Wikipedia talk:Deletion policy/Archive 20

Page contents not supported in other languages.
From Wikipedia, the free encyclopedia
Archive 15 ← Archive 18 Archive 19 Archive 20 Archive 21 Archive 22 → Archive 25

Alternative No 3.

The problem that I, and others appear to have with default to delete is that it's being used on high-visibility bios like the one that started this mess - bios that would be on more than enough watchlists, and not on unwatched shitty bios (where the default is obviously keep - by practice). I would support default to delete on biographies that the closing admin expects would be mostly unwatched. How's that? Hipocrite (talk) 15:04, 29 October 2009 (UTC)

  • Hmmm...I like the way you're thinking, but it runs into the problem that if it's difficult to determine notability, it might be difficult to foresee how many watchers the article might attract. At the very least, I would suggest that out of all of this a WikiProject:Unloved articles crop up, along with a talk page category of Category:Articles in need of watching for those that don't have a certain threshold (5? 10?) of watchers. The beauty of Wikipedia is that there are so many kinds of people here, I could very well see a segment of our community wanting to step in to protect our most unloved biographies. -->David Shankbone 15:13, 29 October 2009 (UTC)
  • Oppose - We cannot ask admins to read the future, and this is as arbitrary as everything. I would endorse, instead, the admin asking (and ensuring if possible) that keep !votes watch the page. --Cyclopiatalk 15:15, 29 October 2009 (UTC)
  • I hear you Cyclopia, predicting the future is hard. But... we ask admins to exercise judgment in all sorts of areas. This would be another (new) area where they'd have to exercise good judgment and if they did not, DRV would be there to say "no, you're wrong this is a highly watched page"... and we could overturn just that part of the close without overturning the rest. I'm not sure how you efficiently make that determination (of how well watched something is) if the tool we have can't tell the difference between 0 watchers and 29 (29 is pretty comfy in my view, while 0?? not so much). Or how you predict the future. But we ask admins to exercise judgment and this is one reason why we pay them so much. support, with the comment that I'm willing to try just about anything to address this problem, so keep those ideas coming everyone. ++Lar: t/c 15:23, 29 October 2009 (UTC)
As an aside, I'll have to dig up the diff, but the 30 watcher limit was added to avoid use of the tool to facillitate vandalism. IIRC the tool originally did tell exactly how many watcher a given page has. Xymmax So let it be written So let it be done 15:51, 29 October 2009 (UTC)
Yep. (I think MZM's subpage for the tool has some discussion of this) I am not saying I think limiting is a bad thing. Just that it makes it hard to predict watched-ness or track. ++Lar: t/c 16:01, 29 October 2009 (UTC)
Here's a relevant link - User talk:MZMcBride/watcher. It may not be necessary or desirable for everyone to have access to a list of unwatched articles or be able to check if an article is unwatched. But we should have a wikiproject like David suggests above to systematically apportion lists of some requested number of unwatched BLPs and other articles to trusted editors.John Z (talk) 18:46, 29 October 2009 (UTC)
  • I am afraid that "expects would be mostly unwatched" would be little more than a wild guess. For some reason I think even well-known scientists - nobel laureates! - are "mostly unwatched" [1] while even obscure porn stars with an internet following are "mostly watched" [2]. Still, I think the trouble tends to lie in the second category of article. The semi-protection proposal one section over us is a better option than this, even though I am leery of limiting the free editing which Wikipedia is founded upon. Given that we have an alternative which does not wind up removing articles without consensus, I oppose this one. Sjakkalle (Check!) 15:38, 29 October 2009 (UTC)
  • oppose Absolutely no way to tell. Would have all the problems of the original proposal and be even more arbitrary. Semi protection is a better idea. We need to stop having people trying to use BLP as an excuse to delete good content. If you can't convince the community to delete, stop trying to run around the consensus. JoshuaZ (talk) 16:14, 29 October 2009 (UTC)
  • Oppose: An admin's opinion as to what BLPs are likely to be watched would be, at best, a wild guess. At worst, it would be the closing admin's proxy for WP:ILIKEIT versus WP:IDONTLIKEIT; that is not how the community wants AfDs to be closed. —Finell (Talk) 19:54, 29 October 2009 (UTC)
  • Oppose This is not a compromise, this is worse than a simple "default to delete". This alternative gives all the deciding power into the hands of particular admins (who, we have witnessed as of recently, are sometimes a little less than perfect).Turqoise127 (talk) 20:56, 29 October 2009 (UTC)
  • Oppose: this is problematic because it also increases the role of admin discretion and because it would mean deciding inclusion on the basis of the Wikipedia community's behavior rather than the subject itself. Everyking (talk) 21:04, 29 October 2009 (UTC)
  • Oppose. This still allows admins to decide to delete an article when there is no consensus to do so, which will be likely used by Lar and others to delete as many bios as possible. Fences&Windows 23:10, 29 October 2009 (UTC)
  • Oppose, the opposition is to giving admins a "default to delete" option in the absence of subject request, period. Regardless of where exactly any given admin may wish to apply a "default to delete" option, deletion is an administrative tool that requires consensus for its use, not simply the absence of consensus against it. If someone is concerned that the bio may be unwatched and feels it needs to be deleted for that reason, they are welcome to bring up that concern during the discussion. They are not allowed to unilaterally decide they're right, when (given a no consensus situation) there is not agreement that they really are. Seraphimblade Talk to me 23:31, 29 October 2009 (UTC)
  • Oppose - as above, worse than the other proposed ideas. Also, kind of logically unsound -- by the time a BLP gets to AfD, it's presumably already had a few eyes on it. Deleting those does nothing to solve the problem of unwatched BLP's that haven't even been AfD'd. So we're solving the wrong problem with the wrong tool. --Bfigura (talk) 02:32, 30 October 2009 (UTC)
  • Oppose - just another version of supervotes for the admins. A note to policy proposers: Content is the domain of the editors. Admins exist for the purpose of contending with editor misbehavior. A good fix for a perceived policy problem empowers editors to fix the articles. patsw (talk) 23:18, 31 October 2009 (UTC)

Alternative to policy change

It seems obvious to me that Lar's proposal is not going to result in consensus. Those supporting it should admit that simply changing written policy is not supported by a huge percentage of the community. But those in opposition need to admit that there is a huge percentage of the community that finds the current deletion policy for BLPs unacceptable, and in a consensus driven project that fact cannot simply be wished away. We are at an unacceptable impasse.

Above I asked about the possibility of a "trial" for the proposed change, and an editor opposing it (Jclemens) offered this reply. Based on that I'd like to propose an alternative to a policy change that would involve us actually doing something rather than simply arguing over a few words. I'm not asking for people to support or oppose, but rather to just discuss. Here (in rough form, and I'm not at all married to this idea) is what I'm proposing:

1) A policy page or WikiProject page is created called WP:Marginally Notable BLPs (or whatever, it doesn't really matter, so long as it's widely announced).
2) A number of administrators (say at least 10, and we probably would not want too many) agree to ignore the current deletion policy at times and, when they feel it is justified by the discussion, close certain no consensus AfDs of marginally notable BLPs as default to delete, and would list themselves at the above described project page as willing to do so. To my mind it would be critical to have admins from different points on the BLP-policy spectrum participate in this. The admins who do participate would agree to the following:
a) Only AfDs that truly ended in "no consensus" would be eligible for deletion (no putting the thumb on the scale when the consensus seems to be closer to keep).
b) It would have to be obvious from the AfD debate (based on significant debate about the subject's notability, or an admission by "keep" !voters that notability was a question) that we are dealing with a case of marginal notability.
c) Specific (i.e. not general) arguments about BLP would need to be present in the AfD. If no BLP concerns are raised, then a "no consensus, default to delete" close is not possible.
d) Admins would log these type of closes at the page described at point number 1 and would implicitly welcome a DRV (see below).
3) Any time an administrator chose to close a no consensus AfD of a BLP as delete, they would log that action at the page mentioned above. All of these closures would be open to DRV, which would proceed as it normally does, and where editors could simply argue that the close was out of process and should be overturned based on existing policy, or could argue that the specific close was a good one even if out of policy, etc. etc. Administrators who are participating in the "ignore policy, sometimes delete no consensus AfDs" project would agree not to close any DRVs of this nature. As to logging, administrators should also log occasions where they felt they could have closed an AfD as default to delete but chose not to, and explain why they did not in the AfD closing statement.
4) This project would run for a set amount of time (perhaps 1 or 2 months), after which the admins in question would cease closing AfDs against policy. At that point the community would evaluate the process and consider certain questions. How often were these kind of closes made relative to the total number of BLP AfDs? What was the overall community reaction to these out-of-process closes in DRVs, were they largely upheld, overturned, or was it mixed? How helpful did this actually seem to be in terms of addressing part of the BLP problem? Is there evidence of abuse by admins or other editors, e.g. trying to delete (or deleting) articles under the guise of this process but really as a way to forward another agenda?
5) After a period of discussion we would again call the question as to whether a change in deletion policy is a good idea or not, but this time we'd have an actual experience of it on which to base our opinions.

There's things here to like and not like for both sides of this debate. For those opposing a policy change, a process like this is limited in time and remains under community control to a significant degree since technically every AfD close could be reverted in a DRV, or reverted after the fact en masse if the community ends up rejecting the policy change. However it is a change from the status quo and if it gained momentum and consensus it could result in a significant change in policy which many oppose on principle. For those supporting a change to how we deal with marginal BLPs, this kind of process would let us act on a change many support but which does not have consensus, and it could sway opinion significantly if successful, while also helping to deal with the BLP problem (even if only in a small way). However the restrictions for deleting are rather tight (this is not at all an automatic "default to delete" scenario as some have proposed), and consensus for a change in policy is still required in the end.

While recognizing that probably everyone will have a problem with some aspect of this, how would folks feel about trying something along these lines? Technically a group of administrators could run off and start this right now, since they don't need anyone's permission to ignore all rules. That might be an interesting step, but it would certainly open those admins up to a possible loss of sysop powers, and more importantly it would be far better (and this effort would have a far better chance of proving useful in the end) if there was some rough consensus to this experiment from folks on both sides of this debate before it began.

As I said I think it would be more fruitful to simply discuss this proposal (and/or alternatives to it) here rather than structure it as a support or oppose !vote. --Bigtimepeace | talk | contribs 18:50, 28 October 2009 (UTC)

Not every compromise is good just because there is a compromise. This is a bad idea. The bottom line is that a handful of these editors have already tried and gone to close AfDs using their out of policy method and as SV documented above and elsewhere the result has been incredibly deletionist with zero evidence that we are actually getting rid of any articles that will actually have substantial BLP problems. This is a bad idea full stop. We don't need this sort of compromise. Maybe it would make sense for something if a majority of users favored the defaulting to deletion, but we don't even have that. We have a vocal minority. The majority of users don't want this. JoshuaZ (talk) 18:55, 28 October 2009 (UTC)
  • I'm not seeing in your proposal what problem this seeks to address, but if it is the same one that Lar, Scott and others mention above then it's a current problem for which an approved solution, WP:FLAGGEDREVS, is due to take effect. FlaggedRevs is current on the German Wikipedia, and I understand it has worked well. In other words, the problem has an accepted solution that just has not come to pass (and...that...has...been...dragging...on...forever...) I'd prefer to wait to see how that solution works (it may not) before we start parsing down content and giving admins any more authority over what stays and what goes. -->David Shankbone 19:02, 28 October 2009 (UTC)
Flagged revs has only been approved on a trial basis (people seem to forget that), and we literally have no idea how or even if it will be implemented permanently. Even if implemented on all BLPs (which, there's a good chance, would not even receive consensus), it likely would not prevent sneaky defamation. The argument is that there may be certain articles (even with flagged revs, and certainly while we don't have it) that we cannot adequately protect and which are marginal enough to warrant deletion. Flagged revs simply is not a solution to that issue.
To JoshuaZ, I take your point that "a handful of these editors have already tried and gone to close AfDs using their out of policy method and as SV documented above and elsewhere the result has been incredibly deletionist with zero evidence that we are actually getting rid of any articles that will actually have substantial BLP problems," but what I'm proposing would, if you are correct, demonstrate that even more concretely. Furthermore, if a majority of editors don't want this change then you have nothing to fear from this proposal since there is still a discussion at the end which requires consensus (not to mention DRVs along the way, which would give us a lot of useful info about how the community feels about marginal BLPs), and in the meantime we will have presumably better demonstrated your point that this is not helpful for BLP and would lead to abuses (I'm open to the possibility that one or both of those are true). If you're sick of discussing this ad infinitum (who isn't?), some sort of trial that results in a more wholesale rejection of this approach is a good way to put it to bed. The "vocal minority" (which is pretty large, you must admit) is not going away anytime soon, and I do not think ignoring their views and shunting them to the side is wise. We could have better dealt with BLP a long time ago if people on both sides of the issue we're a little more open to possible compromises and/or testing out new approaches, but instead we have warring camps shouting at each from firmly entrenched positions, often not even listening to what the other side is saying. --Bigtimepeace | talk | contribs 19:42, 28 October 2009 (UTC)
The problem is that anything that takes power out of the hands of the community and invests it with a smaller group of people is going to cause problems, disenfranchisement and power struggles. It may cause people to leave. FlaggedRevs hasn’t been given a chance, and the German Wikipedia hasn’t needed to implement something akin to this somewhat complex proposal (it would be great to have a German Wikipedian jump in here).

Wikipedia essentially is “society”, or else I have no other way to explain why people who are banned from it or don't get their way then spend years staring at and talking about us, never moving on with their lives. The power structure being focused on the community, and not Admins, at least allows for more voices and less disenfranchisement. Anything that makes Wikipedia and decisions about its content less community-based ought to have extraordinary reasons to be implemented. With the pending arrival—fingers crossed—of FlaggedRevs, I feel that seeking to address the problem FRV seeks to address runs the risk of and disenfranchising the community, including new members with new articles. A lot of people don’t want to contribute to a website where a small power elite continue to remove their voice from important decisions. FlaggedRevs negates any extraordinary need at this moment and trying to double-correct may cause harm. However, pushing for that FRV December deadline to be firm might actually help. -->David Shankbone 20:10, 28 October 2009 (UTC)

The proposal above would have the community participate in overseeing these kind of deletions throughout (in the form of DRVs), and then at the end of a set period the community could say, "this was a disaster, shut this down and restore all articles." If there was strong opposition then as there is now then that's exactly what would and should happen. The entire process would be completely transparent, and I'm asking here for feedback on it before anything even happens. I'm sorry but I don't remotely see where I'm suggesting that we take "power out of the hands of the community" or how this is an example of "a small power elite continue to remove [the community] voice". I'm suggesting we try something for awhile, and if it sucks and people don't like it then we have a strong basis to not discuss it again anytime soon and we simply undo what was done. If people do like it then great. What's the big problem, and where's the power grab? I'm pretty sensitive to those kind of concerns, but I genuinely don't understand how you can read them into what I'm proposing. --Bigtimepeace | talk | contribs 20:36, 28 October 2009 (UTC)
I like this idea. It would allow people to watch the effect of the proposed policy change, which at the moment it's hard to do, because this particular type of BLP deletion isn't being flagged, and hunting down examples isn't easy. I'd suggest a longer trial period, more like six months, but in principle I'd support this. SlimVirgin talk|contribs 20:42, 28 October 2009 (UTC)
I saw a great article earlier today on inertia in decision making, and it applies here: your position is that poor decisions will be reversed at wp:drv, but I'm not sure that's accurate: as a current deletion review that need not be named shows, deletion review is nowhere near a perfect process itself, and the inertia shifts even further against retention there than it would otherwise be in a deletion debate itself. Take, for example, a questioned close: you immediately have the community fractured between those that support the result and those that don't, and get a big portion of the debate becoming a pseudo rerun of the original deletion debate. The same issue at the original deletion debate, no consensus, becomes a very likely outcome at the deletion review, as well. We have policies in place to protect biographies of living people, we need to enforce those instead of trying to come up with new ones. There's technology coming that purports to help us with this, we should see if that ends up being the case. In the meantime, I don't see any consensus for shifting away from knowledge retention towards deletion, whether we call it a policy or a test. user:J aka justen (talk) 20:46, 28 October 2009 (UTC)
That is a bad example of the DRV process overall, which in general is a lot better than the current clusterfuck. Kevin (talk) 00:00, 29 October 2009 (UTC)

This proposal is simply giving SUPERVOTES to the admins. If you want to cement the idea that admins have the right to put thumbs on the scale whenever there is no consensus, then this is it. patsw (talk) 22:15, 28 October 2009 (UTC)

  • Bigtimepeace, I think it might help for the sake of review if this was more in the form of a policy proposal page. Also, I wonder about the wisdom of trying to have the community see any well-meaning proposal independent of this strange moment. Timing might be better if you wait a week? Not sure, but your proposal might be somewhat tainted for prospective readers by Lar's out of policy actions and the proposal above to justify it. That's why I raise the idea of a proposal page and different timing. -->David Shankbone 22:39, 28 October 2009 (UTC)
  • You might well be right about that. I thought it best to throw this out for some initial discussion here while there are a number of people looking at this page. Given the virulent negative reaction expressed already, I'm not going to worry about putting this forward more formally anytime soon. Of course any group of administrators could simply run off and start the kind of thing I'm proposing without seeking permission to do so, though that's far from ideal. However it becomes increasingly difficult to understand how (short of the much ballyhooed flagged revs, the exact nature and ultimate effect on BLP of which are complete unknowns) we will do anything about the BLP issue, given that a decent percentage of editors think there is no problem and another sizable percentage agrees there is a problem but is seemingly unwilling to make any changes to do anything about it. Hopefully we can salvage something useful out of this discussion (I'll offer support for a different proposal below on my next edit), but as someone who occupies neither extremity in the great BLP debate I continue to find these little BLP chit-chats we have extremely disheartening and more than a little embarrassing. It's a critical issue, and the extent to which we do not have our shit together is shameful and quite frankly the fault of all of us collectively. --Bigtimepeace | talk | contribs 01:12, 29 October 2009 (UTC)
  • Absolutely not. Per patsw. You write, BTP, "when they feel it is justified by the discussion", but the whole point is that it is not justified by the discussion. There was no consensus within the discussion. This is to overturn one of fundamental tenets. It puts content control into the hands of administrators. MoreThings (talk) 23:46, 28 October 2009 (UTC)
  • Strong oppose - Unbelievably baroque: it's so complex and prone to every different interpretation that looks like a drama bomb. It creates a super-class of similar-minded editors which would rush to delete as much apparently problematic BLPs as possible using this test period as a temporary green light. It would create distrust by lending credence to often-cited conspiracy theories. Finally, "marginally notable" is a non-concept. I appreciate the intention, but this doesn't look even marginally close to a viable solution. --Cyclopiatalk 00:00, 29 October 2009 (UTC)
  • Oppose, I really don't see any consensus at all for no consensus discussions defaulting to delete in the absence of a specific subject request and clearly marginal notability. I think the proposal below is much more likely to achieve consensus than outright deletion at all in such cases, and be far less divisive if it is implemented and used. Seraphimblade Talk to me 02:37, 29 October 2009 (UTC)
  • Oppose. Admins on this project already have far too much individual power; to suggest giving them even more is preposterous. Everyking (talk) 03:44, 29 October 2009 (UTC)
  • Worth talking about further - that's what I think BTPeace was asking, is this worth talking about further? Yes, it is. ++Lar: t/c 05:30, 29 October 2009 (UTC)
    • How do you figure? There was no consensus to allow default to delete closes without subject request above. How is discussing an implementation plan worthwhile for something that's not, from the looks of it, going to be implemented at all? Seraphimblade Talk to me 13:09, 29 October 2009 (UTC)
      • Well, I could be reading the proposal wrong but here's how I read it. 1) NCDTD is not policy under this proposal and doesn't become policy under it. 2) Some of the arguments advanced against NCDTD make assertions about what might or might not happen if it were, which are untested 3) Experiments often clarify things, so run an experiment 3a) Identify a few admins that will close things that way (as defacto has been happening now) from time to time. 3b) Anything they close is subject to DRV. Since policy hasn't changed, the argument at DRV needs to be sound enough to overcome the fact that policy hasn't changed, and that the article needs to be deleted ("NCDTD is policy so endorse" doesn't cut it!) Or the article gets kept. 3c) Run this a while, then evaluate... did bad things happen because we have some deleteds? Did DRV pretty much every time overturn the deletes? Or did they mostly sustain? The answers to these will tell us more about what policy should be, but as of the end of the expermint, it hasn't changed. ... that's my understanding of the proposal. I could be wrong. But I think it's worth discussing further. ++Lar: t/c 17:36, 30 October 2009 (UTC)
        • Lar's summary of the proposal is completely accurate, with the notable benefit of being far shorter than what I wrote. --Bigtimepeace | talk | contribs 21:27, 30 October 2009 (UTC)
  • My feeling is that admins should not be given supervotes. BLP issues have to do with how the article is written, what we do to defend such articles against misinformation, undue weight, speculation and outright vandalism. There is already a "special enforcement" clause in place to deal with problematic situations where you need to cut through the red tape, and a frequently used G10 criterion in WP:CSD is an effective no-nonsense defense against attack articles. We should not make rules based on the viewpoint that the mere presence of BLPs are inherently Bad Things. Most deleted BLPs are removed due to lack of notability, not a violation of BLP policy. Decisions regarding the notability of a living person should rest with the community to ensure the widest possible range of views, and not be restricted to a limited supercommittee of admins who are "better" than all the others. The role of admins should be one of stewards, not politicians. Hence, I oppose this proposal. Sjakkalle (Check!) 08:54, 29 October 2009 (UTC)
  • Oppose (in case my response above didn't make it clear). The gist of this has been opposed by a majority of the community above, this is a slightly watered down trial version of it being called a "test." The objection is still there, and the reasons for opposing the principles behind this proposal are the same. user:J aka justen (talk) 13:14, 29 October 2009 (UTC)
  • Oppose Frankly I'm seeing admin discretion badly overreaching at the moment and there is no reason to increase that problem. Plus A) it's overly complex and B) it isn't needed and C) we don't need a cabal. Liberal use of semi-protection seems like a good idea though. Hobit (talk) 15:01, 29 October 2009 (UTC)
  • Oppose Seems to have the benefit of being overly complex while also not being any more likely to generate consensus. Agree with Hobit on the infinite semi-protection though. --Bfigura (talk) 02:34, 30 October 2009 (UTC)
  • Oppose on the grounds of not bureaucracy. If people think an article should not be included, all they need to do is show up at the AfD and say why. If anyone think there is an unrecognized BLP problem, there is the BLP noticeboard. DGG ( talk ) 00:16, 4 November 2009 (UTC)
  • Oppose too often BLP exceptions are used when BLP is not the motive Power.corrupts (talk) 09:52, 13 November 2009 (UTC)

Alternative to policy change no.2 : Default to keep+semiprotect

The aim of Lar and supporters of the change to the deletion policy is concern for the subjects of BLPs. However, Wikipedia has the methods to deal with such concerns without arriving at the point of deletion. The problem with BLPs is potential vandalism or conscious defamation effort. This can be countered in a much less drastic and disruptive way than deletion: semiprotection.

I propose that whenever a BLP is closed as no consensus, it is default to keep and semiprotect. This should help reduce drastically the amount of potential anonymous vandalism on the page, and help having vigilance on the page. It looks simple, reasonable and addresses at least some of the concerns of the "default to delete" people. They could perhaps argue that "it is not enough", but better than nothing, and it appears to my humble judgement a much reasonable compromise. --Cyclopiatalk 00:10, 29 October 2009 (UTC)

  • Support I think that is a good compromise. Automatic Semi for No Consensus is simple, and alleviates much of the cause for concern of the significant minority above. Along with Flagged Revs in December, that should give these sensitive articles the necessary protection. If a subject increases in stature, someone can flag ANI. -->David Shankbone 00:41, 29 October 2009 (UTC)
  • I assume this includes the same language about deleting if the subject requests? -->David Shankbone 01:16, 29 October 2009 (UTC)
  • Correct. I very personally have some reserves with that policy section, but I am not going to discuss that at all. It would be simply a matter of saying something on the lines of "AfD on living people which end with no clear consensus should default to keep and the closing admin should, as a matter of precaution, semiprotect the page" - no more no less. --Cyclopiatalk 01:22, 29 October 2009 (UTC)
  • Given how long it is taking to get flagged revisions this doesn't seem like an inherently bad idea. I think that in general, people misunderstand the general impact that semiprotection has. But this doesn't seem so bad if it tides us over until we get flagged revisions. JoshuaZ (talk) 01:09, 29 October 2009 (UTC)
  • Why not, and as JoshuaZ says it would be more of a stopgap until some form of flagged revisions is implemented (if it is—the community has only okayed a trial for now, and we don't know what form it will ultimately take). Arguably no policy change is actually needed here, admins who close BLP AfDs as "no consensus" could presumably start doing this now (probably some have already) per protection policy, ["Administrators may apply indefinite semi-protection to pages which are subject to heavy and persistent vandalism or violations of content policy (such as biographies of living persons, neutral point of view)"], so long as there is evidence of persistent vandalism or violations of BLP. --Bigtimepeace | talk | contribs 01:26, 29 October 2009 (UTC)
What I mean is, semiprotect them even if there is no current evidence of vandalism. That's the concern of the significant minority above who is proposing default to delete. I personally think that pre-emptive semiprotection is a much more constructive way of dealing with these BLP concerns, and even who disagrees on "much more constructive" I think can agree that at least it's something. --Cyclopiatalk 01:32, 29 October 2009 (UTC)
I understand, I'm just saying the main issues (articles that have already been vandalized or had harmful BLP stuff put in) can actually be dealt with now. Also I would suggest changing the language to "default to keep and the closing admin can, as a matter of precaution, semiprotect the page." There are any number of articles where anon editors are making good contributions, and automatically semi-protecting a no consensus AfD of a BLP could cut off article improvement in those situations (this is why flagged revs is better, obviously). So changing to "can" or "as a rule, albeit with some exceptions," is probably better since it allows for some flexibility. --Bigtimepeace | talk | contribs 02:19, 29 October 2009 (UTC)
  • Support, I could go for this as a compromise. Hopefully would alleviate the concerns of drive by vandalism while not resulting in unnecessary, divisive deletions. I like it. Seraphimblade Talk to me 02:25, 29 October 2009 (UTC)
    • To clarify, my support is conditional on this being coupled with a default to keep in no consensus scenarios. What's the purpose to semiprotecting an article if you're going to delete it anyway? It's clear that defaulting to deletion has no consensus from the above discussion, I think this could be an alternative that could help to assuage concerns of those arguing to delete. Seraphimblade Talk to me 13:04, 29 October 2009 (UTC)
  • Support. This is a good idea that gets to the heart of the problem (blatantly malicious or deliberately false editing, which is almost always done by anons or brand new accounts) without undermining Wikipedia's mission of building an encyclopedia. In fact, I think semi-protecting all BLPs is a good idea. Everyking (talk) 03:51, 29 October 2009 (UTC)
  • premature sighted edits will do this better--assuming they work. In any case, not just the closing admin but any admin, can semi-protect when necessary. DGG ( talk ) 04:06, 29 October 2009 (UTC)
The point is that sighted edits still do not exist, and a significant minority of editors is very concerned about that. I think semiprotecting these articles, even in the absence of vandalism/libel evidence, cannot harm, and for sure it helps the people concerned about disputed BLPs to sleep better. We're trying to find a middle ground. --Cyclopiatalk 12:42, 29 October 2009 (UTC)
While I originally opposed semi-protection of BLPs in favor of flagged revs/protection, the implementation has been delayed sufficiently that I can't say any longer that this proposal is premature. I actually think it is fairly surgical in its application in that, as I understand, it applies to BLPs all ready determined to be problematic by going through AFD. Xymmax So let it be written So let it be done 15:59, 29 October 2009 (UTC)
  • Support default to keep, oppose automatic semi-protect: Default to keep is already the policy, and there is no consensus to change it. Therefore, the only proposal to consider is that when an AfD for a "marginal" BLP is closed without consensus, the article should automatically (or presumptively?) be semi-protected. Semi-protecting a large category of articles where there is no evidence of vandalism or defamation (which, in this context, means criticism of the living person without a very reliable source) is a huge change of policy, and is inconsistent with the principle that individuals can edit the encyclopedia without registering. Further, I have not seen evidence (mere worries are not evidence) that vandalism or defamation is a major problem for articles about marginal BLPs; if anything, the problem should be greatest with the most notable (i.e., famous) living person biographies, not the least notable. Wikipedia deals adequately with vandalism and potential defamation with existing policies. We semi-protect articles when anon and new account activity warrants it. Policy is already to be especially sensitive with BLPs. Further, with BLPs there is the additional safety net of office actions, which are quick and aggressive, whenever there is a seemingly legitimate complaint. There is no reason to compromise, because there is no consensus to change the default-to-keep policy. Policy does not not support this proposed compromise. Further, if a semi-protection policy makes sense (and I do not see that it does), why should an AfD be required to trigger it? We should be very cautious about tinkering with fundamental policies, which have served Wikipedia well, in the absence of a compelling need to do so. Changing this policy while the community is strongly divided over one especially controversial BLP AfD is especially troublesome. —Finell (Talk) 05:20, 29 October 2009 (UTC)
  • Support automatic semi-protect, oppose default to keep - Unbundle the semi from the outcome, please. If admin discretion is allowed about whether to keep, or if we get to where consensus is rock solid that we never default to delete, we will in either case have post AfD marginal notability BLPs remaining around. (some of them in the first case, all of them in the second case). Automatic semi is a great idea for those. ++Lar: t/c 05:25, 29 October 2009 (UTC)
  • Support semi-protect in the case of no-consensus AfD's that have been closed as keep - default to keep is not a required part of this proposal as we would not be required to semi-protect non-existent articles (salting aside). Administrator judgement as to the strength of !vote arguments is still important. Pushed further for opinion I would say I am also inclined to agree with Everyking that all BLP's should be semi-protected.--VirtualSteve need admin support? 08:11, 29 October 2009 (UTC)
  • Support automatic semi-protect, oppose default to keep - it's better than nothing. Personally, I'm more in favour of indef semi-prot and indef sysop move prot on all BLPs, but how and ever. Per Lar, I'd like to see this implemented and regarding Flagged Revisions, I've given up holding my breath on that one, too. Whether flagged revs is imminent or not should not be a determining factor in this outcome, IMO, as it's very much an unknown quantity at this time - Alison ❤ 08:31, 29 October 2009 (UTC)
  • Surprisingly, I also support the semi-protection bit, and oppose the default to keep. As the default to keep bit is being debated here, should we move this section to the protection policy talk page? Kevin (talk) 09:01, 29 October 2009 (UTC)
    • Yes, Move to WT:PROT. For continuity it should probably stay here, but for clarity it would be better rebooted there, to effectively unbundle from the "keep/delete in no consensus situations" issue. Rd232 talk 09:24, 29 October 2009 (UTC)
  • Comment: I understand that ideas, once proposed, live their own life. I would like however to clarify: What I am trying here to do is to get a compromise. On one side there is people that want a default to delete because they are concerned with maintainability and possibility of harm made by the BLPs. On the other there is people who think that default should be keep, because no consensus implies "if you don't know what to do, do nothing", deletion is a drastic measure, it doesn't help BLPs, etc.
For these reasons, I think that defaulting to keep AND semiprotecting should be the compromise. The "and" is important. On one hand, we make people who prefer non consensual articles to be kept -and these people seem to be the majority, see above- happy. On the other, we make a constructive contribution towards maintainance and vandalism protection of these articles, which is basically what the other side wants. I think that if we want a middle ground, this should be, indeed, middle.
Of course everyone is free to endorse to split them, but this seems to me contrary to the concept of compromise Just my 0.02£. --Cyclopiatalk 12:38, 29 October 2009 (UTC)
I see your point. I just would rather see this decoupled. If you want I'll give you a weak support for one way and strong the other, I guess. I think the discussion a few sections up is going to end up as "no consensus for NCDTD closes at this time". So this doesn't need to be coupled, necessarily. But yes, I see your point, and full marks for trying to find a compromise here in any case. ++Lar: t/c 15:19, 29 October 2009 (UTC)
  • This is the best option presented so far, since it does not wind up deleting stuff without a consensus to do so. I am not sure if it is truly necessary, and fear that it might stall development of a stub article which faces no BLP issues if a new user is unable to contribute to it. For articles where there is known to be trouble, I can support a widening of the use of semi-protection however, and "no consensus on AFD" BLPs are certainly among the articles where this is worth considering. Sjakkalle (Check!) 15:43, 29 October 2009 (UTC)
Thanks for your support, Sjakkalle. I understand very much the concern about new users/IP users, but I guess they can be told to work their contribs on their user page or on the talk page(s), and ask for a more experienced editor to include them. It's not optimal, I understand, but at least we address a concern of a significant part of the community.--Cyclopiatalk 16:02, 29 October 2009 (UTC)
  • Support Default to keep+semiprotect Whole heartedly. With an obvious division of opinion on this matter, this alternative "gives a little, takes more". It addresses concerns of BLP subjects being harmed and yet keeps the policy/process as it has been all along; default to keep. Honestly, regardless of the fact that this alternative "takes more" as I mentioned, the opposition should be gracious and agree to this resolution, because the actions of some of the members of the "default to delete" ideology have been a bit less than tactful with existing policy changes that are totally against WP:BLP (third line down: Changes made to it should reflect consensus., and this discussion alone shows beautifuly the lack of consensus). Certain editors practically staged a Wikipedia coupe and made decisions in order to create precedent for policy changes as they suit their ideology. This is NEVER the way things ought to be done. Frankly, I am surprised to see this many (what is it, about 40%?) of editors actually wanting the "default to delete" so badly that they are willing to look the other way from a BLATANT disregard for the community and for the way things work around here.

So I ask you kindly, editors who wish BLP to default to delete, Deal, or No Deal? Turqoise127 (talk) 20:53, 29 October 2009 (UTC)

  • Strong support – I think this is a good idea and compromise, at least until Flagged Revisions gets turned on, which IMO when turned on at that time this may not even be needed. MuZemike 22:48, 29 October 2009 (UTC)
  • Question – I'm assuming that this would be indef, the protection? MuZemike 22:50, 29 October 2009 (UTC)
I personally would say "yes", but maybe someone knows better than me what is best. --Cyclopiatalk 22:43, 30 October 2009 (UTC)
  • Weak support - would prefer sighted rev's to this (purely in terms of having the more elegant solution), but it does seem to be a good compromise. --Bfigura (talk) 02:35, 30 October 2009 (UTC)
  • Support default to keep, oppose automatic semi The huge majority of articles where AFD results in "no consensus" have not been and are not targets for vandalism or unsourced negative BLP. They are simply of borderline inclusion-relevance. The protection policy currently states that protection should not be premature, i.e. preemptive without any prior disruption. Such a change would violate the protection policy and introduce a "semiprotect BLPs" system through the backdoor that is currently not supported by consensus. Semi-protect if needed, sure, but don't make it automatic. Nothing that drastically disables editing for a large number of people should be done automatic but manually when it's needed. Regards SoWhy 12:01, 30 October 2009 (UTC)
You have a good point in saying that this conflicts with the current protection policy. Any opinions on that? --Cyclopiatalk 22:50, 30 October 2009 (UTC)
The current protection policy is based around on of the main foundation issues, i.e. "anyone can edit". This foundation requires that we don't restrict editing for IPs or non-autoconfirmed members without a good reason to do so and the PPOL for that reason does only allow semi-protection is specific circumstances. For the same reason the PPOL forbids preemptive protection of a page that is not yet target of problematic editing. Your proposal would contradict this basic principle by requiring mandatory and automatic semi-protection even when there are no issues.

Admins can already apply protection where problematic editing occurs but they can and should only decide to do it on a case-to-case basis. It's unhelpful to require protection of all such articles regardless whether problems really exist (which they often don't). Your proposal is milder than the "default to delete" one but equally fails to address why this should be done with those articles (the majority of them) which were never plagued by such problems. I'm all in favor of protecting BLPs - but it should only protect those where there is a need for protection and not simply all of a specific set. Regards SoWhy 11:56, 4 November 2009 (UTC)

SoWhy, I understand absolutely your concerns and I am too personally inclined to think that such articles are not proven to have significantly higher rates of vandalism than average BLPs. However there is a significant part of editors and admins which happen to think differently, and what we're trying to do here is to find a compromise within the community. The support this proposal has seems to indicate that there is indeed the need of such a compromise, and semiprotecting such articles is one which anyway would lead to the semiprotection of very few articles at a given time. I understand this violates somehow the principle of no preemptive semiprotection, but we have contrasting issues on our table, and we have to adjust these somehow. I guess that if eventually discussion here shows widespread support, we can bring the issue at WP:PROT and see what to do. --Cyclopiatalk 12:18, 4 November 2009 (UTC)
I am not against using semi-protection in those cases where something needs to be done. It's the automatic part that strikes me as problematic since restricting editing to any article should never be done without an evaluation of the situation. Each article is unique and a "protect them automatically" solution will not serve our primary goal, i.e. building an encyclopedia. Regards SoWhy 14:02, 4 November 2009 (UTC)
  • Support - this seems like a reasonable incremental step to take until flagged revisions are enabled. Christopher Parham (talk) 12:53, 30 October 2009 (UTC)


  • Comment: I must say that at this point I am amazed and flattered by the amount of support this little idea has gained -the "split" opinions (support X, oppose Y) seem even to balance each other quite neatly. At this point I have two questions: 1)Should we discuss a wording of this? 2)How long do such discussion go before consensus is declared? Thanks! --Cyclopiatalk 22:43, 30 October 2009 (UTC)
    • I would go more slowly at this stage. This proposal is in the middle of a huge page that many people might find hard to navigate at this point; there might also be some fatigue that is keeping some from getting involved. There does seem to be support for it -- but perhaps the best path is to let things lie for a couple of weeks, let this page get archived, and then start a fresh discussion. I think it will be good to adopt your proposal, but given how things have gone recently perhaps it's best to go slow, to avoid doing something that would lead someone to say, wait I didn't agree to that. Nomoskedasticity (talk) 08:18, 31 October 2009 (UTC)
  • Absolutely agree. There is also the issue of it being at odds with the protection policy. --Cyclopiatalk 12:29, 31 October 2009 (UTC)
  • Support as a compromise. I'd much rather see FPPR and default-to-delete be implemented, but until it is, keep-and-semiprotect is an acceptable compromise if we can't default to delete. Sceptre (talk) 14:18, 3 November 2009 (UTC)
  • Support - seems to be a reasonable compromise that would retain encyclopedic content but minimize risk of libel or other RL harm. Rlendog (talk) 02:28, 6 November 2009 (UTC)
  • Oppose I regret not having read all the arguments on this page, but this little idea needs a lot more debate. Just because a BLP is marginally notable, that says absolutely nothing about whether it is or is not prone to bias or vandalism. If I remember right, blanket semi-protection or blanket flagged protection, without specific reason (ie previous vandalism), has never been allowed. This is a backdoor way to start mass-protecting articles without just cause. Joshdboz (talk) 22:42, 19 November 2009 (UTC)
You will maybe be surprised in knowing that I proposed this, but I actually agree with your basic argument: namely, that a "marginally notable" BLP is not significantly more at risk than any other article. However there is a split community with good arguments on both sides which is very concerned about that, and the thing is beginning to look like, intermittently, as a low-level power struggle. This is not good. We need a compromise that may work. This one seems to have some support, even if it is by no means perfect. I disagree that this specific proposal is a backdoor for mass protecting, because it would apply only to a very specific class of articles after a community debate -namely, BLPs which close as no consensus in AfD, probably a tiny percentage of the total. --Cyclopiatalk 23:24, 19 November 2009 (UTC)
I'll be honest I think it is an interesting idea, and my thoughts on BLP and protection/flagging are evolving, but the current policy (WP:PP) states "Semi-protection should not be used as a pre-emptive measure against vandalism that has not yet occurred, nor should it be used solely to prevent editing by anonymous and newly registered users." If they ever get around to the flagged protection trial we will have a better idea about how this could/should be applied to the BLP issue, and I would be more inclined to support flagged protection in the situation outlined above. I just don't think this is the right tool to bridge the gap until that is in place. Joshdboz (talk) 00:00, 22 November 2009 (UTC)
  • Absolutely Oppose A compromise that compromises one of the wp:pillars is not acceptable. Page protection is reactive. Not proactive. Gigs (talk) 22:22, 21 November 2009 (UTC)
Actually there's no wording in the pillars agains this proposal. The founding principles say: The ability of anyone to edit (most) articles without registration,. There's nothing about reactivity or proactivity. This depends on the protection policy. --Cyclopiatalk 23:10, 21 November 2009 (UTC)
Retrieved from "https://en.wikipedia.org/w/index.php?title=Wikipedia_talk:Deletion_policy/Archive_20&oldid=1142511116"