Wikipedia:Pending changes/Request for Comment 2012/Discussion

Discussion

Click here to edit this section
I'm wondering if I'm understanding something correctly. If a page is level-1 protected, it appears that IP/new attempts to edit will effectively block non-reviewers from editing, until a reviewer comes along and either approves or disapproves the IP/new attempt to edit. Correct? Allens (talk | contribs) 11:06, 24 March 2012 (UTC)[reply]

I don't think so. At least, I once saw two edits to the same article listed in the queue, and if you couldn't edit the page again until the first was reviewed, then there wouldn't have ever been two edits simultaneously waiting for approval.

Also, even if that were true (and it doesn't seem to be), the typical length until review was only a couple of minutes during the trial last year. WhatamIdoing (talk) 13:03, 25 March 2012 (UTC)[reply]

• I'd just like to make a follow-up comment, after my "position 3" comment earlier. Assuming the community goes forward with this, I hope that the users who are working most closely with it will look very carefully at the points that I and others have drawn attention to in section 3, and present the community with a policy that is updated accordingly. I realize that this may be implicit in the way the RfC has been presented, but I thought it would be useful to point it out. --Tryptofish (talk) 20:02, 25 March 2012 (UTC)[reply]

Arguments against Position #2

In bulleted format, each bullet signed, for ease of threading.

• "During and after the trial, PC was shown to be an extremely helpful tool for combatting bad-faith edits while still allowing easy submission of good-faith edits." I do not buy this argument as it relies on the same basic (psycho)logical fallacy that all PC/FR is predicated on, i.e. that every reviewer is knowledgeable about the subject and not a Knight Templar, not to mention it ignores the fact that not all bad-faith edits are obviously so, and those that are not can very easily slip by an unknowledgeable reviewer (of which there shall be many per topic. Not everyone is omniscient). How does this get addressed? —Jeremy v^_^v ^Bori! 21:36, 23 March 2012 (UTC)[reply]

  No, it does not rely on this argument. It does rely on the argument that every reviewer is knowledgeable enough to recognize vandalism, BLP violations, (good faith) deformatting and similar things, but definitely not necessarily in the subject of the article.--Ymblanter (talk) 00:16, 24 March 2012 (UTC)[reply]

  It does, given the arms race that's been going on between vandals (especially dedicated ones) and administrators/editors. We already have issues with sneaky vandalism that unknowledgeable users cannot revert; it is this situation (as opposed to obvious vandalism) this argument is predicated on. —Jeremy v^_^v ^Bori! 03:33, 24 March 2012 (UTC)[reply]

• "Support PC, especially on BLPs to prevent defamation." Last time I calculated number of reviewers with number of BLPs I got a result of 65 articles to one reviewer, which is unworkable logistically. I would imagine the number has only since increased due to there being less active users and more BLPs. Given that BLPs tend to be edited rather frequently how can one address this fundamental flaw? —Jeremy v^_^v ^Bori! 21:36, 23 March 2012 (UTC)[reply]
  • Who says that it has to be used on every single BLP article? "Especially on BLPs" does not automatically equate to "please use PC on every single BLP". Why can't we use it on 1% or 2% of them? WhatamIdoing (talk) 22:05, 23 March 2012 (UTC)[reply]

  Because those 1 or 2 percent are likely the highest-profile ones where PC can't work due to edit volume, and given that much of the arguments for PC that I have seen assume it will primarily be used on all BLPs, that is why I commented about the "all BLPs" bit.—Jeremy v^_^v ^Bori! 03:33, 24 March 2012 (UTC)[reply]

  I wouldn't put PC on a high-traffic article. In fact, I have specifically advocated that it not be used on any high-traffic article, and I believe that my view is generally shared among those who have thought about it for more than a few seconds (e.g., any admin who actually deals with page protection). So your assumption about what is "likely" seems completely wrong to me. IMO it is far more likely that it would be used on a small number of low-traffic BLPs as a less restrictive alternative to semi-protection, or as a way of demonstrating that previously applied semi-protection was no longer necessary. WhatamIdoing (talk) 13:01, 25 March 2012 (UTC)[reply]

  That particular view runs counter to the bullet below, which advocates using it on controversial articles (which are either high-traffic or under sanctions, rendering PC useless or moot). Contrariwise, putting it on a low-traffic BLP is apt to be a waste since there's not likely to be anywhere near enough activity on it to justify PC, and logistics needs to be kept in mind since there's far less available reviewers compared to PC candidates. —Jeremy v^_^v ^Bori! 19:00, 25 March 2012 (UTC)[reply]

• "[...]seems like a perfect mid-way point between page protection and open editing. It will help IPs Be bold and fix our mistakes, even on controversial pages." It will not because controversial pages are generally universally-vandalized in the first place or otherwise difficult to edit. George W. Bush and Barack Obama both were put on PC during the trial and had to be removed from it because the volume of edits was too much. How can you reconcile this point with the actual reality of the matter? —Jeremy v^_^v ^Bori! 21:36, 23 March 2012 (UTC)[reply]

  Pstanton's argument falls under this bullet as well, for much the same reason. High-profile pages are either too busy to use PC on, too partisan to attract anything but edit-warriors and truthers, or are under discretionary sanctions. —Jeremy v^_^v ^Bori! 03:52, 24 March 2012 (UTC)[reply]

• "As a community, we should welcome with open arms anything that offers a solution to our problems of vandalism without totally shutting out new and unregistered editors." Part of the issue is that new editors already feel shut out because of the perceived air of elitism in the place. Adding another userright that amounts in several peoples' minds to "censor" does not help add editors; it helps drive them away. —Jeremy v^_^v ^Bori! 21:36, 23 March 2012 (UTC)[reply]
• "The minor problems encountered in the trial did not demonstrate the unworkability of the system; instead they demonstrated that it basically did work." By what metrics? —Jeremy v^_^v ^Bori! 21:36, 23 March 2012 (UTC)[reply]

  Have you read the metrics that have been collected? For example, did you know that PC permitted us to benefit from more than 200 good edits from unregistered users per day during the initial trial, all of which happened on articles that had previously been semi-protected? WhatamIdoing (talk) 23:50, 23 March 2012 (UTC)[reply]

  Out of how many edits? I have seen the metrics, and they suggest those 200 edits are less than 50% of all edits caught behind the review filter. Compared to unprotection (the metric that should be used) that isn't a better situation. —Jeremy v^_^v ^Bori! 03:33, 24 March 2012 (UTC)[reply]

  You're comparing apples and oranges. The proposal is to change from semiprotection to PC, not from unprotected to PC. If your idea is to remove the semiprotection from all the articles that it is on, and replace it with nothing, that would be a separate proposal. PC is only used for articles that would otherwise be semiprotected or fully protected, and it gives an improvement over those forms of protection. — Carl (CBM · talk) 12:01, 24 March 2012 (UTC)[reply]

  However, by its very nature PC is intended to allow unregistered/new editors to edit pages that would otherwise be semiprotected. Thus, comparing it to semiprotected is pointless (as semiprotected prevents those editors from editing that page in the first place) and the appropriate comparison is to an unprotected state (the only other way unregistered/new users can edit a page). This was brought up in the 2011 RFC. —Jeremy v^_^v ^Bori! 18:50, 24 March 2012 (UTC)[reply]

  We agree that the point of PC is that the pages would otherwise be semiprotected. I don't follow the logic that, therefore, it is wrong to compare PC with semiprotection. Those are the two options for the pages we are talking about, and PC is only used on pages that could be (and would be) semiprotected anyway, so when deciding about the benefits of PC we should look at what would otherwise happen to the pages in question. — Carl (CBM · talk) 20:27, 24 March 2012 (UTC)[reply]

  If you're comparing edits by anon/new users between PC and semi, you're going to get positive results because of selection bias - semiprotected disallows edits by anon or new users. Thus, for a fair comparison, you have to compare it to unprotection, as PC is a middle ground between that and semiprotection. Thus, the arguments should support that PC is as effective as unprotection - the state closest in equivalence - in allowing users to edit. —Jeremy v^_^v ^Bori! 20:33, 24 March 2012 (UTC)[reply]

  No, that seems like a false dichotomy. It's like saying that methadone would have to have as few side effects as no drugs at all in order for people to use it to replace heroin. In reality, when someone is already addicted, methadone is far better for them even if it is far worse than being clean. Similarly PC is much better than semiprotection and that is all that matters, because we are talking about articles that are semiprotected anyway. — Carl (CBM · talk) 20:42, 24 March 2012 (UTC)[reply]

  I see no false dichotomy in my statement, and your drug metaphor is crap because it assumes a position of complete safety (being clean). My argument here is based on anonymous and newly-registered users being able to edit, as opposed to vandalism (where only full-protection grants complete safety). Thus, semi-protection is a poor metric because it prevents that (and even fails your drug metaphor because it's still relatively ineffective at stopping vandalism). —Jeremy v^_^v ^Bori! 20:54, 24 March 2012 (UTC)[reply]

  I'm getting an IDIDNTHEARTHAT vibe here, so I'm out. — Carl (CBM · talk) 20:59, 24 March 2012 (UTC)[reply]

  I heard everything you said. I'm disputing your metrics because they compare a nonzero value against zero, as opposed to two nonzero values. —Jeremy v^_^v ^Bori! 21:05, 24 March 2012 (UTC)[reply]

  There's a bit of "the pot calling the kettle black".
  — V = IR (Talk • Contribs) 01:19, 25 March 2012 (UTC)[reply]

  • I pretty much don't understand this argument. You want the English Wikipedia to be "the encyclopedia that anyone can edit", right? But you object to a system that allows 200 good edits to be made each day, and 400 bad edits to never see the light of day, because only a third of the edits were good? And then you propose, as the preferable approach, that those 400 bad edits be shown to all readers? How is getting 200 good edits and showing 400 edits of obvious vandalism and libel to readers better than not getting 200 good edits and not showing vandalism and libel to our readers? WhatamIdoing (talk) 13:09, 25 March 2012 (UTC)[reply]

  That argument assumes that vandalism remains on an article for a relatively long time (let's argue fifteen seconds). Except for borderline cases (i.e. sneaky vandalism) and deeply obscure pages, this isn't often the case. Not to mention IPs are as apt to revert vandalism as named users. —Jeremy v^_^v ^Bori! 18:54, 25 March 2012 (UTC)[reply]

• "It also will help against self-promoters, allowing established editors in good standing to contribute and sockpuppetters to be blocked." Unless you can apply PC to redlinks, the first is effectively impossible and the second is already done by semi-protection (as most sockpuppetteers make socks specifically to cheat it, making them all the more obvious). Aside from putting all anon/new user edits through a bozo filter (which {{editrequest}} does equally as well) what does PC do that semiprotection cannot? —Jeremy v^_^v ^Bori! 21:12, 24 March 2012 (UTC)[reply]

Editrequest plainly doesn't do it equally well. If you want to just fix a typo, it's non-intuitive and a hassle, and you are not likely to bother. At least, I wouldn't. FormerIP (talk) 22:25, 24 March 2012 (UTC)[reply]

And PC has much the same flaws. Why try to fix a typo if the change isn't live immediately? That hurts the encyclopedia. —Jeremy v^_^v ^Bori! 06:27, 25 March 2012 (UTC)[reply]

No, with PC you edit in exactly the same way you normally would and it takes you exactly the same amount of time and effort. You don't have to find the talkpage, you don't have work out how to explain the change you want in words, you don't have to get into a conversation about it. The only difference between PC and regular editing is that there's a delay before your change will appear. FormerIP (talk) 13:04, 25 March 2012 (UTC)[reply]

And that's a whopper of a difference. You're also glossing over the fact that for said edit to ever appear, yourself or someone else will have to approve it. That's a fundamental, earth shattering change to the way that Wikipedia operates. Yourself and others can gloss that over, and try to dress it up in language about "200 good edits and 400 bad edits", but the fact is that pending changes fundamentally changes the way that Wikipedia works.
— V = IR (Talk • Contribs) 16:07, 25 March 2012 (UTC)[reply]

It does, but only by allowing editors to edit pages that they would otherwise be unable to edit due to semiprotection. No edit that the editor could otherwise make is blocked by PC. — Carl (CBM · talk) 18:04, 25 March 2012 (UTC)[reply]

I think this discussion is hampered by a difference in understanding of what "to edit" means. If it means "submit one's desired changes using the edit tab/text box/save button mechanism", then of course PC enables more people to "edit" than SP does. But if it means "independently alter the content of the article as it is displayed to the world", then PC sometimes allows fewer people to edit. In any case there's nothing earth-shattering about PC, unless it's to be used on many more articles than SP now is, since the differences are mainly technical.--Victor Yus (talk) 18:20, 25 March 2012 (UTC)[reply]

Because PC is only used on pages that would be semiprotected otherwise, I don't know of situations where PC allows fewer people to edit in the second sense. If there was a proposal to change lots of pages from unprotected to PC, that would be different, but as things stand the only change is from semiprotected to PC. — Carl (CBM · talk) 18:31, 25 March 2012 (UTC)[reply]

That is because you are thinking of the hereandnow, CBM, as opposed to the long-term, where PC will (not could, will) be requested on unprotected articles. —Jeremy v^_^v ^Bori! 19:04, 25 March 2012 (UTC)[reply]

Why PC allows fewer people to edit: If there's an unconfirmed user's edit waiting to be reviewed, then I (as a confirmed editor who is not a reviewer) cannot edit (in the second sense). I'm effectively reduced to being an unconfirmed user in that position.--Victor Yus (talk) 19:23, 25 March 2012 (UTC)[reply]

Thanks, that makes sense. I assume that whoever made up this RFC has the statistics on how long it took for reviews to happen when PC was active before? — Carl (CBM · talk) 19:37, 25 March 2012 (UTC)[reply]

Whatamidoing indicates above it was, on average, a "couple minutes". Anything more accurate you'd have to go log-delving for; I do not have the time to do it at the moment but will do so tomorrow or Tuesday (PDT). —Jeremy v^_^v ^Bori! 20:16, 25 March 2012 (UTC)[reply]

• "Editing BLP's irresponsibly actually makes the subject of the biography less free since the subject would become affected by forces (i.e. the irresponsible revisions) outside of his or her control." In related news, it's not certain whether or not the reviewer would suffer legal responsibility for approving sneaky libellous vandalism. I'm unsure whether or not the Foundation has answered that question, but if reviewers are, then they share some of the liability when something like this inevitably happens. —Jeremy v^_^v ^Bori! 19:08, 25 March 2012 (UTC)[reply]
• "Effective tools to combat persistent vandalism, BLP violations, and other unconstructive edits already exist, and they work well; other options have been identified but not actively considered." Since this point was brought up, what's wrong with the edit filter? Where has Cluebot seriously erred? Why are the other antivandalism tools in the arsenal, except semiprotection, not being brought up? —Jeremy v^_^v ^Bori! 19:12, 25 March 2012 (UTC)[reply]
• "[...] [T]he fact that changes will not be visible until reviewed might quench motivation to vandalize." Contrariwise, the fact that changes will not be visible until reviewed might quench motivation to contribute in good faith. You're throwing the baby out with the bathwater. —Jeremy v^_^v ^Bori! 19:15, 25 March 2012 (UTC)[reply]