Jump to content

Wikipedia talk:Deletion policy/redirects/Archive 1

Page contents not supported in other languages.
From Wikipedia, the free encyclopedia
Archive 1

really old talk

My admittedly newbie opinion is that misspelt names should not be allowed in. This is, after all, a reference work. Inaccuracies should be purged. Wayne Gretszky is what sent me searching for a page talking about deleting pages. ATM Gretszky redirects to Wayne Gretzky, but is it really beneficial? Eventually search engines will update and dead links will phase out. --Colin dellow


I have seen that it is very desirable not to delete pages, but only do #REDIRECTs, will be keep doing that, even when a page is some mispelling (especially mistyping, like Wofgang Amadeus Mozart)?

If there is a consensus that some pages should be deleted every now and then, I'd put the previous one on the list.


I think the suggestion to keep all pages came as a means of avoiding dead links in search engines. I've encountered similar mispellings (Martin Scorcese); I would like to do away with them also but I'm not sure it would be a good advertisement for Wikipedia to leave dead links. I guess whatever action is taken depends on how common the misspelling is? It seems that wikipedia does not fuzzy-match search results; is that the case?


I think there are two things discussed here. One is deletion of data. I don't like that. The other is moving data from an unsuitable location (a misspelling) to a correct one. I don't think this is a problem.

If someone creates a new page with a misspelled name and you spot it the day after, I think it's a good idea to move the data and remove the misspelled entry. Of course you should also use search to make sure that any other links to the page are changed. If the misspelling is on an old page so that it can be suspected of being in search engines like [[1]], then I guess a redirect is better. --Pinkunicorn


Hi Pink, a redirect in that case is definitely better. Since ordinary users can't delete pages but only the data they contain, you might as well make the page with the misspelled title point to the one with the correctly spelled title. We do this all the time, and search engines like it (why not give poor spellers a positive result for their trouble? So Wikipedia catches the poor speller traffic--grand!). Colin, it would have been a little better to have asked this question in the FAQ or on Wikipedia chat rather than making an entire new page for it. By this time we have thought through most of the basic issues, so it's not a matter of debate, it's a matter of us informing you what the consensus is. --User:LMS


Ah, my mistake. For some reason I thought the proper system here would be to put freshest comments at the top of the page, hence the misleading positioning of my comment. I didn't create this page, it just appeared to be the most proper forum when I ran a search. Anyway, question answered, lessons learned, one step closer to being better informed. :) --User:Colin dellow


Oh, I thought this page looked familiar! I think it's a really old one. Sorry, Colin. As you can see, I was indeed confused by the position of your comment. Isn't interesting how the place where your comment is put conveys semantic information... --User:LMS


Just to add one more thought, the cost of having a misspelling with a REDIRECT is probably pretty low, and it's not like some is going to have a page on someone else name of Wayne Gretszky. The only disadvantage I see is that it makes search results somewhat noisier; ideally having a way to tag a page as not being searchable, only reachable directly might be nice to have. Though first I want search results to put the best result -- pages with the request name in them -- at the top of the search list. --User:Belltower

Ex-Subpages

Hi, may I ask why we're deleting subpages that have been around for a long time, instead of leaving them as redirects? I just teased Ed Poor for wanting to delete a bunch of subpages of Middle Earth, since these have all been archived by search engines. I guess I was wrong; they're being deleted by the truckload! I remember people being chastised on Wikipedia:Votes for deletion for voting for these pages. Now anybody that does a Google search for the words "Middle-earth" and <fill in the blank with most anything> (look at Wikipedia:Article deletion log and try it for yourself!) will quickly come to the conclusion that following Wikipedia links is a waste of time. Are we really having space issues that these redirect orphans can't continue to bring in traffic from search engines? — Toby 03:19 Jul 24, 2002 (PDT)

Toby, at first I removed all the text from (and requested deletion of) the "subpage" articles. Then, after discussion with LDC and Maveric and Brion, I started just replacing the text of subpage articles with REDIRECTs to the new articles. This has left several dozen blank pages, I'm afraid. But soon Google and other search engines will have re-indexed Middle-earth and every part of it (even Rivendell) will be instantly available. Ed Poor 08:47 Jul 24, 2002 (PDT)

But why are we going out of our way to create broken links in search engines, even temporarily? As mentioned, this seems reasonable is space is an issue, but not otherwise. I've seen this come up several times before, and I never noticed a consensus that we should delete instead of redirect — OTC, popular opinion seemed to go the other way. Did I miss a (public) discussion somewhere? — Toby 11:08 Jul 24, 2002 (PDT)

I forgot, there's another reason not to delete these: We no longer know what their history is (unless we look in http://old.wikipedia.com/). Thanks to Julie's talk page for reminding me. — Toby 11:17 Jul 24, 2002 (PDT)

I should also point out that the "Move page" feature of the wiki software (is this still restricted to sysops?) automatically creates a redirect from the old title to the new title. ---Brion VIBBER
Yes, it's still resticted to sysops — see next item. — Toby 11:38 Jul 24, 2002 (PDT)

I just want to say that I am in 100% agreement with Brion and Toby about not deleting old page titles when their content has been moved to a new page title. My reasoning is in many other places so (See Talk:Middle-earth). I will, however, recreate any deleted page title as a redirect if it is deleted solely because it is no longer the article's title. --mav 19:32 Jul 24, 2002 (PDT)

Good for you! Too bad that that doesn't restore the history (unless it was moved by secrect sysop superpowers). — Toby 13:27 Jul 25, 2002 (PDT)

Misspellings

QUOTE: Do not delete anything that might be a common misspelling of a title. Redirect those pages to the correct spelling. (This can actually result in higher traffic to the website. E.g., if philisophy is created, we might as well just redirect it to philosophy, since "philisophy" is one of the common misspellings of "philosophy.")

No wonder this sight has problems. With this golly, gosh, fellas, if we keep misspellings we will get more hits, I would bet that not one single person who was involved in coming to this brillant conclusion has ever run a SUCCESSFUL business. You don't gain viewers (clients) by offering garbage, you get them by quality and that is particularly true of something purporting to be an Encyclopedia. Note GOOGLE says: Did you mean???? - If the powers that be think this nonsense should continue then at least insert the words: MISSPELLED - REDIRECTED TO: XYZ ....DW

There's certainly a case for an alternate redirect syntax. I've been pondering for a while the idea of having #MISSPELLING foo for these. -- Tarquin 23:59 Dec 18, 2002 (UTC)

Delete the page - PERIOD! This is a learning site, not one to confuse and delay people in their searches. The redirect setup now causes confusion. It is time to get professional. I have no objection to crappy, incomplete articles, (although the laziness tees me off) but they can and will be fixed eventually as part of the building process. Get rid of the redirects: It turns people away when three variations, of which 2 are misspelled, show up as your Encyclopedia's choice? PREPOSTEROUS & AMATEURISH.....DW

Surely deleting these redirects will "confuse and delay people". Somebody searches for "philisophy" now, they find the article they want, and because "philosophy" is spelled correctly within that article, they learn how to spell the word to boot. Delete the redirect and the searcher finds nothing. I don't see how that would be an improvement. --Camembert
I agree. There is no reason to delete common mispellings. BTW DW, it is "site" not "sight". --mav
I agree as well. BTW mav, it is "misspellings", not "mispelling". :) Chas zzz brown 00:48 Dec 19, 2002 (UTC)
BTW, Mav misspelled the plural, not the singular :-) At least correct the correct tense, if you are going to correct it... -- RTC 00:53 Dec 19, 2002 (UTC)
Insert foot here -> O: Exactly why we should support common misspellings. --mav

What a bunch of nonsense! Stop with the little "club" patting each other's back (I've seen it before from you guys), and use your brains, limited though they obviously are. This site (and I always call it "SITE" unless my limited little brain is running faster than my fingers), has not been successful because of its lack of direction from someone with SUCCESSFUL business experience. Morons want to maintain the status quo because of their lack of business acumen combined with tunnel vision and the need to control. Get real! Encyclopedias, none, nowhere, no how, zip, rien, offer spelling variations. It's this lunacy that is sinking Wikipedia and one day soon all the hard work of people with a PROFESSIONAL attitude who make REAL contributions, will evaporate. Rarely does one encounter such blatant stupidity. It is obvious that you have no business training or experience so GROW UP....DW

Very drôle. --Camembert

Camel case

I notice a lot of redirects listed on wikipedia:votes for deletion. But this page says "Do not delete a page which could logically be made into a REDIRECT to an article". Is the policy out of date? Martin

No I don't think so. I think it's just people not bothering to read or understand that a misspelled page may be a useful redirect. Mintguy

What about AbductioN and other camel case redirects? Martin

Lisewise, there's no need to delete them. There's always a possibility that somebody made a link to one of these pages from their own website (or in a newsgroup posting, email, or whatever) back when CamelCase was used, so it's better to keep them to avoid breaking those links. --Camembert
The link likelihood is near zero, as the CamelCase period was very brief. I say they should be deleted, they just clutter up the search. --Eloquence 19:53 Feb 17, 2003 (UTC)
Well, I admit I don't feel strongly about it. But all redirects, not just these ones, clutter up the search, so a better solution to my mind would be to somehow improve the way search results are presented. Of course, that's a damn site harder to implement than simply deleting these redirects, but I don't think they add that much clutter, and if even one link is broken by deleting them, then it's a loss. Incidentally, how long was the CamelCase period? --Camembert
I suggested something to that effect - now at meta:Searches and RedirectsMartin
Yes, redirects will be handled differently in the search, but this still doesn't solve the problem. Most redirects should show up in a search (we will probably eventually want a redirect qualifier to distinguish incorrect spelling from alternative names), it's just the CamelCase redirs that are completely useless. Aside from that, dozens of them have already been deleted because there were no objections on VfD. I challenge you to show me a single website that links to a CamelCase based Wikipedia article. Unless you can do this, I think we should just go ahead and delete them all. --Eloquence 08:55 Feb 18, 2003 (UTC)
But many of these CamelCase redirects were the original homes of current articles and were primitively moved using the cut, redirect and paste method. Therefore the original edit history is at the CamelCase redirect and if we delete that article and its history we violate the GNU FDL by destroying the author attribution. Which reminds me that I have been meaning to bug the developers about possibly creating an interface that would allow Admins to easily move the history of an article without having to do the delete, move, restore routine. --mav

"dozens of them have already been deleted because there were no objections on VfD" -- do we really have to specifically object every time someone votes to delete a valid redirect? :-( Martin

Yes. There is obviously no consensus on the matter, so people need to express their opinions. --Eloquence 23:23 Feb 19, 2003 (UTC)

suggestion moved to meta:searches and redirects

discussion of the Daniel C. Boyer redirects

These are valid redirects, and should stay, in accordance with deletion policy. The wikipedia search engine should be fixed, longterm. Martin 12:31 18 Jul 2003 (UTC)
17? The line has to be drawn somewhere. Unfortunately, unlike other Wikipedians with articles, Daniel Boyer has used the slack he's been given to promote himself beyond a reasonable level. 1994 in film is a case in point. And we don't have a policy to keep redirects, even if our policy documents say we do. A "policy" enforced when it suits us is a newbie-deterrent, not a policy. -- Tim Starling 14:15 18 Jul 2003 (UTC)
1994 in film is obviously bad, and I won't defend that self-promotion.
Redirects cause only one practical problem: in searching. This is very much a technical issue, and can be solved technically. I don't think we should change deletion policy because of this issue - or if we do, then we should be clear that it's a temporary workaround rather than a permanent change.
We do have a policy to keep valid redirects. It's been regularly asserted in practice by comments on VfD, and removing articles from VfD, and in discussion on this page. It's been in bold at the top of VfD for ages. It's as much a policy as you can get on Wikipedia, which admittedly isn't saying much. Yes, some folks have been ignoring it. That is regrettable - it would be better if they explained why it should be changed. Martin 15:21 18 Jul 2003 (UTC)
I don't think searching can be solved technically, because many users search through google, and google is unlikely to magically delist the redirects. I personally support changing the deletion policy so only useful redirects are kept. --Delirium 16:45 18 Jul 2003 (UTC)
Actually, google does magically delist duplicates. "In order to show you the most relevant results, we have omitted some entries very similar to the 9 already displayed. If you like, you can repeat the search with the omitted results included". Martin 19:44 18 Jul 2003 (UTC)
Redirects should only be made for misspellings if they are common and logical. I don't think people mess up the word "boyer" so easily that they will type in one of the redirects. --Jiang 21:20 18 Jul 2003 (UTC)
I am open to the possibility or even probability that this comment is correct; I would just like to point out, however, that in nearly every case the redirects are from a spelling of my name that has seen print (note that I am not saying that this necessarily makes them useful -- I do not object in any way to their deletion). (And don't even get me started on the number of times I've been called "Daniel Blair" or "David Blair" or received mail so addressed! Something about my name leads people to "Blair" me!) --Daniel C. Boyer 17:55 20 Jul 2003 (UTC)
I agree with Jiang's comment. The number and nature of redirects needs to be reasonable. That should be part of the policy. I think search engines are only part of the reason why. I don't think we should depend on Google or future improvements to our own search engine. Even a rudimentary search engine (external or specific to Wikipedia) should be able to do reasonably well finding articles on Wikipedia. Roughly, I would propose: the number of redirects to any one article should be kept to a reasonable number which is roughly defined as the number needed to cover common searches, meet user expectations, and particularly common misspellings (not any misspelling). It is detrimental to Wikipedia to add unnecessary redirects (not to mention that someone could add numerous malicious redirects (like "John 'Genocide' Smith") to express a single POV). --Daniel Quinlan 08:19 19 Jul 2003 (UTC)
I think search engines are only part of the reason why.
What's the other part of the reason?
They lead to confusion (as in the case of Charles Boyer who by any sane standard is someone else),
I would agree that Charles Boyer is "by any sane standard... someone else." I find it offensive that you are revising my position in order to argue against it. The redirect in question is Charles C. Boyer; I have repeatedly explained what the reason for it is. Perhaps the reason does not justify the redirect, and if that is the case, I would not object to its deletion. But the reason for the redirect was generated by someone else, and it clearly was not my intention to cause confusion between myself and Charles Boyer (whom I don't think has ever been called confusedly by this name, though I could be proven wrong). --Daniel C. Boyer 14:14 21 Jul 2003 (UTC)
they are offensive (see below) primarily due to their advertizing and self-aggrandizing nature. They serve no good purpose that outweighs the downsides and abuse. --Daniel Quinlan 06:01 21 Jul 2003 (UTC)
Purely malicious redirects should be deleted, of course. However, note that Chemical Ali, and Comical Ali correctly redirect, despite expressing a POV. If a nickname exists and is not purely idiosyncratic, then it can redirect.
I agree with this, to an extent. I have not, for example, suggested removing the Dubya redirect even though it is POV. I think it's legitimate redirect given the common use of the name. It fits under my proposed modification to the deletion policy regarding redirects. --Daniel Quinlan 06:01 21 Jul 2003 (UTC)

pro-redirect case

Let's make the pro-redirect case: [2]

Most of these arguments are for redirects for pages that have moved or are reasonable. --Daniel Quinlan 06:01 21 Jul 2003 (UTC)
  • Redirects avoid broken links. Note that you can't use google back-links to prove that no links will be broken, because:
    • google only indexes about 1/4 of websites
      • this is actually an argument for being judicious with redirects, if anything. Don't waste google or any search engine's time or storage capacity by making it index/resolve/organize unreasonable or unneeded redirects. --Daniel Quinlan 06:01 21 Jul 2003 (UTC)
    • When google indexes a page, it stays indexed for (at least) a month: deleting a redirect means that google has a broken link to us.
      • these redirects were not added because a page had moved. (We're talking about 17 redirects added by one guy to his own non-user page.) --Daniel Quinlan 06:01 21 Jul 2003 (UTC)
    • people add bookmarks to pages.
      • these redirects were added after the fact, not because the page had moved. --Daniel Quinlan 06:01 21 Jul 2003 (UTC)
    • people email links to each other, or mention them in newsgroups.
      • not these... I think I'm done, you get the idea. --Daniel Quinlan 06:01 21 Jul 2003 (UTC)
    • people write URLS on scraps of paper.
  • Redirects avoid need for cluttering VfD or sysop superpowers. Redirecting scales much better than deletion.
  • Redirects keep history available. Where a redirect has history, this is good because:
    • It may be subsequently useful - example off the top of my head: list of dictators.
    • If content has been moved from one article to another, then it's probably more in the spirit of GFDL to keep the old article as a redirect, in terms of crediting authors. IANAL, so ignore this if you wish.
  • In cases of different spellings, or different ways of titling a page, if a redirect has been created once, then it's because at least one person browsed there. If it's been browsed once, it's likely to be browsed again. A redirect will put a smile on the second visitor's face.
  • It makes editing much easier when there are superfluous redirects, as you can put square brackets round any old word or phrase that you think is significant, and trust that it'll go to the right place. For example, if I don't know Kant's first name, it's not a problem, because the redirect will take the reader there anyway. Searching wikipedia to find the full name would take time.

As far as I can tell, the case against is:

  • clutters up search engine results. (internal and external)

That's a genuine problem, but I think there are ways to fix the search engine (and external search engines, via noindex, nofollow, noarchive, etc) such that we keep the many benefits of redirects. Martin 12:46 20 Jul 2003 (UTC)


I've tabled a suggestion at [3] that I think could circumvent a lot of the problems we are discussing here. I'd appreciate any comments. GrahamN 17:42 18 Jul 2003 (UTC)


I think you're missing a very important argument for redirect deletion:

  • Some people find the existence of certain kinds of redirects offensive

This is related to the search engine problem, because people have to look at the offensive redirects when they come up in searches. But it is not a "subproblem" -- the only way to avoid offence is to make the redirect completely invisible, and if it's completely invisible there's no point having it.

Examples that come to mind are the AKFD redirects, and Wikipedia:Wikicide and (arguably) the Daniel C. Boyer redirects. -- Tim Starling 13:00 20 Jul 2003 (UTC)

I hesitate to get involved in this, but... how could anybody possibly find the Wikicide or Boyer redirects offensive? --Camembert
Perhaps I'm using the term "offence" loosely. -- Tim Starling 00:20 21 Jul 2003 (UTC)


Aside from the search results, when have so-called "offensive" redirects ever caused actual offense? As you say, I see this as a sub-problem of the search engine - if there are examples of people taking offence in other ways, I'd be interested to know. Martin 13:44 20 Jul 2003 (UTC)
Hello, I'm offended... the search engine argument is largely technical for me. It only adds to the level of my offense. --Daniel Quinlan 06:01 21 Jul 2003 (UTC)
The point is that if you don't like seeing AIDS Kills Fags Dead slogan, you're not going to like seeing it wherever it comes up. People can probably tolerate it if they don't see it too often. I could be wrong about this. I just have a gut feeling that people have more than one reason for saying they want certain redirects deleted, but they tend to only say the more rational ones when questioned. -- Tim Starling 00:20 21 Jul 2003 (UTC)
It is a deviation from NPOV to say that we should limit the appearances of the AIDS Kills Fags Dead slogan due to its offensiveness. Sure, I find this slogan highly offensive, but that is a POV. The number of redirects should be based on their encylopedic usefulness and not limited due to their offensiveness, because if the standards of Wikipedia were not to have offensive articles, not only would we not have any article on AKFD, but there would be many more articles we would not have. --Daniel C. Boyer 12:55 21 Jul 2003 (UTC)
I agree that most redirects of this type should also be removed. --Daniel Quinlan 06:01 21 Jul 2003 (UTC)
I find almost all of the Boyer articles and material offensive -- aside from his primary non-user page, Daniel C. Boyer, which could easily include material on almost all of his advertizing articles about his works. It offends me that Wikipedia can be abused for the purpose of advertizing (see my recent edit to Connecticut) as well as self-aggrandizement.
  • self-aggrandizement: The act or practice of enhancing or exaggerating one's own importance, power, or reputation.
  • advertizing: the business of drawing public attention to goods and services
Many of his changes meet the definition of advertizing and/or self-aggrandizement. Advertizing is already against Wikipedia policy and self-aggrandizement should be, if it is not already, because it is itself a form of advertizing (if perhaps more indirect). If I were to, for example, add my own name to many of the Linux and Open Source articles (as I could, with as much justification as Daniel C. Boyer has had), I could be rewarded with additional consulting business just as Daniel C. Boyer's litter attracts attention to his artistic works. --Daniel Quinlan 06:01 21 Jul 2003 (UTC)
So fix it, dear Henry, dear Henry, dear Henry. So fix it, dear Henry; dear Henry, fix it! :) Martin
My name is not Henry... but I fixed Wikipedia:Deletion policy anyway. I tried to make it succinct and capture the things people seemed to agree on. Of course, in specific cases, there will always be disagreement (that's why there is voting). --Daniel Quinlan 10:25 22 Jul 2003 (UTC)
Well, I bet you don't have a hole in your bucket, either. :) I was referring to fixing Boyer's edits, but what you've done is good too. I disagree with some of it - I'll try to edit appropriately. Also, I think much of this should probably be addressed at wikipedia:redirect, just as this page simply points to wikipedia:copyrights for possible copyright infringement. Martin 14:29 22 Jul 2003 (UTC)

Above, I wrote:

When google indexes a page, it stays indexed for (at least) a month: deleting a redirect means that google has a broken link to us.

Daniel Quinlan replied:

these redirects were not added because a page had moved. (We're talking about 17 redirects added by one guy to his own non-user page.)

Note google:donnelly+Wikipedia (for example). #1 result (as of the time of writing) is a broken link, because the redirect has been deleted. It doesn't matter why redirects are added - the point is that if we delete them, we break links. Always. Martin 13:52, 31 Jul 2003 (UTC)

And? It's not our fault that google take a few hours/days to re-crawl wikipedia for changes. Besides, the page isn't gone, you can type http://www.wikipedia.org/wiki/Anything_here, and the page will exist. So, the link isn't broken, it's just that google points to an empty article for a very short period of time. This will always happen, and I don't see it as a problem. MB 16:47, Jul 31, 2003 (UTC)
Google crawls Wikipedia once a month. It's not a few hours or days.
It is our fault. We didn't have to delete this redirect. By deleting it, we've created a broken link. Martin
So what? Even if I cared about "breaking" a vanity link, which I don't, none of these redirects or articles were necessary. He was adding redirects for personal vanity, like a single misspelling on an external website that wasn't even a link, permutations of his name and initials, etc. Most of them were not internal Wikipedia links. And while it may be a short-term inconvenience to delete articles or redirects, that is the only reasonable way to handle advertizing and similar inappropriate content. The google:donnelly+Wikipedia #1 status is merely proof that Daniel succeeded in gaming the system to get better recognition of his works. If you don't like broken pages, let's temporarily redirect all such pages to a "page has been deleted because we're trying to provide better content" page. Daniel Quinlan 17:27, Jul 31, 2003 (UTC)
How the frig is that going to help? It still gives our customers a broken link. Sure, it's got a nice friendly error message, but it's a nice friendly error message that disguises a broken link.
I told you that this would create broken links. You claimed that it wouldn't. You were wrong. Martin 19:56, 31 Jul 2003 (UTC)

Martin, you are a bit confused. There are no broken links. A broken link is a link to a page that doesn't load. The page loads. There is just no content. What is wrong with this? It is either delete it, or keep it. You seem to be suggesting that we should never delete a page that is a redirect, b/c google crawls them. This is simply ridiculous. What if someone creates a redirect page "Martin is a fag", and then redirects it to User:MyRedDice, and google crawls it and adds it to the database? Opps, we can't delete it, that would make it a broken link. We can't make broken links, that would be a bad show to our "customers." It's just absurd. MB 20:17, Jul 31, 2003 (UTC)

Well, in my semantics, when you click on a link that you expect to take you to an article on Daniel C. Boyer, and it takes you to a blank page with no hint as to where to find the content you were looking for, that's a broken link. A matter of definition, of course - what would you call it?
I'm not saying that we should never delete a redirect. I'm saying that deleting redirects creates broken links. Any advantages of deleting a redirect will always need to be weighed against that disadvantage, even if on first sight it may appear that no links will be broken. Martin 20:37, 31 Jul 2003 (UTC)
Martin, I agree with your principle, there is a downside, but I need to rephrase it a bit. It's not the disadvantage vs. the advantage -- there is no advantage per se. Clearly, it would be better to not have to delete any redirects. Ideally, these damaging redirects would not be created in the first place. We're choosing between the lesser of two evils: allowing bad redirects to persist and be published, to be abused, to offend, to mislead, to grow over time, and encouraging further such redirects leaving the redirects in place ... compared against the cost breaking a few pages, pages we agree that we should have had in the first place. This is why I think we need to do the following:
  • Encourage and favor deletion of inappropriate and unreasonable redirects. (And I think sooner is better for these redirects, always. Get 'em before they get indexed, linked, before they affect additional users.)
  • Over time, work on technological methods to lessen the negative impact of deleted redirect via a phase-out strategy. Delete them promptly, but allow them to work for a brief transition period. (details of suggestion to meta:searches and redirects)
I don't think the phase out strategy is really needed since I don't think anyone really needs or uses a Joe is a jerk link in the first place (at least not often or as a preferred mode of operation); and while people might follow a "Donnelly" link off of Google, were they really looking for a Colorado college comic strip by Daniel C. Boyer? Daniel Quinlan 07:18, Aug 1, 2003 (UTC)
If they were searching for, say, google:donnelly+daniel+boyer? Could be.
I like to look at it this way: if someone goes to their address bar and types http://www.wikipedia.org/wiki/daniel_boyer, what's the most useful thing we can provide them? If a redirect is more useful than a blank page, keep the redirect. If the blank page would be more useful, delete the redirect.
Perhaps that's the difference: I navigate primarily by links and URL-hacking: other people here navigate primarily by searching. So I don't care that much about breaking the search engine (especially since it's disabled anyway! :), and presumably you don't care that much about breaking my URL-hacking. Martin 09:37, 1 Aug 2003 (UTC)
Actually, it was never disabled, the link was just removed. If you type in your seach in the search box and then hit enter, the default is not go, it is search. MB 13:53, Aug 1, 2003 (UTC)
Correction, it is disabled now. Oh well. MB 14:21, Aug 1, 2003 (UTC)

Deletion notice for redirects

First, I have a question about the "Listed for deletion" notice, as called for by Wikipedia:Votes_for_deletion, Wikipedia:Deletion_policy etc. It's stated that this has to be added to any page you list on VfD - but does this really apply to Redirect pages? (For one thing, you generally won't even see it unless you manually pull up text of the Redirect page.) I would assume this rule only applies to pages with actual content - if so, could someone make this clear on those pages?

The rule is now optional. Ignore it if you want.—Eloquence
Redirects have modified rules, as they have no content other than a computer-ese line, so no, the "Listed for deletion" notice is not listed --Menchi 06:35, Aug 17, 2003 (UTC)
Well, that's what I thought seemed logical, but that's not what the pages say. Shall I fix them so the exception for pages without content is clear? Noel 17:10, 18 Aug 2003 (UTC)
As if to make it plain that not everyone understands the unwritten policy, I've just noticed that someone recently edited one of the Redirect pages I listed on VfD to add a "this page is on VfD" notice! Unless someone thinks that it's a bad idea, I'm going to modify the policy pages (above) to make an explicit exception for pages with no content. Noel 18:43, 19 Aug 2003 (UTC)
I disagree and I have been adding the notice to redirects. There may have been content on the page before it was a redirect, and the author may still want to be informed that their page is being deleted. There may be no current content, but there could be content in the history. If the policy is going to be that you have to add a notice (which btw I don't think it should be), then I think it ought to apply to redirects as well. Angela 20:35, 19 Aug 2003 (UTC)
OK, how about excepting redirects with history from the exception? (Note that if the old page was moved, the history would have gone along with it.) Noel 00:01, 20 Aug 2003 (UTC)
Ok. The issue of adding the boilerplate message is also being discussed at Wikipedia talk:Votes for deletion by the way. Angela
Since we seem to be done here, maybe this section should get removed from VP and moved over there? I'll fix things to clarify about the notice. Noel 16:31, 20 Aug 2003 (UTC)
I just simply copy the history & content (I've yet to encounter more a stub) to the moving-in-article's Talk page before I delete, since the requesting person usually has good reason. Those redirects deserve to be deleted.
As for those redirects with no hist, a notice is unnecessary and impedes time efficiency. Again, as long as the requesting person provides good reason. --Menchi 20:54, Aug 19, 2003 (UTC)
I didn't mean redirects where there was a new page moving in. Those go through the fast track of VfD anyway, so the boilerplate text would not be used. I mean if a redirect is to be deleted and nothing to replace it. Angela

Second, about deleting Redirect pages which are created by typos. Currently Wikipedia:Deletion_policy calls for keeping Redirect pages, except "when the problems they cause outweigh their advantages"; Wikipedia:Redirect says nothing about real typos, but lists "Misspellings" as one use for them. I would argue that all Redirect pages caused by real typos on the part of people working on the 'pedia should be candidates for instant deletion, no debate needed. They just clutter up the database, and "What links here" pages.

By "real typos" I don't mean things like "Michaelangelo Buonarroti" instead of "Michelangelo Buonarroti"; someone who wasn't sure of the spelling could easily type the former. I mean things like "Michrlangelo Buonarroti", i.e. a real mistake that's just a slip of the finger.

No doubt someone's going to say (they already do!) "but it could be useful". So what? I could create thousands of different wrongly spelled versions of almost any page. There is just no need or use to clutter up the database with them. If we really want to make allowances for people who can't spell/type, we should add some sort of DWIM (e.g. like the one Google has) to suggest what people might have wanted if a search/lookup turns no matches.

Noel 06:18, 17 Aug 2003 (UTC)

  • If you want to "del redir to make way for move". It'll done by an admin within hours if possible.
  • Complete typo ("Presidenté Busch") in redir is also deleted.
  • Similar typo are preserved exactly because:
    1. we don't have DWIM.
    2. alternate and correct spellings, especially for ancient people and non-English personalities.
Anything I didn't cover? --Menchi 06:35, Aug 17, 2003 (UTC)
I'm not after alternative and/or very plausible spellings, I agree we should keep those. I'm only asking about "complete" typos in redirects. I would have thought those fell into the "shoot on sight" category, but several I have listed on VfD are still there, so I can only guess there is no agreed-upon policy about them. Noel 17:10, 18 Aug 2003 (UTC)


Useless Redirects in Talk:

From Village Pump 6 September 2003

I'd like to suggest that when we move pages which have associated Talk: pages, we ought to (in general) not keep a Redirect from the old Talk: page to the new Talk: page when there's no use for that link; i.e. if nothing links to the old Talk: page. I.e. when moving a page from Foo to Bar, if there is a Talk:Foo page it is moved to Talk:Bar, and a redirect to Talk:Bar would be left at Talk:Foo. If nothing actually links to Talk:Foo, that "tombstone" redirect will probably never be used (since clicking on the "Discuss this page" link on Bar will get you straight to Talk:Bar), and just clutters up the database. So is there any problem if we just delete them? (And perhaps someday, when we are knee deep in Developers and all bugs and really needed features have been seen to, the code will be changed to avoid automatically creating them when they are not needed. :-) Noel 23:54, 3 Sep 2003 (UTC)

Rule one of responsible web publishing: never knowingly break links to your pages, no matter how useless you think they are or how sure you think you are that no one's got them stored and is going to use them. Anyone who's deleting redirects really really needs to not do so. (Here, read this.) --Brion 01:00, 4 Sep 2003 (UTC)
What Brion said. I hand-type a lot of links to talk pages: unless you're planning to follow me around changing them all to point to the correct place, then keep them. Thanks. Martin 09:38, 4 Sep 2003 (UTC)
If you will note, I clearly made an exception for cases where "[anything] links to the old Talk: page". I think that covers this case, yes? Noel 18:25, 4 Sep 2003 (UTC)
No it doesn't. If you delete a redirect today, I may hand-type a link to it tomorrow. Hence my suggestion that you follow me around to ensure that I'm not adversely effected by your deletion. Martin 18:53, 4 Sep 2003 (UTC)
Hand-typed article names are in basically the same boat as off-site URL's - and see my reply to Brion about those. Are you also proposing that we never, ever, delete a page that ever had any non-bogus content? Noel 19:57, 4 Sep 2003 (UTC)
Wiki encourages links to articles which do not exist, indeed that is one of the published ways to create an article. In what way is a link to an article which could exist different from a link to an article which no longer exists? If there is no content, is there an article? Of course, links themselves carry information, so before deletion it is nice if someone ensures that the text with the links makes sense without the link. SEWilco 19:40, 5 Sep 2003 (UTC)
Hmm - you asked if there was any problem if you just delete talk page redirects. I'm answering: yes, there is a problem. The only problem you mention in respect of these links is "cluttering up the database", and Brion, who knows more about the database than both of us, doesn't think that's a problem. Thus, in the case of these particular talk pages, they should not be deleted. Martin 20:14, 4 Sep 2003 (UTC)
If so, you should also have the ability to reach through the internet and change people's personal bookmarks, and have a perfect web spider able to check every web page in existence to be sure there are no external links to the page. And be aware of all printed material and scrawled post-it notes in the world to make sure that no one, anywhere, has that URL. Only then should you delete a redirect. --Brion

Look, I understand, and sympathize with, the sensitivity to leaving dangling pointers out there in URL-space. I maintain more than one page of the form "this page isn't here any more, go <here>" because I reorganized some stuff; and I also really hate it when you follow a link and it 404's.
At the same time, delete logs show that Wikipedia clearly doesn't have a policy that "no Wikipedia URL that ever contained valid content (i.e. not just insults, rubbish, copyvio, or something like that) shall ever stop working". A small amount of trolling through deleted articles turned up "Beadwork patterns" and "Gold Faced Pumpkins", both of which contained real content at one point.
I understand why they were deleted, but that's not the issue: the point is that someone out there may have them bookmarked, and now they don't work anymore. In addition to them, I saw a whole series of "Emperor_<foo>_of_Japan" which are now gone too, moved somewhere else, with no redirects left behind (probably because the "What Links Here' page for them was empty, I would assume.) Again, someone might have saved URL's to them out there somewhere.
So, if what you're saying is that you want to have such a policy, that no Wiki URL that ever pointed to non-bogus content shall ever stop working, fine, but those aren't the ground rules that seem to be in place at the moment. Are you proposing such a change? Noel 18:25, 4 Sep 2003 (UTC)

I'm happy enough with the deletion advice currently at wikipedia:redirect, and I was reasonably happy with the prior deletion policy of "do not delete valid redirects", which we had for over a year. Those are the ground rules. Some sysops are either unaware of those guidelines, or choose to ignore them, which is unfortunate. Martin 20:14, 4 Sep 2003 (UTC)
Sorry, but discussing only the redirects doesn't answer my basic point: there are pages with real content - i.e. not junk, not insults, not copyvios - being deleted.
If "not orphaning pointers residing outside the Wikipedia database" (be they mental or electronic) is so important, why do we even have a VfD page? Junk and insults get deleted on sight, and for copyvio we can just blank the page. If not orphaning external pointers is so important, why do we ever delete anything?
I'm not trying to be difficult, I just don't understand what seems to me to be a disconnect between the goals y'all have stated here, and how things actually work. Noel 20:37, 4 Sep 2003 (UTC)


You seem to have conflated two separate concepts here. I apologize if I was unclear above; for clarity let me state my position again:
  • Links to material that Wikipedia continues to publish must continue to work unless there is a real reason they can't (such as, the www.wikipedia.org domain name is taken away from the project somehow and none of us can control what's at that address; or if the link was in fact an invalid URI in the first place due to a software bug and can't be maintained once the bug is fixed). This includes talk pages, and redirects to renamed talk pages are the way this link continuity is maintained in that case.
  • Corollary: links to material that Wikipedia no longer publishes (ie, garbage pages and copyright violations that have been deleted from the database) are free to turn up a blank slate. That is the purpose of deletion: to un-publish material that wasn't supposed to be published in the first place (because it's 100% garbage, or is legally unusable).
If you do see any deleted redirects to existing pages, please restore them or list them on Wikipedia:Votes for undeletion. --Brion 23:35, 4 Sep 2003 (UTC)
People seem to use VfD for purposes outside its ostensible aim; I'd say more than half, maybe more than three-quarters, of the stuff posted there should not be deleted. Many people seem to use VfD as a place to post articles that need work in the hope that someone will be inspired to 'save' them from deletion.
Is the problem that the only 'problem article' page anyone reads is VfD? --Morven 03:39, 5 Sep 2003 (UTC)

New comments can be made at Wikipedia talk:Deletion policy/redirects.