Wikipedia talk:Redirect/Archive 2013

From Wikipedia, the free encyclopedia
Jump to: navigation, search
Archive 2010 Archive 2011 Archive 2012 Archive 2013

Uncertain primary topic


I refer to such link targets as Eidos (edit | talk | history | links | watch | logs). Roughly speaking, a half of users expected to see there the Plato's theory, and another half expected to end with Eidos Interactive. After several edit wars we settled on a disambiguation page, but for 4 years the question was, virtually, remained open. Surely, any experienced user will not link to what currently is a dab or a dab redirect. But what about titles (possible link targets) whose primary topic is unreliable, like Eidos in 2008–2011? Incnis Mrsi (talk) 11:36, 7 January 2013 (UTC)

I'm not sure what you are asking here. Eidos seems like a textbook example of a case with no primary topic so the disambiguation page is placed at the undisambiguated title. Users will typically link to one of two places - 1) where the article currently is, 2) where they think the article is (e.g. where it was when they read it or where they would look for it). In either case, if they are unaware that the article title is in dispute then they will do exactly as above. If they are aware then they'll either point it to where they believe it should be, where it currently is or whatever the stable title looks to be. The best course of action is to link to a title that will always lead to the correct article, either directly or via a redirect. For example if you have articles that could use the title "Example", one a widget and the other a band, and there is a dispute about which is primary then you should link to Example (widget) or Example (band) as even one is later decided to be primary then there is nothing pointing at the wrong place. Does that make sense? Thryduulf (talk) 12:28, 7 January 2013 (UTC)
So, what about to recommend users to avoid problematic redirs, on WP:Redirect? Do not tell what you understood and what you didn't. Say whether was my addition good or bad. Incnis Mrsi (talk) 12:48, 7 January 2013 (UTC)
No it was not a good addition, because it is not at all clear what you mean. It might be that the intention is good, but until we understand what advice you are trying to give we cannot word it better. Thryduulf (talk) 13:16, 7 January 2013 (UTC)

Extension Needed


Is there any option or tool to create "n" number of redirects in single page. For ex. If I want to create a new article means defenitely I know the combinations which could be reach the word.

For ex: If u take the word - Tenkasi the user may search it following options,

  1. Tenkasi
  2. Thenkasi
  3. Tenkaasi
  4. Thenkaasi
  5. TNK etc.

Suppose if one article has more than 10 number of search possibilities means it is time taking to create 10 pages.

But if there is any tool to create "n" number of redirect options in single page means means it is very useful.--Tenkasi Subramanian (talk) 12:35, 7 January 2013 (UTC)

In most cases creating redirects automatically is frowned upon so there is no simple tool that I am aware of that can do that. Wikipedia:Auto Wiki Browser might be able to do it (I don't use it so I am not certain). You would need to list all the redirects you want created, ensuring that they should all redirect to your target - TNK for example already exists as a disambiguation page, so it would not make a good redirect. We do not create redirects from every possible romanisation, just ones that are in widespread use or are otherwise notable or likely search terms. Thryduulf (talk) 13:24, 7 January 2013 (UTC)


Just wondered - is there a way to deal with users who continually ignore this rule despite being asked to stop? Should they receive warnings and eventually be blocked? I'm also assuming that it's not helpful to revert someone who "fixes" such redirects. –anemoneprojectors– 18:50, 7 January 2013 (UTC)

If the redirect is more useful than the piped link, reverting misguided "fixes" is useful. Otherwise, if they ignore the Talk or User talk requests, that might be seen as a disruption and deal-able through the usual channels. There's no way specific to this situation, AFAIK. -- JHunterJ (talk) 19:01, 7 January 2013 (UTC)
Thanks. Some of the redirects in question have been reverted for a different reason. The person isn't even piping the links, but changing the text completely. The uw-disruptive templates say the edit has been reverted or removed, so I will do my own warnings based on them. –anemoneprojectors– 19:34, 7 January 2013 (UTC)
See WP:WARNING#Disruption and issue a {{subst:uw-disruptive1}}, escalating if necessary. --Redrose64 (talk) 19:36, 7 January 2013 (UTC)
So the next question is - should the NOTBROKEN section include some info on what might happen if users continue to edit in this way? –anemoneprojectors– 19:49, 7 January 2013 (UTC)

Edit request on 25 January 2013

Correct to the capitalization of 'James' in the name.

Melvin Anderson (talk) 08:38, 25 January 2013 (UTC)

Not done: this is the talk page for discussing improvements to the page Wikipedia:Redirect/Archive 2013. Please make your request at the talk page for the article concerned. If you are referring to the 'James' on Wikipedia:Redirect then I'm not sure what you want changed. Callanecc (talkcontribslogs) 08:52, 25 January 2013 (UTC)


Is this redirect valid? As far as I know we don't create those cross-wiki redirects. Tbhotch. Grammatically incorrect? Correct it! See terms and conditions. 04:43, 8 February 2013 (UTC)

No, for two reasons: (i) redirects shouldn't be piped; (ii) it's cross-project so won't work in any case. However, a Wikipedia:Soft redirect can be used, see here. --Redrose64 (talk) 12:26, 8 February 2013 (UTC)
Redirects (soft or otherwise) to other languages are not generally useful for English speakers and hide the fact that we are missing an article. Accordingly I'll nominate it for deletion later (I don't have time now). Thryduulf (talk) 13:01, 8 February 2013 (UTC)
I've found {{Interwiki redirect}}, which seems to be used primarily on User: pages. --Redrose64 (talk) 14:31, 8 February 2013 (UTC)

This has now been speedily deleted per WP:CSD#R2, but I don't think that criterion (or any other) applies. I have started a discussion about this at Wikipedia talk:Criteria for speedy deletion/Archive 48#Soft redirects to foreign-language Wikipedias. Thryduulf (talk) 16:22, 10 February 2013 (UTC)

New speedy deletion criterion for redirects

It has been proposed that redirects and soft-redirects to non-English wikis be eligible for speedy deletion under a proposed new criterion R4. Please contribute to the discussion at Wikipedia talk:Criteria for speedy deletion#R4 proposal. Thryduulf (talk) 17:09, 25 February 2013 (UTC)

Hard vs Soft redirect

Could I please get one (or more) people more knowledgeable in the use of redirects to come comment on an active discussion. A bit of history can be found on my talk page here, and the discussion moving forward is here. - TexasAndroid (talk) 13:01, 19 March 2013 (UTC)

redirecting from names not in an article

I recall redirects being deleted because they were from titles not named in a destination article, so since then I've added names directly into the destination article (which is about an organization and names organizational subunits), thus eliminating any need for a redirect that would be deleted anyway. But now there's been a complaint that the destination article is overly detailed, and I think those additional names are good candidates for deletion from the article, which means redirects from those names to the article would be useful, probably with the R with Possibilities template (technically, that's the proper template, but unless sourcing multiplies exponentially the so-called possibilities are unlikely ever to be exploited into an article). I'm thinking of adding comments into the redirects giving sources for the names from which the redirects would redirect. Is that adequate to prevent deletion? Thanks. Nick Levinson (talk) 15:25, 19 March 2013 (UTC)

I see no problem in including references at an article marked as "redirect with possibilities". It's not common practice, but that doesn't mean it's not a good idea; it's compliant with WP:PRESERVE. The usual place to put those references would be at the article target of the redirect, but if that's not a good place then the redirect article itself is acceptable. Diego (talk) 15:40, 19 March 2013 (UTC)
@Nick, I'm not sure I fully understand what you're suggesting, but in general it is a bad idea (c) to redirect readers to an article where there is nothing in that article to indicate why they were redirected to that article. olderwiser 15:47, 19 March 2013 (UTC)
Here's the problem: An article is about an organization. The organization has subunits. Some of of the subunits have multiple names, including official, unofficial, new, old, and possibly erroneous but where we aren't sure they're wrong. A news source cites some name or other. A reader of the news source looks that name up in Wikipedia. The reader should be taken to the Wikipedia article about the whole organization. If the whole-organization article names all the subunits by all the names sources use, the reader will find the article. But then the article is overly detailed and including all those names burdens readers, because that's a lot of names to wade through. Taking most of the names out will make the article easier to read, but then a search in Wikipedia by an obscure name of the subunit won't find the article. Redirects can solve that problem, but that means the redirects will be from organizational subunit names that, although sourced, may be unofficial, old, rarely-used formal names, or uncertainly erroneous, and those names probably don't deserve weight in the article. Names that stay in the article usually don't need redirects (the main exception being when a name is divided in the article, and that's unusual), but we want visitors to find the article from a name they know from a source they read, so we should permit redirects from sourced names that are not in the article. And if a name would be too unimportant for the article itself, the source for that name would also be too unimportant for the article, so, given a redirect's existence, the source should be in the redirect, as a comment. Thoughts? And should this guideline be edited? Nick Levinson (talk) 16:15, 20 March 2013 (UTC)
Again, I'm not completely sure I follow. Redirecting a term to an article that makes no mention of that term (or for which the connection is not obvious, as for misspellings or variants) would seem to be completely baffling result for most readers. A solution might be to have a separate list article for all the various sub-entities with the redirects not mentioned in the main article pointing to the list. olderwiser 16:22, 20 March 2013 (UTC)
Right; I can see misspellings and variant spellings (maybe) not being mentioned in the redirect target, but everything else should be, or else you've got readers trying to find information about "Foo" landing on article "Bar" with no information about "Foo". Deleting the redirect would serve them no worse, and might serve them better (if the search results are relevant, or by indicating that WP has no information yet on the sought topic). -- JHunterJ (talk) 17:24, 20 March 2013 (UTC)
We're talking about cases where the destination article is very relevant and should be displayed to the searcher, but where listing all the forms (other than variant spellings and misspellings) in the destination article burdens readers who have to wade through a large number of them and who (sometimes) complain about article length.
The list being separate from the article, how would it meet the notability guideline? I don't remember ever seeing such a list (I've seen other kinds of lists, but not lists of obscure and uncertainly erroneous names related to a single other article). Do you know of any such list?
The point about confusion is valid for some names, such as abbreviations, but often sufficient similarity would be apparent at the destination.
What do you think of these possible solutions?
  • A plausible similarity would justify a hard redirect even without the destination article stating the name, one advantage being one-click access from a searched-for name to its destination, while many other methods require two-step access, which visitors often dislike to the point of not pursuing the desired information, trying elsewhere instead.
  • If at least two names are similar to each other but not to a destination article's title, a disambiguator may do, since a disambiguator's list items can be annotated to explain relationships.
  • For a name that's too different from a destination or for a disambiguator (since a disambiguator can't link to just one article), create a soft redirect, but only if the soft redirect can display an editor's explanation (I don't know how to cause an explanation to appear, and apparently they'd have to be redesigned to allow this, perhaps in MediaWiki); if a soft redirect cannot be annotated, a hard redirect may as well be used.
  • Allow single-item disambiguators; they wouldn't really disambiguate but would look and function much like one, useful because it could offer an editor's annotation to inform a visitor.
  • Add to a destination article a section listing all the alternative names with sources, placing the section after the notes, references, and External Links sections, probably requiring a change to a policy or guideline, so that weight would be irrelevant and sourcing would be optional (editors could supply citations if available but not tag as missing citations or could apply a lower-priority tag instead); this allows one-click access but may make an article ponderously long.
  • Create a separate list but waive the notability guideline and the weight policy (but, for nearly-weightless items if numerous, perhaps require sourcing); probably it would have to be defined as a special kind of list for a waiver to fit.
Nick Levinson (talk) 14:06, 21 March 2013 (UTC)
This is all very difficult to follow in the abstract. Are you suggesting new rules? I don't think I agree. Anyone can create redirects (and there are undoubtedly many created in good faith at a given point in time that make little sense now after subsequent edits to an article). I think if someone proposes deleting a redirect, it is up to you or whomever see some value in the redirect to present a plausible rationale. olderwiser 14:18, 21 March 2013 (UTC)
A speedy-delete is easy to miss. We need some solution, because creating pages that editors think should be deleted is not a good idea. How about if a redirect includes a comment, such as for a source for the redirected-from name? Would that likely be seen by someone proposing to delete? I've commented in redirects in the past. Nick Levinson (talk) 15:01, 21 March 2013 (UTC)
If a redirect meets the criteria for speedy deletion, I'm not sure that any amount of comments in the redirect could salvage them. If the speedy criteria are being misapplied, you might want to discuss with the editors you think are misusing CSD. olderwiser 15:15, 21 March 2013 (UTC)

Ten changes

The diff isn't very helpful, and of course I must explain directly the changes. I made:

  • the start of sentences and paragraphs state the object of interest.
  • examples the start of paragraphs, as the verbalizations alone were necessarily complex.
  • word-logical contradictions disappear such as between "never" and "rarely", e.g. "no reason" and "few reasons".
  • the key points and the emphasizable points italicized.
  • "change (bypass)" to "bypass or change". We'd just been talking about "bypass", but the bullet list has reasons not to change them.
  • "visible" into "visibility", and added "readability".
  • "(See {{tl:Redirect with possibilities}}) into its own sentence.
  • "page source form" into "wikitext"
  • "Barbaz" and "Foobar" to "A" and "B"
  • "does not work on redirects" to "does not work with redirects"
  • "hint that appears when a user hovers over " to "mouse-over text"
  • "non-piped links" to "piped links" by changing its location in the sentence.
  • "24 hours a day" was added to describe the "what links here" tool
  • the term "label" was applied a few times, where it was never applied

To be fair, I went through the whole section. This should make it seem like a single "style" if one is detectable. (I started out just to fix the grammar "linking to redirects to articles", and ... well, all that.)

I did heavily rewrite the very important explanation about when never to use redirects in the navigational templates, removing "exception to this exception" to do so. But I made the tone consistent with the context there, using emphasis where possible, and starting with the objects of interest.

I left the policy wording and content intact, but I had to remove one bullet that said titles might be moved due to redirect popularity:

  • If editors persistently use a redirect instead of an article title, it may be that the article needs to be moved rather than the redirect changed. As such the systematic "fixing of redirects" may eradicate useful information which can be used to help decide on the "best" article title.

Article titles are chosen by WP:Article titles policy, specifically the provision WP:COMMONNAME. This policy tells us to entitle our articles using the same words and phrases most commonly used by our sources.

Have a good day. — CpiralCpiral 09:35, 3 April 2013 (UTC)

I reverted the change, as on the whole I found it much confusing. And I think it introduces some outright inaccuracies. For example, what since when do piped links make the What links here service run slower? And I find use of the term "label" is extremely confusing. The placement of the last sentence in the edit to the navigational template exception seems to completely reverse the intended meaning. olderwiser 11:29, 3 April 2013 (UTC)
Please admit the current version does violate the title policy, and there are grammar and syntax errors, and that it just might make stylistic sense to edit sections as the whole in some proposal mechanism, rather than in parts. Thank you — CpiralCpiral 18:09, 3 April 2013 (UTC)
I'm not sure I agree there's any "violation" per se. There's certainly possibility for improvement though. I hope you don't mind, but I will add before and after for each of your proposed changes below to make it easier to compare for discussion. olderwiser 19:15, 3 April 2013 (UTC)
Respectfully, I for one will studiously react below. {{RfC}} anyone? — CpiralCpiral 06:44, 4 April 2013 (UTC)

I have to say, I'm not sure I think that any of these changes are improvements. Croctotheface (talk) 07:40, 7 April 2013 (UTC)

Proposals for lead paragraphs

ID Current Proposed
1 There is nothing inherently wrong with linking to redirects to articles. Some editors are tempted, upon finding a link to a redirect page, to bypass the redirect and point the link directly at the target page. While there are a limited number of cases where this is beneficial, it is generally an unhelpful exercise, and it can actually be detrimental. It is almost never helpful to replace [[redirect]] with [[target|redirect]]. Some editors are tempted to do this upon finding a link to a redirect page: to bypass the redirect and point the link directly at the target page, yet with the same link label. Don't bother. It is generally an unhelpful exercise, and it can actually be detrimental, although it is not inherently wrong to label the link to an article with an existing redirect.
2 With a few limited exceptions, there are no good reasons to pipe links solely to avoid redirects. It is almost never helpful to replace [[redirect]] with [[target|redirect]]. Unless the redirect is broken, there is no good reasons to pipe a link solely to avoid a redirect.
3 It is likewise unhelpful to edit visible links for no reason other than to avoid redirects. That is, editors should not change, for instance, [[Franklin Roosevelt]] to [[Franklin D. Roosevelt]] just to "fix a redirect". This rule does not apply to cases where there are other reasons to make the change, such as linking to an unprintworthy redirect.

Editors should not change, for instance, [[Franklin Roosevelt]] to [[Franklin D. Roosevelt]] just to "fix a redirect". It is unhelpful to edit the visibility of a link for no reason other than to avoid redirects, because redirects are designed to enhance readability. This rule does not apply to cases where there are other reasons to make the change, such as linking to an unprintworthy redirect.

Discussion for lead paragraphs

Comments on (1)

  • I would rather lead with the statement There is nothing inherently wrong with linking to redirects to articles. This really should be the first thing people see when referred to WP:NOTBROKEN as some editors seem to assume that links through redirects are somehow "broken". The proposed text implies it only concerns piped links, which is really a sub-topic.
Section 11 of an otherwise holistic structure Wikipedia:Redirects is approached by shortcut WP:NOTBROKEN, OK, that is our premise, as you wish. Then the lead paragraph is the lead. In that case I concede (1) starts correctly with the clause quoted. They should see that.CpiralCpiral 06:44, 4 April 2013 (UTC)
The current phrase "linking redirects to articles" requires that "linking" mean "typing the page name of a redirect as the label in a pipe trick". If "link" can mean "page name of the target", and it can, and if "linking a page" can mean "typing in a page name", and it can, then "linking a redirect" would mean "typing in a redirect page name", which is fine except for the pipe. A pipe changes the link structure to having the label on the RHS of the pipe. It's on the label side and not the link side. "Labeling" a link is what the pipe trick does. This is true because the link is basically the target page name or redirect page name. So "linking" here must mean "typing the page name of a redirect as the label in a pipe trick". You'd better say "piping redirects to articles" or else the current phrase tries to give "linking" a new definition. — CpiralCpiral 03:57, 5 April 2013 (UTC)
Being a lead, it might give a targeted definition to redirect first. I'm not offering that precise definition just yet, but I want to get some terminology done first. A link label is simply the page name of the target. A redirect is "the equivalent page name" to the target. There could be many such equivalent page names, redirecting to the same target. The redirect is a link label. There are many labels for the link to the one target. None of them will need the pipe trick if a redirect is used instead. Redirects reduce pipage, which is an extra load for bots and an extra load for editor entering the wikitext and the editor reading the wikitext. Since Wikipedia could be written by both old professors and school children, redirects allow the latter's slang while reducing pipage and increasing wikitext readability. Redirects are also a single point of correction when the name of a section heading changes and {{anchor}} is not used. — CpiralCpiral 03:57, 5 April 2013 (UTC)
  • As I already mentioned (and will apply throughout), I think using the term "label" is confusing. What does it refer to? Is it the visible text (which could be read as being the visible part of a piped link).
Seen Link label in the 9 yrs? I will apply "What precisely do you mean?" here. — CpiralCpiral 06:44, 4 April 2013 (UTC)using the text that just happens to also be a redirect page name, as a link label".
TBH, no, I had not seen that before now. I'm not sure why we should assume other readers have either. The term is not otherwise used on this page and to my knowledge it is not commonly used on other policy, guideline or style pages. And given you intend that meaning, I think you are misusing the term in some cases here. olderwiser 12:51, 4 April 2013 (UTC)
A "link label" is simple the label of a link. A "link label" is (simply) the page name of the target. A redirect is "the equivalent page name" to the target. There could be many such equivalent page names, redirecting to the same target. The redirect is a link label. There are many labels for the link to the one target. None of them will need the pipe trick if a redirect is used instead.(1)"Redirects reduce pipage, which is an extra load for bots and an extra load for editor entering the wikitext and the editor reading the wikitext." (2)"Since WP is written by old professors and edited by intermediate school children, redirects allow the latter's slang while reducing pipage and increasing wikitext readability." It seems to me almost nothing else need be said except the relation to What links here. This belief arises when I consider that the person who makes the "fix a redirect" mistake simply lacks the knowledge of what a redirect is. I for one lack the knowledge that the mistake is so common that it needs a shortcut. But that is quickly filled by seeing the equivalent to [[page name|redirect]]. — CpiralCpiral 03:57, 5 April 2013 (UTC)
It's always an audience thing, and if you are the audience, perhaps you're right that some English words should not be used on this policy page for it is confusing to you. To me, the true audience because I'm not conditioned (biased) because I'm a newcomer, the current version could use your fullest attention. Other guideline pages have their own style and tone. Subjects bleeds from titles into the style and tone. (Bold is bold; the Five pillars has a simple, screen-sized, sturdy columnar layout; 3RR is warlike: not plain, direct, unambiguous, or specific, (despite policy and guideline style), and starts with and carries on about punishments; the MoS wordiness and length is highly appropriate and acceptable, and ends with six sections of fashion; WP:Copyrights is colorful, like the colorful and dangerous creatures in nature.) And Redirect is hard to get to to edit changes to it, as are redirects themselves. Here where the subject is "If it ain't broke don't fix it" the combination of natural author-bias plus the pre-conscious conditioning of the title, would seem to my task doubly if not triply difficult. — CpiralCpiral 03:57, 5 April 2013 (UTC)
  • I think the second sentence in (1) is less confusing in the current version. Not saying it couldn't be better, but I think the current version is better than the proposed.
Meaning means there is a concrete (less of the abstract) referent. You express intension here. That you have in plenty. Got anything else? Got extension? — CpiralCpiral 06:44, 4 April 2013 (UTC)
I've no idea what you said here. olderwiser 12:51, 4 April 2013 (UTC)
I'm pretty sure it means "please be more specific than 'the current version is better' ". I will try to be also, more specific. — CpiralCpiral 03:57, 5 April 2013 (UTC)
The proposed version's merit is "the discovery of audience". Because I am both a highly experienced editor who is also a newcomer to this page, then that feedback is priceless, because the conditioned mind rarely can escapes its bias for the text they've processed six to eight or more times. (Drafts are put away for years before the final polishing and publishing.)
Now, by starting with a concrete example (ala [[page name|redirect]]), I was able to get that newcomer mind on exactly the structures it sees all the time in its own wikitext. If I try to scan the original version, it's all words, no pic, and the words are all abstractions, a "phraseology" that once I learn, entrenches me (conditions me) in an almost inescapable bias as well. The current version is slow going at first because one either processes an abstract phraseology like that, or skips it. This cannot seem true for the authors themselves unless its been years since they've seen it. It cannot be true for others entrenched "in the know" of a select few. Can you see how "linking redirects to articles" might be too slippery for a lead phrase? — CpiralCpiral 03:57, 5 April 2013 (UTC)
The proposed version's second sentence "Some editors are tempted to do this [keyword "this" referring to the concrete example in the first sentences, that they are all familiar with]", I never have to say again later "unhelpful". To repeat something complex and important is good writing, but check your own premise: "there is nothing wrong with it".
The proposed ":" in the second sentence introduces "phraseology" to repeat what is already understood. That is flow. It is fast, and it works for newcomers more quickly than starting with phraseology. (I've noticed the best template documentation start with examples. Much practical math studies are gained only by the examples, damn the terminology, esp. if I'm scanning and like some neutrino, blasting past material without really needing to understand the nature of human language symbols that are those terms and phrases.) I would just take out the proposal's "Don't bother", and go with it for the reasons stated. — CpiralCpiral 03:57, 5 April 2013 (UTC)
The current use of "With few exceptions, there are no good reasons..." starts out with a hanger, something that taxes the mind to remember until it is explained. It is never explained. So I make a summary sentence starting "Unless a redirect is broken...". There that is the explanation. And there is no need to ever form mental clutter. Furthermore, the repetition is almost too much. It is easy to grasp what part of "no" means. It means don't do what that example in the first sentence said. Anytime you have a one-sentence paragraph, that means that sentence is very important. OK, but why say something that is very important and very simple, over and over (and over) again? Because the reader is dull, or in a hurry, of course. But check your own premise: "there is nothing wrong with it". Exactly why I just do the example, and let the "grave importance" be the authors' own misapprehension. — CpiralCpiral 03:57, 5 April 2013 (UTC)
  • The phrase Don't bother strikes an odd tone for a guideline page.
There are no rules (for tone). Rather, style seem to dictate to circumstance. If you say the entire article's tone is off with that, I can see conceding (1) as a whole. — CpiralCpiral 06:44, 4 April 2013 (UTC)
Ah, well in that case, I don't think that tone is appropriate. olderwiser 12:51, 4 April 2013 (UTC)
You're right. Now I see where that tone was "the mistake of being not familiar enough" on my part. — CpiralCpiral 03:57, 5 April 2013 (UTC)
  • although it is not inherently wrong to label the link to an article with an existing redirect -- in addition to the confusing use of "label", the placement of this phrase makes the entire sentence rather unintelligible. -- olderwiser 20:28, 3 April 2013 (UTC)
Rather less intelligible than...
The "phrase" label: Label is placed there as a verbal (a verb that is a noun). "Don't label me." "--OK, but only if you won't call label me a labeler ever again!" In 1905 Peirce coined the new name pragmaticism "for the precise purpose of expressing the original definition" — CpiralCpiral 06:44, 4 April 2013 (UTC)
Well, no I meant rather unintelligible. It contains words that I recognize, but I am unable to parse the precise meaning. olderwiser 12:51, 4 April 2013 (UTC)

Comments on (2)

  • Grammatically, this really should be are no good reasons not is.
I concede (2). Forget it that way. It's wrong. — CpiralCpiral 06:44, 4 April 2013 (UTC)
  • There ARE exceptions that are not necessarily "broken", so why say there are "no good reasons" to pipe a redirect?

-- olderwiser 20:28, 3 April 2013 (UTC)

Comments on (3)

  • I think the restructuring shifts the emphasis -- as is, it places more emphasis on the DO NOT aspect rather than the parallelism with the current (2).
  • because redirects are designed to enhance readability -- where does this come from? Redirects were "designed" (if that term can really be used accurately with regards to such early features) primarly for ease of navigation. In some cases, they might result in enhanced readability, but that is neither a given nor an inherent characteristic.
We did not design them, we can't say for sure, but we think the link label was meant to "be read", as in "readable", or in general, have "readablility". That's where I'm coming from. — CpiralCpiral 06:44, 4 April 2013 (UTC)
So this is even more confusing. In a piped redirect, the piped text is the link label. So how does the redirect enhance readability? If someone links to a cryptic redirect such as HIRDLS, does that necessarily enhance readability? My point is that readability is not an inherent characteristic of redirects. With the origins of wikipedia in wiki software, I'm pretty sure the purpose of redirects was to make it easy for people to edit pages and make links. Enhancing readability might apply to piped links, not to redirects. olderwiser 12:51, 4 April 2013 (UTC)
I see. You mean "the rendered page will read the same whether or not a redirect exists". You're right. But what if I could have meant "an editor reading the wikitext"? Then we would be right to say "redirects enhanced readability". Redirects are transparent as a concept (or discussion topic?) to readers of the rendered page. — CpiralCpiral 03:57, 5 April 2013 (UTC)
The source of our confusion, and the personal responsibility (here it is a duty) to face the real cause of it, to improve the results of all this work here is this: "Redirect" (and "edit" and "link" and "label" and "reader" and so on) means two different things: one for the software, and then one for the editor. (The editor borrows the software lingo to redefine the same things.) You said "redirects make editing and linking easier". But I know the MediaWiki redirect cannot possibly make MediaWiki linking, or MediaWiki editing any easier, (MediaWiki titles make editing and linking easier because the MediaWiki page name is the MediaWiki database key for the query). The term "redirect" has two two definitions, one for each layer of abstraction, and only two. Terminology itself has three, the third layer being "talking about talking". All this abstraction business is the source of our confusion is entirely habitual conditioning, and hardly ever needs mentioning to produce good product. We are perhaps in terminology here, and if so, we must be careful to discern these levels. We must be distinguished editors.
If you grant the above, then we were both right about readability, just talking about the wrong reader. I hope I have cleared the confusion. — CpiralCpiral 03:57, 5 April 2013 (UTC)

Proposals for reasons not to change

ID Current Proposed
4 Reasons not to change (bypass) redirects include: Reasons not to bypass or change a redirect include:
  • Introducing unnecessary invisible text makes the article more difficult to read in page source form
  • It makes the wikitext harder to read, introducing unnecessary invisible text.
  • Non-piped links make better use of the "what links here" tool, making it easier to track how articles are linked and helping with large-scale changes to links.
  • The "what links here" tool runs 24 hours a day, and piped links slow it down. That tool makes it easier to track how articles are linked, and it makes helpful changes to links system-wide.
  • Shortcuts or redirects to sections of articles or of Wikipedia's advice pages should never be bypassed, as the section headings on the page may change over time. Updating one redirect is far more efficient than updating dozens of piped links.
  • Updating one redirect (or shortcut) to a section of an article or of Wikipedia's advice pages, should that section title ever change, is a far more efficient alternative than bypassing a redirect or shortcut. Section headings on the page may change over time, and that would mean updating dozens of piped links, verses one redirect.
  • Intentional links to disambiguation pages always use the title with "(disambiguation)", even if that is a redirect.
  • Intentional links to disambiguation pages always label the link to include "(disambiguation)", even if that is a redirect.
  • If editors persistently use a redirect instead of an article title, it may be that the article needs to be moved rather than the redirect changed. As such the systematic "fixing of redirects" may eradicate useful information which can be used to help decide on the "best" article title.

Discussion for reasons not to change

  • The phrase 'Non-piped links make better use of the "what links here" tool, making it easier to track how articles are linked and helping with large-scale changes to links.' says nothing at all about speed. Why is it being altered to suggest that there is a performance impact? --Redrose64 (talk) 20:50, 3 April 2013 (UTC)
"Better use of" means "direction" and for a computer, that means work, and work means speed. For example, why can't a category page be updated immediately upon saving a categorization link? Why the incessant warning "this may take a while"? What is "a lot of work" for a computer? Time. Maybe I just misunderstood. It is an opinion, and I appreciate your answer. Thanks. — CpiralCpiral 06:44, 4 April 2013 (UTC)
If you edit a page to add/amend/remove categories, the category pages are updated immediately upon save.
However, if you edit the <includeonly>...</includeonly> portion of a template to add/amend/remove categories, then when you save that template, those pages transcluding that template do not have their categories updated straight away, but are added to the job queue in order to spread the server load. Once each individual page is processed, it's exactly as if you had gone to the individual page and saved an edit (any edit: see also WP:NULLEDIT). Processing of one template in the job queue can take a matter of seconds, more typically it's done in minutes, but I have known it to take some hours (it's only taken more than a day if the job queue has been stopped for some reason). --Redrose64 (talk) 14:14, 5 April 2013 (UTC)
" At Wikimedia, the job queue is, in practice, almost never zero." -- Job Queue
  • The proposal for (6) seems less direct and placing the phrase introducing unnecessary invisible text as a subordinate clause at the end makes it seems almost a non sequitur; I think the current version more clearly explains this reason. olderwiser 00:24, 4 April 2013 (UTC)
non-sequiturs make for "I knew it was going to say that" flow, which is speed (efficiency). Keep. — CpiralCpiral 06:44, 4 April 2013 (UTC)
Hmm, last I knew, non-sequitur meant it does not follow and described illogical statements. Not sure how that improves anything. olderwiser 12:56, 4 April 2013 (UTC)
Your layout is excellent. For one thing, it makes precision easy, but your moderation is heavy. — CpiralCpiral 06:53, 7 April 2013 (UTC)
  • (7) ditto what Redrose64 said and what I said earlier. Further, where does the statement and it makes helpful changes to links system-wide come from? What does the what links tool do this? olderwiser 00:24, 4 April 2013 (UTC)
Comes from my reinterpretation of your "large-scale" phrase, which is, BTW, less specific? — CpiralCpiral 06:44, 4 April 2013 (UTC)
Found MW:pagelinks schema, can't A your other Q, out of my league (yet). — CpiralCpiral 06:53, 7 April 2013 (UTC)
  • (7) However, to be fair, the current version is also rather opaque. The reason was added some time ago after this discussion. But the gist of it is that a link to a redirect, for example DeskSet should not be changed to a piped link such as [[SunView#DeskSet|DeskSet]]. because there is value in being able to differentiate between links to "DeskSet" and links to "SunView" in the what links here tool. olderwiser 00:24, 4 April 2013 (UTC)
About that discussion, and as its content applied to the project of improving the literacy and content of the version now: 1)WP:redirect WP:red link, and WP:dab: Redirects, red-links, disambiguation pages, these all go together in a way that needs some technical understanding that should be "mild repetition" on all three pages. (Just what that is I don't know yet. I'm still reading up. I was only here for the language, and am not prepared for the kinds of issues you bring up.) 2)The only misguided soul who thought "fixing redirects" was an issue mainly needs the definition (plus extras for fussy technical types). (Why? For the same reasons dab pages are exposed here. That's why it could easily and often primarily be perceived "this entire section [existence] seems a bit stupid ignorant". 3)I'm sure all this stuff is simple, but my time to read all these discussions has become extremely limited. I came here for what seems poor language, just superficial, reader-oriented stuff, and I'm getting a history and technology lesson (and politics). Thanks. — CpiralCpiral 06:53, 7 April 2013 (UTC)
  • (8) Sorry, but I find the proposed revision to be far less intelligible than the current version. This section is about reasons not to change redirects, and this leads with instructions on how to do something that while being part of the rationale is only tangentially related. Changes to shortcuts are not that common; the guidance should instead lead with instruction to not change links to shortcuts (redirects). olderwiser 00:24, 4 April 2013 (UTC)
You repeat a general negative, whereas, of course one who hears that who also has a will, will love themselves specifically, and change for specific reasons, not vague ones. I love what I wrote, and I tolerate what is not what I wrote. Where's the third person who reads us both? The audience is the higher middle. — CpiralCpiral 06:44, 4 April 2013 (UTC)
Your love for what you wrote doesn't make it any more intelligible. Sorry. olderwiser 12:56, 4 April 2013 (UTC)
The language is ambiguous to me. I'm currently looking at redirect prefix:Wikipedia talk:Disambiguation/ to try to get up to speed here. — CpiralCpiral 06:53, 7 April 2013 (UTC)
  • Regarding the removed reason, I don't agree that this violates WP:AT. Of course, WP:AT governs, but the chosen title might not be the best for any of a variety of reasons, though that might not be immediately apparent. If an editor comes across a redirect with a disproportionately greater number of links that the actual title, that can be a signal that perhaps the redirect might be a better article title. That alone should not justify a move, only a suggestion that there might be something amiss. olderwiser 00:24, 4 April 2013 (UTC)

For (7), I think the idea is that if you go to look at the "what links here" page for an article (say, the page for my essay), the tool will divide out the pages based on how they "link here." So you can see that most people link to my essay with WP:JUSTAGUIDELINE and not, say, Wikipedia:Follow guidelines. For an article, if most people link there using a certain redirect, that provides useful information. If you pipe those links, that information is no longer present in "what links here." Croctotheface (talk) 07:37, 7 April 2013 (UTC)

Proposals for exceptions

ID Current Proposed
10 Exceptions:
  • The virtual namespace Media: does not work on redirects; that is, if File:Barbaz.jpg is a redirect to File:Foobar.jpg, then [[Media:Barbaz.jpg]] will not work and must be replaced by [[Media:Foobar.jpg]].
  • The virtual namespace Media: does not work with redirects; that is, if File:B.jpg is a redirect to File:A.jpg, then [[Media:B.jpg]] will never work. ([[Media:A.jpg]] has always worked.)
  • In many cases it is preferable to change redirected links in navigational templates, such as those found at the bottom of many articles (e.g., {{US Presidents}} at the end of George Washington). In this case, when the template is placed on an article, and contains a direct link to the same article (rather than a redirect), the direct link will display in bold (and not as a link), making it easier to navigate through a series of articles using the template. There are exceptions to this exception: where a redirect represents a distinct sub-topic within a larger article and is not merely a variant name, it is preferable to leave the redirect in the template.
  • In navigational templates (such as those found at the bottom of many articles, e.g. {{US Presidents}} at the bottom of George Washington) it is preferable to change redirected links. When the template is placed on an article, its own link should display in bold text, not as a link to itself. But when a navigational template link goes to a redirect, then it will look like a link, even when it's a link to itself. However, where a redirect represents a distinct sub-topic within a larger article and is not merely a variant name, it is preferable to leave the redirect in the template. Having that one label be a bold-text font makes it easier to navigate through a series of articles using navigational template.
  • It may be appropriate to make this kind of change if the hint that appears when a user hovers over the link is misleading.
  • If the mouse-over text for the link is misleading, it may be appropriate to make this kind of change.

Discussion for exceptions

  • (10) I don't care much whether the examples are changed or not. I'm fine with either. But for the change in the latter part, I think it is more helpfully phrased as what to do to "fix" the problem if encountered instead of bland assertions about what never worked and what always worked. olderwiser 00:52, 4 April 2013 (UTC)
The section is about the exceptions to not "fixing" links to redirects (that are not broken). i.e. what does not have to be done at all. So saying what "must be replaced"? They were never made. They never have to be unmade. There is no fixing. Or else just say "delete these links to redirects" and don't mention what has always worked. why talk about "fixing" what is not made that is broken? Just say what is wrong, say "Don't even make links to redirects", and be done. Or just say "if you see this delete it". Why put both injunctions? Parenthetical reference to what is right is good enough. Please keep this proposed change, thank you. — CpiralCpiral 06:44, 4 April 2013 (UTC)
Huh? (scratches head). Again I've no idea what you've just said. I'd like to think I'm reasonably competent reader, but honestly I really don't understand your point here or how it relates to (10). olderwiser 13:06, 4 April 2013 (UTC)
(10) is no longer true; since I added that with this edit the MediaWiki software has been changed. I believe that the change occurred in the last few weeks (three months, max) so that the Media: prefix now works through redirects. See for example Media:A-104.jpg which formerly would fail. --Redrose64 (talk) 16:56, 4 April 2013 (UTC)
  • (11) The proposal seems to me to be more confusing than the current version, although there is room for improvement. Specifically, can you really think the language about one thing not looking like a link to itself and that the other will look like a link, even when it's a link to itself is clear for someone not already familiar with the phenomena? olderwiser 00:52, 4 April 2013 (UTC)
(Ignoring the general negative, answering the specific.) I was thinking of an audience unlike us, who is not clear, and who needs clarity from the read. A brand new audience. Now, a link to itself is automatically bold-text. You must say "a virtual link to itself" and that is what I said in other words. — CpiralCpiral 06:44, 4 April 2013 (UTC)
Really? Clarity? I dare you to put that statement to a well-constructed usability test. olderwiser 13:06, 4 April 2013 (UTC)
  • (11) Also, I don't understand why this sentence Having that one label be a bold-text font makes it easier to navigate through a series of articles using navigational template. follows the sentence about not fixing a variant name? What does that one label be a bold-text font refer to? Not anything in the preceding sentence. olderwiser 00:52, 4 April 2013 (UTC)
It was said that the navigation would be better if the bold-text was there. I agreed with that and emphasized that. If the article is one of the links, and if its link is bold, well... that helps. — CpiralCpiral 06:44, 4 April 2013 (UTC)
  • (12) This is a problem with the current version as well -- what does this kind of change mean? Could we perhaps give an example here? olderwiser 00:52, 4 April 2013 (UTC)
That language "this kind of change" means the same thing in both versions, mine and the original? — CpiralCpiral 06:44, 4 April 2013 (UTC)
Right, which is why I said This is a problem with the current version as well. olderwiser 13:06, 4 April 2013 (UTC)

Self-redirects – how to find one

I have a disambiguation notice re the biography Michael Moorcock, User talk: P64#Disambiguation link notification for April 18. That's one week old and another editor has meanwhile fixed my offending link among others.[2] No problem.

According to Dablinks the biography does contain two self-redirects:[3] one to Moorcock and one to James Colvin (pseudonym). Both those pages are untargeted redirects to the biography. As I understand this guideline, both should be eliminated, Wikipedia:Redirect#Self-redirects.

Examining the biography code I find the latter, not the former. "What links here?" at Moorcock confirms the former.[4] What am I missing?

--P64 (talk) 21:01, 25 April 2013 (UTC)

I noticed this myself a while ago... I think the {{Redirect}} hatnote template adds it, somehow. --Joy [shallot] (talk) 23:08, 25 April 2013 (UTC)
Apparently Joy is correct that template {{Redirect}} causes Dablinks to report "1 redirect which point back". This week I have another at Caldecott Medal.[5]
The report by Rdcheck is also confusing to me.[6]
--P64 (talk) 17:35, 5 May 2013 (UTC)

Restoring links to redirects

We have editors removing the piped links to the actual article desired [7]. Was this the purpose of Wikipedia:Redirect#Do_not_.22fix.22_links_to_redirects_that_are_not_broken? And if not should we clarify? Doc James (talk · contribs · email) (if I write on your page reply on mine) 17:17, 4 March 2013 (UTC)

No, piped links are OK, unless the redirect is a "redirect with possiblities". Redirects that might become articles of their own should be used instead of piping. Replacing piped links with links to "simple" redirects (alternate names, for example) is pointless, or at least not the point of WP:NOTBROKEN. Clarification may be in order, yes. -- JHunterJ (talk) 17:40, 4 March 2013 (UTC)
I have clarified the section in question. Feel free to improve further. Doc James (talk · contribs · email) (if I write on your page reply on mine) 08:02, 20 March 2013 (UTC)
So far I only see some edits that were reverted. A first step would be to change the heading from "Do not "fix" links to redirects that are not broken", to "Fixing redirects", and open it with,
"In most cases redirects which are not double redirects do not need to be "fixed". Edits are costly, redirects are very efficient and take little server resources. Fixing redirects should not be done unless another edit is made, and in many cases this is an unhelpful exercise. Replacing [[redirect]] with [[target|redirect]] should not be done if redirect could become the name of the article."
instead of "There is nothing inherently wrong with linking to redirects to articles. Some editors are tempted, upon finding a link to a redirect page, to bypass the redirect and point the link directly at the target page. While there are a limited number of cases where this is beneficial, it is generally an unhelpful exercise, and it can actually be detrimental.
With a few limited exceptions, there are no good reasons to pipe links solely to avoid redirects. It is almost never helpful to replace [[redirect]] with [[target|redirect]].
It is likewise unhelpful to edit visible links for no reason other than to avoid redirects. That is, editors should not change, for instance, [[Franklin Roosevelt]] to [[Franklin D. Roosevelt]] just to "fix a redirect". This rule does not apply to cases where there are other reasons to make the change, such as linking to an unprintworthy redirect." Apteva (talk) 02:55, 15 May 2013 (UTC)
I'm not quite sure I've parsed your contribution correctly, but if I have, I disagree. I would be in favor of strengthening the language to clarify that [[redirect]] is (with very few exceptions) actually better than [[target|redirect]], and while the former should almost never be replaced by the latter, it is often correct to replace the latter by the former. I would also avoid too much performance-related language — maybe just a link to WP:PERFORMANCE to point out that performance isn't a useful criterion. --Trovatore (talk) 03:03, 15 May 2013 (UTC)
I agree with Trovatore. I'd rather see it very explicitly stated that there are very few circumstances where it is appropriate to replace [[redirect]] with [[target|redirect]]. olderwiser 03:12, 15 May 2013 (UTC)
The only exceptions where it is not worth changing [[redirect]] to [[target|redirect]] are listed in the exceptions section, where redirect could be the name of the article some day. We can not say one thing there and another here. From a practical standpoint it is relatively pointless to change [[Franklin Roosevelt]] to [[Franklin D. Roosevelt]], as the article could easily be moved from one to the other, but if an article has been stable at one name it is more practical to correct links, but only in the action of another edit. It is pointless to say either do it or don't do it. Apteva (talk) 03:45, 15 May 2013 (UTC)
Huh? The exceptions section is about when it is OK to replace a redirect by a pipe, not when it is not. Pipes are bad, m'kay? Pipes violate the least-surprise principle and make the corpus of wikimarkup less robust. Redirects are, 99% of the time, better than pipes. --Trovatore (talk) 03:54, 15 May 2013 (UTC)
The difference between Einstein and Einstein is that the second one lets you see the actual name of the article. The second is far superior to the former, but due to performance issues only, does not need to be fixed. It is not an issue of least surprise, it is an issue of removing surprise, because when I put my cursor over the link, I get a pop up that shows me where the link goes, and I can see that it goes to Albert Einstein. The same is displayed at the bottom of my browser with the full URL of the link. The surprise is finding out where the redirect is going to take me when the first method is used. I want to know that the article is actually at Franklin D. Roosevelt, and if someone wants to link from FDR, I would strongly prefer it to be piped, so I could tell where the link was going to take me, but solely from a performance issue it is not practical to add the pipe FDR, unless some other edit is done. Apteva (talk) 04:01, 15 May 2013 (UTC)
You're apparently talking about hovering. I put very little weight on hovering. I don't think most people think to do it. I want to avoid the surprise of clicking on a link and going to a different title, without the little blue notice that explains it.
Pipes are like "goto" statements in a structured programming language. There are times to use them, but they should be avoided when possible, and for many of the same reasons. --Trovatore (talk) 04:14, 15 May 2013 (UTC)
And redirects, which are goto statements are better? Apteva (talk) 04:18, 15 May 2013 (UTC)
Redirects are not goto statements. --Trovatore (talk) 04:20, 15 May 2013 (UTC)
Here's another way of putting it: Pipes are like hard filesystem links; redirects are like symbolic links. If you've ever done any work with these, you know that hard links are a pain in the ass and produce unexpected results. Symbolic links can also produce unexpected results, but being more explicit, they're easier to predict. --Trovatore (talk) 04:22, 15 May 2013 (UTC)
The technical details and software equivalents are not important. To a user, pipes are more useful. Apteva (talk) 06:27, 15 May 2013 (UTC)
No they aren't. The user experience is better with redirects. I'll explain below. --Trovatore (talk) 08:10, 15 May 2013 (UTC)
I might be persuaded that that one falls in the 1%, mainly because Einstein could plausibly be a disambig page. Almost all the time, though, the redirect is better. --Trovatore (talk) 04:03, 15 May 2013 (UTC)
Also, percentages like 99%, and 1% are unlikely to be supported from actual data, and I would not characterize 1% has "hardly never", but instead as "rarely". Apteva (talk) 04:13, 15 May 2013 (UTC)

Basically we are where this section started "Was this the purpose of Wikipedia:Redirect#Do_not_.22fix.22_links_to_redirects_that_are_not_broken?" answer, "No, piped links are OK, unless the redirect is a 'redirect with possiblities'." That there are different opinions on the subject just indicates that the section should be less dogmatic on the subject. Apteva (talk) 04:25, 15 May 2013 (UTC)

The answer "piped links are OK" was wrong. They are to be avoided when possible. --Trovatore (talk) 04:26, 15 May 2013 (UTC)
There is clearly no agreement either way, and no reason to favor one over the other. I would, however, strongly discourage creating redirects just so they can be used instead of pipes, because they are far less useful than pipes. Apteva (talk) 06:27, 15 May 2013 (UTC)
There absolutely is a reason to favor redirects over pipes, though I may not have expressed it optimally. The big danger is coding actual information into a piped link, something that a user who doesn't follow the link (or hover over it) will not see. This must never ever ever be done. Obviously I don't mean the information in the linked article itself; obviously the user who doesn't pay attention to the link will not see that.
But a user who doesn't follow or hover over links, or who reads it in print form, has a right to expect that he hasn't lost anything except the linked articles themselves. He should expect that he can enter any given term into the search box, possibly with allowance for context in the case that the search term leads to a disambig page or a different primary meaning, and get identical results to the user who follows the link.
Now, granted, the difference in the case of [[redirect]] versus [[target|redirect]], is not large; it's just the little blue redirect notice. Still, that's something. More importantly, if this sort of pipe is allowed, editors will come to feel that piping is a normal thing, as opposed to a last resort, and will be more tempted to code information into the links, which is a very very very bad thing. --Trovatore (talk) 08:10, 15 May 2013 (UTC)
Apteva, there clearly is broad agreement that redirects are in almost all cases far better than pipes, otherwise it would not have been codified in the guidelines. Trovatore has already explained some of the advantages:
  • Redirects are much easier to maintain, since they work on logical rather than physical level. Or in order words, whatever physical information is needed to bring a user to the target page (or a section in it), it is coded in the redirect and not in the source page. As Wikipedia develops and grows, the link in the redirect can be changed to a different target or into a full-blown article without any hassle, while reflecting ongoing structural changes in article pipes is a pain as you have to edit many articles (and often introduce cumbersome syntax), when pipes are used. And you will have to find all those dangling pipes first, which is a problem in their own right.
  • Also, redirects are often enriched with additional information ("attributes") which is very useful for Web 2.0 applications, and which simply cannot be put in a pipe. Redirects can carry Rcats (so that bots and humans can make better decisions using the provided semantical information about the type of a link), normal categories (so that an article shows up in categories under a possibly more suitable subject title than its own title, or under different titles in multiple categories - this helps alot to make navigating categories more easy). Finally, a little-known but important feature of redirects is to serve as launch-pads and target anchors for inter-wiki links (even more so after the transformation to wikidata), so that exact equivalents can be paired rather than articles, which happen to have a common nominator but are not about exactly the same topic.
  • Finally, using redirects will help to systematically catch a subject under as many (useful) alternative titles as possible, thereby improving the user experience as the user will be redirected to the information even if the article is named or the information is organized differently than expected by the user. Using these existing redirects in the wiki-text of other articles as well will help to keep the wiki-text clean, but it will also help to evaluate if a target page is actually located under its best possible title or if it should be moved to a different title (if it would turn out that the majority of links to a target page goes through a particular link - and this is not due to a recent move). On the other hand, it gives us more freedom to choose more descriptive subject titles (also longer ones, if they meet our criteria better) while short forms can still be used in the wiki-text.
  • Since redirects are scalable whereas pipes are not, there will be many useful applications in the future semantic web, which can use the extra information coded into redirects to dynamically create semantic navigation maps, making it easier to find related information (mind, that relationships are not necessarily just contents-related relations).
You cannot do any of this with pipes. In fact, there are very few uses of pipes (rather than to use existing redirects or even create new ones). Pipes are intended to solve problems in a local context, for example to place a link on a construct, which is specific to the article (because it is declinated or, due to the sentence structure, uses a different order of words, which does make sense as redirect). Pipes are also still okay in navigation templates which get transcluded into other articles, so that a chosen page gets highlighted (there will be better solutions for this in the future as well).
Regarding, pop-up tooltips in Einstein and Einstein, the former will expand the redirect and show the intro of the Albert Einstein article as well, if you wait just a tiny bit longer, so there really is no advantage in using pipes instead of redirects here.
To come back to Doc James' original question, I'm afraid, JHunterJ's answer was incorrect for the reasons given in the guideline (and explained above as well). It is highly desireable to replace pipes with redirects, whenever a suitable redirect exists already or a useful redirect can be created. --Matthiaspaul (talk) 10:12, 15 May 2013 (UTC)
I agree with Mattiaspaul, pipes are designed for things like syntax and word order changes (e.g. "The unit was [[Decimation|decimated]]", [[Expansion of London Heathrow Airport‎|London Heathrow's expansion plans]]), and elision of full names, titles or disambiguation where the context is clear but just using the shorter form would be ambiguous (e.g.("Princes [[Prince William, Duke of Cambridge|William]] and [[Prince Harry of Wales|Harry]]", "The spacecraft is to make observations of the surface of [[Mercury (planet)|Mercury]]", "[[Steffi Graff|Graff]] won her third title here"). In almost all other cases a redirect is better, including where the short form is not ambiguous or the primary topic. Thryduulf (talk) 17:09, 15 May 2013 (UTC)
Redirects also help to avoid WP:EGG problems, if upon clicking a link they end up on a page that begins "(Redirected from ...)", they have an explanation. With a piped link, you don't get that explanation. --Redrose64 (talk) 18:58, 15 May 2013 (UTC)
EGG problems are created and fixed in the text of the article, and the redirected from redirect is likely not any more explanatory. I see the (redirected from...) solely as an error message, a reminder to go fix the link that created it, a spelling or capitalization error perhaps, or maybe the page was moved, but only fix it if something else is fixed at the same time, unless it is a heavily used link. Reverting this edit was comical (removing middle initial to create a pipe to a redirect instead of a direct link, and tagging it as if it was reverting vandalism). The correct edit would have been to add a pipe to the actual article if the desire was to remove the addition of the middle initial. Apteva (talk) 17:43, 17 May 2013 (UTC)
Speaking of comical, one of the "pro-WP:NOTBROKEN" commenters (User:Matthiaspaul) here has decided to unilaterally revert one of my edits based on his belief that redirects are inherently superior to piped links (per this edit). This was done without any discussion on my talk page whatsoever. I fail to comprehend how someone thinks reverting good faith edits over a guideline that isn't actual Wikipedia policy and that is still being actively discussed is perfectly okay, but a specific usage (style issue) of piped links is not okay. Bumm13 (talk) 22:12, 24 May 2013 (UTC)
Thanks for coming here after my invitation to move the discussion regarding this guideline from my talk page to where it belongs: here. While I understand the motivation and foresight of the reverted editor in Apteva's example, the resulting combinations of piped redirects indeed carried some comic in them at least in this particular case. I would consider them borderline where good arguments could be found for and against them. In contrast to this, your changes of redirects (even to sub-sections) into pipes are not funny at all, they are wrong according to multiple established guidelines and may even have a detrimal effect on the progress of the project, and that you try to make fun of it (or me) is annoying. If you want me to continue to assume your good faith, please stop misrepresenting the facts as you did in your statement above and also on my talk page.
Extended content
I hate my editing time being forced to be bound now to this instead of being available for content contributions, but given the editor's personalized statements above a clarification is necessary, unfortunately:
- Bumm13's change in 5 edits: [8]
- My revert in 2 edits: [9] with proper summary:
"Edit summary for previous revert: We deliberately try to avoid pipes where possible as per WP:NOPIPE. Reapplied Bumm13's improvement of the disambiguation hatnote, but now using corresponding template".
- Apparently unaware of WP:NOTBROKEN and WP:NOPIPE (as well as the corresponding sections in WP:MOSLINK and WP:MOSDAB), the editor posted a question regarding my revert on my talk page ([10]), but then read up the corresponding editing guidelines himself and angrily complained about WP:NOPIPE and WP:NOTBROKEN ([11]) on the Piped link talk page. Then - now definitely knowing that his edits violated WP:NOTBROKEN, which is pointed to by WP:NOPIPE - he posted (in a rather derogative and intimidating tone) a misleading comment on my talk page ([12]) about WP:NOPIPE not being an official guideline and me not answering in "a timely manner while editing WP at the same time". He then changed the article back to his version ([13]) incorrectly stating that my previous edit would have been inappropriate, only to went into a 2-day hiatus himself shortly afterwards. In reality, only half an hour had past since he had issued his question and I had long gone off-line more than 5 hours before. (When looking over his talk page, he does not seem to have been contributing to any discussions on his own talk page for years, which is not a problem per se, but needs to be said to put the "timely manner" phrase into perspective. He removed that bit later on and meanwhile also offered an apology, but given that the editor unfortunately now seems to be out to fabricate a case and escalate what I consider(ed) a non-issue, it must still be mentioned in context.)
- Instead of addressing my detailed reply ([14]) by trying to provide a pointer to a guideline supporting him, the editor just continues to change redirects into pipes against well established guideline WP:NOTBROKEN (f.e. here: [15], [16]) simply by declaring that he would not be bound to it. He also declared that he refuses to correct the original edits in question ([17]) and his various recent edits following exactly the same pattern, while at the same time he tries to keep me from fixing them by trying to intimidate me ([18]), and he continues to misrepresent the facts about what actually happened, as in his comment above ([19]). For example, he wrote on my talk page:
"The specific "It is not necessary to pipe links simply to avoid redirects" statement was added by User:Rockero on June 23, 2006 (per diff). There's nothing authoritative about his statement and it flies in the face of clean, direct linking to article pages",
to which I answered:
"there's nothing more "clean" in pipes than in redirects. They both serve a number of good purposes, and can also be made to work for things for which they do not fit well. Our efforts should always go towards using the best available tool for a particular purpose in a given scenario."
I'm not "pro-redirects" and "anti-pipes", as the editor incorrectly stated above, I'm "pro-using-the-right-tool-for-a-particular-purpose" and "anti-using-the-wrong-tool-for-a-purpose". It just happens, that the editor is incorrectly using pipes for purposes, for which redirects are intended and are technically the better solution (according to WP:REDIR and WP:NOTBROKEN, WP:PIPE and WP:NOPIPE, as well as WP:MOSLINK and WP:MOSDAB, but also in my personal opinion) and therefore, per our rules, affected articles will have to be corrected for as long as the editor does not find a majority to change our established guideline WP:NOTBROKEN to indicate otherwise.
To sum it up, in contrast to what Bumm13 stated above, I did gave a descriptive edit summary for my revert and I did assume good faith. I did answer in the most possible timely manner, constructive and detailed. WP:NOTBROKEN is a long established editing guideline, and besides minor refinements over the years there have been no significant changes to it which would have altered the core of its meaning in more than 8 years. Anybody is bound to adhere to the guidelines, but is free to propose changes to be discussed. Any editor is entitled and encouraged to fix articles according to our established guidelines without being intimidated or sanctioned. At least to me, the wording at WP:NOTBROKEN and the other guidelines is crystal clear and I see Bumm13's discussed edits ([20], [21], [22], and many more) in clear violation to them. Even when deliberately trying to see the edits from his (assumed - as so far he hasn't broad forward anything to explain it) point of view, I am unable to find anything in WP:NOTBROKEN or the other guidelines which would allow his particular edits in question.
Having corrected Bumm13's misrepresentation above now, I hope that the discussion can now continue facts- and contents-based, therefore I outdent:


I for one would like to learn about what an editor finds "cleaner" about pipes which could warrant changing short redirects into long pipes like it happened in the above discussed case with examples like the following:

  • [[FAT12]] -> [[File Allocation Table#FAT12|FAT12]]
  • [[ProDOS]] -> [[Apple ProDOS|ProDOS]]
  • [[CD]] -> [[Compact disc|CD]]

I would also like to learn why an editor would fail to acknowledge the many documented use cases of redirects (WP:REDIR) and artificially narrows their purpose to just one case ([23]):

"Redirects exist as a result of internal link attrition (articles being moved, etc.). I don't care whether having redirects is "some sort of useful tool to gauge [this condition or that]", they exist primarily to deal with internal link attrition, as well as when a linked article's name doesn't fit cleanly into the style of wording of an article that the link originates from."

and declares such edits as "updated/disambiguated wikilink" when in most such cases no disambiguation took place, but a redirect was changed to a pipe.

I mean, preferences and actions have reasons, even if they may not be obvious at first sight. What's the reasoning here?

Also, as I wrote above, I was unable to find anything which would support changes like these ([24], [25], [26] etc.) and wonder what is possibly unclear in the wording of WP:REDIR and WP:NOTBROKEN, WP:PIPE and WP:NOPIPE, and the corresponding sections in WP:MOSLINK and WP:MOSDAB so that it is possible to misinterpreted them.

Thanks. --Matthiaspaul (talk) 18:32, 25 May 2013 (UTC)

Unfortunately, nobody so far has provided his/her thoughts on what might have been the underlying reasons for changes of redirects into pipes against this guideline like in the examples given above. I had hoped that at least Bumm13 would have come forward with his rationale for continuing to carry out these changes., as so far he only declared that "his edits remain as-is" ([27]) because they "didn't break anything" ([28]) and "don't harm", and that a guideline like WP:NOTBROKEN can be ignored (at least by him) because it is "only a guideline" not a policy (for example, see above). Obviously, I don't agree with that and it is my opinion that edits like those given above should not be made (per WP:NOTBROKEN, WP:NOPIPE, MOS:REDIR, MOS:PIPE, MOS:NOPIPE, MOS:LINK2SECT, etc.), and if they have been made, they need to be reverted at least in obvious cases, but I had hoped for other voices to get a broader picture here. --Matthiaspaul (talk) 12:31, 11 June 2013 (UTC)

Statistics on number of redirects

Hi, is there any way of getting the statistics on the number of redirects? I am interested to know how many search including the phrase "primary school" are redirected Wakelamp (talk) 09:39, 16 May 2013 (UTC)

You should be able to work out the number of redirects that include that (or any phrase) by downloading the redirect list from (search for "redirect" on that page) and then doing local queries on it (e.g grep or importing into your preferred database application). Once you have that list you can probably extract the page view statistics for them, but I don't know how to do that asking at Wikipedia talk:Pageview statistics or Wikipedia:Village pump (technical) would be my best guess at where you'll find people who can help you or who know where to ask for help (if you find how to, please could you link to it here as it could be useful for future reference).
The number of searches performed is a different metric and I have no idea if such statistics are kept. At Wikipedia talk:Statistics/Archive 2#Top 100 Searches, user:Richard Arthur Norton (1958- ) seems to have been looking for something related. The toolserver link has now expired, but you might ask on his talk page if he remembers anything about it. Thryduulf (talk) 10:20, 16 May 2013 (UTC)

Some proposed exceptions to NOTBROKEN

I was recently engaged in a lengthy discussion with an editor here who was going around making large numbers of redirect "fixes" in violation of NOTBROKEN. In trying to get him to put an end to these activities I tried to encourage him to address the issue positively by coming up with a set of what he thought might be good exceptions to the guideline. Ultimately we were only able to come up 3 different ideas before the user resumed "fixing" edits and he's now serving out and indef. block. But I told him that I was committed to seeing this thing through, so I'll propose them here for the community's consideration. These are intended to be exceptions to the "Do not 'fix' links to redirects that are not broken" section of this guideline and they are exceptions on which both the other editor and I agreed.

Proposed exceptions
Exception Proposal Example
Spelling mistakes and capitalization errors It's OK to change redirects to correct official spelling/capitalization. Playstation -> PlayStation
Nitnendo -> Nintendo
Consistent consensus-based terminology It's OK to alter redirects to unofficial alternate terms that reflect a community-determined term used in the interest of inter-article consistency. videogame -> video game
Disambiguation pages It's OK to replace a redirect link with a link to the actual page name on a disambiguation page per WP:PIPING "Trinity (wrestler)" -> "Stephanie Finochio"

Now of these three, I think the first may already be covered by the "Typographically problematic redirects" exception already listed at WP:BRINT. Is that true? If so, then never mind that one. The third proposal is new, but it may conflict with WP:DABREDIR now that I think about it. The edit which prompted this proposal was this one, but comparing it to the "James Cary" example at DABREDIR I confess that it seems that it might be actually barred by the MoS. Any thoughts on that? The second proposal, though, I think is entirely new. Please weigh in on these three suggestions. -Thibbs (talk) 05:11, 9 June 2013 (UTC)

There is no problem at all with replacing a redirect by a direct unpiped link to the article, if its title works better in the text than the redirect does. There is no problem with replacing a redirect with a different redirect pointing to the same target, if the name of the other redirect works better in the text than the original one does. The only problem is replacing a redirect by a pipe — that's what NOTBROKEN is mostly about.
I haven't quite understood whether the other editor wants to add pipes in these situations. Nothing in the wording clearly suggests that he/she wants a pipe, and if he does, I'd ask why. If not, then there is no conflict, and these are not really exceptions at all. --Trovatore (talk) 06:34, 9 June 2013 (UTC)
The dab exception is not an exception, as you noted from WP:DABREDIR. The unpiped redirect is used on disambiguation pages, and should not be changed to the piped direct link, if the redirect matches the ambiguous title. -- JHunterJ (talk) 12:36, 9 June 2013 (UTC)
This is misguided, the first item is already covered, the second is a slippery slope contrary to the intent of this policy, and the third is an obvious mistake. --Joy [shallot] (talk) 20:26, 9 June 2013 (UTC)
I don't see the second item as contrary to NOTBROKEN, provided it's not talking about a pipe. Of course you can replace [[videogame]] by [[video game]], if the text video game is considered a better stylistic choice in the particular article. Doing it for "inter-article consistency" is a different matter, and I'm generally opposed to efforts to enforce inter-article consistency for its own sake — but that has nothing to do with NOTBROKEN, and in fact it has nothing to do, except accidentally, with redirects.
Now, replacing [[videogame]] by [[videogame|video game]] would certainly be opposed to the spirit if not the letter of NOTBROKEN, in addition to being just silly. But I don't see any need to mention it, because why would it occur to anyone to do that? --Trovatore (talk) 20:47, 9 June 2013 (UTC)

(To be clear: I certainly don't think point 2 should be added as an exception to NOTBROKEN. My position is that it was never an exception in the first place, so adding it as one would just be confusing. But it might be a good idea to review the wording, to see exactly what misled the unnamed third-party editor into thinking such an "exception" would be needed.) --Trovatore (talk) 20:49, 9 June 2013 (UTC)
Thanks for the input so far guys. I really appreciate it. There's nothing worse than posting a question or proposal and seeing it completely ignored or discovering that nobody watches the talk page. Anyway to supply a few more details for you, Trovatore, the edit that led to the proposal in favor of inter-article consistency was this one. This edit is full of NOTBROKEN vios, but the specific part that was argued to be in the interests of consistent terminology was the "[[Dragon Quest II]]" to "[[Dragon Warrior II]]" edit near the bottom (Note: Not the "[[Dragon Warrior|Dragon Quest]]" to "[[Dragon Warrior]]" edit higher up). I gather that WikiProject Video Games favors the term "Dragon Warrior" (the North American term) over "Dragon Quest" (the international term). If a particular discussion leading to consensus for a term like this can be pointed to, then I assume it would be OK to change instances of redirects in pages to match the consensus-developed terminology, right? But not through piping of course. -Thibbs (talk) 21:08, 9 June 2013 (UTC)
Changing "Dragon Quest" to "Dragon Warrior" might have WP:ENGVAR-type implications, but in my view it has nothing to do with NOTBROKEN. --Trovatore (talk) 21:13, 9 June 2013 (UTC)
Yeah and that may well be the case here. I know this user was having difficulties with that part of the MOS as well and he actually gave no evidence that a consensus-forming decision had taken place on the default term for this series. But for the sake of argument, if there has been a project-wide discussion leading to consensus on a specific term (hypothetical e.g. - a WikiProject Biography decision on whether "Barack Obama" or "Barack Hussein Obama" was to be the default term; or perhaps a WikiProject Politics decision on whether the term "Democratic Party" or "Democrat Party" should be used preferentially), then would it not make sense for this to be an exception to NOTBROKEN? Or is this covered in the current policy somewhere? Or does it fall under a superseding rule? It would make sense to me that we should allow this kind of redirect "fix". -Thibbs (talk) 16:02, 10 June 2013 (UTC)
Well, that just isn't what NOTBROKEN is talking about. You're perfectly free to replace [[redirect]] by [[target]], if the word target is what you actually want to see display on the page. If that isn't clear from the current wording, then maybe the working needs to be fixed, but it's not an exception; it's just not covered by the rule at all. --Trovatore (talk) 17:56, 10 June 2013 (UTC)
Hmm. If I'm interpreting it correctly, though: If the only reason you're replacing [[redirect]] by [[target]] and if the only reason you actually want to see "target" display is because you have an aversion to redirects then that's not a valid reason to make the edit, right? The list of exceptions at WP:BRINT seem to me to be a list of valid rationales for "fixing" redirects like this. And the existence of a consensus-achieving discussion leading to a preference for a synonymous term seems like it would be a valid rationale for "fixing" a redirect. Is it simply too obvious of a conclusion from the "other reasons to make the change" part of the rule? -Thibbs (talk) 18:15, 10 June 2013 (UTC)
You're talking about the FDR example, right? I agree that that's confusing, when taken together with the redirect-from-misspelling "exception". I would propose:
  1. Reword the FDR example so that it's clear that if you want [[target]] to appear instead of [[redirect]], for any justifiable reason that doesn't involve one of them being a redirect, then that change is fine; it's not what the guideline is talking about.
  2. Remove the redirect-from-misspelling "exception", which is not an exception at all; it's correcting spelling, not "fixing a redirect". --Trovatore (talk) 18:55, 10 June 2013 (UTC)

──────────────────────────────────────────────────────────────────────────────────────────────────── I think I mostly agree with Trovatore that the example given of changing Dragon Quest II to Dragon Warrior II is more a matter for stylistics of the article rather than this guideline. In that specific case, since that list article was about Japanese games released in Japan, it stands to reason the list should use the name of the game as released in Japan (Dragon Quest II) rather than the name of the U.S. release (Dragon Warrior II). But I'd be concerned that trying to frame such guidance in the context of this guideline might be problematic. One editor's stylistic consistency (i.e., using Dragon Warrior II as the consistent title) might violate other aspects of consistency, such as using the version of the name that is appropriate in the context of the article. olderwiser 19:08, 10 June 2013 (UTC)

Yeah, Trovatore's suggestion makes good sense to me. I agree that the term "exception" isn't properly applied to the typography issue given the text of the rule. And I think that now deals with all of my proposals above. The first is already clearly covered, the second is redundant as a logical conclusion that would be clearer with the slight tweaking Trovatore has suggested, and the third is improper based on DABREDIR. Thanks again for everyone's help. -Thibbs (talk) 19:19, 10 June 2013 (UTC)
OK, done. I hope that this change meets with everyone's approval; if not, please speak up. --Trovatore (talk) 06:26, 11 June 2013 (UTC)
The one thing that I think the guideline could be clearer about, since some people still seem to get it wrong, is this notion that it's somehow saying that editors shouldn't change redirects ever. The typo exception, or whatever you want to call it, isn't what the guideline is about, but it's unfortunately what some people seem to have thought it was about, even before that language was removed. Croctotheface (talk) 08:23, 11 June 2013 (UTC)
Well, that was what I tried to do with my edit to the FDR example. I definitely don't think it's a good idea to address that point by enumerating lots of "exceptions"; that just supports the idea that the rule does address those cases, so that an exception has to be made. See exceptio probat regulam. --Trovatore (talk) 15:35, 11 June 2013 (UTC)


Is there a guideline for use of the magic word _STATICREDIRECT_ ? For example see Tim Hildebrandt. Reading Help:Magic words I do not understand why this should be used anywhere.

This week I created a few redirects to joint biographies, such as Conrad Buff II, and tagged several others. I did not add Tim Hildebrandt's magic word to any of them. --P64 (talk) 21:38, 9 July 2013 (UTC)

It's mentioned at H:MW#Behavior switches, but I've moved several pages in the past, and have never seen a "Update any redirects that point to the original title" checkbox. I suspect that STATICREDIRECT is an obsolete feature. --Redrose64 (talk) 21:52, 9 July 2013 (UTC)

Problems with redirects to sections

This issue has already been raised two years ago here — Wikipedia talk:Redirect/Archive 2011#Problem_with_redirections_to_sections.2Fanchors — whereby if I have JavaScript turned off, the redirect will take to the page, but not the section. Specifically, if is blocked in NoScript.

So, if I click in the redirect page on the target link, then it redirects to the section, but not if the redirect is automatic, in which case it only redirects to the target page with that section, if is blocked. -Mardus (talk) 14:07, 17 July 2013 (UTC)

(If anyone has replied, please notify me at my Talk page.) -Mardus (talk) 14:08, 17 July 2013 (UTC)

I was the original author of the bug report in 2011 and I can confirm that the problem continues to exist up to the present (for any kind of #hash links, regardless if they go to section headers or embedded anchors), unfortunately. It's quite annoying, as there are many very good reasons to surf the web with JavaScript disabled and we shouldn't force our users to enable JavaScript if they cannot or prefer not to (f.e. because their browser doesn't support it or for security / privacy reasons). The fix appears to be quite obvious; the parser should not strip off a # and anything following it (up to the closing "]]" brackets) in a redirect.
---Matthiaspaul (talk) 17:44, 17 July 2013 (UTC)
Does anyone know if there are any magic words that influence this, like the static redirect one? Would something like Template:Glossary_link be a suitable workaround, in the meantime, or would it be best to use direct section links for noscript support? -PC-XT+ 08:24, 20 July 2013 (UTC)
I have tried various potential workarounds (see original post) but to no avail. (I haven't looked into this template, though). However, since this is only a technical problem, which will be solved sooner or later, I would strongly suggest not to use direct #hash links to sections or anchors (instead of capsulating the #hash link in a redirect). While this would solve the immediate problem, direct #hash links are very difficult to maintain later on and we have several guidelines explicitly stating that we shouldn't "bypass" redirects for various reasons, see for example WP:NOTBROKEN and WP:NOPIPE. If we'd now start to use #hash links in pipes in order to work around this temporary problem, it would create high collateral damage, while for most of us enabling JavaScript for a single trusted site like Wikipedia is just a mild inconvenience. Nevertheless, it should be fixed. --Matthiaspaul (talk) 08:56, 20 July 2013 (UTC)
Far from not using links to sections when they are suited, you should instead use the {{anchor}} and {{anchor comment}} templates at the target so the target is maintained. Thryduulf (talk) 09:03, 20 July 2013 (UTC)
Yes, I use {{anchor}} all the time (sometimes even prophylactually just to make logical linking independent from ever changing section headers. However, this doesn't solve the problem in redirects discussed here. --Matthiaspaul (talk) 10:36, 20 July 2013 (UTC)
Indeed, but we shouldn't really be designing the encyclopaedia around bugs in the software, particularly not ones that only affect a subset of readers. Anyway, it's been reported now so it should (at some point) be fixed. Thryduulf (talk) 11:01, 20 July 2013 (UTC)

Direct links to sections and even links like Beiruit#History work (Beiruit is a redirect to Beirut) with javascript blocked, so there is no obvious reason why redirects to sections don't. I've reported this to the devs as Bugzilla:51736, so hopefully it will be sorted. Thryduulf (talk) 09:03, 20 July 2013 (UTC)

I'd like a clarification please. Does the problem occur:
It occurs only when the redirect page includes a link to the section, e.g. #REDIRECT [[South Devon and Tavistock Railway#Yelverton]]. Thryduulf (talk) 10:18, 20 July 2013 (UTC)
Yep. --Matthiaspaul (talk) 10:36, 20 July 2013 (UTC)

Redirects which are wrong should be fixed

I don't agree with the statement "It is almost never helpful to replace [[redirect]] with [[target|redirect]]."; I find the contrary to be true. Redirects should never be used on purpose; the function exists only to avoid dead links. It is always distrurbing to have the information on top of the page that you came there through a redirect. Lingon of the Dark (talk) 17:24, 1 August 2013 (UTC)

Redirects with possibilities should ever be used on purpose. The “function” not only avoids dead links, but permits for smooth updates of blue links. Incnis Mrsi (talk) 17:30, 1 August 2013 (UTC)

It should not be disturbing to see the redirect message. Yes, redirects should indeed be used on purpose. When a stable redirect exists for the text that you want to appear on the page, then it is virtually always preferable to link that redirect rather than to pipe. That keeps a closer connection between the source code and what displays to the user, in accordance with the least surprise principle. --Trovatore (talk) 17:51, 1 August 2013 (UTC)

Redlink in edited R

Resolved: aha

-DePiep (talk) 22:22, 4 August 2013 (UTC)

From earlier content template {{periodic table (nonmetals variant)}} I created/edited:

  1. {{periodic table (nonmetals variant)/sandbox}} (to have content) and
  2. {{periodic table (nonmetals variant)}} (to be a R)

Mostly manual edit. I also removed the (new to me) <span dir="auto"> tag pair after the R construction (by menu).

Now the #1 not-sandbox page [29] has a red link. What went wrong? Did I leave a bad Unicode char behind? OTOH, the #2 redirect does function. -DePiep (talk) 20:57, 4 August 2013 (UTC) -not sure -DePiep (talk) 21:01, 4 August 2013 (UTC)

You need to specify the namespace, otherwise it assumes mainspace. --Redrose64 (talk) 21:40, 4 August 2013 (UTC)

correct apostrophe?

It seems the "typewriter apostrophe" (') is frequently used in article titles in Wikipedia. However, apostrophe article's unicode section states that the "right single quotation mark (’)" should be the "preferred character to use for apostrophe according to the Unicode standard". (talk) 07:02, 5 August 2013 (UTC)

Please see MOS:QUOTEMARKS. --Redrose64 (talk) 20:30, 5 August 2013 (UTC)

Bots fixing double-redirects

I think we need to come with a decent solution to a problem which I know of 2 occurences of - and there may be many more. A vandal redirects a page, and before the redirect is fixed, some redirect bot "fixes" those redirect which point to it, to point to the vandal's chosen target. Then the vandalism is foxed, but no one knows about the "fixed" reidrects which need to be undone.

My proposal is that:

  1. Bots won't fix a double redirect, unless the target had been stable for at least 24 hours. That would have prevented the results of this edit, but not this one
  2. Bots report every fixed double redirect on some central page; anyone can subsequently unfix those.

I think that those 2, together, should reduce the amount of damage caused by the redirect bots combined with the redirect vandalism. עוד מישהו Od Mishehu 14:19, 9 August 2013 (UTC)

We're not talking about page moves here; we're talking about page redirections - that is, some user replaces an existing page with a redirect. עוד מישהו Od Mishehu 18:19, 10 August 2013 (UTC)
Or where they alter an existing redir to have a different target. --Redrose64 (talk) 18:49, 10 August 2013 (UTC)

Where should Siam redirect?

Where should the Siam page redirect? Please discuss at Talk:Siam. —  AjaxSmack  03:10, 19 August 2013 (UTC)

Talk page reversion

When using the admin tool to close recent AFD Wikipedia:Articles for deletion/Jimmy Neutron (character) as a redirect, the automatic tool created the redirect and on the talk page the tool removed the original page's project listings and discussion and then posted the oldafdful link to the AFD discussion.

A well-intending editor reverted the tool's edits, returning the projects and discussion to the talk page. As the article page no longer exists, this would make the talk page now speediable under WP:CSD#G8. Are there reasons to keep the talk pages of redirected articles? If not, is there an option that could make the editor happy? Schmidt, Michael Q. 18:10, 13 September 2013 (UTC)

I'm not sure that G8 applies, since it excludes "plausible redirects that can be changed to valid targets". The link does exist, even if it's now a redirect to another article. And I understand that the page was moved automatically, but related Wikiprojects have Redirect-Class for a reason. Blanking the talk page, just removes the redirect from those projects, and lessens the possibility of that material being expanded. Fortdj33 (talk) 18:29, 13 September 2013 (UTC)

Redirects to category space

Searching the archives index I find two old threads.

The last contribution (2008-07-16) includes "Redirects to categories aren't that common, but quite a few of them nonetheless exist, and so we should mention how to create them here." Consulting the edit history from 2008-07 and two 2008 versions of the page I do not see that article-to-category links have been covered here.

The main page Wikipedia:Redirect does not cover article-to-category redirects now and I surmise that it has never covered them. I find nothing in the edit history for 2008-07 and nothing in two 2008 versions of the page. --P64 (talk) 16:41, 18 September 2013 (UTC)

Among our 515 Redirects to category space there are only a score apparently intended to be used in articles as unpiped links, more than a dozen near the head of the list, mainly under letter 'A', and this handful: convicted politicians, event venues, hardlines, IEC 60958, softlines. (My judgment is based on skimming the redirect category all 515 listings. Following that handful of links reveals that tThe last four of 5 links named here are redirects to categories from their pagenames in article space.)

Why don't we have many? Such as Irish folklore redirects to Category:Irish folklore (instead its target is an article that seems inappropriate to me, but that is beside the point here).

P.S. That we have so many under 'A' and almost no others suggests to me that a project to eliminate all of them has overlooked 'A'. Eh?

--P64 (talk) 23:25, 4 August 2013 (UTC)

--P64 (talk) 16:41, 18 September 2013 (UTC)

Content on Redirected Page?

I'm still not sure what to do about content on an article that is redirected to another one. Is it copied and woven into the other article? Or deleted? It seems like there should be some process to suggest mergers of articles but I haven't found it on the Wikipedia pages.
There is plenty of information on how to a redirect with a blank article and no guidance on how to do a redirect from one article to another. Liz Read! Talk! 20:50, 27 September 2013 (UTC)

Well, Wikipedia:Merging (which is linked from the section Redirects that replace previous articles) may help. There is also Help:Merging. olderwiser 22:04, 27 September 2013 (UTC)
I've added a title to the section so that it's easier to find. Diego (talk) 06:07, 28 September 2013 (UTC)
Thanks for the reply, olderwiser, and to Diego for the title change. Liz Read! Talk! 12:30, 16 October 2013 (UTC)

Redirects from moves

To Bkonrad:BRD – Let's discuss it, then. As of the 17th, the {{R from move}} template automatically tags redirects left behind from moves. The deletion of that template has been discussed more than once, and the main contention was because editors would use it to immediately make a second edit to the left-behind redirect, thereby ensuring that a quick revert of the move was impossible for most contributors. At the deletion discussions I stressed that any number of Rcats could be used to tag moved redirects, so deletion of template R from move was arbitrary at best. To tackle the "real" problem, I suggest that we now discuss a method of disallowing a second edit of left-behind redirects for some period (7 days?). – Paine Ellsworth CLIMAX! 17:26, 18 October 2013 (UTC)

From my perspective, when I move a page, if I don't immediately go back and add an appropriate R cat template to the redirect, it never happens. I am certainly never going to try to keep track of what moves I might have made seven days earlier. There are other options available for editors to challenge and revert undiscussed moves. olderwiser 17:30, 18 October 2013 (UTC)
Your objection is noted and notable. I arbitrarily chose seven days to give most of the involved editors time to sense the move and to revert it if they deem it appropriate to do so. If it is difficult or impossible to track a move for seven days, a shorter wait period could be used. At present, my larger concern is how such an imposed addition to the guideline could be enforced. What happens if an editor continues to "abuse" (it was called that at the deletion discussion) the delayed-second-edit condition? – Paine Ellsworth CLIMAX! 17:44, 18 October 2013 (UTC)
I really don't see that any enforced waiting period is needed. It's not so much the amount of time, it's more a matter that adding an appropriate redirect template just is not worth the bother of trying to keep track and then later go back. If there is consensus for any enforcement of a waiting period, I expect that many editors performing moves simply won't bother with attempting to categorize the redirects anymore. olderwiser 17:53, 18 October 2013 (UTC)
Also, can you provide a link to the deletion discussion(s) where this came up? olderwiser 17:55, 18 October 2013 (UTC)
I had not thought of that – that many editors would make moves and then not bother to go back and tag the redirects. The most recent discussion was in September and is found at this link. The 2010–11 discussion can be found here. Also, the recent discussion about automating the R from move tag was held on the template's talk page. – Paine Ellsworth CLIMAX! 18:09, 18 October 2013 (UTC)
Also, see Kifisias Avenue station for a "first" example of a redirect with only one edit that was autotagged with R from move. – Paine Ellsworth CLIMAX! 18:15, 18 October 2013 (UTC)
Thanks for the links. For my part, most of the moves I make either 1) fixing malplaced disambiguation pages (e.g., foo is a redirect to foo (disambiguation) -- after checking history and incoming links, I often move foo (disambiguation) to foo and then tag the redirect with {{R to disambiguation}}); or 2) moving a misnamed page and then tagging the redirect with something like {{R from alternate spelling}} or {{R from alternate capitalization}} or {{R from plural}} or the like. An ideal solution, as far as I'm concerned, would be for the move page to allow editors to select one or more R from templates to place on the redirect left after the move. Mostly unrelated, it would also be nice if one could choose to not to leave a redirect for the talk page while leaving one for the article. The deletion discussions seem rather odd to me. First, consensus was to delete the template, but ... only after the software is fixed to automatically add the template and the second appears to have no consensus to do anything specific.

On the one hand, I have no problem with instructions that redirects should not be edited to add THAT specific template, {{R from move}}, after a move. But, an prohibition on adding other templates does not seem helpful -- or rather, it seems to call into question whether any of those R templates should be used at all. I mean, having a general prohibition on editing redirects after a move makes the task of adding those templates rather onerous. olderwiser 18:48, 18 October 2013 (UTC)

No problemo on the links. Maybe more edits to a redirect from a move should be allowed before it becomes more difficult to revert the move? As it is now, one edit to a redirect after a move and the ease of a revert-move for most editors goes away. Is it possible/feasible/desirable to make the move like other edits? i.e., make it so that page moves can be reverted by any editor no matter how many edits have been made to the redirect? Will I lose my barnstars and get my ____ chopped off just for asking this? ;>) – Paine Ellsworth CLIMAX! 19:12, 18 October 2013 (UTC)
The problem is, what happens to the content at that title before the reversion? If it's just a redirect, or a redirect plus tag, then fine, but what if it's something nontrivial? Does it just go into the bit bucket?
That's why I'd like to see branchable histories. Then, if you move a title to a page that already exists, both histories are preserved, and everything can be reversed. --Trovatore (talk) 02:19, 19 October 2013 (UTC)
I'm not sure of the right answer here, but I want to point out a scenario that I don't think is being considered.
To me, the main reason for editing a redirect is when the article foo is written to describe what an editor thinks is a non-primary sense of the word (say, in the context of bar), whereas the primary sense (in the editor's opinion) is as a synonym for baz. So he moves foo to foo (bar), and then retargets foo to baz (ideally, adding a hatnote at the top of baz, pointing out that foo redirects there and that there's an alternative meaning at foo (bar)). Or, if he thinks there is no primary sense, then he moves foo to foo (bar) and writes a disambig page at foo.
In many cases, I think this is a perfectly reasonable action to take, requested moves being cumbersome and having a high probability of ending in "no consensus" even when the situation is pretty clear. It is unfortunate that, for technical reasons, it's difficult to reverse, in case the editor is wrong.
So I wonder if we should be looking for a technical solution, rather than a procedural one? Maybe there should be a way to move an article to a title that already exists, and have both versions available in some transparent way (hopefully not for a long time, but long enough to discuss the issue). I don't have an exact spec to propose, but can we think along these lines, and if we get something reasonable, ask the developers if it's feasible? --Trovatore (talk) 18:16, 18 October 2013 (UTC)
What if the page-move code were updated to allow non-admins to overwrite pages that have multiple revisions, as long as every revision was a redirect? Jackmcbarn (talk) 19:16, 18 October 2013 (UTC)
That's an interesting possibility. It is essentially one thing that a page mover is supposed to check manually before moving a page. And when you give the condition that "every revision [is] a redirect", do you mean that every entry in the edit history is a redirect to the page that is going to be moved over the redirect? If yes, then I'd say that would be just about right. It's a little dicier if you mean that all the edits have simply been a "redirect" and not consider what the target is. olderwiser 19:29, 18 October 2013 (UTC)
I think it would be better if only the targets of the first and last revisions are checked, to fix the situation where a page is moved more than once, and the double-redirect is "fixed". Jackmcbarn (talk) 22:21, 18 October 2013 (UTC)
I dunno. If a page has been moved multiple times, it is almost prima facie evidence that moving the page back might not be uncontroversial and perhaps should not be made too easy to do. olderwiser 22:36, 18 October 2013 (UTC)
I was thinking the case where user 1 moves A to B, then B to C, then someone/something updates the redirect of A to point to C, and now C can't be moved back to A. Jackmcbarn (talk) 02:02, 19 October 2013 (UTC)
Do we really want it to be easy for the page to be moved yet again without some discussion as to what the best title should be? olderwiser 02:12, 19 October 2013 (UTC)
Yeah, I think so. Otherwise you're just privileging whoever happens to act at a particular step. Given that most discussions end in "no consensus", that's a very real advantage to that person. Ideally, any action should be undoable, and if people war over it, then you go to dispute resolution. --Trovatore (talk) 02:16, 19 October 2013 (UTC)
I disagree on both counts. First, it is completely false that most move discussions end in "no consensus". Second, a move war is more disruptive than an edit war. If a page history shows instability, there is a need for discussion and consensus to establish what the title should be. olderwiser 02:32, 19 October 2013 (UTC)
To Bkonrad: Yes, for multiple moves, but for just one move? What makes the present situation unacceptable for some editors is the fact that if an Rcat is "abused" by using it to make a second edit to a left-behind redirect, then the pagename cannot be easily reverted to status quo. I think that's all most editors ask – to be able to effortlessly go back to status quo and let the discussion begin from there. It is at that point, when the pagename has been returned to status quo, that it should be made more difficult to move a page. In other words, editors should be able to return any edit, including a page move, to status quo. How would you have felt if I had done something that disabled your ability to revert the edit I made to this page? After you reverted the part about a delayed second edit, after you returned that part to status quo, we both commenced to do the right thing and discuss it. But if you had been unable to revert it, wouldn't you agree that such might adversely affect the tone of any ensuing discussion on the matter? That isn't always the case, e.g., this recent move I made and the discussion that followed, but then, I didn't add the Rcats just to defeat the revert, I honestly thought it would be an uncontroversial move. And yet, I'm sure that the editor who challenged that move, NYArtsnWords, would have appreciated the power to at least return the page to status quo before having to defend it (the status quo). – Paine Ellsworth CLIMAX! 05:16, 19 October 2013 (UTC)
I think I've already inidicated I'm not that concerned about single moves where the page is moved from A to B and someone later wants to move it back to A and there have been no other intervening renames (i.e., as shown by updates of double-redirects). And I've said elsewhere (Template talk:R from move#Request for comment) that an editor abusing Rcat templates is a behavioral issue to be addressed with that editor. I honestly don't have all that much sympathy for those claiming such great inconvenience at not being able to instantly revert some other editor's page move. There are multiple avenues by which an editor can get a redirect deleted that is holding up a legitimate move. It really isn't that difficult. WP:RMT already addresses reverting an undiscussed move. And {{db-move}} with an appropriate reason will also work. If an editor moves a page in good faith and with a reasonable rationale, and if another editor disagrees with the move and has good reason to disagree, then discussion is appropriate. It might be a request on the first editor's talk page (i.e., perhaps there was a naming convention that was overlooked) or if there are, as there so often are, conflicting guidelines, then a full move discussion or even an RfC is appropriate. I don't see it is essential to be able to immediately revert such a move. Moves are fundamentally and inherently more disruptive than simple edits. I think it is quite OK that move wars are more difficult to engage in than edit wars. olderwiser 12:47, 19 October 2013 (UTC)
In the deletion discussions, I have agreed with your ideas above. And yet some of the arguments to resolve the abuse situation by altering the software scheme were compelling. At least the new gerrit must use the {{R from move}} template to autotag new left-behind redirects. And that template is also still needed for use by us gnomes to tag the old left-behind redirects we find. That Rcat should not see any more deletion discussions. – Paine Ellsworth CLIMAX! 17:14, 19 October 2013 (UTC)
(ec) Well, that helps some, but it doesn't help in the case that you change the redirect to a disambig.
What I would really like to see is a way for the history to branch to a limited extent. You click on the history tab, and at some point you see one of those little boxes with a plus sign in it; not sure what they're called. You click on it, and you see the history for a different branch. There would be an interface to move the top of an earlier branch to become the main branch, and then the previous main branch becomes a branch at that point.
Then the restriction on moving a page to a name that already exists could be removed.
That's just one possible approach; suggested improvements welcome. --Trovatore (talk) 19:33, 18 October 2013 (UTC)
I don't think it's feasible to develop split page history. I think this was suggested once with the pending-changes feature, and it never went anywhere. Jackmcbarn (talk) 22:21, 18 October 2013 (UTC)
Hmm — in what sense is it infeasible? Is it a software issue (e.g. the assumption of a single history is so deeply integrated into the code base that it would be costly to change it)? Is it considered too complicated for users? Or, something else? To me it seems that it would be a useful feature to have in a lot of ways (for example, the complicated and not-particularly-transparent operation of "merging" histories could go away). --Trovatore (talk) 00:42, 19 October 2013 (UTC)

Article to portal redirects

Recently there have been several article to portal redirects coming up at Redirects for Discussion such as Wikipedia:Redirects for discussion/Log/2013 October 28#Current computer and video game events. There are a lot of delete votes at these discussions that merely state it's a cross-space redirect and the policy says to delete them. I proposed that we specifically state that article-to-portal redirects should not be treated as cross space redirects, as both parts of Wikipedia are intended for readers. Ego White Tray (talk) 04:45, 2 November 2013 (UTC)

Redirects for singles/songs to album article

I see Category:John Lennon songs, Category:Bob Dylan songs contains many album songs, presumably so that Users can find them in A-Z using category. Is this practice encouraged/discouraged? In ictu oculi (talk) 03:12, 5 December 2013 (UTC)

i am not sure about general consensus, but MOS:ALBUMS suggests categorizing redirects in the same categories as albums. however, this is a bit of a dead-end as it does not address independent categories. i have a faint memory of some albums guideline advising against making every single song redirect to the album page, but dylan and the beatles are so popular that these might be exceptions. of course, i might be totally misremembering that, and it's not much help with categories. (i think i was remembering wrong) ~ Boomur [] 03:28, 5 December 2013 (UTC)
Category:Middle-earth horses is full of redirects and see its preface which speaks to you directly, if i understand correctly (iiuc). That category is commended at Wikipedia:Categorizing redirects#Categorization of list entries (which is the most pertinent WP page, not this one or MOS/Linking).
IIUC, we are moving toward more complete categorization of redirects with possibilities (which include song-to-album, book-to-series, spouse-to-spouse, and so on --bandmember to band, i suppose-- where the target pages are not simply lists of songs/books/people, so they are not literally covered by the WP section linked above).
For example compare Janet and Allan Ahlberg, Janet Ahlberg and Allan Ahlberg. The joint biography is in two categories only, while the two redirects are in five and three categories (plus the redirect category Redirects to joint biographies which is hidden). For a song example, consider Eleanor Rigby#External links. If that song article were still a redirect to the album article Revolver, the redirect rather than the album would properly be in all those categories. Probably its membership in 20 categories is related to its being an article rather than a redirect.
--P64 (talk) 23:54, 5 December 2013 (UTC)

Pseudo-namespace Redirects

There is currently an RFC opened at the village pump to clarify current consensus and policies about the controversial pseudo-namespace redirects that you might want to participate in. TeleComNasSprVen (talkcontribs) 23:48, 20 December 2013 (UTC)