used when significant new information has come to light since a deletion and the..." Should the next sentence begin "However, if you can simply..."?

   I have grammar my fixed. -Splash 23:32, 13 September 2005 (UTC)

4. In the very last line, add comma after process.

• The proposal is beautiful, and I would strongly support this or a closely related version.—encephalonὲγκέφαλον 
• Question: Someone makes an article about a notable person, but fails to claim notability. It gets speedy deleted. They read the rules, and figure out they have to say "noted game designer" not just "game designer". Now they have to go through VfU, then Afd, just because they didn't know they about a wikipedia idiosyncrasy? Kappa 22:52, 13 September 2005 (UTC)
  • In practice, I don't think a second article with an actual claim of notability would be speedied (of course, claiming notability means more than just slapping "notable" on something though). Andrew Lenahan - Starblind 23:04, 13 September 2005 (UTC)
    • One solution would be for them to recreate the article and immediately place it on AFD, as this appears to be the express procedure for disputing an A7 speedy deletion. I presume it couldn't then be speedied under A7 again. Otherwise I support this amendment to the header of VFU, and the name change, although in all such situations it is preferable to work it out between the involved parties rather than bringing it here. Christopher Parham (talk) 23:11, 13 September 2005 (UTC)
    • Uhm, "game designer" IS a an assertion of notability... This criterion is for things like Sam Toupin. Playing DDR is not an assertion of notability. It would be out of process to speedy an article that claims notability, other then using WP:IAR to delete it anyway, if it's fairly obvious, but doesn't exactly fit into the criteria... --Phroziac 01:45, 14 September 2005 (UTC)

I would support this IFF there were a parallel process, or even the same process were used, to review articles that were kept as the result of a "no consensus to delete" closure of an AfD listing. What this is doing is attempting to go around the AfD process, and is changing the criteria so that if an article gets deleted because of a consensus on AfD to delete, it would only take a majority of votes on VfU to undelete. Change the criteria to require a consensus to undelete as well, or this is just an end run around AfD deletions. User:Zoe| 23:21, 13 September 2005 (UTC)

• • The phrasing says "the outcome of any deletion debate". It is intended to include both deletes and nondeletes in that phrasing. The situation you describe is currently the way things stand, and we shouldn't change more of the current process than is necessary. I don't think I've seen a debate on VfU that overturned a clear consensus on AfD — people are forever reciting the process-not-content mantra and it seems to work qutie well. -Splash 23:29, 13 September 2005 (UTC)
• Oppose. Bad speedies should be restored by an administrator who is able to view the contents of the stub. If there is a dispute then it can be taken to AfD where the article can be views by all and edited during the debate. Requiring a sysop to challenge all bad speedies on VFU would be adding a further unnecessary bureaucratic layer. --Tony Sidaway 01:10, 14 September 2005 (UTC)
  • Ok. What about the other, more significant, part of the proposal: that an admin who wants to amend a AfD (or other _fD) closure comes here rather than just doing it, if they can't persuade the admin on their talk page? The Harry Potter example on the front page is why I revived this debate. -Splash 01:14, 14 September 2005 (UTC)
  • Current practise in my experience is to relist on AfD. It works well and I don't see any point in changing the venue. --Tony Sidaway 01:17, 14 September 2005 (UTC)
  • And I want to make it absolutely plain that I will not accept any major change to policy, particularly any changes permitting articles listed on AfD to be subject to double jeopardy, or limiting sysop powers, or changing the venue for disputed speedies, without clear and loud evidence of a widespread consensus. If the proposed changes are to go through, I think we need a straw poll at least on a similar scope to that which applied to WP:3RR. The proposed change would have ramifications of a similar scale and should therefore not be a step taken lightly.
  • On double jeopardy. I'm concerned with the case of a person who doesn't get the result he wants, or who doesn't agree with a closing sysop's method, bringing cases to DR in the hope of a second chance of deletion. Now presumably the proposed process would stop him just deleting and *then* coming to DR, and that's a good thing. But I'm still uncomfortable with this idea of rerunning the AfD on DR. I don't think it's right to require an article to jump two hoops to escape deletion. We already have a speedy deletion process for obviously damaging articles, and someone who feels strongly can always ask for permission to relist on AfD (or just go ahead and do it if he feels confident enough). So we already have a perfectly good remedy. Now you'll always get sysops who object to a non-delete result who perform out-of-process deletions (as happened with the Harry Potter trolling article), but these are easily remedied by undeleting them pending a properly processed deletion, as policy permits. --Tony Sidaway 01:52, 14 September 2005 (UTC)
    • Afraid of accountability, Tony? We know from your record that you like using your status as a closer to override the wishes of those that voted. This would just provide a well-needed means of giving the community some accountability over your actions. Ambi 03:13, 14 September 2005 (UTC)
• Wholeheartedly support. Tony's past actions illustrate just why it would be wonderful to have better scrutiny of closers, and it seems that this proposal would provide that. Ambi 03:13, 14 September 2005 (UTC)
• I strongly support this with one amendment. The arguments above have convinced me that we should remove the "or speedy deletion" clause. The current rule for a contested speedy-delete (A7 or otherwise) is "find an admin to undelete but immediately nominate to AFD and let the community decide". We should stick to that rule. Rossami (talk) 04:57, 14 September 2005 (UTC)
• Comment: To answer Tony's concern about double jeopardy, remember that the standard used on this page to reverse a decision is even higher than the standard to make the first decision. If someone really wanted to game the system, yes, they could immediately appeal to DR. They would have to do so in the face of an informed and educated user-group which has so far been quite good at discerning ulterior motives and shouting down inappropriate nominations. I believe that the procedural and cultural controls will prevent this page from being abused. However, we may need to rethink the current text on the Undeletion policy page about how to count the rough consensus. For one thing, it's too focused on vote-counting and not enough on discussion. For another, it succeeds in setting a higher standard for overturning a "delete" decision but not for overturning a "keep" decision. Rossami (talk) 04:57, 14 September 2005 (UTC)
• Support, since it provides a venue for disputing a result, but a few further issues need to be adressed.
  1. About how strong support will be needed to overturn a disputed "delete" result? >50% is the norm now, and we might continue with that.
  2. About how strong support will be needed to overturn a disputed "keep" result? Here I think that we need much more than a simple majority, 75% or something like that. We should not have a situation like we saw at Historical persecution by Jews (here and here) where a simple majority (but not a consensus) uses Deletion Review to obtain a deletion which they could not obtain at AFD by means of their majority. Sjakkalle (Check!) 07:07, 14 September 2005 (UTC)
• If Rossami's condition, that "or speedy deletion" is removed, is agreed to, I think that with appropriate safeguards DR could work. If as Rossami suggests we set the margin required to delete on DR even higher than pertains on AfD, at least it answers my objection on VFU. The real trick would be to persuade sysops not to just summarily delete an article if they disagree with a AfD close. If we set up a kind of appeals court then it could work well. It is understood, I hope, that an article whose non-deletion result is under appeal must be kept in an undeleted , editable state during the discussion, and an appropriate template should direct editors to DR.
  • Yes, I'm coming around to the idea of taking speedies out of scope. The mantra of process-not-content is, probably, reversed for speedies where there is basically no process: we can only judge a speedy's correctitude on its content i.e. they are content-not-process and so we have AfD for that purpose. Being ruthlessly firm on the role of DR/VfU being to examine the closure process should help to avoid gaming things in the way you describe — something we must certainly avoid. -Splash 19:41, 14 September 2005 (UTC)
• I would also ask editors in this debate to refrain from personal attacks. -Tony Sidaway 07:09, 14 September 2005 (UTC)
• On the numbers: we already have too great a preoccupation with numbers and votes. It should be pretty clear when a consensus to delete exists, and as long as the person closing the DR debate isn't the same person who closed the AfD I don't think we need to put numbers on it. A "strong consensus" to delete should be required (implying something more than the kind of marginal stuff that leads to these cases). For undeletion, I assume that an article would be undeleted unless there is consensus support that the deletion was correct. If this isn't what happens now, perhaps we should review this. Loading the odds in favor of keep/undelete would be in keeping with Misplaced Pages deletion policy: if in doubt, don't delete', which I sometimes think we sysops should have stenciled across our screens. --Tony Sidaway 07:13, 14 September 2005 (UTC)
  • On the numbers: I'm inclined to think we should load the odds in favour of not reversing the closer's decision. We should work on the presumption they likely got it right, whether they deleted/notdeleted. So we should scrap the present simple majority that the undelete policy has. If this process is to work well as an 'appeals court' it should also hope to be less subjective than AfD closure or it will need an appeals court of its own! The bulk of present VfU debates are pretty decisive (see the main page), but not all of them are. Perhaps if we phrase it something along the lines of "...requires a strong consensus, usually about three-quarters of the participants concurring..."? We can allow the usual socks to be thrown away, and we should also be brutal at dismissing any 'vote' that speaks to the content not the process except insofar as the closure may have judged some of the comments against the content. To the end of simplicity and objectivity, we should also retain the current up-or-down system i.e. you can't say "undelete and merge", you can only say "undelete (and relist)". -Splash 19:41, 14 September 2005 (UTC)
    • I agree with Splash. In a VfD/AfD the "keep" is the default state because you have to have a good reason to overrule the article's creation. In DR "keep deleted" needs to be the default because you need a good reason to overrule closure. If there was no doubt in the closing admin's mind regarding consensus then I don't see a reason "if in doubt, don't delete" would apply. Consensus was followed. There is no such guidance for VfU and should not be for DR - it would fly in the face of the consensus that was followed. It would be abused into an "extension" of AfD for every deletion closed with consensus to delete. - Tεxτurε 19:53, 14 September 2005 (UTC)
      • I tend to agree that the default should be "leave the result unchanged" (note that for a keep decision on XfD this means the default is keep). I don't agree that a 75% strong consensus is needed. If a majority, or perhaps a 60% majority, think the process was in error (after discounting socks and comments based on content rather than process) then reverse the decision or restart the process (i.e. relist) to obtain an untainted result. Perhaps when a "keep" is overturned by a narrow margin the articel involve shoukld be promptly relisted, with a note that a different admin will cloe the debate, insted of being promptly deleted. Note aslo that if a "merge" or "redirect" consensus was clear on AfD but was not followed by the closer, DR could and IMO should direct that the AfD consensus be followed, thus it could ahve a "merge" result. But ONLY if such a result was the clear consensus on the original fD discussion, after correcting any procedural errors. DES 20:13, 14 September 2005 (UTC)
        • Yes, what a DR is really deciding is was did the decision follow consensus. This should not be a forum to revote the AfD but rather a vote that consensus was not followed. - Tεxτurε 20:28, 14 September 2005 (UTC)
          • Or other problems with process. Issues of vote forging, sockpuppetry, vote solicitation, and votes that should or should not be discounted could also be brought up on DR, with possible results of "outcome valid", "relist" or "outcome reversed". DES 20:58, 14 September 2005 (UTC)
            • I'm convinced. You should add to the proposal that the only valid votes and closing consensus are "outcome valid", "relist" or "outcome reversed". (Or I suppose this could be handled later.) - Tεxτurε 21:06, 14 September 2005 (UTC)
        • Yes, I meant that the default for a "keep" closure is "keep kept"(!) and the default for a delete closure "keep deleted" etc. I can see the utility in having a purgatory kind of outcome at narrow margins, although it would be nice to keep things as simple as possible. Still we could say that between 50%+1 and about 75% it goes back to AfD (undeleting first if necessary) but above about 75% the decision is simply reversed, or amended as appropriate. And yes, now I think about it I agree that DR should be able to mandate the appropriate outcome, although only as far as a {merge} tag since merges can be difficult sometimes. -Splash 20:31, 14 September 2005 (UTC)
          • I agree that a "mrege" decision on DR need mean no more than putting a proper merge tag on the relevant article(s), (plus a note on the proper talk page that the merge was a result of a consensus decision with a link to the relevant discussion) anoyne so inclicned can then do the merge. A redirect should just be done, if no merging is involve. I still think that 75% is a rather high bar, and 66% could be regarded as decisive, but i could accept your formulation, Splash. Also, a "relist" outcome might be explicitly determined when the DR consensus is that the process was so damaged that no clear consensus can be determined (for example if there was disruptive vandalism to the XfD page, or if an article was rewritten late in the process and it is probable that the opnions expressed do not refelct the current state, etc) DES 20:58, 14 September 2005 (UTC)
• It should be made clear that DR is not to be a "second chance" to delete an article that had no conssnsus to delete on AfD, nor a method to make an 'end run" around the AfD consensus requirements. DR is not to be used by people who simply displike the AfD result. Those who try to use it to chanmge a consensus result (and there will be such people) should be refused and rebuked. DR should be for reviewing claims of incorrct procedure, and correcting such errors when found. When a clsoer ignores or clearly incorrectly determines the consensus on an AfD (or one of the other XfD pages), or when an improper speedy deletion is made, or when there is soem other error in procedure during an XfD discussion that affects the result, DR should be the place to review the events andf take corrective action as needed. DES 14:52, 14 September 2005 (UTC)
• Support. I've read over the arguments carefully, and although I do certainly understand the fear of creep, I cannot believe that any device through which the deletion process can gain more accountability and transparency can be bad. There can be no doubt that this will lead to wider criticism of admins, and this is both the whole point of the new institution and a detriment. Admins should welcome more constructive criticism in all cases, but admins should also back up and protect other admins when criticism is excessive and/or unwarranted. My two bits. Fernando Rizo T/C 15:45, 14 September 2005 (UTC)
• "Not reversing the closer's decision" goes against the fundamental tenet of deletion policy: if in doubt, don't delete. If we want to reverse that principle, let's get a wiki-wide consensus. Otherwise it's a non-starter. Whatever we do here must be compatible with deletion policy.--Tony Sidaway 20:49, 14 September 2005 (UTC)
  • I just re-read deletion policy. i don't see "if in doubt, don't delete" enshrired there as a "fundamental tenet". it is a good rlue of thumb for AfD closers, but does not trump other policies and guidelines. DES 21:03, 14 September 2005 (UTC)
  • Read Misplaced Pages:Policy: WP:DP is part of official policy: policies that are widely accepted and that everyone is expected to follow It isn't a rule of thumb, it's our policy as a wiki not to delete pages where there is some doubt. This does trump all guidelines, although it is subject to the key policy: Misplaced Pages:Copyrights. In the latter policy, related to copyright questions, we adopt the reverse principle: we delete unless we have clear evidence that we have a right to use copyright material under the GFDL. --Tony Sidaway 13:19, 15 September 2005 (UTC)
    • I do not think there is a particularly close relationship between the phrases "don't reverse the closer's decision" and "if in doubt, don't delete". They deal with two quite separate concepts: the first works on the presumption that our admins get things right the vast majority of the time, the second is to help them in doing so. -Splash 19:01, 15 September 2005 (UTC)
    • I have read both Misplaced Pages:Policy and WP:DP fairly throughly and I thgink i understood them quite well. Please re-read what I wrote above. I didn't say that WP:DP was a rule of thumb, I said that one phrase from that page was a rule of thumb. I am contending that that phrase is not the heart or central meaning of that policy page. I supported that by mentioning in what context the phrase was used in the policy page. Of course policy is to be adhered to, and I never argued otehrwise. I would like to see just what aspects of current policy you or anyone things this proposal would go against. Of course, a newly adopted consensus can amend an existing policy, either explicitly or implicitly, and then, in general, the newer consensus trumps the older one, unless the older one had a much wider or stronger consensus, or is for soem reason more fundemental, for exmaple the WP:NPOV policy tumps most others with which it might possibly be in conflict, IMO. But my main point is that one phrase from WP:DP deos not override all the resat of that policy page, nor the undelwtion policy, nor other policies adopted by consensus. Note that WP:DP also says "You can expect administrators to follow the process detailed below to aid them in their judgment." and "Hence, the decision to permanently delete an article is not taken lightly, and the deletion process is followed." thus indicating that proper process is important. I could as easily claim that those statements are the key to the deletion policy. In any case, I would argue that the propsoed DR process does not violate the maxim "if in doubt, don't delete". It is merely a method of helping to determine when there is significant dobut, and when, in fact, there is not enough doubt, but rather it is clear that an articel should go. I am not proposing, and I do not belive that anyone is arguing, that any page should be delteded without a clear consensus. DR should have a delete result only when there is agreement that there was in fact a celar consensus to delete on XfD (or a clear consenus in advace via the speedy deletion criteria) and the relevant closer somehow made an error in determining that copnsensus, or violated policy by ignoring it. Of course no one should ever violate policy, but soem people do, for various reasons. DR is one method of correcting that, or (more often I hope) correcting well-intentioned but clearly mistaken attempts to apply the deltion policy. DES 19:47, 15 September 2005 (UTC)
• As someone who just had a closure reversed (and felt slightly hurt by not being approached for discussion, but feels all better now) I think that this is a good idea, but should not be the first course of action. The first course of action should be that if someone disagrees with an AfD closure, he or she should approach the admin who closed the vote, giving the latter the chance to say "hey, you're right, guess I can't count" or "I didn't notice the sockpuppetry" or whatever. There is no need for byzantine procedures when simple, polite human interaction will solve the problem 95% of the time. moink 21:33, 14 September 2005 (UTC)
  • I don't think this proposal changes an admin's right to undo their own mistakes. - Tεxτurε 15:36, 15 September 2005 (UTC)
    • You're right, Texture, but I'm somewhat worried that it will change behaviour in practice. I think the description should say, not only that you don't have to use this process if the admin goes back and corrects it, but that it should not be the default process. The default process should be a polite informal discussion that assumes good faith. Only if that fails to procure agreement should this process be used, in which case I think it would be useful. moink 18:12, 15 September 2005 (UTC)
• Support. We need this yesterday. Too often the fate of an afd'd article is more influenced by the closer's wikipolitics than the comments on the afd subpage. VFU has worked exceedingly well for improperly deleted articles; currently, the only recourse for improperly kept articles is a disruptive and bureaucratic afd renomination. —Cryptic (talk) 14:33, 15 September 2005 (UTC)
  • Strongly support now. The most vocal opponent says above that disputed keeps should be brought back to AFD, but labels such renominations as bad faith and calls for censure of the disputant. I find it impossible to assume the opposition is in good faith. —Cryptic (talk) 11:37, 18 September 2005 (UTC)
• Support. The current way of disputing a bad keep decision is too disruptive and acrimonious. I like this. android79 15:15, 15 September 2005 (UTC)
• I Support this well-thought out, balanced, and consensus-driven proposal. Android's point that the current dispute process is failing is a good one. Nandesuka 19:23, 16 September 2005 (UTC)
• Strong support. I go away for a week, and you guys change everything... gee! :P Now, a little bit more seriously. The only thing I would amend is this particular sentence: "This page is not intended to be a new deletion discussion. This page is about process, not content." Why? Simply because those new to VfU don't know it. --Titoxd 17:37, 18 September 2005 (UTC)

Mechanics

There appears, to me at least, to be a broad support for making a change towards the boxed proposal above. There are concerns about the detail of the process, however. As I read the discussion above (and not the discussion below, since it relates to a fundamentally different approach) we are consider something like 3 options:

1. Simple: A strong consensus (e.g. three-quarters) to amend a closing admin's decision i.e. a delete turned into a notdelete or vice versa, or a straight keep turned into a redirect/merge, or a keep turned into a delete. Also, a simple decision to "relist" when things are very unclear.
   • There are suggestions of a lower notion of consensus: at present VfU needs a simple majority. This is too low a threshold imo, since we should start out by trusting our closers — there is no reason to defy tradition and remove the element of trusted judgement implied by a successful RfA.
2. More complicated: A kind of purgatory. If the numerics are in the 50-75% range (with the usual emphasis on process not content demanded from participants) then the article returns to AfD (or _fD). If above three-quarters the decision is simply mandated and implemented.

Some sort of a merge of the two might be appropriate. What do people think? -Splash 18:33, 16 September 2005 (UTC)

• I believe the complicated method is more effective. After all, right now we're supposed to relist things on xFD when we decide to undelete. But even 75% seems low IMO. Maybe a quasi-unanimous discussion (discounting sockpuppets, meatpuppets, etc.)would be required to simply overrule a lower decision. Titoxd 17:42, 18 September 2005 (UTC)
  • Quasi-unanimity means that the process can be held to ransom by a single editor (e.g. the closing admin) and a friend. That just won't do; consensus is not unanimity insofar as those in the minority are able to accept the decision of the numerical majority because it is so sizeable. That does not need unanimity. -Splash 21:57, 18 September 2005 (UTC)
    • That's why I said quasi-unanimity. Admins are smart enough to recognize when there's an attempt to delay the process and when someone thought of something that hadn't been thought of before. But I agree, that makes it very ambiguous and that's what we're trying to get rid of. Maybe 75% isn't as bad as I think, but 80% or 85% should be more resounding. After all, we're making VfU the equivalent of an appelate (Supreme?) Court in the US, so if we are going to downright reverse a lower court, we might as well be sure of it, since there won't be courts with appellate jurisdiction over VfU. Titoxd 02:24, 19 September 2005 (UTC)

Here is a proposed wording for the mechanics, similer but not identical to numer 2 above. What do people think of it? Obviously it is merely a suggestion, and is subject to changes if anyone offers ideas for them DES 21:35, 18 September 2005 (UTC)

• In the deletion review discussion, users may opt to either overturn or endorse the previous deletion decision. Those opting to overturn should also specify whether the page should be relisted, kept, merged, redirected, or deleted. (Note, a "merge" result may mean no more than approperiate merge tags being applied to the article and any sugested merge target, plus a note referencing the DR discusion on the appropriate talk page. After that the merge can be performed via normal editing. No closer is ever required to actually do a merge.)
• If a majority, but less than a strong consensus, favor overturing the deletion decision, then the article will be relisted (or initallly listed, if it was initally speedy-deleted) for discusion on the proper XfD page, with links to the DR discusion and any previous XfD discusion.
• If there is a clear consensus (say 70% or more) then the decision may be overturned directly, and the consensus result applied. However, the consensus may be to relist, in which case that will be done. If there is consensus to overturn a previous decision, but not on what the result will be, the item in question will be relisted as above.
• Anyone expressing an opnion should also provide reasoning, at least in the form of a shorthand expression. Opnions that rely primarily on the contnet of the item, rather than the process being reviewed, will be ignored. However, a claim that valid resons were simply ignored during the deletion discussion is about process. So is a claim that the deletion policy was ignored or improperly applied durign the deltion process.
• Exception: users stating that they were not aware of the deletion discussion may indicate that they had valid reasons which they would have expressed at that discussion. Such users should normally opt for Overturn and relist so that their reasons can be more fully evaluated in a proper deletion discussion. (obviously users who commented in the previosu deletion discussion cannot honestly claim to have been unaware of that discuussion.

Any coments on the above? DES 21:35, 18 September 2005 (UTC)

I may have more comments as I think about it more but right off the top of my head, I recommend removing the clause about speedy-deletions. Leaving speedies in scope will inevitably draw us into content discussions, not process discussions. Remember that the community has not yet expressed an opinion on the content in a speedy decision. A speedy is executed on the opinion of a single admin. Speedy-deletes are only supposed to be for non-controversial deletions. If a speedy is contested in good faith, the article should immediately be restored without bureaucracy or discussion. The restored article should, however, be immediately nominated for a full deletion discussion. That restored article could be kept (as a "no consensus") if as few as a third of the discussion participants want it. That process and standards are at odds with both the process and the standards proposed above for overturning a full AFD decision. Easiest just to put speedies out of scope and stick with the current review process for them. Rossami (talk) 00:06, 19 September 2005 (UTC)

Afd challenge

• An alternative proposal: Afd challenge
  • Already, policy dictates that disputed speedies go to Afd.
  • I suggest that all disputes to deletion and AfD results go to AfD using the existing process, with the following additional rules:
    • A sysop who disputes a deletion in process can undelete and relist on AfD (he can already undelete without further process if the article in question isn't a speedy or AfD candidate; this is filling in the gray areas)
    • A non-sysop who disputes any deletion can ask a sysop to undelete and relist on AfD.
    • If no sysop agrees to undelete, the article can be relisted on AfD in its deleted state. If it was an out-of-process deletion, existing policy says that it can be undeleted at will by any sysop.
    • Only one challenge is permitted. Once an article has been through two AfDs, recreation of the same content can be speedied and any further attempt to challenge will be summarily removed.
    • Any person who disputes any non-delete close can relist (this is already allowed)
    • During the dispute period the article will not be deleted.
    • After a dispute of a non-delete that results in deletion, there is no further appeal.
  • There would be no need for:
    • VFU: appeal against deletion would take place on AfD.
    • Separate close procedures: the same close would be performed on AfD for primary listings and challenges.
  • I think this would work better than having a separate venue. In the interests of transparency, challenges to AfD results should occur in the same venue as AfDs. --Tony Sidaway 20:49, 14 September 2005 (UTC)

• • • • I strongly oppose this "alternate proposal". I think that the status quo is a better choice than this. Why? I think there should be two distinct fora because there are two disntict sets of standards. VFU/DR should be about process, it should be, as Tony said above, a "court of appeals". Even now we have some people incorrectly basing decisions on content issues on VfU. Putting reveiws on AfD will just encourage this. Different fora for different procedures. I might add that Tony is incorrect that "policy dictates that disputed speedies go to Afd.". My understanding is that this is true only for A7 (non-notable bios). Can you point to any general policy statement that says this for other types of speedy deletes (granted this is the most commonly disputed speedy type). Also this proceduree IMO over-emphasizes the role of the admins. Currently any editor can bring a deeltion to VfU and it is treated in much the same way whether that editor is an admin or not. Under the above proposal it makes a huge difference if an non-admin can get an admin's support or not. i think this is unwise and unwiki. DES 21:12, 14 September 2005 (UTC)
        • I based my statement that "policy dictates that disputed speedies go to AfD" on the following:
          • Misplaced Pages:Undeletion policy: Exception: If the page was obviously deleted "out of process" (i.e. not in accordance with current deletion policy), then a sysop may choose to undelete immediately. In such a case, the sysop who deleted the page should be informed of the undeletion and the reason for it. If deletion policy dictates that the undeleted page is a VfD candidate, please list it there. If you are unclear as to whether it was out of process or not, the normal undeletion procedure should apply.
        • I interpret it as follows: Sysop 1 speedies an article. Sysop 2 decides that it's been speedied out of process (no speedy criterion covers it). Sysop 1 objects, so sysop 1 and 2 have a dispute. It can't be a valid speedy (WP:DP: if in doubt, don't delete; two sysops in disagreement is a reasonable doubt) but it may still be deletable, so it is a AfD candidate.
        • DES is wrong to claim that VFU is purely for procedural objections--far from it! In the Misplaced Pages:Undeletion policy: Reasons why an article might be requested for undeletion: "Misplaced Pages would be a better encyclopedia with the article restored". So VFU is a valid place to dispute a properly closed deletion discussion, if one simply wants to make points that one feels were ignored during the debate. Nothing we decide here can contravene official Misplaced Pages policies. If we want to change policy, we should do so with an appropriate level of consensus. --Tony Sidaway 13:05, 15 September 2005 (UTC)
          • I believe that consensus already exists, Tony. The policy may state that article content is reviewable in VfU, but that doesn't reflect reality. Appeals to article content are ignored by most VfU participants, except in the case of radical content change during an AfD, in which case it's the process that failed to take the new content into account, anyway. Article content should not be reviewable in VfU or any replacement scheme, or it will simply degenerate into a "second-chance" AfD where anyone, admin or no, will be able to dispute a validly-executed delete (or keep) decision (and in the case of current VfU, interpreted strictly, this could be done ad nauseum). android79 15:08, 15 September 2005 (UTC)
            • If you believe the undeletion policy needs to be changed, you should endeavor to have it changed. If your belief that there is a consensus to change policy is based solely on current practice of a majority of editors in VFU, you are mistaken. Misplaced Pages is not a democracy. I've just noticed that there is a fashion for citing "Valid AfD" as a reason to keep deleted. This is contrary to the undeletion policy. Accordingly, I have edited the head of the page, which was at variance with Undeletion policy, so that it correctly quotes it. --Tony Sidaway 19:58, 15 September 2005 (UTC)
              • I never said that this was a democracy. The established practice of a sizable number of people, over a sizable period of time, is one way that consensus can be established. Isn't it? I also note that I am not claiming that the undeletion policy is wrong, nor that the practice I cite is at varience with it. That is your claim. I also think that changing a long-stable instruction page (which arguably represents a level of consensus itself) to favor your views, during an active debate over policy and procedure, without discussing the change, is not the proper way. DES 20:58, 15 September 2005 (UTC)
              • If you think the original proposal is at varience with the current policy, do you not agree that a discussion seeking consensus is a proper way to modify that policy, by the way? if not, why not? or is there some beauracratic rule that says such a discussion must happen in a particular place? Thsi discusssion is not secret -- it has been widely announced in several proper places. DES 20:55, 15 September 2005 (UTC)
        • On the objection that the proposal makes a distinction betweeen admins and non-admins, this is becasuse there is a difference between admins and non-admins. A normal editor cannot undelete an article during the challenge process, although in the case of a non-trivial challenge the article should be undeleted. So this simply provides a mechanism by which an editor can try to convince an admin to undelete during the challenge. It could be merely formal; most reasonable admins wouldn't have a problem undeleting an article so that it could be edited and examined during the challenge. --Tony Sidaway 13:25, 15 September 2005 (UTC)
          • There should be no difference between an admin and a non-admin. Both should follow policy set by consensus. Just because an admin can do something does not mean that it is proper to do so. Any admin can delete the VfD and cause massive database impact. Does that mean they have the right to do so as your statement implies since they can do it while a "normal editor" cannot? The ability does not imply the right. Any admin can now undelete a properly or improperly closed VfD/AfD. Deletion Review seeks to make it policy that no such unilateral overturnings should happen. Let's make the admins follow the same rules and be "Administrators" and not "Gods" - Tεxτurε 14:43, 15 September 2005 (UTC)
          • If the editor in question can find an admin who can be convinced they should also be able to convince the community at large, since admins should be in touch with the community feelings. I would also fear friends-of-AdminX-clubs who know they'll do just about anything they're asked to, when it would be better if everyone just sat around the table together and spent a few days thinking about things. -Splash 19:01, 15 September 2005 (UTC)
      • This also leaves no method to deal with procedural erroes on CfD, TfD, IfD, MD, or SfD. These are less common than on AfD, but are currently in scope for VfU. DES 21:15, 14 September 2005 (UTC)
        • The idea can be extended to work in the relevant forums. I do think it would improve transparency to have challenges of this kind in the original fora, and not tucked away unnecessarily on a page that is disjoint with the one where the original decision was made. --Tony Sidaway 13:05, 15 September 2005 (UTC)

The thing I like least about this is that it strongly entrenches a philosphy that "admins can when they feel like it", but ordinary editors "must ask nicely". The original proposal treats everyone equally, and I don't fully see the benefit in enshiring inequality. And lets fact it, with nearly 600 sysops you'll be able to find one who'll undelete just about anything. -Splash 22:11, 14 September 2005 (UTC)

• The original proposal, if it contravened Misplaced Pages policy, was wrong. Change Misplaced Pages policy if you think it's wrong. Misplaced Pages is not a democracy. --Tony Sidaway 13:05, 15 September 2005 (UTC)
  • The proposal seeks to change Misplaced Pages policy as you suggest we should. As for "Misplaced Pages is not a democracy" - I need to restate that admins should not be "Gods" and were once described as "Administrators" because their task was intended to be only a maintenance position and not a god-like aristocracy. - Tεxτurε 14:51, 15 September 2005 (UTC)
  • I don't see this proposal as significantly changing wikipedia policy (although inso far as it is contrary to current policy it is a proposal to change that policy, of course). IMO this is a proposal to change the procedure by which policy is implemented. it does not seek to change any of the criteria or standars in the deletion policy that indicate what should be deleted, nor any of those in the undeletion policy that say why a deletion decision should be reversed. It only seeks to chnge some procedure, to say when and how a decision shall be considered for reveral. Specifically it seeks to extend the current process for reconsidering decisions to delete to also reconsider decisions that did not result in a deletion. it also seeks to make explicit that in reconsidering such a decision, valid results include "merge" and "redirect" as well as "delete" and "keep as an article", IF such a result is the clear consensus of thos who discussed the matter. Following consensus is hardly a violation of policy here. DES 15:04, 15 September 2005 (UTC)
  • Part of the problem the original proposal would resolve is the lack of any policy to guide behaviours in the event of a disputed AfD closure that is not a delete. It does not override any of the deletion policy and with a firm focus on process-not-content (which VfU most certainly does have) there will be no risk of double jeopardy, especially if we seek a strong consensus as has been suggested further up. -Splash 19:01, 15 September 2005 (UTC)
  • Tony Sidaway makes much of the phtrase "If in doubt, don't delete". That phrase does occur in the current deletion policy. it is at the end of a paragraph discussing repeated recreation of deleted articles by multiple editors, and similarly repeated nominations of an article for deletion. it suggests that in each case the repeated action may indicate consensus, but modifies that with "If in doubt, don't delete". I don't see this as a central tenant of the deletion policy. It certianly doesn't apply to copyvios, where the mantra is basically "if in doubt, delete" -- unless there is clear evidence of a proper release, wikipedia doesn't want any possible copyvios. Tony also says that VfU is not primarily for process. Note that the instructions on the VfU page itself say "Please don't list articles for undeletion just because your position was not endorsed on Votes for Deletion". The undeletion policy gives as possible reasons for undeletion that an article was 1) Deleted "out of process" (I take this to mean arbitrary deeltion in no way justified by any policy); 2 ) The article was wrongly deleted ( 2a)Admin not aware of dicussion on AfD; 2b) Never listed on AfD; 2c) proper objections to deletion were made but ignored); 3) temporary undeletion; 4) histroy only undeletion. Of these only 2c is soemwhat about content, and even that is significantly about process, it deosn't say that valid objections could have been made but noew one raised them, it says that they were raised and should have been heeded, but they weren't. In short I think the current VfU instructions and the Undel policy strongly indicate that VfU is curently primrily about procedure, not content. Surely this is the current practice on VfU, and thus the clear community consensus. If the policy doesn't follow that, then perhaps the policy has been silently changed and needs to be formally adjusted to match. DES 15:20, 15 September 2005 (UTC)
    • I certainly think it is the current process on VfU. Admins need not fear that they are going to be overruled in terms of demanding a higher or lower level of consensus. The question asked on VfU is, in effect, "was the admin within their discretion to close the debate the way they did". If they were, it's usually pretty obvious. If they weren't, that is usually pretty obvious also and it seems only natural to have a means of correcting the process.
  • Tony Sidaway also says that incorrect speedies should not be brought to VfU, but should be listed on AfD (or XfD) after uinilateral undeletion by any admin who disputes them. He basews this on two policies. WP:CSD says that under A7, if notability is disputed, the article should go to AfD instead of being speedied. But this is speeking at least as much about disputes that occur after an articel is tagged but before it is actuially deleted. In fact i read it that if ther is a dispute before an A7 is deleted it should go to AfD, but afterwards it should first come to VfU under the general undeletion policy, and only if undeelted be placed on AfD. Noter also that the undeletion policy says "If the page was deleted via AfD or Md rather than via Speedy Deletion....", thus clearly indicating that the poliy, and VfU, is intended to apply to speedy deletions as well as XfD deletions. Tony also relies on the "Eception" section of the undel policy. This says "If the page was obviously deleted "out of process" (i.e. not in accordance with current deletion policy), then a sysop may choose to undelete immediately." and goes on to say what should happen thereafter. i read this as primarily applying not to speey deltions whose basis is disputed -- those are not "out of process" because the admin who deleted them used the speedy deletion process, although it may have been used incorrectly. IMO this exception is designed primarily for deltions that followed no process at all, where an admin simply felt that the deletion was a good idea, with no policy to rtely on. The recent deltion of the VfD page comes to mind, as does the compltely out of process deletion of the proposed GNAA policy pages. Similatly an admin who comited what amounts to deletion vandalism or just bypassed policy could have the deeltion instantly undone under this exception. This is said to be an "eception" -- it is supposed to be a rare case, i should think. Seedy deletions are not rare, and debates about speedies are not all that rare, while admisn acting compeltely out of process are i am gald to say, quite rare indeed. So I think that Tony has misinterpreted the policy and VfU is cuirrently the proper place to debate allegedly improper speedy deletions, and DR as proposed would be the proper place to discuss allegedly improper speedies in the future. DES 15:36, 15 September 2005 (UTC)
    • To a certain extent, I wouldn't mind so much if disputed speedied were taken to AfD instead: the content of a speedy matters after all, since there is no debate to evaluate. The content of a disputed AfD closure on the other hand is a dispute about precisely that: not about the content of the article itself (except insofar as the closer may have had to refer to the content in determining what to do with some of the comments). -Splash 19:01, 15 September 2005 (UTC)

"Purpose of the page" section

Tony, I find it deeply distasteful that you make such changes as you just did to an established process when your proposal has thus far received no support. You told us to change policy if we didn't like it. That's what we're talking about doing. You on the other hand, have simply decided to reshape procedure you don't like on a whim. I could as well have changed the header, moved the page to a new anme and proclaimed the issue settled. But I figured discussion was a better way to go. Thus far, everyone apart from you seems happy that discussion is indeed the way to go. -Splash 21:53, 15 September 2005 (UTC)

I also wanted to say, before everything just died, that I realize you took the wording out of the undelete policy. But it seems to have been done with the fairly clear intention that the mechanics of VfU be peremptorily changed, given your current interpretation of what they are. -Splash 22:25, 15 September 2005 (UTC)

So, he was trying to make a WP:POINT? - Tεxτurε 22:31, 15 September 2005 (UTC)

Are you asking me or Tony? I don't think it was a WP:POINT, particularly. -Splash 23:17, 15 September 2005 (UTC)

• I reverted Tony's changes to the "Purpose of the page" section on the process page. Tony has asked me, on my talk page, to revert my reversion. I have declined, pending discussion here. Does anyone think that the current section should be changed? Does anyone think that tony's version is an improvemet, or that some other version should be used? (you can see the relevant changes in the page history, and particularly in this diff.) Note that the instruction section currently points users to the deletion policy and the undeletion policy and instructs people to read those policies (and implicitly to follow them) before using the VfU page. DES 16:10, 16 September 2005 (UTC)

• I note that Tony just reinstutited his changes, in a version a bit more through than before, and i have re-revertted to the longstanding text. I will not revert the VfU page again anytime soon, no matter what further changes are made to it. I hope that people can start discussing the issue here. DES 16:16, 16 September 2005 (UTC)
  • One way of determining that the community is ok with a process is that it has stood for some time. Just because one person comes along suddenly and changes it without discussion does not undermine the strength of the acceptance imparted by the passage of time and the long-standing behaviours of those regularly involved in the process. One person instituting a change after a good discussion is, of course, another matter. -Splash 18:48, 16 September 2005 (UTC)
• Given that the page already indicates that users should read the undeletion policy, I didn't think Tony's paragraph added anything (in addition, it read somewhat poorly). I removed it on that basis. Tony, if you'd care to explain what problem you're trying to solve on the talk page before adding text like that, that would be helpful. Nandesuka 16:51, 16 September 2005 (UTC)
• I have also made a relevant suggestion at Misplaced Pages talk:Undeletion policy#Process or content. I urge people interested in this issue to read and comment there as well. DES 16:56, 16 September 2005 (UTC)

Simply: the purpose of the page section should be replaced by the boxed proposal above, subject to us working out the details of the mechanics. The consequential changes should then be made to the undeletion policy. -Splash 18:26, 16 September 2005 (UTC)

Clearly it is right to include the precise wording from the undeletion policy at the head of VFU. That it was removed, I find inexplicable--whether or not people disagree with Misplaced Pages official policy, we are still bound by it, so it's only fair to warn editors coming here that if they see "Valid AfD, keep deleted", they will know that those votes are contrary to undeletion policy. I am happy to see that there is now at least some kind of discussion on undeletion policy itself. If you can get a Misplaced Pages-wide consensus to change that policy and reverse its meaning, good luck. It isn't going to happen. --Tony Sidaway 02:26, 19 September 2005 (UTC)