Wikipedia talk:Redirects for discussion

Retargeting a WP redirect

If a redirect in the Wikipedia namespace has lots of incoming links (I stopped at 1500), is it possible to retarget it to a more appropriate target? If not, I won't submit that for discussion. ―Mandruss ☎ 01:26, 21 November 2019 (UTC)

I'm assuming this is a shortcut. Is it used as intended? If there's a large number of incoming links where the intended meaning is something other than the current target, then that would be a strong argument for doing something about this redirect at RfD. On the other hand, if it's used correctly and it has already caught on, it's rarely sensible to attempt to retarget it, as that would entail changing the following: incoming links from pages (laborious, though not impossible), habits of editors who've become accustomed to the shortcut (borderline impossible), and incoming links from edit summaries (properly impossible). – Uanfala (talk) 01:50, 21 November 2019 (UTC)

Kind of what I assumed, just wanted to be sure we didn't have some magic solution. I spot-checked a few and they are used as intended. It was WP:ATP which in my strong opinion should have targeted Wikipedia:Talk page guidelines or Wikipedia:Talk page guidelines#How to use article talk pages, consistent with WP:UTP. Ah well. Thanks. ―Mandruss ☎ 01:58, 21 November 2019 (UTC)

Deletion without nomination?

Wikipedia:Books/Mathematics, which was redirected to Book:Mathematics ten years ago, was deleted on 3 November 2019 by JJMC89 following this discussion (opened by Black Falcon). The page was not actually nominated in that discussion, nor apparently in any other deletion discussion, based on what currently links to that title. This is worrisome. Are deletions becoming so routine that they can't even wait for the affected pages to actually be nominated? Note that Portal:Mathematics was still linking to there (through one of its "component" subpages) at the time of deletion, which is how I learned about it. Although the idea of breaking links was considered in the discussion, apparently it wasn't taken seriously enough for anyone to actually check each page (sufficiently carefully, anyway) for that possibility. To all relevant parties involved (including UnitedStatesian, who argued all similar redirects should be deleted without further discussion), I say: please be more careful when deleting pages (even lowly redirects), and do not delete an entire "class" of pages without linking to and checking all of them to see how deletion might affect "reader-facing content". I know that's annoying to do for hundreds of pages, but if something is worth doing, it should be worth doing right. - dcljr (talk) 10:03, 28 November 2019 (UTC)

My original intention was to nominate all ~600 Wikipedia:Books/... redirects in a series of nominations, specifically so that incoming links could be checked in smaller, more manageable batches. While I support the subsequent decision to speedily delete all similar redirects on the basis of efficiency (nominating hundreds of redirects is quite time-consuming), I also agree with you that it is important to check incoming links to each affected page even when deleting hundreds of pages at once. Tedious, to be sure, but still important. -- Black Falcon ^(talk) 20:48, 30 November 2019 (UTC)

That strikes me as highly irregular. Yes, occasionally IAR could be invoked to delete a class of redirects without a deletion discussion, but I'd normally expect the "precedent" to have had a decent quorum, something of the order of an RfC, or a lengthy and well-attended RfD. If I understand correctly, the redirects concerned are all redirects from moves, and that makes it necessary to have a workflow for fixing incoming links before any deletion is carried out. But I think we really need to hear from JJMC89. – Uanfala (talk) 02:21, 10 December 2019 (UTC)

I agree that all of these redirects should have been individually tagged for deletion (doing so via AWB or similar is fine, it doesn't have to be done manually) and would support their undeletion. This should probably be at WP:DRV rather than here though. Thryduulf (talk) 10:58, 10 December 2019 (UTC)

Shorthand redirects to XfD daily log pages

Per Wikipedia:Redirects for discussion/Log/2019 December 8#Wikipedia:RFD/2018 November 18, would it be good practice to routinely create redirects of the forms Wikipedia:RFD/2018 November 18 and/or Wikipedia:RFD/2018Nov18 as pseudo-shortcuts to daily log pages for XfDs? As of this writing, these two are the only ones in existence, and the consensus at the linked RfD is shifting in favor of keep. As several users suggest that there could be value in using these redirects as shorthand in edit summaries or perhaps even as shortcuts (to avoid typing long titles), I'm bringing this here for a broader survey. Should there be consensus in favor of mass creation, a new bot task could be requested for its implementation.

@BrownHairedGirl, Utopes, and Thryduulf: pinging the participants of the linked RfD.

I'm at weak support right now; the idea and its rationale look reasonable, but I'm not sure how necessary it is. ComplexRational (talk) 21:06, 9 December 2019 (UTC)

• Support. As I mentioned at the RFD, this is something which I have been meaning to propose for ages, for the three CXFD venues which use a daily log page: RFD, CFD and TFD. It would allow much easier linking to XFD discussions.

The examples above were both created by me to allow linking to the discussion in edit summaries of a few thousand AWB edits which I made as a result of the deletion of the redirect, e.g. [1]. This was needed because AWB allows only 156 characters in edit summaries, compared with the 500 chars in the default source editor. This is a fairly common situation, and having these redirects in all cases would be very helpful.

If there is consensus to do this, I suggest that there should be a one-off bot task to do it for all the historical cases, and an ongoing bot task to do it for all new daily pages. --BrownHairedGirl (talk) • (contribs) 01:46, 10 December 2019 (UTC)

• Oppose, I don't see the need. I don't mind these being created on a case by case basis, like your bot task, but situations like this don't pop up every single day. -- Tavix ^(talk) 01:55, 10 December 2019 (UTC)
• If an editor has created a few redirects to use in certain situations (like stringently short edit summaries in AWB), then sure, let them do that. However, I don't see a benefit in this type of redirect in the general case. To begin with, RfD doesn't work by precedent and so situations where you'd need to regularly refer to an older RfD are generally rare. And if you want to link to one, what you'd normally do is locate the discussion and then copy-and-paste the link. I don't think you would normally be able to type it off the top of your head (short or not) as that would imply remembering both the exact date of the discussion and the exact section header. Even if these redirects make it a little bit easier to type out a link, I don't think this would be a practice we'd want to be encouraging: typing stuff out of memory has a high potential for error that you don't get with copy and paste. – Uanfala (talk) 02:15, 10 December 2019 (UTC)
• Support the existence of. I said in the RfD that a redirect such as the one presented allows for easy reference to a past RfD result. I'm not suggesting that a shorthand redirect is added for every month, but I can definitely see the worth in having one around, even if it is just for personal use. If there is a significant result that a user wants to link to multiple times, I wouldn't be opposed to them making a shorthand redirect for discussion purposes, as long as they all follow the same naming convention. To that effect, the redirect has a usage and cannot be misinterpreted. To Uanfala's point, I wouldn't suggest encouraging the creation of these redirects, but WP:R#KEEP point 5 says that we shouldn't delete a redirect if somebody finds them useful. And the only reason that these redirects would exist would be "to be useful". Utopes (talk) 04:04, 10 December 2019 (UTC)

DumbBOT down

Looks like DumbBOT is down. (RFD subpages not being created or transcluded.) Tizio, any status on this? Steel1943 (talk) 21:04, 15 December 2019 (UTC)

• ...And looks like the bot is running again. Steel1943 (talk) 21:06, 16 December 2019 (UTC)

Links that have existed for a significant length of time

WP:RFD#KEEP says that we don't delete "Links that have existed for a significant length of time".

My question: Is there an agreed-upon understanding of what "a significant length of time" means?

Concern for breaking internal and external links was added by User:Angela in August 2004. It was updated once or twice soon after (e.g., to include the old subpages – most editors won't remember when AliceExpert was the "correct" title for an article about Alice Expert, and that AliceExpert/talk was what Wikipedia used instead of Talk:Alice Expert, because the idea of namespaces hadn't been invented yet), and it reached its current version in this edit by User:Thryduulf.

Given that the first edits to the English Wikipedia were made about 3.5 years before the creation of the rule, I don't think it's unreasonable to guess that this was meant to discourage deletion (though not changing the target of) of "unnecessary" redirects that something in the range of months to a couple of years old. Do we have anything like a modern understanding of what "a significant length of time" means to us? If we have an estimate, then it'd be a kindness to all concerned if we actually wrote it down, instead of making editors guess whether "a significant length of time" is usually taken to mean approximately a week, a month, a year, or a decade. WhatamIdoing (talk) 20:34, 30 December 2019 (UTC)

There isn't really a consensus. Those people like me who take the attitude that redirects should generally be kept unless it's clear they are harmful or very clearly useless, generally regard being old as a factor tending to indicate a redirect should be kept ("old" here meaning more than, generally, 6-12 months). Others, Tavix for example, generally recommend deleting redirects that don't have clear utility, and for them age is generally not regarded as anywhere near as significant.

There are a host of exceptions however. For example, if a redirect is the result of a page move shortly after creation from a clearly incorrect original title (e.g. a typo) and the author(s) of the page before the move are (or should be) aware of the new location then it will almost certainly be deleted regardless of age. Thryduulf (talk) 21:12, 30 December 2019 (UTC)

I don't appreciate you misinterpreting my redirect philosophies. -- Tavix ^(talk) 00:06, 31 December 2019 (UTC)

Tavix, I'd really like to know how you interpret "a significant length of time" in the context of this criteria. Do you agree with Thryduulf that 6–12 months is approximately the right range, or would you pick a different length of time? WhatamIdoing (talk) 00:37, 31 December 2019 (UTC)

I decline to give generalities because it depends on the context. -- Tavix ^(talk) 01:07, 31 December 2019 (UTC)

• It really surprises me that length of existence should have been stated as relevant at all. Of course, I'm not referring to either extreme of the scale: redirects created in 2001 should be part of our hallowed heritage, no matter how silly they might be, while redirects created just now are immediately eligible to get summarily deleted if implausible. But otherwise, I can't see why length of existence should get into the equation. Sure, if a redirect has been around for a long time, it's possible that it might have had a higher likelihood of becoming entwined into the navigation network of Wikipedia and the wider internet, but then we have more direct proxy measure for that (like incoming links or traffic statistics). An argument can conceivably be made that the longer a page has been around the more likely it is to have gained some sort of tacit approval by the community, but this is applicable (within limits) to articles, and not to redirects (most of which tend to remain forever off peoples' radars: redirect gnoming is really a small corner of our garden). – Uanfala (talk) 01:26, 31 December 2019 (UTC)
  • I think that the longer a URL has existed, the higher the likelihood of it being linked from somewhere – but "higher likelihood" is not at all the same as "actually happened in the particular case". Traffic statistics will probably show links from external sites (at least higher-traffic sites), but we have no easy way to find links in edit summaries or previous versions of an article. WhatamIdoing (talk) 22:08, 31 December 2019 (UTC)
    • Do we ever strive to keep links alive in older revisions or in edit summaries? Whether a link is blue or red doesn't really change the communicative value of the text (and anyway, from a redlink, any logged in user should be able to get via the deletion log to the relevant RfD discussion). I take it we're not talking about retargeting (rather than deleting) WP space redirects (like shortcut for policies and guidelines), in which case a change of the status quo will obviously be a bad idea for redirects that have existed for any non-trivial period of time. – Uanfala (talk) 22:27, 31 December 2019 (UTC)
      • Yes. In fact, WP:RFD#HARMFUL indicates that it is one of two major goals for RFD: "if a redirect is reasonably old (or a redirect is created as a result of moving a page that has been there for quite some time), then it is quite possible that its deletion will break links in old, historical versions of some other articles—such an event is very difficult to envision and even detect." WhatamIdoing (talk) 00:43, 1 January 2020 (UTC)
        • Whoa! We should chuck that bit out. It's not that I don't care for preserving old revisions (quite the contrary: I've often argued at Templates for Discussion that templates that have formerly been used widely should always be kept for this precise reason, and I've almost always been a lone voice there), but that just doesn't make sense. The only difference that the existence of a redirect will make on the pages that link to it is the colour of the link, and that's beyond trivial. – Uanfala (talk) 01:26, 1 January 2020 (UTC)
          • If you keep an old redirect, then anyone reading will see a blue link and be able to reach the target article (either by clicking on it or by using WP:NAVPOPS and similar tools). If you delete an old redirect, then anyone reading will see a red link, and instead of reaching the article, if they click on it, they'll be taken to a wikitext editor to re-create the page. (There's a handy red link at the top of WP:Red link if anyone wants to try it out.) If you delete a redirect, a red link is the result, but the difference between a blue link and a red link goes well beyond the visible color. WhatamIdoing (talk) 06:24, 1 January 2020 (UTC)
            • I agree completely with WhatamIdoing here. If someone sees a link and understands from the combination of context, their memory and experience what the target is and why it was referenced then all is fine. However, if any one of those four is missing or if they are unsure then with a blue link they can find all the necessary understanding with a single click. With a red link they will be taken to either (a) an edit screen to create the page after a summary of the deletion log (if they have the necessary permissions to do so), or (b) a page explaining there is no article at that title, and that they do not have permission to create it. The message invites them to search for a page with a similar title, or as the last of about 6 options, search the deletion log to see if it was previously deleted. If someone does choose to view the deletion log, there is no guarantee that the message will give any hints about the previous content meaning the reader is not helped. These are only some of the reasons why deleting redirects can be harmful and why I recommend keeping so many redirects that have evidence of potential use. Thryduulf (talk) 10:12, 1 January 2020 (UTC)
              • Well, if preserving the clickability of links in old revisions is so important, then imagine what grave consequences there might be when retargeting redirects. So a link that previously went to one article, now goes to a completely different one, and the meaning of the original text is turned on its head. If old revisions are an argument for not deleting redirects, there are all the more an argument for not retargeting. If this logic is followed through, then redirects that have existed for a certain amount of time should never ever be edited again! – Uanfala (talk) 14:23, 1 January 2020 (UTC)
                • Yes and no. We do always need to be careful when retargetting (and especially so for shortcut links, which are referenced far more often than they are read) but when a page is retargetted the link remains blue and hatnotes/disambiguation pages can give links back (with context) to the previous target and a user is never confronted with an apparent dead-end. Thryduulf (talk) 21:53, 1 January 2020 (UTC)

Rewriting WP:RFD#HARMFUL

Currently, WP:RFD#HARMFUL reads:

The major reasons why deletion of redirects is harmful are:

• a redirect may contain nontrivial edit history;
• if a redirect is reasonably old (or a redirect is created as a result of moving a page that has been there for quite some time), then it is quite possible that its deletion will break links in old, historical versions of some other articles—such an event is very difficult to envision and even detect.

Additionally, there could exist (for example) links to the URL "https://en.wikipedia.org/wiki/Attorneygate" anywhere on the Internet. If so, then those links might not show up by checking for (clicking on) "WhatLinksHere for Attorneygate"—since those links might come from somewhere outside Wikipedia.

Therefore consider the deletion only of either really harmful redirects or of very recent ones.

Now, the second bullet point and the paragraph that immediately follows it are all really about the same thing (breaking incoming links), so I think it's reasonable to merge them. Additionally, I think it goes to a greater than necessary length explaining external incoming links. And the presence of really in the last sentence seems to imply – using the common everyday meanings of the words – that we should only delete really harmful redirects, but keep redirects that are just a little bit harmful, which I take to be obviously bad advice. I'm proposing the following version:

The major reasons why deletion of redirects is harmful are:

• a redirect may contain nontrivial edit history;
• if a redirect is reasonably old (or a redirect is created as a result of moving a page that has been there for quite some time), then it is possible that its deletion will break incoming links (if such links come from older revisions of Wikipedia articles or from the internet outside Wikipedia, they will not show up in "What links here").

Therefore consider the deletion only of either harmful redirects or of very recent ones.

— Preceding unsigned comment added by Uanfala (talk • contribs) 14:48, 1 January 2020 (UTC)

• I certainly see the rewritten section as an improvement. I'd also remove very for the same reasons you removed really. Perhaps it may be a good idea to pipe break incoming links with WP:EXTERNALROT (or is there a better location?) in case someone does want more detail on that. -- Tavix ^(talk) 15:38, 1 January 2020 (UTC)
• I agree with Tavix that this is an improvement and that the "very" is not really needed (although I'm not opposed to it remaining). I would though expand the final sentence to: "...if such links come from older revisions of Wikipedia articles, from edit summaries, from other Wikimedia projects or from elsewhere on the internet, they will not show up in "What links here")". It might also be worth mentioning offline sources as well, but I can't think how to elegantly do this. Thryduulf (talk) 21:59, 1 January 2020 (UTC)

Encourage new article creation

In a recent RFD folks argued that #KEEP clause 7 is obsolete and incompatible with #DELETE clause 10: Please delete the keep clause for clarity. –84.46.52.210 (talk) 04:22, 7 January 2020 (UTC)

Please remove keep clause 7 as obsolete, it is not always compatible with delete clause 10, and while it is in theory still true, in practice deleting a redirect makes it more obvious on other Wikipedias that enwiki does not yet have a corresponding article. "Obsolete" as in "since ILLs are managed on WikiData and not more a hopeless mess tackled by bots". –84.46.53.221 (talk) 21:11, 14 January 2020 (UTC)

 Not done: The page's protection level has changed since this request was placed. You should now be able to edit the page yourself. If you still seem to be unable to, please reopen the request with further details. Sakura Cartelet^Talk 21:35, 14 January 2020 (UTC)

Resolved

 – I anyway like DIY or {{sofixit}} better. –84.46.53.221 (talk) 21:48, 14 January 2020 (UTC)

Hi IP84, thanks for being bold! Editing policies is a valid way of testing and building consensus. I disagree that KEEP7 is obsolete, and the linked discussion doesn't give me reason to believe that consensus has changed. I totally understand your problem with K7 and D10, but disagree that it's one or the other. Neither of them is an absolute truth, in every case we need to weigh the benefits of deleting and keeping an article (you may find this essay interesting). I don't think the discussion shows that editors came to a decision on whether K7 or D10 is better in this case, so I asked Rosguill to reconsider the close. It may well be that K7 has outlived its usefulness, but I'd want wider consensus than an RfD with 3 participants before removing it entirely. — Wug·a·po·des​ 23:23, 14 January 2020 (UTC)

You should have something on public record that supports #Encourage new article creation as valid delete reason overruling #K7. I don't care how it's arranged, I was only surprised by this "obviously wrong" argument, and folks convinced me that the ILL argument is in fact better than what I recall from 2006…2011. –84.46.53.221 (talk) 23:55, 14 January 2020 (UTC)