Sounds good to me. :)
I really think people should be able to vote delete, not only keep as alternatively proposed. That way one can see how much checking the listings have gotten. People will also be more inclined to help check when they can document their checking contributions. Thue | talk 18:20, 22 Oct 2004 (UTC)
Why don't we just make a policy that if an article on VfD has unanimity for deletion after three days, we close debate and delete it? That would obviate the need for yet another page for those of us involved in this sort of thing to monitor. -- Jmabel | Talk 19:01, Oct 22, 2004 (UTC)
- Beats me. I do recall that some inclusionists bitched about this when some people removed discussions from VfD before the duration was up, even though it was clear the article(s) concerned wouldn't be kept. I guess anything would be good at this point. I'm just fed up of hearing the same old extremist inclusionist crap day after day. I want to solve this so we can get on with work instead of getting unanimous votes on VfD and constant moaning about how inappropriate articles were speedy deleted. Johnleemk | Talk 19:06, 22 Oct 2004 (UTC)
- I objected to discussions being removed before the voting period was over, because in general those doing the removing were those who were opposed to VfD in the first place, and because there was no policy change which allowed such an action to occur. RickK 04:55, Oct 23, 2004 (UTC)
- That would work too IMO. But there is some value to 1) Making VFD shorter 2) collect all the real debates one place and all the obvious stuff another place. I prefer seperate pages. Thue | talk 19:43, 22 Oct 2004 (UTC)
The mechanism in this proposal can be defeated by one person who is willing to vote "keep" on every candidate for preliminary deletion and so refer every candidate to VfD. Gdr 22:56, 2004 Oct 22 (UTC)
- The mechanism is meant to deal with articles which are clearly delete, but not speedy deletes. Look at the criteria: jokes, vanity, dicdefs. If somebody votes keep to all articles just for spite and at odds with common sense and policy they will be just reverted as a vandal. On the other hand, if they vote keep with a good reason then the article will rightly be moved to VFD. Preliminary Deletion is NOT meant to delete articles whose deletion is in any way contentious. Thue | talk 10:54, 23 Oct 2004 (UTC)
- So in other words you'll simply ignore the keep votes by claiming the voter is a vandal? What's the point then? Why not just allow yourselves to delete what you like, when you like?Dr Zen 05:08, 29 Oct 2004 (UTC)
- I will remove/ignore votes if they are obviously cast in bad faith, in blatent violation of policy and against common sense. With suitable respect for erring on the side of caution. That is not the same as not respecting other people's valid opinions. Thue | talk 06:52, 29 Oct 2004 (UTC)
- A policy I very much admire is Assume good faith. The reason there are VfD is to allow "common sense" the chance to reign. If you personally are going to judge what "common sense" is, then you are privileging your POV above everyone else's. This policy will almost certainly be carried, but although I do think a mechanism to aid the deletion process one way or the other is a good thing, I don't think simply allowing particular admins to delete what they think should be deleted and to disallow any discussion by simply describing it as "not common sense" or "not policy" is in any way a good thing.Dr Zen 00:34, 30 Oct 2004 (UTC)
- Yes, assuming good faith is important, and should be practiced very widely. But there is a crossover point where it bad faith is obvious. If you do not believe that point exists then take a look at fx User:10101 :). Thue | talk 19:05, 5 Nov 2004 (UTC)
Well, I appreciate seeing the idea revived. I do wish to warn about one very great danger to open voting. If only one "keep" means VfD, and if voting is open to all users, then presumably authors of spurious articles would show up to cast those "keep" votes. That's why I had had admin-only voting. However, rather than just cursing the darkness, I have two candles to light:
- 1. A mechanism for guarding against suspect people voting (early and often) "keep" on their articles would be to have not admin-only, but a registered voter pool. Anyone may sign up, so long as they have an account that has been in existence on the site for over 4 weeks. However, they may not take part in the Purgatory voting until 2 weeks after registering to be voters, and no one may remain a registered voter, except administrators, for more than 4 weeks without taking one week off. That way, people can't register and vote "keep" as soon as they see their article named, nonce accounts can't take part, and yet everyone gets a chance to participate.
- 2. What I just said will be opposed by some folks as involving a new layer of beaurocracy. No way around that, that I can see: it does introduce a new layer of beaurocracy, but the alternative is closed voting or a zero-sum process.
- 3. Supposing that the voting is altogether open. I predict that it'll do not much good. However, it might still knock out the IP-created articles that are in whole sentences but of no value.
I'll keep watching as this progresses, and I'll certainly support it, even if it's open voting, but I really think some measure of guarding would help. Geogre 03:38, 23 Oct 2004 (UTC)
- I don't really support such complicated ideas. I think they create unnecessary instruction creep. It sounds damn complicated even to me; I believe it would turn off too many people from voting. Johnleemk | Talk 06:28, 23 Oct 2004 (UTC)
There are just too many inclusionists who will vote "keep" on everything. I can't see this ever leading to anything being deleted. RickK 04:54, Oct 23, 2004 (UTC)
- Ok, I'll handle these worries one by one. First off, from how I interpreted anthony's case on RFAr, voting "keep" for every article is not against the rules; but how you vote definitely is covered by the rules (anthony confirmed this on a post on the mailing list). So if some inclusionist votes keep on all articles, but provides no reason/an obviously invalid reason then the vote won't be counted. I think I'll amend the policy to fit this in. Secondly, Rick, even if inclusionists find a reason to "keep" every article, once it goes to VfD, depending on the article's delete-worthiness, it still has a chance of being deleted. Johnleemk | Talk 05:53, 23 Oct 2004 (UTC)
- Can we then change the rules so that articles listed here only need to stay on VfD for two days after the three days they spend on Preliminary Deletion? Mark my words, nothing will get deleted as a result of this page. RickK 06:19, Oct 23, 2004 (UTC)
- Uh...that's already in the rules. It's been there since my first draft of the policy yesterday. Have you read the policy thoroughly? And I don't really see how one keep vote will hold everything up on VFD. On this page, it will (but I've drafted an alternative already), but since after three days we move it to VFD... Johnleemk | Talk 06:28, 23 Oct 2004 (UTC)
- Mark my words, nothing will get deleted as a result of this page. I don't think that is true. Lots of articles go through VFD without any keep votes. Thue | talk 12:40, 23 Oct 2004 (UTC)
- But following this proposal, an article can only be deleted if there are zero keep votes. I don't see that happening. RickK 19:57, Oct 23, 2004 (UTC)
- I took a look at the current VFD. Of the article which would be covered by this proposal, few had keep votes from their author. If the authors don't vote in the current VFD then I don't see why they should vote under Preliminary Deletion. I guess most people respect our policy when they are just told what it is. Thue | talk 23:31, 23 Oct 2004 (UTC)
- I'm not talking about the author. Have you met Netoholic and Anthony DiPierro? RickK 04:09, Oct 25, 2004 (UTC)
- Har, har, har. I've clashed swords with anthony, but Netoholic is so trollish he isn't worth clashing with. However, if their reasons for keeping aren't reasonable, i.e. "Keep, because Wikipedia is not paper," then I think it's at the admin's discretion. People are free to counter their points for an additional 12 hours under a proposed amendment to this proposed policy (see the new section I added a few days ago). Besides, keep in mind that there have been a great many debates on VFD which had 100% delete votes. If suddenly people are voting keep for the sake of it thanks to this proposal, I think disciplinary action should be taken. Johnleemk | Talk 07:56, 25 Oct 2004 (UTC)
- Implying that I wouldn't respect policy? Please RickK, end the thinly veiled personal attacks. anthony 警告 15:08, 26 Oct 2004 (UTC)
- What is the point of this? Why bother having a page if you allow the proponents of deletion to simply disregard votes they don't like? Will you have a list of Wikipedians who are not permitted to vote on the page? I thought cabals were disallowed. This seems to me a way to give admins ungovernable power to delete as they see fit, whereas at least with VfD they must round up the posse and chase the articles out, as it were.Dr Zen 05:13, 29 Oct 2004 (UTC)
- Hmm. Not all admins are deletionist in intent or practice, nor are all admins adherents of the cabal. As an admin of extremely long standing, I can tell you from deep experience that the cabal is a very thin and tenuous construct in the imagination of a number of people, (including a few admins who do actually constitue what they conceive of as a cabal). For every admin who is liable to bend the rules to their own whim, I think you will find that there are easily another dozen who will not put up with such abuses. Fact. As a matter of record, I am strongly opposed to further extensions of the power base of admins, and I am deeply suprised that this vote was not more vehemently opposed than it was; I certainly voiced my reservations about it and I know a number of other admins also feel as strongly on this matter as I do. The point of Wikipedia is eroded by a strong policy-oriented lobby that seek to impose a spurious and completely unnecessary set of rules which will undermine productivity and the quality of work, both of admins and contributors. Sjc 18:55, 14 Dec 2004 (UTC)
- So now this proposal is the spawn of Satan? It seems like people who oppose efforts to reform the deletion system view it as giving more power to sysops. Understand: The discussion on this talk page is not part of the proposal. It is discussion about possible ways of implementing this proposal. It was because I'm very averse to making subjective judgements affect rulings that I had to complicate the proposal even more. If I intended to let admins discount votes based on voters' reasoning, I would have placed that in the proposal. But I didn't, because I know that that would leave too much margin for error and would lead to too much flamewars. So I don't understand what the big fuss is about; are all proposals to reform the deletion system really an attempt by the top sekrit sysop cabal to deter the unwashed masses? Johnleemk | Talk 04:02, 15 Dec 2004 (UTC)
- It is more unnecessary and unwanted policy on top of a system which is already top-heavy. VfD crept in as far as I am aware by the back door while I was taking a sabbatical; prior to VfD the policy of deletion was, by and large, managed amicably and sensibly with very little of the prevalent merde which now seems to have permeated WP as it has become a part of the mainstream.
- The votes in these foisted referenda are invariably unrepresentative on a number of counts. Moreover; they are deeply undemocratic since only a small, nay, tiny, percentage of the active WP population actually vote in them, usually, in my considered opinion (having studied the monotonously predictable list of all the usual suspects), only those who are hell bent in ensnaring us in a web of policy and those few who have time and inclination to actively seek to stop the whole process grinding to an inexorable halt. Usually most sensible people are busily working away at articles of a constuctive and productive nature and are not up to their necks promulgating some spurious proposal or other. I am probably going to promote a vote which will seek a moratorium on unnecessary and unwanted policy reform indefinitely. The system ain't broke, and is most definitely not in need of fixing, if anything it needs a good and rigorous rollback. Abolish VfD, abolish arbcom, abolish all the rest of the trite nonsense in which the process is inexorably becoming gridlocked. Sjc 19:18, 16 Dec 2004 (UTC)
On authors voting keep to their own articles: That doesn't happen that often af VFD. Why not just take it in stride when it happens, and the vote the article into the ground at VFD if it really clearly should be deleted? I think that is a more elegant solution that introducing extra rules. On the other hand, if the keep vote argumentation is not in line with policy as at Wikipedia:Votes for deletion/Rootian english then it seems fair to not move it to VFD. Thue | talk 12:40, 23 Oct 2004 (UTC)
- So will people who are voting 'delete' for invalid reasons have their votes discounted too? 'WINP' is NOT an invalid reason. Mark Richards 17:08, 26 Oct 2004 (UTC)
- It's at the sysop's discretion, just like VFD. For clarification, this is not intended to be some medium for changing VFD. It's supposed to be just like VFD, except that it will hold articles that meet the criteria listed in the policy. Otherwise, there will be little policy change related to the handling of deletions. WINP is an invalid reason when it's the only one, because it means we should accept any old prose, even unencyclopedic ones such as material suited for other projects like Wiktionary, Wikisource or Wikiquote. Johnleemk | Talk 14:09, 27 Oct 2004 (UTC)
Johnleemk, I sympathise thoroughly with your attempts to clean up the VfD process. I would support this proposal as one step in that direction, but I am afraid that it fails to address the structural problems with VfD. As Wikipedia is growing, the rate of new article creation is also growing, and exponentially. Some proportion of these are bad articles - arguendo a roughly constant proportion, but the actual number of bad articles will also be increasing exponentially. VfD is presently straining to cope with a deletion rate that is arguably already far too low. Various streamlining proposals will help, but will only speed things up by a constant factor; the exponential will soon catch up again. An additional problem is that there is a sort of evolution, or arms race, going on between vandals and deleters. The vandals are becoming more and more sophisticated, and it is taking increasingly large amounts of work to do a fair assessment of some VfD candidates - so really, we should be extending the voting period! It seems to me that rather than streamlining VfD, we should be making it harder - much harder - to create bad articles. Something like:
- Anonyms should not be able to create new articles. This is because new articles will not be on anyone's watchlist, and Recent Changes is too fast to keep up; consequently, many new articles are never assessed at all, and Wikipedia's consensual "security" mechanism is bypassed. (Anyway, with a million articles but more than half of them stubs, we should be encouraging newbies to expand stubs rather than create new ones.) To avoid vandals creating throw-away accounts just to make a bogus article, I can imagine at some point introducing even more restrictive conditions on article creation (e.g. title must be pointed to by at least 2 existing articles), but just restricting anonyms would be a good first step, and less controversial.
- We need tools to quickly spot the new patterns of sophisticated vandalism. One such sign is the creation of several new articles that are all interlinked but have few interlinks from the rest of the 'pedia. There are graph connectivity algorithms that could trawl the links database looking for such patterns, so we could provide a display feature for such article "islands" similar to current orphan article pages and so forth. Or Recent Changes could have a feature where new articles are listed with the number of inlinks from articles older than n days.
- Related to your Wikipedia:Managed Deletion proposal: as soon as an article gets a VfD notice, it also gets a robots.txt entry to stop it getting mirrored, or cached by Google.
Just a few thoughts, anyway. Securiger 12:35, 24 Oct 2004 (UTC)
BTW, out of curiosity, how many articles do people tend to have in their watchlists? I have 436, and that's just barely manageable. Securiger 12:41, 24 Oct 2004 (UTC)
How long are articles listed on preliminary deletion? anthony 警告 15:04, 26 Oct 2004 (UTC)
- Ah, there it is, 72 hours. anthony 警告 15:12, 26 Oct 2004 (UTC)
My biggest problem with this is that any votes made before listing on VFD should be removed if the article has significantly changed after the vote was made. anthony 警告 15:26, 26 Oct 2004 (UTC)
- I don't really think that's too big an issue. Articles change constantly while on VFD. The votes are at the sysop's discretion. Currently, I don't know of any policy/guideline that votes made before a major rewrite should be discounted. I don't think such a change would be within the purview of this proposal. Johnleemk | Talk 15:40, 26 Oct 2004 (UTC)
- I'm not talking about changes while it is listed on VFD, though, I'm talking about changes made before it is listed on VFD. Without such a clause, I guess I have to object to this proposal. The purpose is supposed to be to reduce the size of VFD. If an article is fixed during the preliminary deletion process (and the reasons for listing an article here suggests that many of them will be), we shouldn't be wasting people's time by then putting it through the VFD process. anthony 警告 19:35, 26 Oct 2004 (UTC)
- Whether it's fixed is subjective. I remember withdrawing a nomination for VFD once after it was cleaned up only to have the article renominated by somebody else — the article was eventually deleted. If we could work out a way to fix this, it would be great, but this policy is supposed to preserve deletion policy status quo unless otherwise necessary. Johnleemk | Talk 14:09, 27 Oct 2004 (UTC)
- Sometimes it's subjective. Sometimes it isn't. But counting people's votes who haven't even looked at the page doesn't make sense. We can work out a way to fix this. I just presented one. anthony 警告 18:59, 27 Oct 2004 (UTC)
Let me clarify my problem, though, because maybe this was misunderstood. My understanding is that people will be able to vote Keep or Delete during the Preliminary Deletion phase. Then, after this phase has completed, those votes will be transferred to VFD. My problem is what if the article is improved during the Preliminary Deletion process? Those votes made during that process should not be transferred to VFD. anthony 警告 19:39, 26 Oct 2004 (UTC)
- Your understanding is correct. However, what would happen if such a thing happened to a current article on VFD (I have seen this happen many times before)? How to count the votes should be left to the sysops' discretion. Often the point is moot because if an article gains many delete votes, it is likely to be almost certainly unencyclopedic. If it doesn't, two keep votes should be enough to prove a lack a consensus. If the sysop mishandles this, then it's an issue with the individual sysop, not the policy. Johnleemk | Talk 14:09, 27 Oct 2004 (UTC)
- What happens when such a thing happens to a current article on VFD is outside the scope of this proposal. What happens to an article on Preliminary Deletion isn't. Nothing should be up to the admins' discretion. Admins merely perform the deletion, they don't decide whether or not to delete. anthony 警告 19:04, 27 Oct 2004 (UTC)
- As I said, we don't need this anyway, because an article will get two or three keep votes once it has been rewritten appropriately, enough to hold up the deletion (and in borderline cases, enough to prevent consensus to delete), and because sometimes the reason for deletion is not related to the article content — for example, tell me how a vanity article about my best friend could ever become encyclopedic (unless he became a celebrity overnight)? Johnleemk | Talk 20:40, 27 Oct 2004 (UTC)
- Two or three keep votes is not always sufficient to keep an article from being deleted, and besides, the point is more to waste less of people's time. tell me how a vanity article about my best friend could ever become encyclopedic It could be rewritten by someone not related to your friend. But in any case, I never argued that all articles listed here could be fixed. It just seems that most of them could. And I don't think you can argue against the fact that some could. anthony 警告 15:30, 5 Nov 2004 (UTC)
- Without some sort of incentive to improve an article before it gets to VFD, this proposal is useless, and I will oppose it. anthony 警告 19:04, 27 Oct 2004 (UTC)
I would like to suggest removing all reference to 'notability' from this. It has no objective meaning, and is purely a value judgement. Mark Richards 16:49, 26 Oct 2004 (UTC)
Adding a further level will give more to argue about. Is that what we really need? Then there's the added work in transferring the material. Of course the shorter period will improve the chances that nobody will notice that something is there, and enhance the chances that it will be deleted. But I guess that that's all part of the scheme. Eclecticology 01:14, 2004 Oct 27 (UTC)
- No offense, but assuming there's some scheme behind this reeks of the cabal accusations some trolls like Lir have hurled. Johnleemk | Talk 14:09, 27 Oct 2004 (UTC)
- I don't think that's fair. Suggesting there is a scheme to get more articles deleted and more quickly is hardly trolling you, John. It's precisely the outcome you're looking for. There's nothing wrong with that. It is a worry, though, that you suggest that the sysop should disregard keep votes at their own discretion. This would rather allow a group within the community to ensure the rapid removal of articles it doesn't approve of -- if such a group can be said to exist.Dr Zen 05:25, 29 Oct 2004 (UTC)
- I don't know how you can judge whether such comments are trolling when you yourself accuse such an outcome of being the one I'm "looking for". And vote-handling is something the sysops are entrusted to do. After all, if sysops are allowed to disregard delete votes at their own discretion as anthony has suggested, why can't they disregard keep votes at their own discretion as well? All in all, a problem of deciding how to weight a keep vote would easily be solved if people don't maliciously troll the system or put forth a decent reason for their vote. Johnleemk | Talk 07:56, 29 Oct 2004 (UTC)
- I'm awfully sorry. I thought the idea was to clear the VfD page and bring in another means of deletion for deserving candidates. If it isn't, then I've misread, and yes, it would seem to be just more bureaucracy. The problem of weighting votes is not going to be helped by the insistence that people are trolls, surely? Dr Zen 08:47, 29 Oct 2004 (UTC)
After all, if sysops are allowed to disregard delete votes at their own discretion as anthony has suggested I'm not sure what you're referring to here. Am I the "anthony" you're talking about? I've said that I think all votes should be thrown out when the article has significantly changed, but I never suggested admins should have any discretion and certainly not that they should be able to ignore delete votes at their own discretion but not keep votes. In any case, this proposal is a long winded version of "if no one has voted keep after three days, delete the article". There's nothing of substance other than this, so I see how someone could say that this will accomplish anything other than to get more articles deleted and more quickly. Mostly just the more quickly part. If this proposal provided anything to give people incentive to create rather than destroy, such as my proposal to throw out votes made during preliminary deletion when an article is fixed during preliminary deletion, I'd be in favor of it. But instead, we'll have someone making an article saying "Bob Dole is a student at Greenville School", it getting listed here, receiving 8 delete votes, someone will fix it into our current Bob Dole article, it'll get two keep votes, and then we'll actually list it on VFD where it will sit for two days just because some troll wants us to follow some ill-conceived rule like this one. (Yes, it won't be Bob Dole, as we already have a Bob Dole article, but you know what I mean). anthony 警告 01:59, 30 Oct 2004 (UTC)
- I'm not sure what you mean. As I pointed out, often the problem with the article in the first place isn't the content; it's the topic. The only criteria under which this would apply would be substubs; vanity pages, original research, etc. — no matter how well they're rewritten, they aren't encyclopedic. Throwing out the votes just because the article was rewritten makes no sense at all if the issue with the article isn't its content, but its topic. Johnleemk | Talk 04:43, 30 Oct 2004 (UTC)
- The only criteria under which this would apply would be substubs; vanity pages, original research, etc. — no matter how well they're rewritten, they aren't encyclopedic. Huh? When you expand a substub, it's no longer a substub. When you rewrite a vanity page or one that is original research, it's no longer vanity or original research. anthony 警告 00:49, 31 Oct 2004 (UTC)
- What I was saying was that the only reasonable case where throwing out votes might actually be justified would be substubs. As for original research, some topics are just so *original* (i.e., have not even been peer reviewed nor made known to the scientific community; something a college student just cooked up and threw onto Wikipedia) that no matter how well-written, they'd still be original research. And for vanity pages, kindly read Wikipedia:Vanity page. The definition of such a page makes it clear that if a page really is a vanity page, it has no chance in hell of ever making the cut as encyclopedic, no matter how well-written. After all, I could write several pages worth of information about me, but that's not the point — it'd still be a vanity page, because such an article about me would meet all the criteria for vanity pages listed there. Johnleemk | Talk 10:20, 31 Oct 2004 (UTC)
- I named two other categories. Vanity pages and original research. Maybe some topics are "so original" (as far as my guess as to what that's supposed to mean) that you can't write a good Wikipedia article about them. But I guarantee you these won't be the only articles that are listed here. As for "vanity page" that page you point to (which only has 5 different editors and is primarily the work of a single editor) is by no means the only definition of "vanity" which people here on Wikipedia use. If you look at VFD probably less than 5% of articles listed there as "vanity" qualify as vanity according to that page. Sidney Morgenbesser was listed on VFD as "vanity", for instance, and it fits none of the definition. If preliminary deletion was limited to vanity pages about which there is no verifiable information available, then you might have a point. But that's never going to happen. anthony 警告 15:40, 5 Nov 2004 (UTC)
A lot of people think this policy would be a good idea, but problem X would nullify any effectiveness/create more problems. So why don't we hold a trial run? I'm not sure what a good length would be, maybe two months? Anyway, I added a second question to the voting to reflect this, although it conspicuously doesn't contain a length for running the process. Suggestions? If I don't get any, I'll probably put it as two months the night before the vote. Johnleemk | Talk 16:57, 27 Oct 2004 (UTC)
- I'd agree to a trial of a week, perhaps, but two months is way too long. anthony 警告 19:05, 27 Oct 2004 (UTC)
- How about a month? I think we'd need at least two or three weeks before being able to make any serious judgement on this. Johnleemk | Talk 09:41, 28 Oct 2004 (UTC)
- I'll have to reread the proposal, but I guess three weeks won't hurt. I'm not really sure what the improvement is, and there are some things which I see as a drawback (more complicated being the biggest). But maybe having preliminary deletion can convince admins not to speedy delete as much. anthony 警告 13:14, 28 Oct 2004 (UTC)
I think that simply disabling the ability for anon users to create new articles would probably halve the number of VfD candidates. This would be a lot less effort, and actually tackle a root cause of the problem, rather than patching up the result. Shane King 04:50, Oct 29, 2004 (UTC)
Another idea would be to just make deleting/undeleting like reverting: something anyone could do. If you get into a delete/undelete war, you take it to the talk page. If that fails, you go through the usual dispute resolution process. That removes all need for any VfD process, and once again, tackles the root cause rather than the symptom of the problem.
For these two reasons I vote no to the proposal: it creates more work for no real benefit. 04:54, Oct 29, 2004 (UTC)
- However, disallowing creation of articles by anons is anti-wiki. I've seen anons create useful articles as well, so it doesn't seem fair to punish all anonymous contributors. Besides, anyone determined to vandalize or push POV on Wikipedia would simply register.
- I'm not sure, but I bet that deletions/undeletions put more load on the database than reversions. Also, deleting an article deprives users of its content, at least on the live, main Wikipedia. Deletion wars would be even more of a mess than revert wars are.--Slowking Man 04:59, Oct 29, 2004 (UTC)
- I'm not worried about vandals, vandalism is already a candidate for speedy deletion. I'm worried about people with good intentions, but who don't know what they're doing. I don't see how requiring someone to register is anti-wiki. If they take the time to register, there's more chance they'll read the editing guidelines. Also, there's more chance if they learn the ropes on existing articles (which I think should still be anon editable) they'll know how to create good new articles.
- Anyway, I'm starting to lean towards the second proposal as the better solution: make deleting and undeleting a normal action anyone can do. It solves all the problems as far as I can see, and is no more liable to vandalism than things currently are (anyone can already blank and article they don't like, what does it matter if they delete it?) Shane King 05:33, Oct 29, 2004 (UTC)
- Hell, no. Forcing users to register has been discussed a billion and one times before, and the only result that we've come up with is that doing so provides little benefits and many problems. Deletion and undeletion? Hell, we have enough problems alreayd with people whining about admins speedy deleting things they shouldn't have. Do you expect all users to conform to this if even admins can't do so? Making deletion a unilateral process won't help the problem much, if at all. Johnleemk | Talk 06:25, 29 Oct 2004 (UTC)
- If undeletion is open to all, it doesn't matter if a page is deleted in error. Any user can bring it back with a few clicks of the mouse. It would remove the current heavy cost of deletion (that it's painful to do and undo) and bring it more in line with normal edits. It would remove the need for votes for deletion, as only truly contentious deletes would end up requiring votes to gather consensus (through the usual dispute resolution procedures). Furthermore, it's very pro-wiki: it's an open process that priviledges no one user over another. I don't see what's not to like about it. Shane King 09:22, Oct 29, 2004 (UTC)
- Regardless, that's not under the scope of this policy. Perhaps you could bring this up on the mailing list or the village pump. Johnleemk | Talk 09:56, 29 Oct 2004 (UTC)
- I agree it's not under the scope of this policy. I posted it here because I felt I should explain my no vote, and the page specifically requested no comments on the vote page. I'm not subscribed to the mailing list at the stage, maybe I should be. Shane King 11:28, Oct 29, 2004 (UTC)
- I am surprised that there is not already a "hide article" feature, as a midpoint between existence and deletion. - KeithTyler 19:24, Nov 5, 2004 (UTC)
Hi there. I have a question about the corollary for handling a possible caveat of the preliminary deletions policy. It says "that each article must receive at least two "keep" votes before going to VFD." Later, it says that "if 90% or more votes are for delete, then the article will not be listed on VFD until the percentage of delete votes drop below 90%."
So what happens if the article has two votes (which should normally send it to VFD), but the percentage of delete votes is over 90%? If no more "keep" votes are added, and the percentage of delete votes never drops below 90%, what happens to the article? Is it declared that consensus hasn't been reached? Is it deleted? Is it sent to VFD anyways? Will votes by striked out? I tried asking on #wikipedia, but nobody seemed to know :(. Thank you for your anticipated response. -[[User:Frazzydee|Frazzydee|✍]] 03:53, 30 Oct 2004 (UTC)
- Thanks for your query. In the case you mentioned, then the article would be deleted, because if there's more than 90% delete votes, then there's a clear consensus for delete — note that this is even higher than the standard for consensus we use to elect admins, which is 80%. If there's a lack of a clear-cut consensus, then the article goes to VFD. Hope this cleared the matter up. (btw, I will be reposting your message and my reply on Wikipedia talk:Preliminary Deletion). Johnleemk | Talk 06:38, 30 Oct 2004 (UTC)
(note: the messages above have been copied from User talk:Johnleemk and User talk:Frazzydee)
90% delete votes is by no means a "clear-cut consensus". It may constitute "rough consensus", but to be a "clear-cut consensus" it would need to have no serious objection. anthony 警告 15:44, 5 Nov 2004 (UTC)
- True, I used the wrong terminology. Johnleemk | Talk 16:06, 5 Nov 2004 (UTC)
- The proposal itself seems to use neither term, using the phrase "unanimous agreement for deletion". anthony 警告 20:50, 5 Nov 2004 (UTC)
- That's in the original unadulterated proposal. We're currently discussing the alternative that adds precautions to prevent abuse. Johnleemk | Talk 08:27, 6 Nov 2004 (UTC)
- I still don't see the term "rough consensus". anthony 警告 15:43, 8 Nov 2004 (UTC)
- Because the alternative in question wisely avoids calling it a consensus of any form, instead merely providing a definition of 90% and two keep votes or more being the threshold for moving the listing to VFD. Johnleemk | Talk 15:45, 8 Nov 2004 (UTC)
I would have expected dicdefs to be among the categories valid for this method. A dicdef that has no room to expand into an encyclopedic article should be quick-deleted. This is another category of article that we have a lot of. - KeithTyler 19:22, Nov 5, 2004 (UTC)
- Do we really? Most of the dictdefs I see can be easily expanded. anthony 警告
- I base that on the frequency which dicdefs get submitted for VFD, and more often than not get deleted i.e. transwiki'd to Wiktionary. The preliminary deletion process could speed up the movement of idle dicdefs to Wiktionary. - [[User:KeithTyler|Keith D. Tyler [flame]]] 22:16, Nov 5, 2004 (UTC)
- Most of the dictdefs I've seen on VFD were quite easy to fix. They might have ultimately been deleted anyway, but whether or not something gets deleted by the VFD process has little bearing on whether or not it could have been expanded. The preliminary deletion process would speed up the process of moving dictdefs to Wiktionary, thereby decreasing the number that get fixed. anthony 警告 02:20, 6 Nov 2004 (UTC)
Predictably, suggestions for purgatory policy are turning the process into something more and more convoluted, until it's something that requires you to first page through about 20 caveats, exceptions and what-ifs before you find out what you can and can't do with your purgatory vote, and what's likely to happen with it if you cast it.
Purgatory isn't going to work, for two main reasons:
- An inclusionist idiot can vote "keep" no matter what. Some idiots undoubtedly will.
- Demanding that said idiots provide a reason is no solution, because then deletionist idiots can cast every reason into doubt. Is "Wikipedia is not paper" a valid reason? No? Why not? Because you don't like that line of thought? Shall we have a few heated discussions every time it's used, then?
Please read "zealot" instead of "idiot" if the above offends you. If it then still offends you, tough luck. :-)
In short, purgatory would just become VfD except that the arguments would be about what votes should and shouldn't be considered, instead of what articles should and shouldn't stay. Instead of lifting the load of VfD, it would just burden people even more. Adding more red tape doesn't change the fact that the idea is fundamentally broken.
I have voted No on both this proposal and a trial run for it, because I think even a trial run is going to damage the deletion process. Thinking about the problem, I've got a different idea. Unfortunately, "preliminary deletion" is already taken, so to prevent confusion, I'll refer to my idea as "countdown deletion". I'm mentioning it here and now to give people a better chance of catching it in context. Note that this idea is not a fully rounded-out proposal (yet), is not directly related to this vote, and I am not voting "no" so I can ego-trip on a private suggestion (I genuinely feel the original proposal will not work in any case, for the reasons stated above).
I need quite a lot of space to explain what it's all about, so it's on a personal subpage. Here is the summary:
- Articles can be put on countdown deletion instead of speedily deleted and before VfD.
- Articles on countdown have one week to be improved. If there are no edits, the article is just deleted. If there are edits, there's a vote whether the article is significantly improved. If there's a majority vote that it's not, the article is deleted. Otherwise, it lives and can go on to VfD if that's still necessary.
I'm still fairly new here; if I'm committing some major policy faux pas with this suggestion, correct me. If you think this is not worth reading because I'm still fairly new, bite me. :-) JRM 00:28, 2004 Nov 8 (UTC)
- Er, well, have you read the proposed alternative for avoiding inclusionist rampages? People have brought up this obvious problem before. I should probably take out the provision encouraging a reason after voting, because that's rather completely irrelevant to the process (I mean, "encouraging"? It's not even mandatory! I must have been on crack when I wrote that). I don't see why we should avoid discussion; discussion is good. Even if it may waste a little time, it's helluva more productive than edit warring or trolling or just plain ignoring other users. Discussion is the first step towards building consensus. The proposed alternative isn't red tape; rather, all it does is step up the requirement to two keep votes or more, the total of which must be at least 10% of the total votes on the article, before the article may be listed on VFD. I'm thoroughly confused as to how this could be construed as extra bureaucracy or red tape. As for your proposal, I think it's good, but it still doesn't fundamentally address the larger issue of VFD being crowded with articles that not even inclusionists bother to vote "keep" for. Thinking about it logically, Preliminary Deletion addresses a significant set of articles. Countdown Deletion addresses a subset of the set Preliminary Deletion handles. Logically, it would be better if we found a way to sufficiently improve Preliminary Deletion instead of creating a new policy which will handle even less articles than Preliminary Deletion. I wouldn't oppose Countdown Deletion if it came up for a vote, though. Johnleemk | Talk 10:02, 8 Nov 2004 (UTC)
- Yes, discussion is good. Discussion also doesn't scale. Discussion is what we now have on VfD, and it has a tendency of rehashing the same arguments over and over again, without ever coming to consensus. There is middle ground between inclusionists and deletionists, I'm sure of it. I just don't think VfD, or Preliminary Deletion, will ever bring it to light, or be able to make use of it effectively. Countdown Deletion won't, either, but it's not a stopgap solution or a replacement. It's just an alternative track. VfD is good. VfD is for when you need discussion over whether something should be deleted. But do we always need discussion over whether something should be deleted?
- At first sight, it seems that Preliminary Deletion solves this by just putting in a vote. But as far as I can tell, this is just VfD before VfD (in fact, the proposal says as much). It's a census poll of whether inclusionists or deletionists care more at any given moment. PD tries to keep people honest by appealing to policy, but unfortunately it's the same policy that VfD discussions are always mired in. No real hope of speeding things up.
- The "red tape" remark was badly worded, as the provisions (well, most of the provisions I've seen people suggest) wouldn't slow down the process itself, just make it harder for people to estimate its impact and usefulness. I also made the mistake of making it appear as if this criticism was levelled against the proposal as-is. It's not. I shouldn't have identified unrelated suggestions. I'm sorry for being unfair. What I meant was that, IMO, PD would just be a no-change buffer before VfD: an article would be discussed on exactly the same merits as it would on VfD (perhaps a bit more streamlined) and if PD lightens the load at all, it won't make any real difference. (Come to think of it, perhaps I was too hasty voting down a trial run—no, personally I still don't want one. If the majority wants it, I'll be interested in the results, though.)
- You remark that Preliminary Deletion > Countdown Deletion is quite accurate. However, Countdown Deletion requires practically no active input to work (other than improving the article, which can hardly be bad). When it does, it's based on whether you think the article has improved. This should finally steer things away from the pointless bickering (and it is bickering, not discussion) over whether policy does or does not permit "notability" and "Wiki is not paper" as reasons for deletion/inclusion, for those cases were we can do without it because people are (not) willing to improve the article.
- "...it still doesn't fundamentally address the larger issue of VFD being crowded with articles that not even inclusionists bother to vote "keep" for." I beg to differ. In a sense, that's exactly what it's intended to handle. Articles that not even inclusionists care about are the most likely to get axed by CD. (Of course, I am assuming that people would use CD, instead of just blindly putting everything on VfD—but notice that PD implicitly counts on that, too.) Instead of "hey, inclusionists, vote Keep here or the article will get deleted by the evil deletionist hordes" you'd get "hey, everybody, improve this dubious article or it will get deleted for lack of caring". And if you can't improve the article because it is irredeemable crap, according to you, that's fine, leave it alone. If everyone thinks its irredeemable crap, the article goes.
- For the rest of the world, I feel it's necessary to stress here that CD and PD don't exclude each other. I personally think PD won't work, but that doesn't mean it's no good. There's always a sizeable chance that I'm wrong. :-) Vote on the proposal on its own merits. JRM 10:49, 2004 Nov 8 (UTC)
- The chances of there being a debate are likely to be quite low, considering that inclusionists don't raise up a fuss about a good deal of those 100% delete articles on VFD. If there's a sudden rise in this due to Preliminary Deletion, disciplinary action ought to be taken. This isn't VFD before VFD. This is a trimmed VFD for articles which would get no keep votes on VFD, but because they're prose, are ineligible for speedy deletes. Things will be sped up, if you haven't noticed, because after 72 hours, either the discussion moves to VFD, or the article's deleted. Notability discussions, etc. are very unlikely to appear on this page (if they do, there's a gigantic problem), because inclusionists are willing to let what deletionists call "non-notable" articles be deleted if the information isn't verifiable. This page isn't for handling the controversial school articles; this is for handling the garage bands or articles about dead cats, etc. — things which are clearcut delete cases as defined by policy, yet cannot be speedied because they're prose. Come to think of it, I was probably on crack when I said that CD wouldn't handle the same thing as PD as efficiently, so I retract that (although I think it's probably true to a certain extent, I'm too sleepy to think of an example). Regardless, it seems to me that your proposal is just a different way of approaching the same problem, albeit one that would be a bit too radical for the tastes of most editors (and somehow I get the feeling that hardcore deletionists and inclusionists will both oppose it; it's not like they didn't oppose Preliminary Deletion either, though). Johnleemk | Talk 15:16, 8 Nov 2004 (UTC)
- OK. I think I get PD more clearly now. I'm still not quite convinced it'll run as smoothly as you hope, but at least I'm willing to concede that it could. If it does, it would help. I still don't think it should be policy, but I am reverting my No vote on the trial run question to Yes. (I can do that, right? Sure I can. :-)
- "Come to think of it, I was probably on crack when I said that CD wouldn't handle the same thing as PD as efficiently..." That's the second time now. You definitely need to lay off the illegal substances. Like, chill out, you know? :-)
- Regarding the radical nature: yeah, it's probably radical. I think we need radical. I've noticed a lot of people are opposing PD because they see it as "more of the same", so maybe radical will have more success. But that's neither here nor there. JRM 16:11, 2004 Nov 8 (UTC)
- I think this whole proposal/reject process is wacked. The proposal should be for a kind of contest between proposals, each of them written collaboratively. Its not so much that we should choose whos idea is best, but since we all seem to agree on the generals, we need only to work on the specifics. So, the idea of using the Category scheme to list pages on VFD was a good one, but it seemed to fail because it wasnt articulated well - simply, without all the blah blah... - SV 02:29, 10 Nov 2004 (UTC)
In response to Johnleemk's request on the mailing list, I'd like to start this section so we can have a dialogue on why some users opposed this proposal. Rhobite 04:35, Nov 13, 2004 (UTC)
- I'll start. I voted no because I believe problem resolution is already too complex here. When a user has a problem on Wikipedia, there are a dozen or more places it can be listed. Some of these places result in little or no action, but one of these places, VfD, is known for "getting the job done." This is why it's overused and abused. Here's the list: VfD, Speedy Deletion, RFC, RFM, RFAR, PR, Cleanup, Current surveys, PNA, and the pump for good measure. I won't vote yes on another problem resolution section until one of these existing sections is eliminated. I still don't know how Cleanup and Pages Needing Attention are fundamentally different, maybe we kill one of those. And it's hard for me to see the value in three separate deletion paradigms. Maybe we should take a step back and think of realigning both VfD and Speedy. Be bold, right? One more thought, the reason VfD is so effective at cleaning up articles which should never have been listed, is the time limit. VfD is one of the only problem resolution pages with a sense of urgency. Rhobite 04:35, Nov 13, 2004 (UTC)
- I won't vote yes on another problem resolution section until one of these existing sections is eliminated. But how would PD dreadfully overload things? This would be easier to understand if you think of Preliminary Deletion as a part of VFD, except on a different page. It's merely doing the same job as VFD for articles that crowd VFD. If you reread the proposal (I've rewritten it to include a section answering nine questions/objections), you'll find that in a listing comparing the purposes of every page handling problems, deletion of articles, which has a workload comparable to discussion on the pump, has only one page for handling it (speedies don't count because they're more of a category instead a page, and are handled very fast); the pump has seven — one main page and six subpages; any of these pages can be read separately and remain coherent (while VFD has subpages, they are all centralised on one page; the pump has six subpages for placing a discussion; article deletion has one for listing articles for deletion). Preliminary Deletion is necessary because VFD is overloaded as it is. I still don't know how Cleanup and Pages Needing Attention are fundamentally different, maybe we kill one of those. Whew, I thought I was the only one who thought that. And it's hard for me to see the value in three separate deletion paradigms. It makes little sense lumping articles of dubious encyclopedic value and articles definitely lacking encyclopedic value together on one page. It stifles discussion for both types of articles. One more thought, the reason VfD is so effective at cleaning up articles which should never have been listed, is the time limit. VfD is one of the only problem resolution pages with a sense of urgency. As you can see, I agree; people argue VFD is only for deletion, but I disagree. It's the easiest way to get a page cleaned up; I think that we need to reform cleanup so that VFD is no longer so necessary for this; it's simply easier to find an article you can fix on VFD, than trudging through cleanup. Johnleemk | Talk 09:20, 13 Nov 2004 (UTC)
- Thanks for the great explanation. I was picturing a new user looking at some Pokemon fan article or a garage band article, and wondering "Do I Speedy it? Do I VfD it? Or Do I PD it?" It's very confusing. If there's a way to present full VfD and PD as two components of the same "system" it would be much better. I forgot to mention RFE in my first list. While they all have subtle differences, I really think Cleanup, PNA, PR, and RFE are redundant. Since the acts of cleaning up, peer reviewing, and expanding are all subsets of attention, I think people should consider eliminating PNA, or realigning those four sections down to three. Rhobite 17:28, Nov 13, 2004 (UTC)
- Perhaps we could create an "article deletion toolbox", ala Template:Featuredtools, to cover this. I agree PNA needs to go, since I've never understood the difference between it and Cleanup. Peer Review is fundamentally different from Cleanup/PNA, though, because it's for improving articles that are likely close to featured level. Cleanup is for the other end of the spectrum, borderline articles. Johnleemk | Talk 14:07, 14 Nov 2004 (UTC)
- I voted no because I don't think that VfD is that bad right now. The only major problem I see is that the page does get too large, but that's something that can be fixed without a seperate deletion process. In my very limited experience, VfD pretty much works, obvious deletion requests rack up the unanimious votes against it fast, and obvious keep candidates get the unanimious keep votes fast too. After that people who don't know anything in particular about the topic can just skip over it, and those who do know something still have time to chime in. I really like the five to six day waiting period. I think it's a good thing for all pages, even ones that seem obvious, because I've seen a few cases that went from an "unanimious delete" to a cleanup to an "obvious keep". - Lifefeed 20:11, Nov 24, 2004 (UTC)
- Your basic argument is that this is a solution looking for problem. As you admit, though, the problem is very real — VFD is just getting too big. You say we don't need a separate deletion process. This isn't. It's supposed to be viewed as part of VFD; it's just housed on a different page. This is a method of categorising VFD nominations. The problem right now is that we have a "third class" of articles; this class of articles is hardly enyclopedic, yet it does not meet the standards for speedy deletion. We get a lot of these (seriously, I've counted two or three within the space of ten minutes on RC patrol). VFDing these is an absolute waste of time, because of Wikipedia:What Wikipedia is not. No matter how well written, by the standards of our policy on vanity pages, a vanity page can never be encyclopedic. And so on. Now, we get a lot of these, and they are crowding VFD. The obvious solution is to house a specific part of VFD designed for these articles on a separate page. This is designed to be not very different from VFD, except that it's for handling articles clearly not encyclopedic, and that there's a shorter turnover time (three days should be enough for someone to dig up proof of an article's encyclopedic value). Johnleemk | Talk 06:43, 27 Nov 2004 (UTC)
- Let users delete and recreate articles freely? Imagine the delete wars!
MeatballWiki and WikiWiki both allow normal users to delete and undelete articles (the exact methods differ), and have not suffered as a result. We would probably want to adopt a version of Wiki:DeleteOnceRestore. Martin 15:53, 14 Nov 2004 (UTC)
- I think you mean Wiki:DeleteOnceRestoreOnce. —AlanBarrett 10:46, 20 Nov 2004 (UTC)
- Meatball:WikiPediaIsNotTypical. We have a much too contentious and popular community for typical wiki deletion. Andre (talk) 18:24, Nov 16, 2004 (UTC)
Does anyone have any plans to implement the policy now that the vote passed?
Don't look at me, I don't feel like doing it right now, I just want to use it :).
Thue | talk 22:03, 22 Nov 2004 (UTC)
- Um, no, the vote did not pass. The "consensus" was very tenuous, if it existed at all. So, this won't be implemented any time soon. Johnleemk | Talk 05:32, 23 Nov 2004 (UTC)
- The vote was 114-41, or 73% in favor (adding the yes votes together and ignoring accusations of sockpuppetering). That is not consensus, but it is pretty good. I consider it a good enough mandate to implement, but if you feel we should get a wider consensus before a new policy is implemented then I respect that. Thue | talk 19:22, 23 Nov 2004 (UTC)
- What percentage do you mean by consensus? I view this resolution as having passed and have been quite mystified up to now as to why no action has been taken on it. I think it's very unrealistic to expect a better voting margin than this. However you phrase a proposal there will always be those who disagree with it. I view 73% as an overwhealming majority. If we don't accept even such wide margins we will find ourselves paralised. Barnaby dawson 11:48, 2 Dec 2004 (UTC)
Speedy Deletion, as I've now learned to understand it, is a request for "summary judgement" on a page where the content is undeniably and objectively inappropriate.
This "inappropriateness" is not one that is supposed to be defined in a POV way. It is not based on agreement or disagreement with content, or distaste over content's presentation, or like or dislike of the topic, or like or dislike of the contributor. It is one that, in theory, could probably be automated.
What are the current reasons for speedying?
1. No meaningful content or history (e.g. random characters). See patent nonsense.
- The definition of "nonsense" that is intended here is admittedly a hard one to recognize (I made incorrect inferences myself at first), but "patent nonsense" is not defined as "imaginary topics" or "theories that will never work" or some other value judgement or prediction. Rather it is, strictly speaking, content that is not intelligible. Random characters, "greeking", etc. are what applies here.
- This is pretty easy to define. If it can't be read, or can't be converted to something with some meaning, it's nonsense.
2. Test pages (e.g., "Can I really create a page here?").
- In other words, pages that have absolutely no informative content. This is not the same as "pages that only include content that everyone already knows". It includes pages that do not include any information on their topic, on other topics, etc.
3. Pure vandalism (see also dealing with vandalism).
4. Very short articles with little or no context (e.g., "He is a funny man that has created Factory and the Hacienda. And, by the way, his wife is great.")
- This one is not fully defined. I think that it is a bit controversial, because a value judgement on "very short" and "little content" has to be made. Perhaps that really is all there is to say about Mr. Hacienda. To illustrate this, the CSD page goes on to urge that the page should try to be saved:
5. Reposted content that was deleted according to Wikipedia deletion policy.
- Nice and clear cut, and again, automatable. If it failed VfD before, it can be speedied for all eternity. Again, CSD includes caveats here.
6. Articles created and edited solely by a banned user after they were banned, unless the user has been unbanned. This is slightly controversial!
7. Foreign language articles that already exist on another Wikimedia project, as a result of having been copied and pasted in to Wikipedia after their creation elsewhere, or as a result of having been moved via the transwiki system.
8. Temporarily deleting a page in order to merge page histories after a cut and paste move.
CSD is very specific and narrow, because it is intended to be fast and unchallenged. As a result, it has to be clearly and specifically defined beyond any shred of POV. Just as with summary judgement in a court of law, the reason for summary deletion must be undeniable and unarguable.
Expanding CSD is an idea on the side of deletionism to codify a particular POV of what is allowable on Wikipedia into a process where there is no room for objection, no audit trail (at least not as there is with VfD), and therefore little room for appeal.
Preliminary deletion, most importantly above all, retains the consensual process, but attempts to expedite it for cases where a consensus is expected. Currently such cases must go to VfD, where unnecessary effort is wasted on subjectively obvious deletions, as opposed to objectively obvious deletions (what Speedy is for). If the cause for deletion of an article is subjective, then there needs to be room for an oppositely subjective opinion on the article to be introduced into the process. There is no room for this in speedy.
Expanding CSD removes the consensual process for a class of deletion cases which are not objectively inappropriate. Therefore, it is not an appropriate Wikian solution to the problem that Preliminary Deletion is trying to solve.
[[User:KeithTyler|Keith D. Tyler [flame]]] 20:16, Dec 9, 2004 (UTC)
Does this page still need to be listed at RFC? Maurreen 05:52, 6 Jan 2005 (UTC)
- Well, I'm still receptive to comments... Johnleemk | Talk 16:15, 9 Jan 2005 (UTC)
- OK, thanks. I've been checking on a lot of old RFCs, and you're one of the few people who have answered. Maurreen 17:38, 9 Jan 2005 (UTC)