Wikipedia talk:Please, put Pandora back in the box

On the subject of the search function (and other stuff that came to mind)

A previous revision of this essay, as it came up in Wikipedia:Redirects_for_discussion/Log/2024_March_21#Time_(Magazine) said that it is permissible to delete redirects if they are unneeded due to the search function. It was later revised again to provide the example of "sentence" redirects like "Who was the 15th president of the United States?". My position on the subject is that such a redirect would be deleted per WP:PANDORA and there is not really another valid reason to delete (meaning that this essay inadvertently admitted that WP:PANDORA has value). Looking at historical examples like Wikipedia:Redirects_for_discussion/Log/2019_July_8#What_is_a_museum and the RFDs linked from that one, we see two arguments for deletion. One is WP:NOTFAQ (often veiled as Not WikiAnswers) and the other is WP:PANDORA. The problem with the first one is that, as some editors point out, WP:NOTFAQ only applies to articles. Just because an argument reaches the correct conclusion that doesn't mean that it's a correct argument.

Even the first (now only) paragraph of the final section finds me wanting more. You say that implausible redirects should be deleted but your definition of implausible is too restrictive. TIME (MAGAZINE) (and judging by your keep !vote on the RfD above TIME (mAGAZINE)) is a plausible redirect by that definition since forgetting Caps Lock on can happen. At least the previous revisions dealt with this possibility by deferring to search but your choice to remove the second paragraph makes this a genuine problem.

On the subject of this essay's core message, I think a fundamental problem with it is that it views each redirect as a separate entity, completely independent and unaffected by the existence of any other redirect. WP:PANDORA is written from the opposite standpoint. The existence of a redirect affects other redirects, namely by motivating their creation or their continued existence. If one never sees a redirect of a certain type, one is far less likely to create it. Also, to answer the criticism of WP:OTHERSTUFF we read in that essay (actually that's in WP:When_to_use_or_avoid_"other_stuff_exists"_arguments, sorry Nickps (talk) 01:26, 23 March 2024 (UTC)) While these comparisons are not a conclusive test, they may form part of a cogent argument; an entire comment should not be dismissed because it includes a comparative statement like this. WP:PANDORA is part of WP:COSTLY. WP:COSTLY#Relevant factors 1 and 2 apply here. Opening the Pandora's box will increase the workflow of editors since redirects might end up pointing to the wrong targets, vandalised or as content forks. And to bring up a relevant example, an incorrectly capitalized disambiguator provides all of these headaches without any benefit other than the kinda flimsy argument that is WP:EXTERNALROT (which taken to its logical extreme disallows deleting any redirect other than those listed in WP:RDELETE including every sentence redirect that was ever created).

Admittedly, since you say that we don't rightly care what will happen if we "let this bad redirect stand"-- what we care about, is if it's actually a bad redirect or not everything I typed above is superfluous. A WP:COSTLY argument doesn't apply when my interlocutor doesn't seem to accept its premise (not saying you're wrong. Just that I disagree). My question to you then is, how do you know a redirect is bad unless you look at the results it brings about. WP:RDELETE cases are bad because they make WP worse either by introducing usability problems (e.g. 1) or by violating policy (e.g. 3). There is no Platonic ideal of a "bad redirect" that we can determine, the redirect is bad precisely because something bad will happen if we let it stand.

To close this mess on a light tone, I just wanted to say that while I understand your attempt at a catchy name, Pandora was never put into her box so the title misses the mark since you can't put her back in the box. You don't need to do anything about that if you don't want to, it's just something that annoys me personally. Nickps (talk) 20:33, 22 March 2024 (UTC)

Well, let's go point by point.

...this essay inadvertently admitted that WP:PANDORA has value

• The last section's purpose was to attempt to find non-Pandora-related reasons to delete the examples shown in WP:PANDORA. I do feel like the redirects that WP:PANDORA points at are bad redirects-- but WP:PANDORA itself is a monumentally bad argument, and I'll once again go as far as to describe it as a non-argument. You said it yourself-- Just because an argument reaches the correct conclusion, that doesn't mean that it's a correct argument.

You say that implausible redirects should be deleted, but your definition of implausible is too restrictive.

• *'My definition of implausible' in this case is WP:RCAPS and WP:MIXEDSCRIPT. Both of those handle the subject far better than Pandora does.

On the subject of this essay's core message, I think a fundamental problem with it is that it views each redirect as a separate entity, completely independent and unaffected by the existence of any other redirect.

• That's not a fundamental problem, that's a design goal. Per WP:OTHERSTUFF, that's how ANY article, redirects included, needs to be viewed-- on its own merits. Bringing up other articles and redirects only muddies the issue.

How do you know a redirect is bad, unless you look at the results it brings about

• That's exactly the problem-- you CAN'T tell a redirect is bad simply by looking at "the results it brings about". Simply going off of WP:PANDORA's text, there is nothing in its arguments that tells me why the listed examples are bad-- which means that the only thing from WP:PANDORA that we really have to go on, as for what redirects it actually wants us to delete, are the examples. This is why, despite WP:PANDORA and WP:MIXEDSCRIPT both targeting Cl0ck, WP:PANDORA is a bad argument, and WP:MIXEDSCRIPT is a good one. That's not saying I don't agree with the premise of WP:COSTLY as a whole-- everything under WP:COSTLY#Relevant factors are very good arguments... for needing to delete bad redirects in general. The issue comes with identifying which redirects are bad-- and for that, you cannot simply ask the question, "will this encourage other redirects just like it?", and if yes, delete. (Which, for the record, seems to be what WP:PANDORA implies, if you don't take into account the examples it uses to pinpoint what redirects it actually wants to delete. That's the argument actually being made.)

Pandora was never put into her box, so the title misses the mark

• :P

𝔏𝔲𝔫𝔞𝔪𝔞𝔫𝔫🌙🌙🌙 𝔗𝔥𝔢 𝔐𝔬𝔬𝔬𝔬𝔬𝔫𝔦𝔢𝔰𝔱 (talk) 11:11, 23 March 2024 (UTC)

I guess I'll start with my comment about the definition of "implausible". I was using if a human being would logically never type a string of characters (in parlance, an "implausible" redirect) which implies everything I said it does since forgetting Caps Lock on can happen. If you instead want to use WP:RCAPS, that's a much better definition and I welcome the change but I can't read your mind.

Per WP:OTHERSTUFF, that's how ANY article, redirects included, needs to be viewed-- on its own merits This is true and PANDORA invites us to do so with its first word, the redirect must be "unhelpful" for PANDORA to apply. However, OTHERSTUFF doesn't completely disallow making comparisons. However, such an argument may be perfectly valid if such can be demonstrated in the same way as one might demonstrate justification for an article's creation which applies to PANDORA since if someone proposed the creation of e.g. a mixed script redirect, it would be reasonable to respond that mixed script redirects are generally not made and then make an argument about why such redirects are bad most of the time. Yes, you might land in the 1 in a million case where something has a mixed script name (like Bayer designations which MIXEDSCRIPT mentions) but the proposer will easily be able to prove notability in that case and the argument would still be true in general.

In any case, some of the examples in WP:PANDORA can be deleted per WP:RFDO but that on its own only says that we should follow precedent. Not only is this not an airtight argument since WP:CCC but in one of the examples linked in WP:MIXEDSCRIPT we even read In either case, they are not needed and should be deleted to discourage the creation of similar pages. PANDORA helped set the very precedent we are now using to supersede it. It's true that we could replace every PANDORA argument with "the redirect is unnecessary and therefore we should delete per WP:COSTLY". However, PANDORA actually helps us more accurately estimate the cost of keeping so in that sense it's a good argument. It isn't only saying "will this encourage other redirects just like it?", and if yes, delete it says, this redirect is not good (that is, it has a cost that outweighs its benefit even when examined in isolation) and it will encourage other redirects like it which makes that cost even bigger.*

Finally, I have to address you CAN'T tell a redirect is bad simply by looking at "the results it brings about". How do you tell then? WP:MIXEDSCRIPT says something about implausible search terms and links to RTYPO which says essentially the same. Why is it bad to have a redirect if it's an implausible search term? In fact, if we ignore the results, what is wrong with redirecting Adam B. Smith to Andrew B. Smith? Confusion is a result that was brought about by the creation of the redirect. In that sense, you saying that you accept the principle behind COSTLY is puzzling. The cost of a redirect is a result of its existence and this is what WP:COSTLY#Relevant factors attempt to determine. You make it sound like the reason a bad redirect is bad is metaphysical.

*This is something I messed up in Wikipedia:Redirects_for_discussion/Log/2024_March_21#Time_(Magazine) where instead of addressing the notion that the redirect "costs nothing" (put forth by the first keep !voter who evidently upgraded WP:CHEAP to WP:FREE) by pointing out WP:COSTLY#Relevant factors #1 and #2 I only mentioned PANDORA. What I should have said is, "It's not true that it costs nothing because it's a vandalism/POV fork target etc etc." and maybe mention the possibility of other redirects at the end. In my mind the way I went about it was fine since PANDORA represented such a cost but I could have chosen a less controversial one. There's also a second reason. Since PANDORA uses the word unhelpful, I thought it implied all of the above not thinking that, since the other editor said that the redirect "costs nothing", they didn't have COSTLY in mind at all. Nickps (talk) 14:57, 23 March 2024 (UTC)

• ...implausible stuff

Those two definitions are meant to be the same. If you can point me at where I screwed up when listing the definition of 'implausible', feel free to guide me there so I can fix it.

• How can you tell, then?

On its own merits.

A redirect needs to do one thing, and one thing well-- it needs to take a reader who types in something into the search bar, or follows a link left by an editor, to the information they actually want, with as little WP:SURPRISE as possible. That is the core, explicit purpose of a redirect. And RfD's role is to make sure that all redirects on Wikipedia play that role. If a redirect is implausible (i.e. there is little to no chance it would ever be typed by a human), it needs to be deleted. If a redirect is mistargeted (i.e. it goes to the wrong place), it needs to be retargeted and/or dabified. If there IS no proper target, the redirect is not helpful and thus should be deleted. That, summed up, is the basis behind RfD. That is why we are here, that is what metric we judge redirects on.

Or, as summed up by WP:RGUIDE:

• ...If someone could plausibly enter the redirect's name when searching for the target article, it's a good redirect.

WP:PANDORA does not do that. Pandora makes up an entirely separate metric to judge redirects on, one that flouts WP:OTHERSTUFF, and it doesn't even give us the luxury of telling us WHY the examples it gives go against its metric. THAT is what my issue is with Pandora. 𝔏𝔲𝔫𝔞𝔪𝔞𝔫𝔫🌙🌙🌙 𝔗𝔥𝔢 𝔐𝔬𝔬𝔬𝔬𝔬𝔫𝔦𝔢𝔰𝔱 (talk) 17:44, 25 March 2024 (UTC)

About implausible, my objection is simple. wikt:implausible says the word means "unlikely" not "impossible", so even if a redirect gets non-zero use, it should still be deleted if that use is low enough. That doesn't fit with if a human being would logically never type a string of characters. The word never is too strong. On the other hand, there is little to no chance it would ever be typed by a human which you used in the comment I'm replying to is just fine.

• WP:PANDORA does not do that. Pandora makes up an entirely separate metric to judge redirects on, one that flouts WP:OTHERSTUFF, and it doesn't even give us the luxury of telling us WHY the examples it gives go against its metric.

WP:UNHELPFUL (as it's also called, probably a better/more accurate name even if PANDORA is cooler) is concerned with the creation of similar redirects. The examples "go against its metric" because they are easily generalisable. Formulating article topics as questions is easy, trivial even. Coming up with mixed script redirects is easy by using a list of confusables. Also, I'm not so sure about it being made up. It easily follows from the premise that redirects are costly. If redirects are costly then we shouldn't encourage the creation of needless redirects which means that we should not allow such redirects to stand since their existence leads to the creation of more useless redirects.

• Now THAT is a good line of questioning. That leads my brain directly to WP:NOT, the source of WP:CRYSTAL-- is there ANYTHING in WP:NOT that might be helpful here? And indeed, there is. Right on WP:NOTFAQ:...

Hopefully you won't mind me replying to this here. Anyway, even if NOTFAQ applied, which I strongly believe it does not per my original post here, what's the harm in leaving in question redirects for some high view articles per WP:CHEAP? Ah, yes, that it will warp user expectations leading us down a slippery slope. That sounds familiar! And what about the other examples? Mixed script is arguably RDELETE #8 so we'll leave that aside but informal abbreviations are not necessarily novel or obscure.

Finally, I disagree with the idea that WP:UNHELPFUL only applies to the examples it gives (which you implied by ... the examples it uses to pinpoint what redirects it actually wants to delete). UNHELPFUL is meant to be a criterion that applies to more than just the examples given. Here's an idea: it applies to all unhelpful redirects. Otherwise it would explicitly give a list of cases. I'm saying this to make it clear that just finding other justifications to delete the examples misses the point. The point is whether the argument itself can help us decide which redirects to delete. If you think it can't, then you'd make a better case for your side if you found a case where the argument fails. To make this point clearer, "unhelpful" does not mean "implausible". WP:UNHELPFUL makes the case that even if some redirects might be used by some people (the few readers who stumble upon them), there are still reasons not to keep them. I'm not forgetting WP:RKEEP #5 exists, we still have to scale how useful they are to those few readers vs. the adverse effects they have on WP.

P.S. I've also subscribed to the thread. Nickps (talk) 15:37, 26 March 2024 (UTC)

I've mostly replied to everything here in my most recent reply to Brusquedandelion, but this brings up a point I didn't cover there--

• To make this point clearer, "unhelpful" does not mean "implausible".

Then Pandora needs even more of a serious rewrite. If you're willing to tell me that, by 'unhelpful', it means anything OTHER than 'implausible', then WP:UNHELPFUL could do with more than literally two words explaining itself before it launches into the WP:ATA#CRYSTAL/WP:OTHERSTUFF-ridden mess of an attempt at an argument that gave it its other label, WP:PANDORA. 𝔏𝔲𝔫𝔞𝔪𝔞𝔫𝔫🌙🌙🌙 𝔗𝔥𝔢 𝔐𝔬𝔬𝔬𝔬𝔬𝔫𝔦𝔢𝔰𝔱 (talk) 17:16, 26 March 2024 (UTC)

I'm willing to say that, since we agree here that some of PANDORA's examples are not implausible but are still deleted. It follows that unhelpful must have different meaning. Here's what it is. List of Seinfeld eps is not a serious link target since ep is too informal to be used on WP. So, the only reasonable way someone ends up on it would be via search. Google Search finds the WP article as first result so that's fine. What about our search? Autocomplete comes up with the correct title for that, therefore, there is no reason to redirect. That's how I understand the word unhelpful. It serves no function, because, even if someone were to try and use it, it's non-existence wouldn't be felt. Even Варtism is dealt with by search (WP search that is. Hilariously, google finds WP:COSTLY). It's pretty good. Whether PANDORA needs to be rewritten to reflect this, I don't know. I think it's obvious enough that it doesn't.

Oh, and by the way, WP:POLSHOPing per WP:IAR is a completely puzzling idea to me. If you believe a redirect should be deleted but you can't back it up with policy you should invoke IAR directly and actually convince the other editors that deletion would improve the encyclopedia/make it easier to maintain. That might actually result in the policy you need getting written. Because that's your current problem. The policy you're trying to use doesn't fit the purpose so you appropriate stuff that's "close enough". You ask Brusquedandelion how they know WP:NOTFAQ doesn't apply to redirects. The text says 5. FAQs: Wikipedia articles should not list frequently asked questions (FAQs). Instead, format the information as neutral prose within the appropriate article(s). And the section it's in starts with Wikipedia articles should not read like:. How would you infer it applies to redirects? The rules concerning redirects are much looser than those of articles. Most of WP:NC for example doesn't apply to redirects and any parts that do say so explicitly so why would this apply implicitly? Nickps (talk) 18:54, 26 March 2024 (UTC)

Whether PANDORA needs to be rewritten to reflect this, I don't know. I think it's obvious enough that it doesn't.

I'm on the complete other side, and believe that this is not only proof that it needs a rewrite... but proof that such a rewrite might be possible. Keep in mind that as it stands, if you ignore the examples given and the flouting of WP:OTHERSTUFF, WP:UNHELPFUL is a grand total of two words long. Extracting that sort of meaning out of those two words requires paragraphs of explanation that renders the citing of the essay moot at best, even if it doesn't end up kneecapping your argument because people focus more on the WP:OTHERSTUFF than the argument you're trying to make.

So... why not take those paragraphs of explanation, and put them in the essay? ...And also, remove the WP:ATA#CRYSTAL and WP:OTHERSTUFF problems while you're at it... edited 20:24, 26 March 2024 (UTC)

If you believe a redirect should be deleted but you can't back it up with policy, you should invoke IAR directly, and actually convince the other editors that deletion would improve the encyclopedia/make it easier to maintain

That's good for a case-by-case basis, not for what I'm trying to do here-- which is write an essay. Either way, I'm honestly at my wit's end and liable to give up. I can't find a good method of excluding these redirects...

so I won't. I'll rewrite the ending section to be a little more general, and will leave finding an alternate method of arguing against such redirects up to the reader-- or, as it might be, whoever decides to finally rewrite WP:UNHELPFUL to shut Pandora's box for good. 𝔏𝔲𝔫𝔞𝔪𝔞𝔫𝔫🌙🌙🌙 𝔗𝔥𝔢 𝔐𝔬𝔬𝔬𝔬𝔬𝔫𝔦𝔢𝔰𝔱 (talk) 20:19, 26 March 2024 (UTC)

The last section's purpose was to attempt to find non-Pandora-related reasons to delete the examples shown in WP:PANDORA. I do feel like the redirects that WP:PANDORA points at are bad redirects

And yet you have yet to give us a reason why. The person you're replying to correctly pointed out that, as a result, you are indirectly admitting the utility of WP:PANDORA-style arguments. By not offering a substitute in its place, and instead dismissing WP:PANDORA as a non-argument (source: WP:IDONTLIKEIT) and thus unworthy of consideration, you have only further underscored the need for WP:PANDORA.

That's not a fundamental problem, that's a design goal. Per WP:OTHERSTUFF, that's how ANY article, redirects included,

That last bit which I bolded is doing a whole lot of heavy lifting. I'm not seeing anything in WP:OTHERSTUFF to imply that. Redirects are fundamentally unlike articles. They form part of a coherent user experience for users of Wikipedia, and the existence of one redirect may very well imply the existence of other redirects. Also, While these comparisons are not a conclusive test, they may form part of a cogent argument; an entire comment should not be dismissed because it includes a comparative statement like this. Nuff said. Brusquedandelion (talk) 02:37, 25 March 2024 (UTC)

• You are indirectly admitting the utility of WP:PANDORA style arguments

No, I am not. There is a difference, between admitting that an argument is arguing for the correct action, and admitting that an argument is structurally sound. This is a correctly aimed, but structurally EXTREMELY unsound, argument. As for not offering a substitute in its place, that's... again, literally what the last section of this essay was attempting to do.

• I'm not seeing anything in WP:OTHERSTUFF to imply that

...Are we reading the same essay? WP:OTHERSTUFF (and the essay it's from, WP:ATA)) isn't trying to say how we should write articles-- it's saying how we should conduct XfD discussions. Not specifically AfD, but XfD-- hence the title of the essay saying "deletion discussions" instead of "Articles for Deletion". I do understand that at RfD we tend to flout certain parts of WP:ATA-- like WP:PERNOM, WP:MAJORITY, and WP:JUSTA-- but I don't see a good justification to flout WP:OTHERSTUFF (or its brother against Pandora, WP:ATA#CRYSTAL) at RfD. 𝔏𝔲𝔫𝔞𝔪𝔞𝔫𝔫🌙🌙🌙 𝔗𝔥𝔢 𝔐𝔬𝔬𝔬𝔬𝔬𝔫𝔦𝔢𝔰𝔱 (talk) 17:25, 25 March 2024 (UTC)

There is a difference, between admitting that an argument is arguing for the correct action, and admitting that an argument is structurally sound.

Sure, but if you don't give us a better argument... anyways I wasn't aware that (you think) that's what the last section of this essay was attempting to do. so let's get into that. The last section is short:

I'm glad you asked! When arguing against a redirect, try to keep in mind how well it actually serves the function of a redirect- being, a good redirect will catch common searches, and redirect them to the page the user was actually looking for. Thus, if a human being would logically never type a string of characters (in parlance, an "implausible" redirect), the redirect is pretty useless and should probably be deleted.

Who was the first president of the United States? is entirely plausible, so that argument fails. The only other argument I'm seeing is the popularity argument: is this actually a common search? First, a procedural issue with this: AFAIK, there's no way for us to check what terms people are searching for in what frequency, so this can't possibly be a good, general-purpose litmus test. But suppose there were such data. Per your argument, if enough people are searching for Who was the first president of the United States?, then we should have a redirect for it. That is, there is no principled reason, from you, as far as I can tell, why such redirects are bad; at best, they are bad only due to empirical happenstance, which could change for any number of reasons. Since the existence of such a redirect would surely increase the amount of people using similar queries, it would inevitably lead to Wikipedia becoming WikiAnswers by means of redirects, if we keep applying this principle. Oh look, we've reinvented WP:PANDORA, and we are still left without a principled reason for why such redirects shouldn't exist. Please feel free to provide one.

P.S. For future reference, I have already subscribed to the thread; no need to ping :) Brusquedandelion (talk) 06:47, 26 March 2024 (UTC)

Who was the first president of the United States? is entirely plausible, so that argument fails.

...Yep! Which means, if it is indeed plausible, that the redirect should be kept, assuming it has a valid, unambiguous target-- and it does, George Washington. Behold, the reason why I removed my attempt at an argument against "sentence" style redirects-- I couldn't originally think of an argument against their plausibility.

Except you didn't stop typing there, let's work through what else you said...

AFAIK there is no way for us to check what terms people are searching for in what frequency

There actually is-- we can check the usage stats of the page in question, perhaps comparing them to the target page! For example, Here's the usage stats of WP:PANDORA. This doesn't work if the redirect doesn't exist yet, and/or if the redirect is brand-new, but for a redirect that's been around for a while, it's a perfect way to gauge EXACTLY how plausible the redirect is.

Since the existence of such a redirect would surely increase the amount of people using similar queries, it would inevitably lead to Wikipedia becoming WikiAnswers by means of redirects, if we keep applying this principle...

Now THAT is a good line of questioning. That leads my brain directly to WP:NOT, the source of WP:CRYSTAL-- is there ANYTHING in WP:NOT that might be helpful here? And indeed, there is. Right on WP:NOTFAQ:

FAQs: Wikipedia articles should not list frequently asked questions (FAQs). Instead, format the information as neutral prose within the appropriate article(s).

I do think that that's pretty much what we were looking for. This is an FAQ style redirect, and thus, as per WP:NOTFAQ, we shouldn't have it! 𝔏𝔲𝔫𝔞𝔪𝔞𝔫𝔫🌙🌙🌙 𝔗𝔥𝔢 𝔐𝔬𝔬𝔬𝔬𝔬𝔫𝔦𝔢𝔰𝔱 (talk) 07:10, 26 March 2024 (UTC)

I chose the Who was the first president of the United States? example from WP:PANDORA. You previously implied you agree with the conclusions of WP:PANDORA, but not the means by which it got there. You are now, however, fundamentally disagreeing, and claiming that a Who was the first president of the United States? redirect (and presumably other, similar ones) should be kept, assuming it has a valid, unambiguous target. It is unfortunate you wasted our time by pretending otherwise until this time, but if you sincerely believe this, I don't think we are likely to convince each other. We fundamentally disagree about what Wikipedia should be, and what is a manageable workload to expect of editors.

There actually is-- we can check the usage stats of the page in question, perhaps comparing them to the target page!

Ok. Please tell me (1) how many people searched for Who was the first president of the United States? on Wikipedia in the last year and (2) how many people searched for George Washington in the same time period.

Oh you can't? Probably because usage stats are not the same thing as search stats, not least because people can arrive at a page in any number of ways besides search.

This doesn't work if the redirect doesn't exist yet, and/or if the redirect is brand-new, but for a redirect that's been around for a while,

So it doesn't work—as I said. Got it. The limited set of cases when it does work is only after a redirect has been allowed to exist for a while—actually, this is also not true, for the reasons stated prior (links exist!) but let's roll with this. However, the mere fact that a redirect (or a similar redirect) exists will drive search traffic, because people know it exists from having typed it before, or because other similar redirects exist. We can add to that a third reason, that people will use the redirect as a link, since your method does not, actually, tell us about search traffic, only foot traffic.

Right on WP:NOTFAQ:

Sigh, I knew you would say this; somehow, everyone who thinks they know better than WP:PANDORA manages to reinvent the same (debunked, defective) wheel. WP:NOTFAQ is about articles, not redirects. It tells us why articles like Who was the first president of the United States? shouldn't exist, but not why redirects of this nature should not exist.

I said earlier it is unfortunate that you previously implied you agree with the conclusions of WP:PANDORA, but not the means by which it got there, when in fact you do not agree with the conclusions either. It now appears you have changed your mind again, and actually do believe redirects like Who was the first president of the United States? shouldn't exist, for WP:FAQ reasons, as deficient an argument as that is. All of this is a sign that the internal algorithm you're using to determine if a redirect should exist or not is chaotic in a mathematical sense, that is, sensitive to tiny variations in the input. This is, ideally, not something we want from a decision-making process. I suggest revisiting and reconsidering your beliefs on this until you have a rock solid, stable set of criteria on what sorts of redirects should and shouldn't exist; at a minimum, it would make life easier for the people you are disagreeing with, if your opinions could actually be pinned down.

But even putting all that aside, WP:PANDORA has other, non-FAQ examples: Pres of United States and List of Seinfeld eps. Any particular reason your methodology would delete these? Or are they keeps, for you? Brusquedandelion (talk) 15:54, 26 March 2024 (UTC)

...I am feeling like we're talking a bit past each other, and/or you have misread/misinterpreted my goals/beliefs/methodology here. Thus, I'll restate everything, in an attempt to clarify.

• I do not believe that WP:PANDORA is a sound argument.

I am not attempting to attack its conclusions, I am attempting to attack its methodology. I believe that using WP:PANDORA is a bad method of arguing against these redirects due to it not properly giving an explanation as to why these particular redirects should be deleted, and that if it is continued to be used, it will continued to be parried by accusations of WP:WHATABOUT and WP:ATA#CRYSTAL, which cause arguments that needlessly clog up deletion discussions.

• I am basing my 'internal algorithm', as you put it, on the first and second statements from WP:RGUIDE.

If I may expand that out, this means the following:

• A redirect's title must be plausible. Pick whatever definition you want to use here, I'm not interested in waffling on for fifty more paragraphs about oh, Lunamann used the wrong definition of plausible.
• A redirect must go to the correct place. If multiple such places exist, it should be dabified. If no such place exists, it should be deleted as per WP:REDLINK.
• A redirect must satisfy WP:NOT as well as other such guidelines and policies-- it being a redirect and not an article does not give it blanket immunity from following the rules of Wikipedia.
• A redirect is WP:CHEAP, and can be freely created, kept, retargeted, expanded, dabified, or deleted without ruffling too many feathers.

Any perceived 'chaos' comes from the following:

• I am, currently, rather blatantly flouting WP:POLSHOP as per WP:IAR.

I have, as previously, stated that I do not believe the examples given by WP:PANDORA should not be kept-- that conclusion has not changed at any point, despite appearances. The issue is, I am having issues finding policies that give coherent reasons why, and when conceding that a particular policy does not exclude something, I have repeatedly made statements such as, "...which means this redirect should be kept," with the intended- and clearly not-picked-up-on- subtext, of "...which means we need to find a different angle of attack." That is an error in clarification, and I apologize.

With that out of the way, let's go over my current thoughts on the different parts of WP:PANDORA:

• Unhelpful titles whose existence might encourage the few readers who stumble upon them to assume that there exist redirects of the same type for other targets as well (opening a "Pandora's Box" of user expectations)

This entire line is just one big wash. The ONLY part of this line that has any help for us is specifically the first two words, "unhelpful titles"-- so, is this an attempt to target the implausibility of the titles in question? That's the only way that the title of a redirect can be unhelpful, after all-- if it fails to grab a useful amount of incoming searches/links. All other functionality of a redirect is handled by the targeting, not the title. edited at 17:25, 26 March 2024 (UTC) The entire rest of this can just be tossed in the trash.

• titles that turn articles or article subtopics into questions, like Who was the first president of the United States? or What is the capital of France?

This one's the problem one. Unlike WP:PANDORA itself, seemingly enough, I'm not interested in trying to target the implausibility of these titles. They are, in fact, plausible. I am more interested in trying to find some other avenue of attack. Sooo... going back to WP:NOTFAQ, you seem intensely adamant that it doesn't apply to redirects. Might I ask why that is?

• titles with informal abbreviations, such as Pres of United States or List of Seinfeld eps

Very implausible. You're willing to tell me that (as per WP:RTYPO) leaving out two letters is implausible, but leaving out an entire "ident" off of "President" is plausible? Not buying it.

• titles with incorrect mixed use of non-Latin or otherwise inappropriate script, such as Варtism (where the Вар is in Cyrillic script) or Cl0ck (as a redirect to Clock)

Already covered by WP:MIXEDSCRIPT and WP:RCAPS. 𝔏𝔲𝔫𝔞𝔪𝔞𝔫𝔫🌙🌙🌙 𝔗𝔥𝔢 𝔐𝔬𝔬𝔬𝔬𝔬𝔫𝔦𝔢𝔰𝔱 (talk) 16:59, 26 March 2024 (UTC)

• You're willing to tell me that (as per WP:RTYPO) leaving out two letters is implausible

RTYPO wouldn't fly at all. wikt:Pres and wikt:ep exist. They're not typos. They're informal abbreviations. So, no, they are not implausible by any reasonable definition. Nickps (talk) 17:25, 26 March 2024 (UTC)

Mmm. ....Then these are plausible titles. And unlike the sentence redirects... I'm a lot less willing to spend an ungodly amount of time trying to WP:POLSHOP to find a way to exclude THESE redirects. 𝔏𝔲𝔫𝔞𝔪𝔞𝔫𝔫🌙🌙🌙 𝔗𝔥𝔢 𝔐𝔬𝔬𝔬𝔬𝔬𝔫𝔦𝔢𝔰𝔱 (talk) 17:27, 26 March 2024 (UTC)