Wikipedia talk:Manual of Style/Linking/Archive 16

Page contents not supported in other languages.
From Wikipedia, the free encyclopedia
Archive 10 ← Archive 14 Archive 15 Archive 16 Archive 17 Archive 18 → Archive 20

Overlink overdone

I'm concerned that some editors are excessively rigorous in applying WP:OVERLINK. In one recent case, the word "sculptor" was unlinked in the articles's opening sentence:

Bill Smith was a British portrait sculptor.

and likewise in the |occupation= field of the infobox.

I suggest that, in such a case, words which are strongly linked to the reason why the subject is notable should be linked on first use, and again on first use in an infobox.

In other cases, places like "Great Barr, Birmingham, England" are reduced to "Great Barr, Birmingham, England" in the location or birthplace field of an infobox.

In these cases, I suggest that the first use in an infobox should be linked, even if it is not linked in an article body. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 13:45, 2 September 2013 (UTC)

Although no context is given in the example of what article X is or is likely to be, I really fail to see how chain linking that is in the example can be justified, even if it is restricted to the infobox. Using the given example, Great Barr would seem to be relevant to the subject, and would justify a link that testifies that primary relationship. I fail to see the point in linking Birmingham, which is already one step removed from the subject compared with Great Barr. One would legitimately expect a link to the former from the Great Barr article because of its proximity and relevance to it. Interestingly, the Great Barr article presents the subject as

Great Barr is a large and loosely-defined area which straddles the boundaries of Birmingham, West Bromwich (Sandwell; including the Great Barr with Yew Tree Ward) and Walsall (including Pheasey). The area was historically in Staffordshire, and the parts now in Birmingham were once known as Perry Barr, which is still the name of an adjacent Birmingham district.

It would thus seem that even juxtaposing Great Barr to Birmingham may not be appropriate depending on the context of the subject of article X. England is commonly known, and even if we were to agree that a link to Birmingham was desirable from article X, which is yet to be demonstrated, England is even one step further away from the subject and as such the term does not warrant a link. Furthermore, it may distract from the more important links elsewhere. -- Ohc ¡digame!¿que pasa? 04:27, 3 September 2013 (UTC)
"Sculptor": if you read English, you're meant to know what the word means. Years ago people were plastering links in for "author", "writer", "poet", "painter", and just about every occupation you could name. Watering down the linking system weakens it for the readers. "Great Barr, Birmingham, England"—exactly: why would a reader want to click to "England"? Why would they want to click directly to "Birmingham" rather than going to the specific target, "Great Barr", which itself contains a prominent link to "Birmingham"? MOSLINK constrains bunching and recommends specific over general links. Tony (talk) 05:41, 3 September 2013 (UTC)
Irregardless of the simplicity/commonality of the word, any article that starts "X is a Y" in the lead (which is most of them), should have Y - or the most narrowest aspect of Y that WP covers, like "British author" instead of just "author" as to avoid chain linking - linked, for categorization/classification purposes. Past that, yes, very common English words should not be linked, but that single link at the topic can be helpful if a reader is not sure of their search target but knows related terms and thus can do a six-degrees-type of approach to find the right topic without having to scan down the entire article. --MASEM (t) 05:50, 3 September 2013 (UTC)
  1. I know US maps well, England maps not so. I would rarely follow even the first link of Akron, Ohio, US or USA and I would usually follow the second link of Gateshead, Tyne and Wear, England, UK. Our metro article Birmingham disappoints me because I must scroll to see the map! And I might write Birmingham, West Midlands (county) (in prose, more in {infobox} or {persondata}) or select its second link if another editor had provided it.
  2. What should be the target of 'British author'? British literature or one of the Lists of British writers or that super-category (which happens to be incomplete, lacking the lists of British children's writers)?
  3. Children's novel is one of many redirects to 'Children's literature'. Today one enthusiastic editor at Double Act (novel) unlinked it in both the lead sentence and {{infobox book}}. Since then I have added the portal bar with indirect link that displays the words "children's literature", but the page includes no link to children's literature as I write. Should there be one? in every article on a children's book or writer? or two, one in prose and one in box? (I have provided one or two universally, and commonly by one of the redirects that do not display the L-word.)
  4. Within uses of {{infobox book}} and {{infobox writer}}, it may be useful that country or nationality (or language) links to a national literature article (or literature by language article). As I understand my watchlist, many piped links to United States/England/Britain/UK and to English language have been deleted this summerspring or summer while piped links to American/English/British people/literature have been retained. I don't know whether that follows consensus.
--P64 (talk) 20:35, 4 September 2013 (UTC)
This is a very old debate, for those not aware. There is also no consensus for such sweeping delinking, often done by script through hundreds of pages a day – indeed it's still the same one or two people who always turn up to justify it, against a succession of people who have passed by to query it over the years, even though they can never point anyone asking questions to where such consensus was established or ever explain how they can say with such unerring confidence things like, "No one's ever going to want that link, and we shouldn't even give them the option of being able to click it". Nor will they ever grasp that linking is not just about explaining what a word means to a reader who might not know but about offering the option to them of navigating, immediately or later, to relevant and related pages that might contain information of interest (they're links to other WP entries after all, not to simple dictionary definitions). While WP:OVERLINK does indeed, correctly in my view, deprecate the addition of endless links to basic English words mentioned in passing in text, keeping links to terms that are genuinely relevant and related to the topic at hand is fine, both by the guideline and by any common-sense reading of the concept of user-friendliness. I'm personally ambivalent about always linking to individual's professions, but I'd certainly not support a default practice of removing them, especially from infoboxes. N-HH talk/edits 20:58, 4 September 2013 (UTC)
I normally unlink British people, American people, etc, as being too vague and not helpfui in almost all anchor contexts. Tony (talk) 21:00, 4 September 2013 (UTC)
(edit conflict) I agree with Andy and N-HH. Links have purpose simply as a navigation aid. The fact that England is well-known is irrelevant when it reminds me that I wanted to look up some other fact about England when reading an article about a British subject. I don't agree that they distract the user, lessen the quality of, or clutter the text. Even if users are distracted by links, they can turn off the differential rendering of them in several ways.
With regard to Great Barr, the lede paragraph doesn't even contain the word "England", much less a link to it. Unless you're worldly enough to recognize the other names as British placenames, you could easily think it's in Birmingham, Alabama! England is currently linked in the Infobox, but if the link-deleters out there got wind of it, that link would be removed (as it would in Birmingham) on the grounds of it being a well-known place.
Further, who is to decide what "well-known" is anyway? Maybe England, United States, Germany, and India are obvious, but what about Belgium, Luxembourg, Andorra, Bangladesh, Tanzania, Ghana, Fiji, or American Samoa? Surely it's POV for any of us to decide which of them is "well-known" enough to not be linked. That aside, we're back to the fact that, when reading an article about Junior Seau, I'm just as likely to want to look up something about American Samoa as I am to want to look up something about India when reading an article about Bengal tigers. —[AlanM1(talk)]— 21:27, 4 September 2013 (UTC)


I live in Great Barr, and in Birmingham. Many things are correctly and officially in "Great Barr, Birmingham". Other things may be in "Great Barr, Sandwell". But the specifics are not the issue here, which is why I dind't give such context. The same could be said of St Paul's Cathedral, London, England, which is unambiguously there. Note also that the Great Barr example was specific to infoboxes'.

On the issue of "if you read English, you're meant to know what the word means"; I think that's a generalisation. What about young people? What about people learning English as a second or other language? What about people forced to use English Wikipedia through an imperfect translation service, because the equivalent article is not available in their language? Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 11:47, 5 September 2013 (UTC)

Right, so if you want to go to less specific geographical locations, like "Birmingham", you can do it via the most specific link. Otherwise the text would be flooded with links. This is editor perspective, not reader reader perspective. Tony (talk) 18:31, 5 September 2013 (UTC)
I'm not sure that's not back to front. Surely enabling readers to get to pages they might well want to get to as quickly as possible rather than forcing them to do it in multiple steps is genuine "reader perspective"; by contrast a small group of editors declaring, without any justification or evidence, which links are "better" for those readers and which ones should not even be available to them, is very much "editor perspective". As ever, we are not talking about flooding text: this is not about linking every word and is often anyway about links in infoboxes. N-HH talk/edits 18:36, 5 September 2013 (UTC)

<i">"I don't agree that they distract the user, lessen the quality of, or clutter the text".</i"> I'm very much of the opinion that they do distract the user, that they do lessen the quality of, or do clutter the text, in a very big way.--Lubiesque (talk) 19:46, 5 September 2013 (UTC)

We should probably ask more users, rather than simply all project our individual views onto the readership as a whole, surely. What constitutes "quality" or "clutter" is a pretty subjective call after all. As for distraction, I've never understood why this comes into it. All sorts of things potentially distract, from pictures to layout. If we're saying that people might be somehow tricked or forced into clicking links that they really didn't want to and will regret doing so, that just seems to suggest not only a pretty low estimation of the general reader's faculties and judgment but a rather high estimation of our own entitlement and ability to keep them on the correct path. And when it comes to aesthetics, let's not forget that we're just talking about a very small number of words that either will or will not go slightly blue. No one's calling for a link to every single word, even if that were possible. N-HH talk/edits 20:42, 5 September 2013 (UTC)
  • Although I'm tempted to agree with you on some points you have made along the way (but not the ones immediately above), I have still not heard from you as to where do you draw the line after many months of discussing this. Your argument relies on equally if not more subjective notion as to what can be considered "helpful". It's easy for any third-grader to come along and stuff the 'pedia full of links because he/she thinks they may be helpful, but in truth it is potentially damaging to the integrity of the structure, just as excess floor loading can weaken the structure of a building. The inescapable conclusion of your above logic of "links do no harm but are in fact 'helpful' because it gives the reader choice to click it if it is interesting" is link saturation, which is a situation we have now in many articles. We have a number of excellent, albeit extreme cases of overlinking at Wikipedia:SILLIWILI that have somehow survived for years. And "slightly blue", as you are a Brit, I'll just assume that is an understatement for which Brits are well-known... ;-) -- Ohc ¡digame!¿que pasa? 01:54, 6 September 2013 (UTC)
Really? To take but two points, you don't agree that we should ask users more widely, or that what constitutes "clutter" is a subjective thing? As for "slightly blue", that was actually a serious reference to the point that there are, I believe, ways of making linked words appear much paler than the default blue. Anyway, as I'm sure I mentioned a while back, I've always been tempted to set up a silliest wikilink removal of the month, and there are usually plenty of candidates for that too. And as for drawing precise lines, my whole point is that that is pretty hard to do (and in many categories, such as professions and nationalities in opening sentences, I really don't know the answer) – but that is part of the reason why I object so much to blanket removals of the same links from every article. A lot depends on the context and judgments that are better made in the course of ordinary copyediting. And the latter of course will often spot and clean out more genuinely redundant and repetitive links on a page than a script which simply removes links to, say, "France" or "Europe" across the site, even from the page on Germany. N-HH talk/edits 09:15, 6 September 2013 (UTC)
Children's literature (3, above) and genre, continued

Yesterday's enthusiastic editor of Double Act (novel), namely User:Ohconfucius, is back on my watchlist with the script-assisted undoing of 'fantasy' links (fantasy, fantasy fiction, fantasy fiction, ...) --as well as 'children's' links that I mentioned yesterday-- from the lead sentence, {{infobox book}}, or both at Elidor, Northern Lights (novel) and Over Sea, Under Stone --three book articles on British children's fantasy novels.

Elidor may be the best example (revision by Ohconfucius). Its previous edit, four months ago, is work that I mentioned yesterday (number 4 above) --unlinking the display words American/British/English and United States/US/England/UK/... only when their targets are English language and country/nation articles, not when they link a national literature or national people article.

Elidor does indirectly link children's literature via that related portal, however, whereas Over Sea, Under Stone does not have that much after the unlinkage of its infobox book. Northern Lights (novel) sensibly links both children's literature and fantasy fiction indirectly, via the portal bar. I'll remedy that for the other two without restoring the lead sentence and infobox genre links under discussion. --P64 (talk) 17:28, 6 September 2013 (UTC)

Our genre article historical novel provides another example. Our book article The Mark of the Horse Lord formerly linked there from both its lead sentence and infobox genre. After script-assisted strip of both links by OhConfucius -09-10 (diff) it no longer links to our coverage of historical fiction* novel or otherwise. (This one remains linked to our children's literature article (Genre: Children's historical novel) and portal (See also).)
* Links to our historical fiction may have been stripped in the same run, but I have no example and I don't know how the scope of script-assisted unlinking is defined.
Our biography of John Newbery --John Newbery (1713–1767), called "The Father of Children's Literature"-- no longer links to children's literature after script-assisted strip from later in the lead and from the infobox by OhC yesterday -10-04 (diffs). County name was unlinked at the same time (1, above).
--P64 (talk) 20:54, 5 October 2013 (UTC)
Aside from the fact that "children's literature" is self-evident in meaning, the article that was linked starts: "Children's literature or juvenile literature includes stories, books, and poems that are enjoyed by children.". Gee whizz. And it's got a copy-edit/mess tag at the top. If there's a specific part of that article (by section) that is relevant to the point, maybe a section-link. Tony (talk) 03:05, 6 October 2013 (UTC)
"Gee whiz!" You equate internal links with consulting a dictionary, and even the dictionary is frequently consulted on familiar words.
"This article may require copy editing for grammar, style, cohesion, tone, or spelling." It's one of millions.
A section link may be more valuable? The table of contents is near the top; many visitors to our article children's literature will see it on the first screen. If a specific section of such a general article is unusually relevant to a biography subject, as section Children's literature#1700s is for John Newbery, that is reason to include a section link in the biography text at a specific point, while a general link is appropriate in the lead paragraph prose and/or infobox Genres field.
If this were a matter of unlinking a keyword in one article we would be on its talk page or a user page. But these three points --lead sentence is platitude or boilerplate; text needs copy editing; section link might be better-- are advanced to justify script-assisted unlinking in dozens or thousands of articles without reading them. --P64 (talk) 19:24, 9 October 2013 (UTC)

Non-English-language sites

Hello. The Non-English-language sites section is at odds with Template:Language_icon/Notes and Template:Language_icon (changed recently). Compare

  1. Placement:
  2. Usage:
    • (Non-English-language sites) "When using one of the above templates in references that use a {{cite}} template, make sure you place the {{language}} or {{XX icon}} template outside of the {{cite}} template, like this: <ref>{{cite web ...}}{{es icon}}</ref>"
    • (Template:Language_icon) "For citations, the language parameter of the various citation templates ({{cite web}}, {{cite news}}, {{cite journal}}, etc) should be used instead."

Moreover, the rendering of this information in {{language icon}} and the language parameter of the various {{cite}} templates use a totally different typographic convention. Compare:

  • {{cite web}} with language argument
    • Doe, John (12 October 2013). "Perfecto ejemplo de sitio web" (in Spanish). Retrieved 13 October 2013.
  • {{cite web}} with {{language icon}}
  • Plain external link (by the way, can someone please point me to any guidelines for using the plain external link as opposed to {{cite web}}?)

I think the fact that we have two different formats is particularly unprofessional. (My preference would be the more sober one chosen for the {{cite}} templates, but that's not important.)

Do you agree that we have a problem?

I'm open to other ideas, but I propose we

  1. Deprecate {{language icon}} (apparently external links is what it's used for)
  2. If {{cite web}} is considered inappropriate for External Links, create some {{elink}} template, to be used as follows
    • (optional use, at least for now, for English language) {{elink | title=Perfect example of website | url=http://www.example.com}} (renders in the same way as [http://www.example.com Perfect example of website])
    • {{elink | title=Perfecto ejemplo de sitio web | url=http://www.ejemplo.es | language=es}}, which enforces some convention to be decided (placement and appearance), and can be easily extended in the future to incorporate whatever other metadata attribute we need.

Thanks. 219.78.115.45 (talk) 13:05, 14 October 2013 (UTC)

Official site question

The MoS isn't clear on this issue. I recently got into an edit war where an editor was suggesting that the following is correct:

I believe that this is the way the entry should be displayed instead:

Any assistance in the matter would be appreciated. Feel free to comment here or on the article's talk page. Walter Görlitz (talk) 23:51, 22 October 2013 (UTC)

I disagree, as per comments made here Talk:IPad_Mini#EL_section. Additionally, "official site" is NOT the title of the page being linked to, so as per citations "title=" we avoid adding WP descriptions into the link itself. Jimthing (talk) 00:35, 23 October 2013 (UTC)
This exchange concerns WP:External links. That page is a content guideline and its hatnote points here (MOS linking) only for internal links.
This exchange also concerns use of template {{official}}, whose documentation says of the linkname, "it should not include the article subject's name as the reader reasonably expects that all external links pertain to the subject."
Let me continue at the original location only, Talk:iPad Mini#EL section.
--P64 (talk) 20:22, 31 October 2013 (UTC)

Repetition of links with-in tables

I may have been overzealous User_talk:Comp.arch#A_note_about_overlinking. I know that tables can repeat links from the articles themselves. But I assumed repetition within tables is bad. Should it be repeated in separate table? Or Even within them? And please make this page explicit about the result. Before (really started a little before but it was reorganized): [[1]]. After (still more could be done if): [[2]]. I use search/replace with a little fix-up afterwords so (almost) all or (almost) nothing would be best. I might not bother changing with more complicated rules.. comp.arch (talk) 14:15, 30 October 2013 (UTC)

OVERLINK only applies to prose. The reason linking is limited is because it can impact on readability, but this doesn't really affect tables. It's very rare a reader will read through an entire table anyway, and will often just read a specific row. For instance, if you want to know who won the Best Actress oscar in 2011, you would scroll down to the year at Academy Award for Best Actress and it is convenient to have the Meryl Streep link available to you, rather than having to search back through the table to find when she was first nominated (1981). Betty Logan (talk) 14:28, 30 October 2013 (UTC)
Betty makes some assertions above as though they are fact, which I don't at all see as such. Where is her support for her assertion that it only applies to prose? That it does not affect readability in tables? That "It's very rare a reader will read through an entire table anyway, and will often just read a specific row." Those have the smell of someone making up "facts" that support their personal view, where the facts may very well not exist at all. I, for one, don't share her assumptions.--Epeefleche (talk) 00:41, 31 October 2013 (UTC)
It's been discussed before (you'll have to check the archives) but it's summed up by the statement "Generally, a link should appear only once in an article, but if helpful for readers, links may be repeated in infoboxes, tables (even many times in each table), image captions, footnotes, and at the first occurrence after the lead." When it comes to table, usually this advice is because many tables are sortable, and thus linking once will not necessary leave the link as the first occurence; it also helps with uniformity if there's a lot of different terms. You aren't required to repeat links in a table, but it would be improper to remove such links if they are already there without seeking consensus. --MASEM (t) 00:47, 31 October 2013 (UTC)
(Noting that that "(even many times...)" part, yes, wasn't in there a few days ago and it is clunky, as epeefleche reverted). --MASEM (t) 02:37, 31 October 2013 (UTC)
That instruction sounds very much like a nonsense ratchet in favour of overlinking – people are allowed to flood a table with repeat links, but once there you need consensus to remove. -- Ohc ¡digame!¿que pasa? 02:39, 31 October 2013 (UTC)
Just like how date formats and citation formats should be handled? I do agree that if the editor creates the table and purposely does not link all entries, someone coming along to add links is breaking that "first-editor" concept, and instead should seek consensus. But in the same manner, if the first editor included most of the links in a column, removing those would be a problem. I totally agree that drive-by "link filling" is not appropriate. We simply want internal-article consistency. --MASEM (t) 03:09, 31 October 2013 (UTC)
In our environment, it's impossible to determine just whether that intention genuinely exists. -- Ohc ¡digame!¿que pasa? 03:25, 31 October 2013 (UTC)
No, not really. You have to use Wikiblame ([3]) that lets one search on specific text and if we're talking the addition of a table, "{|" should be sufficient. I just texted this on The Wolf Among Us, an article I know I added a table to a week ago, and it hit the exact revision that the table was added. From there, you can hit back the history to see what the short term fate was.
Irregardless - if you come across a page that clearly has not linked anything in the table at all, you could bring it for discussion on the talk page, or you could be WP:BOLD and add the links once but if that's reverted you'll have to fall back to the talk page discussion to add. Same with removal: discussion or do one BOLD edit followed by discussion. The same with how citation formats and date formats are handled. --MASEM (t) 03:44, 31 October 2013 (UTC)
  • Common sense, please. Just minimise linking so that the system attracts more clicks: flooding with repeat links will do the opposite. And minimise the risk that readers will think the blue is part of the colour design of a box or table. That said, if there's an aesthetic issue in a sortable table, there might be a case for repeat linking. One alternative is to link (once) the items in the adjacent main text, or in a "See also" section at the bottom, where you won't need to pipe but can show readers exactly where they might go. Tony (talk) 05:50, 31 October 2013 (UTC)

Would "Generally, a link should appear only once in an article, but if helpful for readers, links may be repeated in infoboxes, tables (in each table is ok), image captions, footnotes, and at the first occurrence after the lead." be ok? I had added that (without the "is ok") and then a less conservative advise. Maybe both got reverted by mistake. comp.arch (talk) 09:11, 31 October 2013 (UTC)

Well what does "in each table is ok" even mean? It sounds like you are restricting each table to just one link. The guideline is ok as it is. Linking in prose has a restriction placed on it by the MOS, and linking in tables, infoboxes and footnotes is left to editorial discretion. Betty Logan (talk) 15:12, 31 October 2013 (UTC)
I think what the change was going for was saying "...tables (both within a table, and between tables in the same article)..." Now, that's nowhere as elegant, and I'm not sure if necessary (I would consider the advice implicit). I do, however, agree that it might be wise to add what Tony, OC, and Betty point out, that full linking (or full delinking) is optional and an editorial decision akin to dates/citation formats. --MASEM (t) 16:52, 31 October 2013 (UTC)
  • I support what Tony wrote ("minimise linking so that the system attracts more clicks: flooding with repeat links will do the opposite. And minimise the risk that readers will think the blue is part of the colour design of a box or table....) and OhC wrote above (contra "flood[ing] a table with repeat links"). The rationale for linking once only [in text, or table, or image caption], as a general matter, is what we should focus on. It is the same for any part of the article, though we allow a repeat of it in the table or caption if it is in the text, because people who read the table or caption might not read the text. And that link-once rationale has consensus -- that's the reason we have the basic rule in the first place, and have had it in place for the longest time. We don't forget the rationale suddenly in corners of the article, such as tables and captions, and go for a sea of blue in those corners.--Epeefleche (talk) 04:17, 1 November 2013 (UTC)
The suggestion that we should minimise the number of links to attract more clicks is actually a little odd. First, why are we aiming to necessarily attract more clicks, and secondly who says that will be the effect anyway? As for consensus, it's long been agreed, as far as I can recall, that we are indeed more liberal in tables than with prose, for all the reasons suggested here and previously (aesthetics, sortability etc). More generally, as ever, I also fail to understand the apparent obsession some editors have with removing links, based seemingly on nothing more than, ultimately, their own personal preferences and idiosyncrasies lathered with some assertion-heavy post-event rationalisations. Isn't there anything more important to worry about? N-HH talk/edits 11:31, 1 November 2013 (UTC)
The reasons to minimize linking in prose makes sense, because I will completely agree that lots of blue links while trying to read an english sentence can be distracting. But blue links have to be considered like data - and similarly, you wouldn't load down a sentence with many different numbers and units to make it hard to read and confusing too. That all changes when the data is placed into a table which is normally not designed for sequential reading. The reader might even skip over a table, knowing its there as useful data in the future to continue on to higher-level details later. That same reader may come back after some time and immediately jump to the table to find a piece of data, and there if we mess with linking, it can screw up that type of search. So to try to remove linking from such tables by comparing it to prose is just trying to turn the consensus that currently supports this.
I will argue that this is for linking terms that are direct data elements in a table (eg if one column is a Location field, I expect the locations below to be linked). Many tables have a Notes or Comment or some other section that is written in prose, and I would agree that in these sections, the linking style applied to normal prose (avoiding duplicate links) has to apply here. --MASEM (t) 14:32, 1 November 2013 (UTC)
Agreed: people don't, surely, usually "read" tables as such. Instead, they will often go to a specific point in them to find a specific piece of information. Not only is there no readability issue but we're actually making perfectly reasonable navigation around the site more difficult by insisting that a link can only occur once, and hence in one specific place, in a table. With sortable tables, the choice of that place becomes even more arbitrary, because we're not even talking about the definitive "first" occurrence. There are obvious downsides and no clear upside to removing links quite so obsessively here. N-HH talk/edits 14:48, 1 November 2013 (UTC)

Species names

Hello all :) I was wondering what the correct style guideline was for species names: should their common name be linked, the scientific name or both? E.g. should it be:

  1. Silaum silaus (Pepper saxifrage)
  2. Silaum silaus (Pepper saxifrage)
  3. Pepper saxifrage (Silaum silaus)
  4. Pepper saxifrage (Silaum silaus)
  5. Silaum silaus (Pepper saxifrage)
  6. or finally Pepper saxifrage (Silaum silaus)

I have no idea, and as I am editing quite a lot of plant-related articles, it would be nice to know what the official guideline is (if one exists). Thanks, Acather96 (click here to contact me) 22:41, 5 November 2013 (UTC)

I'd eliminate 2, 3, 4 and 6 because they do not follow the assertion in the target article's title that the Latin name is what it should be known by on WP. I would then prefer 1 over 5 because it doesn't need a pipe (|). ~KvnG 15:41, 9 November 2013 (UTC)
My thought is that because species often have many different common names, it's probably best to put the Latin name first and link that, so I'd go with #1 to eliminate piping, though #3 would probably be my second choice. –anemoneprojectors– 12:18, 10 November 2013 (UTC)

Linking "United States"

When I de-linked "United States" in an article about a church that happened to be in the US, I was reverted and it was linked once again.

My thoughts, and those of the re-linking editor (who thinks it appropriate under wp:overlink to link the United States there), are found here.

We don't agree -- perhaps others have thoughts on this.--Epeefleche (talk) 06:25, 12 November 2013 (UTC)

In an article about a local church, the most germaine geographic term is the town that the church is in, so linking the state or country is unnecessary (within prose). --MASEM (t) 07:03, 12 November 2013 (UTC)
Perhaps (I should have mentioned) it would be helpful if -- whatever editors' views -- they leave them on that talk page, where the discussion of the issue is ongoing. Many thanks.--Epeefleche (talk) 08:04, 12 November 2013 (UTC)

Protocol-relative URLs

I would like to alter

The URL must begin with http:// or another common protocol, such as ftp:// or news://. If no protocol is used, the square brackets will display normally – [like this] – and can be used in the standard way.

to

The URL should begin with http:// or another common protocol, such as ftp:// or news://. For URLs beginning http:// or https://, the protocol may be omitted provided that the double slash is retained; this produces a protocol-relative URL to which the MediaWiki servers will add either http: or https: according to the protocol that the user's browser is using - for example, [//www.example.org/ example.org] → example.org. If neither protocol nor double slash is used, the square brackets will display normally – [like this] – and can be used in the standard way.

Protocol-relative URLs have been available for a little over two years now. --Redrose64 (talk) 00:24, 24 November 2013 (UTC)

Repeating links in different sections

The guide states that links should generally not be repeated within an article except in certain cases:

"...if helpful for readers, links may be repeated in infoboxes, tables, image captions, footnotes, and at the first occurrence after the lead."

However, when reading a long article, one may skip down to a desired section, and come across a related topic they want to look into. How does it benefit them to have to go back to the top of the article to comb it for the first usage of the term, or use a search function to find the link? It seems to me that having a link repeated up to once per section of the article would be reasonable.

Pruwyben (talk) 19:57, 3 October 2013 (UTC)

Sections can be quite short, so it comes down to how far apart the links would be. --Musdan77 (talk) 00:12, 7 October 2013 (UTC)
The default assumption is that readers read through an article, so that just like an abbreviation, which is commonly presented in expanded (and abbreviated) forms on first occasion only, generally it's not a good expenditure of "blue capital" to repeat. The analogy with "capital" is used to stress that the less dense the linking, the more likely readers are to notice (and maybe, once in a blue moon, click). There may be the odd reason to repeat-link in the main text, but it should be a good one and the links distant from each other. Tony (talk) 02:06, 7 October 2013 (UTC)
I would argue that repeat linking only becomes reasonable when the article text size has passed a certain level - 50k of characters or so, and at that point we're talking about repeating links only in the back half of the article. --MASEM (t) 02:18, 7 October 2013 (UTC)
As ever, surely the best way is to make a reasonable judgment in the context of the page in question? I'd certainly be against any outright bar on repetition. I agree that the size of the article and the distance between the occurrences counts, as does, surely the pertinence of the term in question to the section(s) it appears in. For example, it would surely handicap readers and their ability to navigate WP as a whole to limit the links to penicillin on the Alexander Fleming page to one occurrence? Currently the guideline even deprecates repeating a link after the lead. That just seems wilfully unhelpful and restrictive for its own sake.
Where btw was the "default assumption" established that "readers read through an article"? Any evidence for that claim? I'd suggest that in fact that is very unlikely in many cases. And the dilution/distraction argument has never been properly explained or justified: if the links are there, any rational user can decide for themselves which ones they are interested in, regardless of how many there are (and, as ever, we are not talking anyway about saturating a page with links). Also, the comparison with the use and explanation of abbreviations, while valid to some extent, slightly fails because of course explaining abbreviations adds text and trips up the reading process. Adding a link, and hence making a word go blue, does not; and anyway I'd argue that abbreviations often can and should be re-explained within lengthy article divided into multiple relatively discrete sections. N-HH talk/edits 08:59, 7 October 2013 (UTC)
N-HH seems reasonable. The existing guideline makes it seem very strict, we should add language pointing out that for loner articles, or where particular ocntext is needed, additional links are at the sound discretion of the editor. I don't think a particular cutoff size is useful. If anything, a more useful guideline is that a single term should not be linked more than once or twice in a page of text printed on A4 or letter, and should almost never be linked twice within 3-4 paragraphs that are in the same section. Dovid (talk) 18:51, 19 November 2013 (UTC)
I've just reverted Dovid's change, which I do not believe reflects consensus. Tony (talk) 15:14, 1 December 2013 (UTC)
Yes you did. There's clearly a consensus that the existing text is no good, though. Of the four parties in the discussion above, three agreed that the existing standard is overly strict, and at some threshold, a repeat of a link is reasonable. The only lack of consensus was over what threshold to use. You were the lone dissenter to any sort of change, based on an two arguments: 1) That it is analogous to the "initial abbreviation" in documents, and that it "diminishes capital." The initial abbreviation idea is not nearly as relevant online (and in fact, many electronic documents that use abbreviations do have a subtle UI for expanding the abbreviation throughout the document). The capital argument I think is misstated. It is not relevant to multiple links, it is relevant to the overall density of links. That said, the argument against repeated links is really a tool for finding a way to reduce link density, but it is a very broad tol, applied equally to articles that would benefit from link reduction and those that do not need it. So I'd like to offer the following ideas for compromise:
  1. . Explicitly state that editors should aim to have sparse links, usually 5-10% of the text of any paragraph.
  2. . Where there is a need to exceed the above, efforts should be made to reduce the visual impact of linking, using the following techniques:
    1. . Link text should be shortened; where one or two words adequately express the connection to the linked article, additional words only serve to obscure the break between links and reduce the visual value of links
    2. . Give priority to the most important terms that need linking, which include non-trivial subjects that are important to understanding the article containing the links. More common terms should get lower priority, as should terms that are less important to understanding key information in the article. Eliminate lower-priority links until the link density is reduced to about 10% of the text of any paragraph.
    3. . Links to several related articles will also have lower priority, when one link will suffice. One method is to replace several specific links to one broader link. For example, in a discussion of fruits, where a sentence links to oranges, lemons, and tangerines, instead of linking to all three, change the text have to one link for [[citrus fruits], such as oranges, lemons, and tangerines. Where introduction of the broader subject is not reasonable in the text, it is acceptable to link to only the first related term, and let the others be figured out by the reader from context or by through links found in the second article. For example, link oranges, but not lemons or tangerines. The reader can figure out that the others represent a similar idea.
    4. . Multiple links to the same article should always be avoided where the user is likely to have already seen the first link, even when reading only parts of the article. They should never appear closer than within half a page of printed text (A4 or letter), unless separated by at least one section. Even where they are allowed because they are distant form each other, the additional links to the same subject should have ower priority than any new terms, when fixing an overlinked paragraph.

Dovid (talk) 19:29, 1 December 2013 (UTC)

Redirects for singles/songs to album article

I have asked a question at Wikipedia_talk:Redirect#Redirects_for_singles.2Fsongs_to_album_article which may be relevant to this page too. (the question is: "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?" ) Cheers. In ictu oculi (talk) 03:14, 5 December 2013 (UTC)

Section links

I would suggest to rewrite section "Section links" to give preference to anchors. Reasons:

  • In terms of typing, anchor is less work than even a single "links here" comment
  • Unlike the alternative, anchors make it easy to rename the section.
  • Comments was a kludge to resolve the problem with lost linkage after renaming; anchor is clear mechanism

- Altenmann >t 00:21, 13 January 2014 (UTC)

Question (possibly suited for an RfC)

From the WP:OVERLINK portion of this page:

  • "the names of major geographic features and locations; languages; religions; common occupations; and pre- and post-nominals"

Should this also include "(major) ideologies" (such as capitalism, communism, feminism, liberalism, etc.)? Toccata quarta (talk) 18:52, 16 December 2013 (UTC)

As far as I understand, overlinking deals with linking to pages only tangentially related to the article subject. (When you read Doctor Who article, I don't think you (would want to) be wiser by clicking the "doctor" link.) On the other hand, when a page mentions "feminism", there is a much higher chance it is related to the subject. E.g., while reading a bio of Liz Lilly people would interested to know more about the subject of her devotion than about the state of Virginia where she broke her leg while skiing. - Altenmann >t 00:37, 13 January 2014 (UTC)
Depends on the context, I'd say. I agree about "feminism" in most contexts, first occurrence. Capitalism is more widely known and shouldn't need to be linked in some contexts, but if it's in relation to other systems or ideologies ... maybe. Got examples? Tony (talk) 05:27, 13 January 2014 (UTC)

Poor example in the Link clarity section

While I generally agree with the arguments presented in the section, I think the Mozart example is not well-chosen. It is proposed that for clarity both words in "his Requiem" is included in the linked text in the phrase "When Mozart wrote his Requiem". My question is: who else's requiem could he possibly have written? So, if the context is Mozart (only), I think only "Requiem" should be linked. However, if we're speaking of requiems in general, and we have mentioned another composer's requiem earlier, then some kind of emphasis is probably needed, like "When Mozart wrote his Requiem", and in that case, I think both words should be included in the link.

So my conclusion is that the linking "strategy" depends on the context.

HandsomeFella (talk) 09:38, 21 January 2014 (UTC)

I think the Mozart example is well-chosen. The problem with When Mozart wrote his Requiem is that the user sees the blue-linked "Requiem" and expects an article on the topic of Requiem, not an article on the topic of Mozart's Requiem. When the linked text is "his Requiem" the user expects an article on the topic of Mozart's Requiem. I would not change this section. —Anomalocaris (talk) 12:08, 27 January 2014 (UTC)
I don't agree. "Requiem" is not a proper noun, and so is normally not capitalized. The capitalization here indicates that this is indeed about a specific work of music, namely Mozart's own Requiem, which carries a name, and thus is capitalized. Had the link led to the more general article on requiem(s), it would not have been capitalized. HandsomeFella (talk) 12:21, 27 January 2014 (UTC)

Discrepancy between guidelines and example article

The guidelines at Wikipedia:Manual of Style/Linking#An example article uses this example:

For example, in the article on supply and demand:
  • almost certainly link "microeconomics" and "general equilibrium theory", as these are technical terms that many readers are unlikely to understand at first sight;
  • consider linking "price" and "goods" only if these common words have technical dimensions that are specifically relevant to the topic.
  • do not link to the "United States", because that is an article on a very broad topic with no direct connection to supply and demand.
  • definitely do not link "wheat", because it is a common term with no particular relationship to the article on supply and demand, beyond its arbitrary use as an example of traded goods in that article.
  • Make sure that the links are directed to the correct articles: in this example, you should link good (economics), not good.

Despite the above, the word "wheat" has been linked in the stated article since 2004 (the above was incorporated into the guideline in 2009 but may have been moved from elsewhere). I'm leaving this note here and at Talk:Supply and demand#Please pay attention to appropriate linking: this article is used an example to draw attention to this and encourage cross-checking to avoid such discrepancies recurring. —sroc 💬 12:56, 2 February 2014 (UTC)

Discussion at WP:FILM

Any input would be appreciated at Wikipedia talk:WikiProject Film#Linking to film year page. BOVINEBOY2008 22:55, 4 February 2014 (UTC)

Categorization and inconsistencies between "cite" and "language icon" templates

Hello,

I noticed that use of {{cite web}} with language=el parameter places the article in non-existent Category:Articles with Modern Greek-language external links (note Modern), while related {{el icon}} places articles in Category:Articles with Greek-language external links which currently has 3000+ pages. Besides, there seems to be a number of other red non-English-external-links categories here.

I also noticed that there is a recent unanswered question about inconsistencies between use of the language parameter in {{cite}} templates and the {{language icon}} templates.

If these two mechanisms serve the same purpose but conflict with each other, can there be a discussion to merge/reconciliate them? Can anyone see the source of the categorization problem, and/or have a take on the duplicate process issue? Place Clichy (talk) 14:47, 8 February 2014 (UTC)

The "recent" (October 2013) unanswered question was asked just before Wikipedia:Templates for discussion/Log/2013 October 12#All language icon child templates, which closed as "keep." However, it was far from unanimous and there was some good discussion from both sides that is worth reading. davidwr/(talk)/(contribs) 15:42, 8 February 2014 (UTC)
The CS1 cite modules, including {{cite web}}, render |language= in one of two ways. If something other than a valid two-letter code is used, the text is displayed as is. If a two-letter ISO639-1 code is used as the parameter value, it is replaced by text in a list at Module:Citation/CS1/Configuration (do a search for "639" in the code). It appears that "el" is listed as "Modern Greek" there, in accordance with http://www.loc.gov/standards/iso639-2/php/English_list.php and in order to disambiguate the language from Ancient Greek.
The {{el icon}} template has the "Greek-language" category built right into the template's code. Actually, it sends "el" to {{Link language}}, which sends it to {{ISO 639 name}}, which sends it to {{ISO 639 name el}}, which displays "Greek" instead of "Modern Greek". Therein lies the difference.
It appears that reconciling the problem involves either changing Module:Citation/CS1/Configuration, which appears to be using a reliable source to supply "Modern Greek", or changing {{ISO 639 name el}}, which appears to be ambiguous. Note that {{ISO 639 name grc}} also exists, which displays "Ancient Greek".
Is "Articles with Greek-language external links" ambiguous? Unless there is something about the usage of "Greek" versus "Ancient Greek" that I do not understand, I would recommend changing {{ISO 639 name}}. – Jonesey95 (talk) 15:47, 8 February 2014 (UTC) [modified and clarified to show "ISO 639 name el" by Jonesey95 a couple of hours later.]
As I understand it, {{ISO 639 name}} exists simply as an error message template; for Greek, the actual template is {{ISO 639 name el}} and so on for all of the language codes – one template per code.
I agree with your related post at Help talk:Citation Style 1 that there should be a common source for language names. As part of the Scribunto extension's language library there is a mw.language.fetchLanguageName() function that appears to return language names when given ISO 639-1 codes. I tested this just a bit with this line of code:
local name = mw.language.fetchLanguageName( Language:lower(), "en" );
Because of the pending update to the live CS1 module I have disabled the test until a later date. The test, by the way, returned Greek for the |language=el code.
—Trappist the monk (talk) 16:59, 8 February 2014 (UTC)
To answer the question "Is "Articles with Greek-language external links" ambiguous?", I believe the answer is no. By that I mean that in the context of a web page (and most other contexts), the word Greek with no other indication can pretty safely be assumed to mean modern Greek, as ancient Greek is not very prevalent in Internet communications. To answer another concern below, I'm not sure it would be appropriate to review all external links to English-language web pages (or French-language web pages) to determine the most precise variety of the language in which they are written, and I think that this concern can be skipped for Greek as well. Place Clichy (talk) 17:37, 8 February 2014 (UTC)
I wouldn't be so sure about that. A web page that consisted of a transcript of Ancient Greek text (e.g. Homer, parts of the New Testament, etc.) would be in something other than Modern Greek. Whether it was "tagged" as Greek or not would depend on the whim of the editor who added the link and whether the page had non-Greek content. davidwr/(talk)/(contribs) 18:29, 8 February 2014 (UTC)
The code that processes |language= is used by all CS1 citation templates, including {{cite book}}. I can imagine that there are books whose text exists in Ancient Greek, and that saying that those books are in "Greek" could be ambiguous for a reader. – Jonesey95 (talk) 19:13, 8 February 2014 (UTC)
(e/c w/ Jonesey95) Strong recommendation The first interested editor who has the time should step up and volunteer to audit all of the cite-, language- and language-icon templates and their resulting categories for inconsistencies then make a recommendation of how to fix them. There will be cases of "keep this in favor of that" or "keep that in favor of this" when "this" and "that" serve the same purpose. The volunteer will get the honor of whether to keep "this" or "that" when he drafts his proposal. The only hard requirements for the proposal are
  • It reduce or eliminate inconsistencies
  • It be consistent with the consensus(es?) from the discussions from last October that are linked above
  • It be "merely" a "clerical change" so that it's extremely unlikely to get any serious opposition.
My hope is that someone can put a proposal on the table then after a pro forma week-long discussion period there will be unanimous or near-unanimous support with no issues raised that show the proposal to be anything more than a "clerical change" then we can just make it happen.
Once this is done, we can talk about issues that are not non-controversial in nature, such as "when would it be appropriate to revisit the issues raised in the October discussions, assuming that a lot of editors (disclaimer: myself included) think that 4 months is too soon."
So, who wants to volunteer to identify all of the language-related purely-technical (i.e. non-controversial) cleanup that needs to be done with these templates? Given the number of languages that need to be at least glanced at, please don't volunteer unless you can put in 10-20 hours on this over the next week or two, and please keep us updated on when you expect to have a complete proposal ready. davidwr/(talk)/(contribs) 15:57, 8 February 2014 (UTC)
A side note: In cases where the current system is just plain ambiguous, like when one template provides for two different languages (e.g. Ancient Greek and Modern Greek) and another only has one language (e.g. "Greek" being "Modern Greek") it will be controversial to "assume" that the existing category assignments are correct. In your proposal, consider keeping or creating a "deprecated" ambiguous category (e.g. Category:Articles with Greek-language external links) for current use but having new uses go into the proper categories (e.g. Category:Articles with Ancient Greek-language external links and Category:Articles with Modern Greek-language external links). This might mean recommending that both Template:el icon and Category:Articles with Greek-language external links be put into a new category, Category:Pages with unclear Greek-language categorization, or some such. Similar work may be needed for other languages where the 2-letter code contains multiple languages. As a side-note, there are at least 7 "Greek languages" according to the infobox in Greek Language (as of the current edit, from earlier today). davidwr/(talk)/(contribs) 16:14, 8 February 2014 (UTC)
Frankly, I don't see a reason why an editor would have used {{el icon}} or the el language code for a text in Ancient Greek instead of grc. If they did, they made a mistake, and Wikipedia is not protected against editor mistakes. Ancient Greek is not Greek, it's Ancient Greek, like Old English (ang) is not English (en) and Old French (fro) is not French (fr). WP has plenty of links to Ancient Greek web pages, and they are correctly categorized in Articles with Ancient Greek-language external links, which today has 40 pages and yesterday had 6000+. Most of the ones I saw yesterday were pages from the Greek–English Lexicon linking etymology, which is understandable as plenty of English words have Greek etymology, directly or through French or Latin. If you're not convinced that el categorization is not ambiguous, I suggest to take a sample (say the first twenty) of these pages, and if one of them is actually in Ancient Greek, then go ahead with your process. Otherwise just drop it. Place Clichy (talk) 22:50, 9 February 2014 (UTC)
An editor would use "el icon" instead of "grc icon" because "el icon" shows as "(in Greek)", which may be ambiguous to some editors. If "el icon" showed as "Modern Greek", the editor would have seen a problem and fixed it. I'm not saying that "(in Greek)" is wrong for Modern Greek, but ambiguity may result.
See Ptolemy, section "Primary sources"; Plato, section "External links" for examples that appear to use "(Greek)" to mean "(Ancient Greek)". – Jonesey95 (talk) 00:27, 10 February 2014 (UTC)
In common usage, I'm not sure that's wrong per se, just imprecise. We also have some external links to Old French texts that are marked "(French)" (ISO code fr, intended for Modern French) even though they are properly "(Old French)" (ISO code fro). I don't think the best solution is to explicitly mark modern languages with rendering like "(Modern French)" or "(Modern Greek)", but rather just to recategorize texts in old versions of languages to the more precise categories, as they're found. I suspect this comes up more often with el/grc than fr/fro just because the ratio of how much English-language readers talk about each language is much more lopsided in fr/fro, so confusion happens less, but the situation isn't really inherently different from ISO's perspective. --Delirium (talk) 21:25, 12 February 2014 (UTC)

Whether to wiki-link place names ("City, State" or "City, Nation") as one or two links?

I'm not finding any specific advice within the MoS as to which method is preferable (and why). When linking within articles, should the name of the municipality be linked separately from the greater region (with a second link for the region), or should they be linked as one? As an example, should it be Portland, Oregon, or Portland, Oregon? Thank you, startswithj (talk) 01:01, 13 February 2014 (UTC)

Hi, generally the best practice has become to link only the most specific geographical item, given that the less specific ones will be linked very prominently in that target article. Providing a direct link to "Oregon" at the top of a BLP would be of little use. However, in an intrinsically geographical article, such as on Portland, yes, link Oregon (preferably once, in the lead). There's probably no rule about whether it should be combined because it's hit and miss whether the article includes the state. I believe there's objection somewhere else on en.WP (can't think where) to unnecessary piping to remove a state from a city link (and then the insertion of a separate link to the state). Tony (talk) 02:16, 13 February 2014 (UTC)

Linking non-proper-noun "the"

I've been under the impression that we don't include the lowercase "the" in Wikilinks, i.e., "the Olympic Games" rather than "the Olympic Games". Yet I can't seem to find a guideline that states this. Does anyone have more information? --Tenebrae (talk) 18:41, 31 January 2014 (UTC)

Good question. Grammatically it would be more logical to include "the", since it's part of the nominal group; but unless the owners of a proper name are particularly fussy about the/The in their title, I couldn't care less in terms of linking. There's an advantage to making less rather than more of our text blue for each link—for one thing a narrower link display is less likely to judder against another close by. (See bunching advice in MOSLINK.) I don't think this is worth making a rule about. Thanks. Tony (talk) 03:26, 2 February 2014 (UTC)
I don't think grammar really covers links. Anyway, you need to exercise caution here, as adding "the" to a link can change the destination article (e.g., Queen vs The Queen; Guardian vs The Guardian). I wouldn't feel obligated to extend the link to the whole quoted text: link whichever text is the WP:COMMONNAME for the intended name, but feel free to include "the" if it still points to the correct article. —sroc 💬 13:04, 2 February 2014 (UTC)
Wikipedia:Naming conventions (definite or indefinite article at beginning of name) --  Gadget850 talk 11:45, 15 February 2014 (UTC)

Names within names

I raised this issue in April 2012; there was only one response at the time, and nothing came of it. I propose an addition to the section "What generally should not be linked": names within names. If there is a building, street, airport, park, other geographical entity, tool, or pretty much anything named after someone or something else, do not link to the name within the name. For example, all of the following links should be avoided: Hancock Airport; Wilshire Boulevard; Maimonides Medical Center; Mount Edith Cavell; Hoover High School; Mahatma Gandhi University; Humphrey–Hawkins Full Employment Act; Esquipulas Peace Agreement; Rio Conference; I Love Lucy; The Lion in Winter; The Pink Panther; Cascade Elementary School; Phillips screwdriver; Ferris wheel; Teddy bear. Whether the outer name would be a red link or blue link, the link should be to the outer name or not at all. I look forward to a constructive discussion here to generate consensus and refine the wording. —Anomalocaris (talk) 23:54, 13 January 2014 (UTC)

  • Those examples would count as nonsense-linking, or linking for its own sake, in my book. I usually see these where there are no "suitable" or immediately relevant links, and some editors create the links, seemingly out of desperation for some article to link to, resulting in the contrived or contortionist links we occasionally see, and these links are rarely germane. -- Ohc ¡digame! 01:52, 14 January 2014 (UTC)
I raised the issue here because of the discussion at Talk:Highway 4 (Israel)#Street name links. If you agree that links to names within names should be avoided, we need a rule on it, otherwise editors will continue to make (and revert removal of) illogical links like in Highway 4 (Israel)#Junctions and interchanges, which include, among others, Bar-Lev Blvd., Ben Gurion Blvd., Menachem Begin Blvd., Aluf Sadeh St., Hadassim Youth Village, Weizmann St. and many more. —Anomalocaris (talk) 11:52, 27 January 2014 (UTC)
Agreed, such links are nonsense.
  • If the topic is notable it should have its own article and the whole thing should be linked (Ferris wheel.
  • In a few cases, the topic may only be notable enough to warrant a mention or section in another article, which may well be the article of the person for whom it is named, but the whole name should be linked (pointing to the relevant section, if possible), not just the name within the name (Wonkavator, not Wonka-vator).
  • If the topic is clearly not notable at all, it shouldn't be linked at all, in which case it might be useful to include a brief description that links to the notable person or reference (Reverend Frederick Cavell, father of World War I nurse Edith Cavell).

—sroc 💬 13:49, 2 February 2014 (UTC) [revised 14:03, 2 February 2014 (UTC)]

So, three of us agree that such links are bad. Is this a consensus? Shall I edit the style guide? —Anomalocaris (talk) 05:16, 25 February 2014 (UTC)
I agree, definately. Not much more to add to what you have already brought up earlier :D Jayaguru-Shishya (talk) 18:58, 6 March 2014 (UTC)

Suggestions for additions to WP:OVERLINK

Doing a lot of work on sport articles, I run across a lot of uses of wikilinks and many feel like overlinking. I am interested in seeing if any of the following have been discussed before or if we could consider strengthening language around the cases:

  • Did WP:OVERLINK once include nationalities in the "What generally should not be linked" section? It seems like it once did. At any rate, I am interested in asking why nationality is not treated the same way as country (generally not linked). It seems like linking "Australian" in a bio to either Australia or Australians is about the same as linking the country. Not sure if this was removed and if so why, or if we could consider adding (re-adding?) it?
  • There is a section under Piped links that gives an example of linking "apples" as either [[apple]]s or [[apple|apples]], suggesting that the first is preferable. Any reason why this language isn't strengthened in this section to say that the first example is how this should be displayed? It produces the same result and marginally reduces the size of articles, of benefit to WP servers.
  • Is there a reason that the practice of breaking a city article up into two links (e.g. - [[Waltham, Massachusetts]] becomes [[Waltham, Massachusetts|Waltham]], [[Massachusetts]]) isn't specifically addressed? I can see a few cases where the separate links might be called for, but in most this is a complete waste, adds nothing for the reader and doubles the transclusions present.

I am interested in understanding the history here (I am sure this is not the first time for any of this to be discussed) and, if possible and desired, amending this section of the MOS to address these items. Thoughts? Rikster2 (talk) 13:57, 26 November 2013 (UTC)

Linking the nationality of a person is much akin to country linking, but on the other hand, if there is a "<nationality> <profession>" list or a more explicit categorization that can be made, then that's reasonable, eg: "John Smith is an American composer...". The city/state/country break-up is part of chain linking and strongly discouraged. As for the plural example, we're under no storage issue requirements (the net result will be the same amount of HTML delivered to the end user), so it seems too bitey to try to require the shorter version in wikicode. --MASEM (t) 14:30, 26 November 2013 (UTC)
I agree with re-adding it to the "What generally should not be linked" -section. Jayaguru-Shishya (talk) 20:16, 6 March 2014 (UTC)
Thank you for the responses. One question, where is chain linking as it relates to city/state/country strongly discouraged? I did not see anything on the page that came across that way. Are you referring to the section that states "When possible, avoid placing links next to each other so that they look like a single link, as in [[Ireland|Irish]] [[Chess]] [[Championship]] (Irish Chess Championship). Consider rephrasing the sentence, omitting one of the links, or using a more specific single link (e.g. to Irish Chess Championship) instead?" It seems odd that the city/state example would not be specifically addressed since it occurs so often. Thanks. Rikster2 (talk) 15:46, 26 November 2013 (UTC)
I didin't see anything on the page either that'd have came across that way. I definitely agree that those should be part of the "not to be linked" -list. Jayaguru-Shishya (talk) 20:16, 6 March 2014 (UTC)
Taking your Waltham, Mass example - the way that I understand it is that if somebody doesn't know here Massachusetts is, they're unlikely to know where Waltham is either. If they're interested in finding out where Waltham is, they'll click the Waltham link (whether that be [[Waltham, Massachusetts]] or [[Waltham, Massachusetts|Waltham]]); and if they still don't know where Mass is, there is bound to be a link from the Waltham page. But it's less likely that they'll click on [[Massachusetts]] first, then say "ah, that Waltham - not the one in Maine then". So I consider that making a separate link for Massachusetts is superfluous. I have seen it stated, but I don't recall where. --Redrose64 (talk) 17:11, 26 November 2013 (UTC)
I have to disagree with you, Redrose64. As wikilinks aren't actually references to support any particular claim, they are mere additional readings for background information. This means that they are not 1) refrences (to support claims), nor 2) passive references, that are including a set of assumptions widely affecting to the topic (Kivilä, S., Lindblom-Ylänne, S. & Mäntynen, A. 2007. Tiede ja teksti: tehoa ja taitoa tutkielman kirjoittamiseen [Science and text: efficiency and skill to writing a scientific paper]. Helsinki, Finland: WSOY). Jayaguru-Shishya (talk) 20:16, 6 March 2014 (UTC)

Re the prose recently added to the Overlink section: Would it be appropriate here to wikilink A4 paper and Letter (paper size)? I've personally always been leery of including links to main space articles in guidelines, policies, etc., finding them too transitory to rely on. However, I suspect that US editors might be taken aback by the term "A4" without a link. --Chaswmsday (talk) 12:30, 1 December 2013 (UTC)

I don't see much of the "apples" issue, but certainly common (stand-alone) nationalities should not be linked. This was a near universal practice seven years ago, but started to wane when people realised that a sea of blue wasn't helping the wikilinking system one bit. Generally, try to avoid bunching related links, as MOSLINK says. The usual practice is to link only the most specific item in a geographical string (on first occurrence, and if it's not obviously well-known like New York City or Paris), on the basis that the less specific items will be linked to at the start of the more specific geographical article. I'm a bit wobbly about the advantage of linking to "American composer": almost always a more specific link can be found, or none is necessary at all. Is there something not obvious about the item? Tony (talk) 14:09, 1 December 2013 (UTC)
By "prose recently added to the Overlink section," I take it that Chaswmsday is referring to this reverted material? Flyer22 (talk) 21:11, 1 December 2013 (UTC)
  • The "prose recently added to the Overlink section" is distinctly creepy. -- Ohc ¡digame! 01:39, 2 December 2013 (UTC)
Uh, yeah. That diff was what I meant. Sorry I was unclear. Now that it's been reverted, my question is moot. --Chaswmsday (talk) 13:20, 2 December 2013 (UTC)
Why do yo uthink it is creepy? I think it is consistent with the overall goals: linking where useful but not overlinking. Seems on point. Which is why I wrote it. Dovid (talk) 10:29, 5 December 2013 (UTC)

Propose explicitly adding nationalities to OVERLINK - I'd like to suggest nationalities be added to the list of items that generally should not be linked. Example:

  • the names of major geographic features and locations; nationalities; languages; religions; common occupations; and pre- and post-nominals;

How about it? Seems consistent with what people in this discussion think the guideline means anyway. Rikster2 (talk) 14:15, 7 December 2013 (UTC)

Counterproposal – I'm uncomfortable with that loose major, as it is, and further uncomfortable about loading more onto it, without prior bounding. We have some convenient metrics on countries upon which 'major nations' might be prescribed, e.g. list of countries by population, list of countries by GDP (nominal) and list of countries and dependencies by area. How about we:
  1. Separate major countries (from major geographic features and locations), into a dedicated (prior) clause, with major defined by a simple combination of available metrics.
  2. Add major nationalities to the same clause, with major similarly defined (without a GDP exclusion, I suppose – since the nationality is likely to be complex / informative for small in-egalitarian countries).
There may be further scope to further tighten major where more becomes implicit within a narrowly scoped article.
I think this is an issue where WP:CSB aspects are due for consideration (e.g. my GDP reservation, above).   – Ian, DjScrawl (talk) 15:42, 7 December 2013 (UTC)
I agree there's a problem (and always has been) with defining the qualification "major". I'm not so sure thought that there's a sensible solution to it. Nor am I sure we should be using such a distinction at all: a geographic location, or nationality, is either relevant to the topic at hand, and hence worth linking, or it isn't. Whether it's "major" or not is surely a secondary point, and one rooted in the misconceived rationale that linking is only about helping explain things to notional readers that they might not otherwise know what they are or understand rather than also being about facilitating intra-site navigation to other topics of likely interest. Also, if this proposal is a bid to not ever link certain nationalities, eg in opening sentences, it's not going to have that effect, precisely because of that "relevance" exemption. N-HH talk/edits 15:52, 7 December 2013 (UTC)
I think the word "generally" means that adding it wouldn't mean "never." I think some good points are being made (though I think there are issues with using GDP - "brand awareness" is more the question). I just think "nationality" should absolutely follow the same standard as "country" and "language" - whatever that is. There are certainly times when linking nationality makes sense, but I'm not sure footballers and artists are "generally" those times. Rikster2 (talk) 16:54, 7 December 2013 (UTC)

interwiki linking to other languages within body text

Is there a guideline concerning linking to other language wikipedias within an article body when there is no equivalent English article? For example, the paragraph on liquor in the Jerusalem artichoke article discusses a German liquor called "Topinambur" that we currently have no article for. It links instead directly to de:Topinambur without any indication that the reader will be sent to German wikipedia:


I'm not intensely opposed or in favour of this practice, but it's not one I've run into before, so I thought I'd ask. - 86.101.7.169 (talk) 10:00, 21 January 2014 (UTC)

I don't know the policy, but I think this should be discouraged. It essentially stifles article creation as many notable people/places/things don't yet have English Wikipedia articles. If an editor sees a redlink that is a signal to create such an article. If they see a bluelink they might get the mistaken idea that the article exists and not create it in the English version. Maybe someone else knows what the actual policy is, but again I think it is a bad practice. Rikster2 (talk) 12:57, 22 January 2014 (UTC)
I agree totally. Besides, what's the use if an non-German speaking reader is redirected to a German language wikisite? He wouldn't understand it anyway. I am strongly in favour of prohibiting linking to other language articles. Jayaguru-Shishya (talk) 20:16, 6 March 2014 (UTC)
see Help:Interlanguage links. There are suggestions there on how//when to use them, as wikidata link, or inline links--☾Loriendrew☽ ☏(talk) 13:36, 22 January 2014 (UTC)

Thanks to you both. Would it be worth including a reference to that page in the MOS/Linking ? 86.101.7.169 (talk) 13:36, 30 January 2014 (UTC)

Proposal: Do not use links to highlight words or ideas

My understanding has been that wikilinks are solely meant to provide easy access to further information about the linked terms. However, I occasionally see, especially from novice or infrequent editors, links being used to visually highlight certain terms in the text, in a manner akin to the way boldface, underlining, or italics might be used. To my knowledge, there is no explicit statement in the MOS or in any other guideline that wikilinks should not be used in this way, so I propose that we add such a statement to the MOS/Linking page, perhaps in the WP:OVERLINK section: 

Do not create wikilinks in order to highlight or draw attention to certain words or ideas in an article. Links should be used to help clarify the meaning of linked words, not to place emphasis on the words.

Or something to that effect (we can discuss the precise text to be used). Thoughts? Dezastru (talk) 21:30, 16 February 2014 (UTC)

Good idea. Cases like this one, which I just encountered, can only be explained – it seems – by inexperienced editors completely failing to grasp the purpose of links. --Florian Blaschke (talk) 03:17, 25 February 2014 (UTC)
-> Nice idea, I like it. Jayaguru-Shishya (talk) 20:16, 6 March 2014 (UTC)

In-text wikilinks possessing a serious threat

I am new to take part in this discussion, so hello there to you all! First of all, I'd like emphasize the importance of this article and how glad I am that this problem has been taken seriously in the Wikipedia Community. In my own experience as a contributor, however, it isn't just the navboxes (like strongly emphasized at WP:LINKCRISIS) that are causing a problem. It is the very regular in-text wikilinks.

Excess linking does not only lead to a quadratic explosion in the total number of links, but reduces the quality of articles making them hard and unpleasant to read. Excess linking also makes a disfavor for the contributors as they might not really describe the terminology they are using, but just throw a wikilink instead.

I personally think it like this: if you are writing an article on paper, how many times and in which occasions you would like to add brackets referring to an external source? E.g. Volkswagen is a German (see Bayer, H.: World Geographic Atlas, 2013) automobile manufacturer headquartered in Wolfsburg, Lower Saxony, Germany (see Dickinson, R. E.: Germany: A General and Regional Geography, 2011). Volkswagen is the original and top-selling marque (see Kotler, P.: Principles of Marketing, 2006) of the Volkswagen Group, the biggest German automaker and the second largest automaker in the world.

I am not sure if the above-mentioned serves as the best possible example, but I think it demonstrates quite well the situation with excess linking. If one sentence has even up to three wikilinks (and many times even more), and this keeps recurring from sentence to sentence, the text would be absolutely ludicrous! In a paper-published article you couldn't naturally do that. What I think, is that the same that applies to publishing a paper article, applies to Wikipedia's articles too.

I am optimistic about the future though, and a lot of good work has been done in order to fix this matter. WP:OVERLINK and WP:LINKCRISIS are already raising awareness on this matter. Much is still to be done. For example, here at WP:OVERLINK, the names of major geographic features and locations; languages; religions; common occupations; and pre- and post-nominals are mentioned in the list of "not to be linked". This is just a tiny fraction of the problem, but perhaps the aforementioned could be extended to geographic locations in general (towns, cities, regions, countries). Well, that's just an example, so better not to grasp too much to it.

Anyway, great thanks to everybody working in order to fix this problem and to improve the Wikipedia project! :) Jayaguru-Shishya (talk) 18:47, 6 March 2014 (UTC)

I don't believe the analogy to citation in print publication doesn't provides any guidance here, regarding how much wikilinkage is appropriate. By analogy even to the most heavily referenced print publications, we wouldn't link even Wolfsburg in the example lead paragraph about Volkswagen.
--P64 (talk) 18:10, 26 March 2014 (UTC)
Actually, all that was from an existing Wikipedia article, Volkswagen :) I just wanted to illustrate how it would appear on a print publication by replacing the wikilinks with in-text references to some external sources.
I agree with you P64, we shouldn't link even [[Wolfsburg]], but unfortunately we do at the moment. So the point was, they do not reference so heavily in the print publications, but unfortunately we do here in Wikipedia. Therefore it'd be a good idea to take example from the print publications, where the use of in-text reference (equivalent to our wikilinks) is much more restrained. Well, that's my opinion Jayaguru-Shishya (talk) 19:33, 26 March 2014 (UTC)

An exception for a duplicate link

Often times when I come across a linked topic I'll pass it by because I'm not interested at the time. Typically I'll come across the same topic several sections away and I'll be wanting the convenience of the link without having to scan and search for it about the article. In cases where the topic is important to the article subject and it occurs far away from the initial linking, it would be a big convenience to this reader (and I'm sure the average reader who could care less about a duplicate link) if the topic were linked twice in a practical location. Duplicate links are already allowed in captions, info boxes, etc for this reason. Proposal:

  • A duplicate link (i.e.one more) is permissible in cases where the topic to be linked is a key topic to the article-subject and when it occurs in a different section and/or far away from the initial occurrence of the link, however discretion should be used.

A good example would involve an extra link for the commander of a ship, when the article is about the ship, or vise-versa. -- Gwillhickers (talk) 22:13, 26 March 2014 (UTC)

Disagree. Unfortunately, I don't quite agree with you. Excess linking is already a problem serious enough in Wikipedia, and in my humble opinion, no new policies that encourage increased linking are needed. Jayaguru-Shishya (talk) 08:13, 27 March 2014 (UTC)
The problem is not serious from my experience. Content disputes, POV pushing, copyright violations, etc, are serious problems. I prefer to extend good faith and treat fellow editors like they have their own minds allowing them a bit more discretion in matters of simple editing. If and when an over linking problem occurs it can and has always been remedied. We certainly need rules and guidelines, but when they get in the way of practicality and bettering an article then Wikipedia allows us to deal with it. There are plenty of rules and guidelines for linking already. Be nice if we incorporated an exception when appropriate and not 'treat pedestrians like they need to wear crash helmets'. One of Wikipedia's biggest problems is that for every one writer we have ten editors on a 'mission to clean up' who do little more than spend their time playing wiki-cop going around hounding editors over "serious" problems and imo the overly restrictive linking policy only gives them one more reason to do so. -- Gwillhickers (talk) 16:52, 27 March 2014 (UTC)
@Jayaguru-Shishya and Jayaguru-Shishya: How is "excess linking" a serious problem? In certain articles it is annoying to have excessive links to irrelevant or trivial material, but a draconian "no duplicates!" policy does more harm than good in making it harder to for readers to find relevant material. Mandating a quantitative limit on the number of links does nothing to address irrelevance. ~ J. Johnson (JJ) (talk)

What is "major" for the purposes of overlinking?

WP:OVERLINK says not to link "the names of major geographic features and locations".

This is a list from an existing article:

Florence, Italy, Australia, Benin, British West Indies, China, Costa Rica, England, France, Japan, Kenya, Mexico, Spain and Zambia.

It could easily be argued that some of the linked places are not "major", but where do you draw the line? Does it mean "likely to be able to be pointed to on a map by the average reader"? Doesn't that depend on the place of residence, level of education, and experience of the reader (something that varies widely)? Aren't we being asked to make a decision that could be taken to mean value or importance to the world?

This guideline results in a significant number of edits and disagreement among editors. Is this really worth doing, just to avoid a few extra links in an article?

If I believed in this rule, I could easily unlink all of these as being "major", yet I would link Saint-Quentin, Aisne, which I've never heard of (but is a fairly large commune), which might seem stupid to many Frenchmen, who may feel quite justified when they come along and unlink it as being "major". —[AlanM1(talk)]— 03:58, 26 March 2014 (UTC)

Good question. I would also approach the matter by asking, "how relevant is the content of that wikilink to the actual article?" Perhaps it could be put some other way instead of using a wikilink? For example, in biographical articles, a person might indeed have lived in - let's say - Saint-Quentin, Aisne at some point of his/her life, but how relevant is the content of [[Saint-Quentin, Aisne]] as a whole to the article? Could it be possible to say, e.g. "...Saint-Quentin, a commune locating in the Aisne Department in Northern France; ..."? We are on a very slippery slope if we only think about about one's assumed level of knowledge.
I'm not sure if this is helpful but in my last thread on this Talk Page, In-text wikilinks possessing a serious threat, I was comparing the use of wikilinks with the use of in-text references in a paper-published articles: would you really add an in-text reference to an external source after every single geographical location (town, city, region etc.)? Wouldn't it appear somewhat ludicrous?
In my humble opinion, I concluded that when researching a paper-published article, indeed one wouldn't be able to do that. On the contrary, the author should phrase and write the text in such a manner that each piece of special terminology that is being used, is also being described to the reader. So forth, no disfavors would be made and the author would stay alarmed to keep the text easily-readable to the readers. What holds with the paper articles, also holds with wikiarticles.
Aside from an individual article, there are also few points that we should notice considering Wikipedia as a whole. Excess linking leads to:
  1. quadratic explosion in the total number of links
  2. decrease in the quality of articles making them hard and unpleasant to read
Trying not to talk too much, I just add one more citation from what I wrote earlier. I hope it'd demonstrate my point even just a little =P It goes as follows:
"I personally think it like this: if you are writing an article on paper, how many times and in which occasions you would like to add brackets referring to an external source? E.g. Volkswagen is a German (see Bayer, H.: World Geographic Atlas, 2013) automobile manufacturer headquartered in Wolfsburg, Lower Saxony, Germany (see Dickinson, R. E.: Germany: A General and Regional Geography, 2011). Volkswagen is the original and top-selling marque (see Kotler, P.: Principles of Marketing, 2006) of the Volkswagen Group, the biggest German automaker and the second largest automaker in the world.
I am not sure if the above-mentioned serves as the best possible example, but I think it demonstrates quite well the situation with excess linking. If one sentence has even up to three wikilinks (and many times even more), and this keeps recurring from sentence to sentence, the text would be absolutely ludicrous! In a paper-published article you couldn't naturally do that. What I think, is that the same that applies to publishing a paper article, applies to Wikipedia's articles too."
Anyway, thanks for your post AlanM1, you are bringing forth a serious topic here! Jayaguru-Shishya (talk) 10:11, 26 March 2014 (UTC)
The patchy effect is regrettable, although inescapable in some contexts (usually not as bad as that). Question: are links to any of those countries helpful in the context? It would be fine to link all if they are, I believe. Yes, there's a grey area where we end up having to make judgements as to how much our readers—first- and second-language speakers—are familiar with. Tony (talk) 11:08, 26 March 2014 (UTC)
If the links were rendered as in-line parenthetical clutter or even footnotes, I might agree that they are distracting and should be more limited, but when they are simply a slightly different color, ...? I quite often have reason to navigate to articles (especially those on places) that are just tangentially related to the subject, and am annoyed when those links are not there, especially when they were originally there and then removed by someone, forcing me to open a new tab, nav to wikipedia, type the name in the search box, and then select it from the results. —[AlanM1(talk)]— 03:38, 17 April 2014 (UTC)
They're not just "a slightly different color" from the black default text: they're a garish blue—a hue that Tim Starling told me at a WM dinner years ago could well have been chosen by one of the early technical people on a whim. I'd have chosen a blue two shades more like the black (see my user page for how to alter the colour), with an option to make it the current garish blue in prefs, if your vision really wants the disruptive effect. Better still would have been plain black that underlines and is clickable when you hover over it—and every single word linked automatically. This is the case for my Encarta desktop dictionary: you get zero visual obstruction and maximum linking, all in one. The disadvantage, and it is a big one, is that it removes the current service we give readers in isolating our suggestions for the most valuable links. I'm afraid that one-click links to targets that "are just tangentially related to the subject" would litter the text, under the current system, with blotch-blue. Look at many articles in foreign-language WPs for that—the French, the Italian, the Russian, for starters; they seem to have no serious system for using wikilinking intelligently (i.e. selectively), even though most have some form of guidelines, usually as vague as water. Editors on those sites who argue for more disciplined wikilinking are usually howled down: it's a great pity. Tony (talk) 06:12, 17 April 2014 (UTC)
I agree with you Tony (talk) . In Finnish Wikipedia: the more links, the better. For example, if there are dates October 5th and 2004, they are both linked as October 5th, 2004. There is also actually an article for breakfast[4], where words, such as bread, porridge, fruit, berries, yogurt, cheese, and eggs are all linked :) I'm not kidding!
AlanM1, how does linking a geographic location help you to gain a better understanding on the main article anyway? Okay, you don't know for example where Nicaragua or Saint-Quentin, Aisne are located. Then you click the link and some 9000 words will pop-up in front of your eyes telling you everything about the natonal/regional economy, administration, population, transport, politics etc. (call me a no-life, but I just copy-pasted the whole deal into a words-calculator). Now, how relevant can you consider that wikilink? How much does that linked article itself help the reader to understand the main article he was reading?
Also, would you add to a printed article an external reference after such location? If we truly want Wikipedia to compare with distinguished encyclopedias (where we have some prominent results, so far), the exactly same applies with the conduct of writing wikiarticles. Please also bear in mind that there will be a quadratic explosion in the number of wikilinks if we indulge in excess linking. Jayaguru-Shishya (talk) 07:30, 17 April 2014 (UTC)

Cities, states, provinces?

I was about to ask the same question, but my question has to do with cities and states. Some editors consider all states as "major locations", and some consider some large metropolitan cities (New York, Los Angeles) as "major". I don't like when an MOS is so vague. --Musdan77 (talk) 19:29, 12 April 2014 (UTC)

My problem is that I don't know how we can justifiably call "all U.S. states" major but not all English regions/counties. What about Indian states? Nigerian states? A large number of people would consider each of these 1st-level admin divisions as "major" and unworthy of links, yet most of the world may not have heard of them:
—[AlanM1(talk)]— 03:38, 17 April 2014 (UTC)
As a "rule of thumb" if having a page linking to the state or other top-level entity would help the reader of the article in which the location is mentioned, then linking is appropriate. If it's only tangentially related to the topic and your average reader will at least have heard of the place, then no linking is required. I would link the 8 examples above in almost all articles, but I would not necessarily link London in an article about a person who happened to be born in London, England unless there were details in the city's article not known to the average reader which would help in understanding the person whose biographical article I was editing. Vague? Yes. When in doubt, I generally link such places in their first usage. davidwr/(talk)/(contribs) 21:22, 19 April 2014 (UTC)

What generally should not be linked?

Generally, a link should appear only once in an article, but if helpful for readers, links may be repeated in infoboxes, tables, image captions, footnotes, and at the first occurrence after the lead.[5]

I'd like to open discussion about the last addition with boldings. I don't think it is necessary to repeat a link after the lede. For example in many movie or band articles, there are usually members and castings listed in the lede. Aside from that, they are usually mentioned in various infoboxes also. If we now stick to the practice of linking everything that has already been linked in the article lede, we will end up with 3x the same link in total.

Most of all, we should always remember that every time we add a wikilink in the article, we send the reader a message: "There is something really important you should read about behind this link". However, in most cases the actual linked content is just tangentially related to the subject. Now, to link even once needs to have good reasons behind it; it seems highly unlikely that we would have good grounds to make such a claim to the reader thrice already.

Therefore, I'd like to suggest that the aforementioned passage is changed to:

''Generally, a link should appear only once in an article, but if helpful for readers, links may be repeated in infoboxes, tables, image captions, and footnotes." Jayaguru-Shishya (talk) 07:48, 22 April 2014 (UTC)

I strongly disagree with that proposal. The important part is if helpful for readers, and it's a good point. There are many times, mostly in lengthy articles, when it's convenient to link to a complex subject a second time. Sometimes a topic is mentioned in the lead, and then not again until 100,000 bytes later; if the subject is of significance to an article, I would prefer the option of occasionally creating a second link. As for the implied importance of a wikilink, we should not assume that readers are incapable of managing their own interests. Wikilinks provide more info for readers at their own option. We are not sending the message that they should read about it, merely that they can read about it if they so choose. Grayfell (talk) 07:08, 24 April 2014 (UTC)
Thanks for your response Grayfell! I may disagree with you, but I strongly appreciate your opinion! :D Jayaguru-Shishya (talk) 18:35, 24 April 2014 (UTC)
Likewise. I appreciate the sentiment, and agree that overlinking is detrimental to the project. I think the they way to address it is to follow the current standard, rather than changing it to be more limiting. Grayfell (talk) 20:26, 24 April 2014 (UTC)


For lengthy articles 3x links are appropriate. The word vital is wikilinked in another section of the article far way from the history section. The word "vital" is also unfamiliar to most readers. See WP:UNDERLINK: articles explaining words of technical terms, jargon or slang expressions/phrases—but you could also provide a concise definition instead of or in addition to a link. If there is no appropriate Wikipedia article, an interwiki link to Wiktionary could be used. User:Jayaguru-Shishya, I don't understand the reason why the wikilink was removed when you consider it fixes a technical word that is underlinked. QuackGuru (talk) 20:21, 26 April 2014 (UTC)

That sounds pretty much the wrong way around. I'd say only link if relevant. If explanation is needed, then make it plain on the face of the article or gloss to wiktionary for comprehension. Avoid links that detract from encyclopedic understanding of the subject and dilute the effectiveness of the linking mechanism. -- Ohc ¡digame! 20:41, 26 April 2014 (UTC)
QuackGuru, this is not the right forum to discuss edits of some certain article. Please take discussion to the place where it belongs to. Jayaguru-Shishya (talk) 15:25, 8 May 2014 (UTC)
QuackGuru: we expect readers of the English Wikipedia to know what "vital" means. Convenient desktop dictionaries are now free for all computers, too. Tony (talk) 15:55, 8 May 2014 (UTC)

The iceberg junk code

This recent addition strikes me as something that needs to be removed, and certainly does not reflect WP consensus or normal practice:

  • Items within quotations should not generally be linked; instead, consider placing the relevant links in the surrounding text or in the "See also" section of the article. Alternatively, immediately after the quotation and its references, consider adding an {{Efn}} template for each term you're linking and explain its meaning (usually by getting it from the linked-to article's first sentence); and before the References or similar section add a Notes section with a {{Notelist}} template. Example:
Classical arts reviewer Chris Ng wrote, "the mural shows icebergs with great sensitivity."[39][40]{{Efn|[[Iceberg]], a large piece of ice that floats freely after it broke from a glacier or ice shelf}}
== Notes ==
a. [[Iceberg]], a large piece of ice that floats freely after it broke from a glacier or ice shelf

I don't think I've ever seen such an absurdity in any article here, and this kind of abuse of complicated footnote code "bloat" is far worse than any excessive bluelinking problem.  — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  00:47, 14 May 2014 (UTC)

Oh my. Yes, that is awful. Let's delete it (or at least decently bury it) before it propagates. ~ J. Johnson (JJ) (talk) 22:45, 14 May 2014 (UTC)

I'm going to WP:BOLDly just delete that nonsense.  — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  02:13, 16 May 2014 (UTC)

Links on/off button?

I don't know whether this is technically possible. If it is, I don't know if it's politically possible. But it's worth at least throwing out there and getting shot down, since it would render most of this discussion moot.

What if there were a button to toggle the links off and on?

The page would be rendered initially as it is today. If the links were hurting my readability, I could choose to turn them off.

Whether that would mean making them non-links, or simply changing the color of the text, would be a purely technical consideration, and I think the latter would be better if it's possible.

What if I then wanted more information about a word or phrase in the text? That would depend on the implementation. If only the color were changed, then the links would still be there and functional. I could hover over the word/phrase; if it became underscored, and my mouse pointer changed to the "hand", and the tooltip popped up, I would know it was a link and I could click on it as usual. If, on the other hand, the implementation was to change it to a non-link, then it would be necessary to toggle the links back on.

It might or might not be necessary to re-render the page when the button is pressed, I guess that depends on the capabilities of HTML. But I wouldn't see that as a serious problem, since the majority of pages render fairly quickly. Mandruss (talk) 09:37, 12 May 2014 (UTC)

You can apply a local stylesheet (in Special:Mypage/skin.css, or locally, in your browser) to style links however it suits you. You could also use a (browser-speciifc) plugin to disable all links, on a page-by-page basis. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 11:06, 12 May 2014 (UTC)
Thanks, Andy. I guess I shouldn't have used the first person personal pronouns. This isn't about me and my readability issues, it's about a general solution to the problem of wikilinks and readability. Mandruss (talk) 11:13, 12 May 2014 (UTC)
You can certainly tone down that default garish blue: see my userpage ... scroll down. Tony (talk) 11:49, 12 May 2014 (UTC)
Allow me to suggest that we limit ourselves to solutions that will work for all Wikipedia readers. First, they need to be aware that a solution is available. Then they need to be capable of using it. Mandruss (talk) 12:00, 12 May 2014 (UTC)
The solution to overlinking is to remove excessive links, if there are any; but my response was based on the fact that what you (or any individual editor) prefer to see is not what others do. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 13:03, 12 May 2014 (UTC)
Agree on your first statement, but let's consider the case where there's no significant overlinking, just a lot of appropriate links. It happens a lot, and it's what at least some of this page is about. It's what this solution addresses. I hope you meant reader rather than editor; I don't think anyone discussing the issue of readability is talking only about editors. And regarding your last statement, I think this solution allows everyone to have it their way --- as well as change "their way" with the click of a button. Mandruss (talk) 13:19, 12 May 2014 (UTC)

Some interesting issues are intersecting here. Let's think about some extremes:

  1. We could go the way of my desktop Encarta dictionary (which incidentally has a button taking you to the WP article of the same title you've typed into the dictionary/thesaurus); in Encarta every single word functions as a link, but professional appearance and smooth reading are not degraded by patchy blue spattered all over the place—you only see a word turn (light) blue and underlined when you hover your mouse over it; until then, it's plain-sailing black. What Encarta sacrifices—if it was appropriate (which it's not)—is the opportunity to display intelligently selected and rationed links to readers. It's not an idea that appeals to me in relation to our environment. This selecting is what wikilinking does so well on WP, when not used to excess and when skilfully done.
  2. We could abolish linking. No one is thinking of that, I hope.
  3. We could go back to the old days of linking anything in sight, as many (most?) other language WPs still do, today. I think the en.WP community has spoken on that one, now that people have become used over the past eight years or so of gradual taming of the linking system. I get a lot of echo-thank-yous, from editors I've never heard of, for gnoming out overlinking nowadays; it surprises me.
  4. We could give readers on/off buttons, yes; that proposal would need careful thought, and I'm too attached to the maintenance of a good linking system to think this is a viable option. If linking is carefully applied and not too disruptive of the reading process, perhaps that wouldn't be necessary. See No. 5 ...
  5. We could float the idea of a button or two that offer toned-down link hues. Perhaps: "Light-blue link color ◌ "

    I've become so used to the colour patch/thingy I installed about six years ago that when I go to other WMF sites, links are like neon signs. See possible gradations here. But I'm such a tech dummy I can't work out how to update the advice at the bottom for vector (it's still says "monobook"). BTW, many people wouldn't be happy with my choice of minimal difference between the link colour and the default black text, but there are gradations slightly sharper gradations that are a little closer to the current default. I'm interested in people's opinions generally about pref. buttons for link colour—and link colour in general. Apparently the colour was chosen without much thought by a WMF engineer one day back early this century (according to Tim Starling, when we spoke a long time ago about it). Tony (talk) 15:02, 12 May 2014 (UTC)

I wouldn't have a problem with your option 5, with just one little change: the option to go all the way black. Is there a reason I'm missing why that would be a bad thing? Mandruss (talk) 16:00, 12 May 2014 (UTC)
Greetings! I am glad to see that the issue is being discussed here. I have to say, however, that I agree with Tony1 on this one. It's more about displaying intelligently selected and rationed links to readers. In one word, relevance. Moreover, in most cases the actual linked content is just tangentially related to the subject, so there isn't really any reason to link such content. In my humble opinion, the main issue here is good linking skills. Jayaguru-Shishya (talk) 16:17, 12 May 2014 (UTC)
I think that convenience (to the reader) ranks with relevance as the essential criteria as to whether some link should be added. However, that wasn't the initial topic of this section; should we open a new section for that? ~ J. Johnson (JJ) (talk) 19:59, 12 May 2014 (UTC)
Links are for convenience. But the selectivity (based on editorial skills/judgment) is a large part of their utility for readers. Just maximising "convenience" would result in every single word being linked, and then you lose the selectivity benefit. Comment on Jayaguru's post: not relevance alone, but utility, which is a tougher test. Part of utility is the obviousness of link pipes (misleading pipes are a major problem I sometimes have to fix when gnoming); and specificity: fixing vague pipes gets to be too much for gnoming, since often you have to research sections and offspring articles to identify a more specific target). So quite a few additional criteria are involved in the balance sheet of whether to link, and if so, how. It's a skill like prose-writing that wikis have made possible ... indeed I would say they demand it for high quality. Tony (talk) 03:42, 13 May 2014 (UTC)
By convenience I was thinking in terms of how many places a link should be replicated such that a reader doesn't have to scroll back several lengthy sections to find the link. A link should not be inserted anywhere if it is not relevant, and even a relevant term should not be linked on every use. The interpretation of "only once per article" is being used to override editors' considerations of other criteria, wherefore I think we should explicitly allow "once per section". But we have wandered off-topic; perhaps this should be discussed elsewhere? ~ J. Johnson (JJ) (talk) 23:04, 14 May 2014 (UTC)
Let's do discuss that though; I concur that "once per page" is too rigid a rule, though "once per section" is too loose; typically once per page is enough, but once per page as a default is fine, if we permit re-linking in subsequent sections if it seems particularly pertinent to do so, especially in longer articles.  — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  02:13, 16 May 2014 (UTC)
I think a clear guideline is necessary since otherwise there would be a link on every mention. I appreciate your opinions but personally, I am in favour of once in an article. Jayaguru-Shishya (talk) 13:29, 16 May 2014 (UTC)
Okay, I am going to work up a proposal; will bring it up in a new section in a day or two. ~ J. Johnson (JJ) (talk) 21:16, 16 May 2014 (UTC)
  • But my default assumption is that readers start at the opening and read through. If they want to poke around patchily, they should expect lots of the same item to be blued out: they should be assume to instinctively know that skimming back is requred—or they can type into the search box what they want. The slippery slope would return us to the link-anything days, whereas it's been a lot of work to get the system pared back to where each link really means something. Tony (talk) 03:43, 17 May 2014 (UTC)

Let's continue this discussion below at #Proposal to allow links once per section. ~ J. Johnson (JJ) (talk) 20:37, 18 May 2014 (UTC)

Proposal to allow links once per section

This was a remarkably fractious and bad-tempered discussion which included confrontations and allegations that seemed rather unnecessary. I see that at the end of the discussion, some editors feel that this discussion has not resulted in a consensus. I beg to differ. I think it has reached several conclusions, and I'll enumerate them in the following paragraphs. Before I do, a word about my closing methodology: I gave no weight to the many ad hominem or ad feminem allegations in this discussion, and in particular, I gave no weight to the many allegations about J. Johnson's motives; and in several cases where the same argument was repeated several times in different wording, I gave it weight only once. The substantive, closable part of the discussion is a great deal shorter than it might seem at first glance.

Taking into account the fact that where there is no consensus, the status quo continues to apply, this RfC has reached the following conclusions:-

Q: Should we make the specific change to the MOS wording mentioned at the start of this RfC?
A: No.

Q: Should we allow one link per section?
A: No.

Q: Does WP:OVERLINK mean that a given article can only ever have one link to another given article?
A: No. The word "generally" in WP:OVERLINK does allow for occasional exceptions. These exceptions should only arise on the basis of good editorial judgment. This is what article talk pages are for. There is a presumption that duplicate links would be removed in most circumstances, and GA and FA reviewers are often particular about it. They are entitled to be particular when assessing what's supposed to be Wikipedia's best content.

But there is a genuine consensus to increase the frequency of links. Not necessarily to the level of one per section, but editors do express a feeling that relevant links are helpful to our readers and where an article is long, it should not be necessary to scroll from one end of the article to the other in order to find the link you want.

Q: How should this work in practice?
A: Although I can see no consensus on that point in the discussion below, the common sense answer is that repeat links to carefully-selected relevant articles can be used to reduce scrolling. While I was preparing this close, I spent a little while looking for examples of where I think this is done appropriately. The first one I found was Second Battle of El Alamein where Bernard Montgomery's name is linked at least four times in appropriate places. There is a consensus to allow this practice.

Q: Should we edit the Manual of Style in some way to make this more clear?
A: Not at this stage. The current language does allow for this interpretation but I think it's fair to say that it's less than crystal clear. Editors are encouraged to seek agreement on an appropriate wording in a more focused and collegial discussion to take place after the dust has settled from this close.

I do hope this helps. If anyone has questions, commentary or personal abuse relating to this close, it should be directed to my talk page in the first instance.—S Marshall T/C 02:02, 10 August 2014 (UTC)

The following discussion is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.

This proposal arises from a discussion that developed above at #Links on/off button.3F, and a discussion at WT:Manual of Style/Layout#See Also section: is our guidance realistic.3F that came down to a question of overlinking.

At WP:OVERLINK the MOS says: "Generally, a link should appear only once in an article..." (emphasis in the original). Some editors have ignored the "generally" and interpret this strictly, using this as grounds for removing all "duplicate" links, including those in a separate "See also" section. Many editors find the strict "only once" interpretation too rigid, while others are concerned about (to quote Tony) a "slippery slope [that] would return us to the link-anything days". A reasonable compromise is to limit links to only once per section. Therefore I propose that the last two sentences at WP:OVERLINK be replaced with the following, which slightly relaxes the rigid interpretation, and clarifies some related aspects:

In general links should be made only at the first instance in the text of an article or section (not counting instances in infoboxes, tables, image captions, or footnotes). Where a link is proper at different places in an article it may be repeated as a convenience so the reader does not have to scroll back through the article to find it. Links that are otherwise acceptable should not be removed solely because they are duplicated, but only if duplicated excessively, such as multiple times in a given section.

"See also" sections are intended as a convenient guide to articles on related topics that a reader might be interested in, independently of whether the such articles have been mentioned or linked in the current article. Links are not precluded from a "See also" section because they have appeared earlier in an article.

~ J. Johnson (JJ) (talk) 20:47, 18 May 2014 (UTC)

Support/Oppose

  • Support as proposer. ~ J. Johnson (JJ) (talk) 17:36, 24 May 2014 (UTC)
  • Oppose—I think this is very unwise. Tony (talk) 03:14, 19 May 2014 (UTC)
  • Oppose It is specifically discouraged to have a link in the 'See also' section that is already in the article itself. -- Ohc ¡digame! 03:31, 19 May 2014 (UTC)
  • Oppose as a terrible idea. Specifically ruled out by WP:SEEALSO. --John (talk) 10:00, 19 May 2014 (UTC)
  • Support the first paragraph, oppose the second. As a reader I really hate it if I have to go link hunting, i.e. if I have to manually scan the article text to find a link just because the term was already linked somewhere else. E.g. an article about a composer might mention specific works at different places, and then it would be really inconvenient if only the first place were linked. Tobias Bergemann (talk) 10:17, 19 May 2014 (UTC)
  • Oppose This would take Wikipedia back to the Stone Age of linking. Even the current policy is quite a flexible one. Jayaguru-Shishya (talk) 16:30, 19 May 2014 (UTC)
  • Oppose - if people are ignoring part of what's currently there, you should discuss the matter with them. The current wording is flexible enough to allow relinking where it makes sense to do so; the proposed wording is far too loose. Nikkimaria (talk) 20:45, 19 May 2014 (UTC)
  • Support the first paragraph for the reason given above by Tobias Bergemann. Many readers jump to a section and will not have had an opportunity to benefit from an earlier link. (So far all the "oppose" arguments seem to be "we've always done it that way." Not particularly persuasive.) Butwhatdoiknow (talk) 21:34, 19 May 2014 (UTC)
  • Oppose - I didn't think Wikipedia was a democracy! 64.134.232.217 (talk) 18:34, 24 May 2014 (UTC)
  • Support - the policy of linking once only is a nonsense, particularly in lengthy articles - linking within sections makes much more sense. Readers will often jump to a specific section (which is what the TOC is designed for) so why force them to go hunting through each other section to find a previous occurrence of the word(s) to find a link to the relevant article? The mobile apps of WP present an article with all the sections closed up and actively encourage section jumping and so exacerbate the problem. It is also an extremely poor experience for users of screen readers who will find it very laborious to use link nav to scroll up and down the page. Bladeboy1889 (talk) 10:26, 26 May 2014 (UTC)
  • Support first part - much more helpful to the reader. Consider also mobile view, where sections are collapsed, and a link in, say, section 1 may not be seen by someone who has expanded, say,only section 4. Oppose second part. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 13:47, 27 May 2014 (UTC)
  • Oppose The issue seems to be the implementation of the current policy, rather than its actual wording, and I am nervous about how the altered wording would be applied in practice. Hchc2009 (talk) 14:36, 27 May 2014 (UTC)
  • Oppose Is the proposal a solution to a nonproblem? 166.147.88.28 (talk) 02:47, 6 June 2014 (UTC)
  • Support This change would address two annoyances I have frequently seen in longer articles. The purpose of links is to help a reader quickly finding associated information without having to guess the exact article title and enter it into the search box. Complex articles aren't read linearly. Users might skip chapters by clicking the TOC, or they might arrive at the article via a link from another article or redirect to a section header or embedded anchor. A strict rule for no repeated links defeats the very purpose of why links exist in the first place. From the usability point of view, a reader cannot be expected to hunt previous chapters for links - a reader might waste alot of time scanning through information outside his/her scope of interest trying to find a link without even knowing if it exists. This is a very bad user experience. Therefore, the question, if a link should exist or not, should be answered from the viewpoint of a reader arriving at a section header without knowing the previous part of the article. In a very short section or article, it might still be okay to scan the whole article for a link, while in longer and more complex articles it is not. Similar for See also sections. While it is generally unnecessary to repeat already existing links in the See also section, sometimes See also sections contain well-organized lists of orthogonally related topics and if some of the links, which would belong into such a list, cannot be found there (because they are already used further up), it could easily be interpreted as if the subject isn't covered at all in Wikipedia. In such cases, not repeating the link creates inconsistencies and hinders easy navigation. --Matthiaspaul (talk) 20:45, 8 June 2014 (UTC)
  • Oppose It opens the door to large scale overlinking and repetition of large number of links in the see also section that are already in the text.--SabreBD (talk) 23:14, 8 June 2014 (UTC)
You didn't provide a rationale for your comment. My request for further comment was moved to the bottom of the page; perhaps you would comment there. ~ J. Johnson (JJ) (talk) 00:14, 11 June 2014 (UTC)
  • Oppose Yes it opens the door for unnecessary linking and since it is irrelevant to the subject it shouldn't be linked anywhere. Sometimes the word is so common that there is no need of introduction. OccultZone (Talk • Contributions • Log) 04:39, 8 July 2014 (UTC)
What the heck are you talking about? Any link that is "irrelevant to the subject", and links to common words, are already excluded by the existing rule, and this proposal makes no changes to that; your comment irrelevant. "Unnecessary" linking is not the same objection as "large scale overlinking", and seems very weak. Links that may not be strictly necessary can still be useful. The possible inclusion of an unnecessary link seems a much lesser offense than the exclusion of a useful link. ~ J. Johnson (JJ) (talk) 20:53, 9 July 2014 (UTC)
@J. Johnson: You cannot link to very common words, geolocations, professions even once. That one link is only for linking a important subject. You cannot link Barack Obama two times and it is about whole article. OccultZone (Talk • Contributions • Log) 11:13, 15 July 2014 (UTC)
As I just said: links to common words are already excluded, and (did you miss this part?) that is not changed by this proposal. But if you need to discuss this further we should do it below, lest we stimulate JS to further removals. ~ J. Johnson (JJ) (talk) 23:27, 15 July 2014 (UTC)
  • Yes and no, mostly no. The second part is bunk; the way see-also sections current work is just fine, and they're terribly overused as it is. Usually there should not be one, when there is one, more often than not all the links in it are redundant, or can with very little effort be worked into the text. That said, the first part is onto something we've all know for a long time, that the "once per page" thing is not actually viable, especially for long and/or dense articles. But "once per section" isn't useful; many sections are only one sentence. The proper solution is to clarify existing wording a little to be clearer relinking is okay where it makes sense to do so, such that users do not have to unreasonably link-hunt. "It already says that" isn't good enough; if it were being reliably interpreted that way, this proposal would not be happening and attracting so much support for its top half.  — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  06:20, 10 July 2014 (UTC)
I mostly agree (except re See Also). Any suggestions? ~ J. Johnson (JJ) (talk) 21:55, 10 July 2014 (UTC)
  • Oppose for a few reasons. First, sections are an implementation detail, not a clear division of content like chapters in a book. A section may be as small or as large as the content requires so for some sections it may be appropriate to link multiple times and for others it may be appropriate to not link across many sections (even if we define a section as something under a level 2 header). Second, although I understand the frustration that comes from policies built around some conception of shared common sense--in this case the word "generally" should lead an editor to link where it might be appropriate just as the admonition to link once would lead another editor to remove the same link--we should be very hesitant to replace broad guidance with narrow proscription. While narrow proscription is of course the purview of MOS (I kid, I kid), it rarely results in the outcome desired. This is especially true where we're creating a narrow rule to clarify a more general one or where we're aiming for a broad outcome (getting roughly the right amount of links on a page) using a specific decision rule. Protonk (talk) 18:13, 15 July 2014 (UTC)
  • Support one link per section in article body, Oppose in "See also." — In the typical GA, sections are often detailed enough to be read independently, and it's not unlikely that a reader will look at the TOC and jump to the most interesting section. I think the proposal provides a genuine service to readers who jump into an article at a section and do not always read from the top down. Not a joke at all: I think of this proposal as particularly helpful to readers with disabilities such as ADHD. The proposal for one link per section helps make WP more accessible. On the other hand, I think there's a serious risk of bloat in "See also" sections, which are too often abused (especially in C-class and lower articles) as a poor substitute for incorporating relevant matter into the article body. Lwarrenwiki (talk) 20:54, 18 July 2014 (UTC), updated 21:09, 18 July 2014 (UTC)
  • Oppose - Generally speaking, the problem that most Wikipedia articles are experiencing is not too few links, but too many. Duplicate links withing relatively short articles of 750 words or fewer do not need the same word linked once in the lead and then once in each of three relatively short sections that follow. This is what I see on a daily basis. There is no doubt a formula for a revised MOS section on linking that would provide enhanced advice on when repeated links are appropriate within an article's main body text, but this, in my opinion, is not it. I see this proposal as codifying a more permissive standard than I believe is advisable, and will only contribute to the present OVERLINK problems, rather than reducing them. Dirtlawyer1 (talk) 13:25, 4 August 2014 (UTC)

Discussion

Invoking the "not repeat" language at WP:SEEALSO is meaningless as it only repeats the "only once" langauge here (from which it is derived). It most certainly does not rule out making changes. Nor does it rule out "duplicate links": the supposed prohibition is given (in both places) as a general rule. The problem is that quite a few editors routinely remove such "duplicate" links, without consultation, as being flatout prohibited by WP:OVERLINK (end of discussion, editorial judgment and common sense be damned). The current language is nominally general and flexible, but the result is not.

You opposers seem to fear some immediate and overreaching kudzu-like fate for en:Wikipedia if ever a link be repeated in an article. Note that this proposal in no way relaxes the requirement for relevancy, nor what should not be linked. There has been no showing how duplicating a link in "See also" is in any way objectionable, except for violating an arbitrary and rigid prohibition. If this is such a terrible and unwise idea please show how that is so. ~ J. Johnson (JJ) (talk) 00:18, 20 May 2014 (UTC)

We don't "fear", we know. Less is more. Links are valuable when properly used; they detract when in excess. Although overlinking has reached its peak, it would be a shame to let those gains in parseability/readability be eroded by removing provisions in the guideline that have served the project well. In the same way as WP:MOSFLAG allows the use of flag icons in clearly defined but limited circumstances, editors have naturally gone beyond what is allowed or optimal, and linking practices are no exception. Links far enough away from the first one are not prohibited as it seems the nom implies. Thus the rationale for allowing an existing link to be repeated in the See also section is mighty weak, and is likely to turn it into a repository of repeat links. -- Ohc ¡digame! 01:24, 20 May 2014 (UTC)
Unfortunately, this started out as two proposals: One to make it clear that links may be repeated in articles. Another to say that links can be repeated in See also. Are you two discussing both proposals or only one? And if one, which one? Butwhatdoiknow (talk) 12:29, 20 May 2014 (UTC)
  No, it is really only one proposal. The second paragraph clarifies a special case. The existing OVERLINK text says once per article, which implies the whole hog, and then exempts infoboxes, tables, etc., without addressing "See also" sections. The proposal specifies the text, which would not cover a SA, but Wikipedians being so various in their interpretations I thought it best to explicit address SAs. I could have proposed merely adding "See also" to the list of exemptions, but it seemed to me better to go for the most general case.
  Ohc: Setting an arbitrary limit ("only once"!) on the number of links without regard to their quality or relevance has a serious failing: it gives a free pass to link ANYTHING (with a few exceptions) once. Provide any example you want, and identify every instance of what you deem overlinking, and I suspect that many of those links (most?) will be 1) non-duplicated links, and therefore pass under the existing rule, and 2) better handled (i.e., removed) on the basis of relevance and quality.
  As to turning a SA "into a repository of repeat links": do you mean like repeating every link in an article? The way to avoid that is not by absolutely and universally prohibiting repetition, but addressing that problem at WP:SEEALSO. ~ J. Johnson (JJ) (talk) 22:57, 20 May 2014 (UTC)
"It gives a free pass to link ANYTHING"? No, not really. It's been described at sections What generally should be linked and What generally should not be linked. In my humble opinion, the latter should also be made more precise even: it's still missing many problematic areas that should be added on the not to be linked list. For example, I hope WP:OVERLINK could be extended to geographic locations in general (towns, cities, regions, countries). A small quote from my post at What is "major" for the purposes of overlinking?:

... how does linking a geographic location help you to gain a better understanding on the main article anyway? Okay, you don't know for example where Nicaragua or Saint-Quentin, Aisne are located. Then you click the link and some 9000 words will pop-up in front of your eyes telling you everything about the natonal/regional economy, administration, population, transport, politics etc. (call me a no-life, but I just copy-pasted the whole deal into a words-calculator). Now, how relevant can you consider that wikilink? How much does that linked article itself help the reader to understand the main article he was reading?

.
Jayaguru-Shishya (talk) 00:18, 21 May 2014 (UTC)
  • What u talking about, Dude? WP:UNDERLINK says

    An article is said to be underlinked if words are not linked that are needed to aid understanding of the article. In general, links should be created to:

    *relevant connections to the subject of another article that will help readers understand the article more fully (see the example below). This can include people, events, and topics that already have an article or that clearly deserve one, so long as the link is relevant to the article in question. (emphasis is mine)

    That links need to be relevant and useful to the reader's understanding is unambiguous, and the converse being that anything that isn't strictly relevant and useful to the understanding of the subject would constitute overlinking. -- Ohc ¡digame! 03:39, 21 May 2014 (UTC)
That's exactly my position: that relevancy (or usefulness to understanding) is the proper criterion of whether a given link is in, not whether it's already been used. The current rule (as interpreted) is that no matter how strongly relevant, useful, etc., a link is, sorry, you can use it "only once" (emphasis in the original) in the entire article. And if editors won't assess links on qualitative relevance and usefulness then, yes, any link can get a free pass once. ~ J. Johnson (JJ) (talk) 21:05, 21 May 2014 (UTC)
  • And if editors won't assess links on qualitative relevance and usefulness then, yes, any link can get a free pass once – Your fears are overblown and based on a false hypothesis. I see plenty of anecdotal evidence, on a daily basis, of editors applying the "relevancy test". -- Ohc ¡digame! 00:19, 24 May 2014 (UTC)
What are you talking about? If you're saying that editors do assess links on the basis of relevancy, then fine, we don't need a fixed, quantitative limit of "only once". On the other hand, it is a fact that "only once" is being applied by some editors strictly and without regard to relevancy (resulting in non-hypothetical aggravation). I'm saying the proper criterion for assessing links is, as you said, "relevant and useful to the reader's understanding". At best "only once" is just a crutch to avoid having to look at relevancy. ~ J. Johnson (JJ) (talk) 18:14, 24 May 2014 (UTC)
Relevancy deals with the quality of the link. About the quantity, it is said that:

Generally, a link should appear only once in an article, but if helpful for readers, links may be repeated in infoboxes, tables, image captions, footnotes, and at the first occurrence after the lead

I don't see any reason to loosen the current one. There are editors out there who would like to link a specific term on almost every single mention, and I don't find it a good idea to encourage that via WP guidelines. Jayaguru-Shishya (talk) 13:57, 25 May 2014 (UTC)
  JS: Not eactly. What the page says is "... only once" (you left off the emphasis). And I have shown you the reason for changing this: some editors intepret this as absolutely only once, regardless of quality. Quantity, to the measure of only once, trumps quality, relevance, helpfulness, etc.
  Your objection to linking "a specific term on almost every single mention" is ill-formulated. If a term is mentioned only once (and assuming a link is relevant and useful), would you object because every instance (albeit only one) is linked? What is the precise problem you envision with linking "on almost every single mention"? ~ J. Johnson (JJ) (talk) 00:53, 26 May 2014 (UTC)
  • As clearly stated in the guideline, a second appearance is not prohibited, but seems to be strongly implied by JJ; the assertion of only once is not borne out by any evidence – I have not seen editors removing a second occurrence outright when there is reason to keep same (for example, where an occurrence is in a section far removed from the first). OTOH, the proposal opens the door to linking of relevant words as many occurrences as there are sections in any given article. From my own observations about past linking practices, I do not consider this to be a desirable state of affairs. -- Ohc ¡digame! 01:42, 26 May 2014 (UTC)
"[O]nly once" (with emphasis) is from WP:OVERLINK (second to last sentence), as just quoted by JS. Yes, it is supposedly qualified by the leading "generally", but in practice it is strictly interpreted by some editors as sufficient basis for removing all "duplicates". Evidence? see (e.g.) this edit and this edit for "outright removal" on the sole basis of being duplicates. ~ J. Johnson (JJ) (talk) 21:31, 26 May 2014 (UTC)
Ohconfucius: Do our different views here resolve (in part) to your "I have not seen editors removing a second occurrence ...", whereas I have seen them? You asserted that "the assertion of only once is not borne out by any evidence". I have provided two contra-examples where links have been removed on the sole basis of being duplicated; do you require more? ~ J. Johnson (JJ) (talk) 20:59, 4 June 2014 (UTC)
J. Johnson: You seem to have created this proposal as ex-post justification for edit-warring with another editor over their removal of overlinking. I have no problem with those removals from the See also section. I am only surprised that more links were not eliminated, but perhaps the editor was only working based on WP:SEEALSO. You just have to search around the rest of the article and count how many times the links are repeated after the 'See also' links had been removed to see what I mean. -- Ohc ¡digame! 08:00, 5 June 2014 (UTC)
You said you had "not seen editors removing a second occurrence", I supplied two instances, and on that basis you make a wild accusation of edit-warring? If that is the best response you can make then I think it is adequately demonstrated that "duplicate" links are in fact being removed on the sole basis of being duplicate. Moving on, I suggest that the question is now whether this is how WP:OVERLINK should be interpreted. ~ J. Johnson (JJ) (talk) 23:11, 6 June 2014 (UTC)
The examples you gave are not second instances, but nth instances. The Seattle Fault article isn't very long, "Tacoma Fault" is already linked once in the middle of the article, and once in the navigation template at the bottom, so the See also link is not warranted. User:Brianhe's edit was perfectly reasonable and also in line with guidelines. See this evidence of your edit warring to reinstate the link to Tacoma fault.

In the Puget Sound faults article, I found five links to Seattle Fault in the version you apparently object to, plus one instance using the {{main}} template. If these two examples are the best you can do to contest my assertion, I think we can close this discussion straight away. -- Ohc ¡digame! 02:51, 7 June 2014 (UTC)

Geez, if you wanted more/better examples you only had to ask. (The ones I previously supplied were just the easiest to find.) Fine, here are some examples that cite WP:OVERLINK as authority for "only once" where I have no involvement:

  • Edit summary at Plate tectonics: "Alfred Wegener's name is already linked twice in the article - so for strict compliance, one of those links should be removed, ...."
  • Comment: "That was a duplicate link, removed per WP:OVERLINK".
  • Comment: "Sorry, but every city is already linked, and that's why they're not supposed to be linked again. Per WP:OVERLINK, a link should really only appear once in an article unless the two links are a long way away from each other in the article."
  • Comment: "According to WP:OVERLINK— A link should appear only once or twice in a page."

And there is a whole trove of "There are a couple of duplicate links which should be removed per WP:OVERLINK" comments from a script being used on GA articles, with subsequent removal. E.g.:

And many more instances of second instances being removed merely for being a second instance, as explicitly stated. Your assertion that "a second appearance is not prohibited" is contradicted by the evidence. ~ J. Johnson (JJ) (talk) 21:21, 7 June 2014 (UTC)

Okay, there are other examples, but are these examples of a serious problem? 173.160.49.206 (talk) 21:29, 7 June 2014 (UTC)
Well, Ohc has just accused me of edit warring (hardly a collegial attitude) simply because I and another editor have different interpretations of OVERLINK. Whether an absolute "only once" rule is helpful or not is frequently questioned, most recently here and here. ~ J. Johnson (JJ) (talk) 22:18, 7 June 2014 (UTC)
And you have not accused others of edit warring? 24.9.99.196 (talk) 23:38, 7 June 2014 (UTC)
An entirely irrelevant comment. With one exception (64.134.232.217) all of the IP addressees that have appeared in this discussion (166.147.88.28, 173.160.49.206, 24.9.99.196, and 166.147.88.48) have a common source (Denver, Colorado) and pattern of edits. If anyone else wants to discuss whether the "only once" interpretation results in a problem, fine. But I am disinclined to engage in an irrelevant discussion with an anonymous, wiki-hounding sockpuppet. ~ J. Johnson (JJ) (talk) 23:39, 8 June 2014 (UTC)
  • @J. Johnson: Just to recap: I said: "I have not seen editors removing a second occurrence outright when there is reason to keep same (for example, where an occurrence is in a section far removed from the first)". I stand by that. Note that I'm not saying second occurrences need to be removed in an absolute sense. But when there's a third (or more) link to the same article from another, chances are at least one can be removed.
    1/ Plate_tectonics: As far as I can see, this does not involve link removal, and there seem to be an adequate number of links for navigation. Arguably, the link locations could have been swapped, but it's otherwise irrelevant.
    2/ User_talk:Sitush: The article referred to was Narendra Modi, and it seems that "The Emergency (India)" may have been a necessary and relevant link, once. A second instance, only three sentences later, was superfluous and correctly removed.
    3/ User_talk:Imzadi1979: That comment is completely in line with guideline and what I am to preserve. The inexperienced editor was placing links in the next line and here omitting the necessary disambiguation. In fact, he/she had been attempting to alter wikilinks to mine metadata for an external site.
    4/ User_talk:Moonriddengirl: Er, seems to illustrate a humour deficit on your part. Thanks for providing the comedy, amigo.

    You have not disproved my above (repeated) assertion. As for me, I don't have any more time for this nonsense. -- Ohc ¡digame! 04:17, 9 June 2014 (UTC)

  I can't answer for what you can, or cannot, see. The "assertion" I address is where (on the 26th) I said that "some editors interpret this as absolutely only once", and then you said "the assertion of only once is not borne out by any evidence." What I have shown is multiple assertions by various editors that WP:OVERLINK means "only once".
  You seem to have overlooked that this proposal would still prohibit duplication within a section, and would not prevent removal of links duplicated "only three sentences later", or "in the next line" (cases #2 and #3 above); there would be no lessening of authority to do so (a point I made previously). Editors would have to give up this "more than one" quantitative crutch, but there would still be adequate bases for blocking "large scale overlinking". ~ J. Johnson (JJ) (talk) 21:10, 9 June 2014 (UTC)

Discussion continues

John: how is this "a terrible idea"? ~ J. Johnson (JJ) (talk) 20:33, 23 May 2014 (UTC)

John: your bare opinion, lacking any explanation or reason, amounts to no more than WP:I just don't like it, and warrants no consideration. Can you provide any basis for your opinion? Or should should we ignore it? ~ J. Johnson (JJ) (talk) 21:03, 4 June 2014 (UTC)

Yes, and please yourself, respectively. --John (talk) 22:14, 4 June 2014 (UTC)
If you have some basis for your opinion that might inform the discussion I should like to hear it; perhaps it would even change my views. (And I allow that even feelings and speculative considerations can be relevant, albeit not compelling.) ~ J. Johnson (JJ) (talk) 22:32, 4 June 2014 (UTC)
Let's be polite. — Preceding unsigned comment added by 166.147.88.48 (talk) 02:27, 6 June 2014 (UTC)

The implementation issue dervies from the "absolutely only once" interpretation, which is based on the existing actual language of "only once" (emphasis in the orignal). ~ J. Johnson (JJ) (talk) 23:03, 6 June 2014 (UTC)

As explained in previous comments: there are problems trying to find a link when it is buried in the article's text. There are also problems when editors clash on whether the policy allows any exceptions to "only once". ~ J. Johnson (JJ) (talk) 23:03, 6 June 2014 (UTC)

Sabrebd, regarding your "oppose" (above):

Why do you think this "opens the door to large scale overlinking and repetition"? ~ J. Johnson (JJ) (talk) 21:14, 9 June 2014 (UTC) (added context to comment that JS moved. J. Johnson (JJ) (talk))
J. Johnson, you are misinterpreting the guideline. There is no such thing as only once restriction. If somebody still misinterprets the guideline, you can sure notice him/her about that per WP:MOS/Linking.
Anyway, I have already quoted the current guideline, and you have shared your own opinions about it. If there is nothing more to add to the discussion, I'd suggest to not go over the same arguments over and over again. Jayaguru-Shishya (talk) 22:22, 9 June 2014 (UTC)
I suspect we keep going "over the same arguments over and over again" because we are not really on the same argument, have yet to find a common view from which to proceed. Like, you say that I am misinterpreting the guideline. Only I am not interpreting it (mis- or otherwise). What I am saying is that other editors are interpreting the guideline (as shown by the edits and comments I have pointed to) as an absolute "only once" restriction. Whether this is a mis-interpretation is a separate issue; my point is that the guideline is, in fact and as demonstrated, applied as an "only once" restriction. If you don't see that (or can't explain why I should see it differently) then there is a problem. If you do agree that some editors mis-apply "only once", then we can proceed to the next sub-issue. ~ J. Johnson (JJ) (talk) 00:28, 11 June 2014 (UTC)
I don't see how the words "only once' can be interpreted to mean anything other than 'only once'! - if the MOS is meant to mean that it is acceptable to link to relevant pages in multiple sections then it should be amended to say that to remove any misinterpretation. And if that is the intention of the MOS then I fail to see why there is anyone opposing this change - no one has suggested relaxing the outline of what should be linked, merely that if a link is relevant in the first section of an article there is no reason not to re-link the same name / word in the fifth section of the article if it comes up again.Bladeboy1889 (talk) 07:48, 11 June 2014 (UTC)
Yes, my point exactly. Though I can see multiple interpretations, others don't, and therefore clarification is needed. As to a reason for not duplicating links, the concern seems to be that any such permissiveness would be a "slippery slope" leading to "link-anything". Why that should be has not been explained; perhaps some opposer would explain that. I wonder if the WP:Overlink crisis sensitized people to such a prospect. ~ J. Johnson (JJ) (talk) 21:20, 11 June 2014 (UTC)
Then you can notify the users who violate this, this is the wrong forum to discuss it. The current guideline is very clear: a link should generally appear only once unless it's repeated in infoboxes, tables, image captions, footnotes, or at the first occurrence after the lead. Conversely, if a duplicate link does not fall under the first occurence after the lead, footnotes, image captions, tables or infoboxes, then it may be removed. If somebody still removes a link that does fall under one of the aforementioned, then you can notify the user about the guideline.
There is no reason to keep going over the same arguments over and over again. Jayaguru-Shishya (talk) 07:30, 12 June 2014 (UTC)
The whole point of the RFC was to suggest that relevant links should be viable in multiple sections of an article not once and once only to aid uability and accessibility for the reader. The objections have either been "no - because it means people will link things like dates and other non relevant stuff" (which the RFC doesn't suggest) or "no - becuase that's already what the MOS means" (which is patenetly not clear due to the wording.) So neither of those objections hold any weight because they don't actually address the issue. Bladeboy1889 (talk) 07:39, 12 June 2014 (UTC)
JS: note that we are not talking about the exceptions ("infoboxes, tables, image captions, footnotes") in OVERLINK, or mis-removal of same. The issue is whether "duplicate" links are allowed anywhere in the text of the article. You say that such duplicates "may be removed", simply for being duplicate. However, some of us disagree, and (skipping over why the guideline is not clear) argue that an absolute, quantitative limit of "only once" per article can be a disservice to the reader, and that where links may be "viable in multiple sections" (the other requirements of OVERLINKK still applying) a limit of once per section is more sensible, and more useful. ~ J. Johnson (JJ) (talk) 21:06, 12 June 2014 (UTC)
The point is moot, you have utterly failed to demonstrate anyone going around removing useful duplicated links. But you have demonstrated instances of editors removing links in 'See also' sections that repeat the links already made more than once in the text, and you seek to change this guideline because you disagree with a rule in WP:SEEALSO. Most bizarre. -- Ohc ¡digame! 00:37, 15 June 2014 (UTC)
  Talk about (below) WP:IDHT: I maintain that I have demonstrated removal of "useful duplicated links" on the explicit basis of being duplicated. You reject some of those instances of the basis of a personal conflict (implied bad faith??), or that they are SEEALSO issues, or that they warranted delinking. None of which disproves my point, which is the reliance on "only once". However, even in examples where those objections are not applicable it seems you still can't see them. So let's look at one up close.
  From Talk:Little Boy/GA1 we have an explicit statement of removing duplicate links on the sole basis of being duplicate:
  • There are few duplicate links in the article which should be removed per WP:OVERLINK. Those are: nuclear fission (in the lead), cordite, and Trinity nuclear test.
    • checkY Removed duplicated links. Hawkeye7 (talk) 12:58, 30 September 2013 (UTC)
  Actual removal was done here and here, with edit summaries of "rm dup link" and "rm dup links". This is only one of many similar examples to be found at GA. I don't know how much clearer this (or anything!) can be shown. (Why can you not hear/see this?) It is certainly more definite and more solid than anything you have shown us as to why allowing "duplicate" links will necessarily lead to "large scale overlinking." At any rate, even this single instance just demonstrated utterly explodes your statement that I have "utterly failed to demonstrate...." Whether you can see (or hear) that could still be debated, but your statement is disproven. ~ J. Johnson (JJ) (talk) 21:07, 15 June 2014 (UTC)
  • The link to "nuclear fission" that was removed was indeed a duplicate of a link that occurred in the previous paragraph within the same section (the lead, commonly the most densely-linked). Changing the guideline to what you want – permission to link in other sections where it may be relevant – "a link is proper" (your words) would not have prevented its removal. The link to "cordite" removed here was actually the third such link in the article. The relevant duplicated link still existed (in the "Design" section) after that removal. I don't see how it has any greater relevance or utility where it was removed. Your examples are either careless, or contrived, and show that you are obsessed with linking to the detriment of parsibility. Looking at your proposed revision, it seems virtually all "relevant links" can be duplicated in the 'See also' section. In practice, if relaxed, that will mean a large number of links will be duplicated there. I am certain it is not a desirable state of affairs. -- Ohc ¡digame! 04:33, 16 June 2014 (UTC)

Request for explanation of opposition

I have yet to see any explanation or argument as to why this proposal is "very unwise" (Tony), "a terrible idea" (John), "take Wikipedia back to the Stone Age of linking" (JS), "far too loose" (Nikkimaria), "opens the door to large scale overlinking" (Sabrebd). I point out that simply not liking something, without explanation or support, is not a helpful or useful argument; please explain yourselves. If you all cannot, or will not, explain why this proposal is unwise, or would lead to large scale over linking, etc., then such mere statements of opinion, in failing to contribute to understanding, are properly ignored. ~ J. Johnson (JJ) (talk) 20:47, 14 June 2014 (UTC)

  • Links, like medicine, are useful when used in moderation. Too many of either leads to side effects – in our case the detraction from other more pertinent and useful elements in the text, links or otherwise. Less is more. Careful selective use enhances; overuse devalues. Human nature being what it is, such loosening would lead to undesirable large scale overlinking, much like there is overuse of flagicons (but that is despite our guidelines). So your not accepting that as a rationale looks to me like WP:ILIKEIT. -- Ohc ¡digame! 00:43, 15 June 2014 (UTC)
You seem to be making a case for not linking at all. The guidelines as written do not provide the ability for "Careful selective use" at all - it says "first appearance and never again after that" which means "this is rigidly non-negotiable. If you're reading the fifth paragraph of the sixth section of an article and want to find out more information about someone mentioned then we're not going to help you - you've got to read the whole article from the top until you find mention of them previously in the second paragraph of the third section and we don't care about how usable or accessible that makes it." Bladeboy1889 (talk) 07:26, 15 June 2014 (UTC)
Ohc: I tend to agree with all of your general comments, except for two points. 1) "Less is more": If "only once" is better than all the higher numbers, why isn't zero even better? (Which goes to Bladeboy's point, above.) 2) Your implicit position that allowing even a single duplication necessarily leads to "undesirable large scale overlinking": this is not demonstrated. What I do not accept "as a rationale" is your failure to provde a rationale. ~ J. Johnson (JJ) (talk) 21:10, 15 June 2014 (UTC)
  • JJ, I appreciate that you care about the navigational aspect, the networking potential of a wiki. But as OC says, the density of links needs to fall within a window of optimal functionality. Perhaps you're not as oriented as many editors have become to the negatives in linking beyond a certain point. Many coders/programmers have objected to restraints on linking, because their profession cares a lot about navigational aids. But over the years we've learnt that readability and focus on high-value links is the other side of the equation. Tony (talk) 02:56, 15 June 2014 (UTC)
Accessibility and useability should be the key imperative and should not be dismissed just because they don't fit with a rigid adherence to a rule. The current guideline doesn't allow for "high-value" linking as you claim - it says "first mention only, no deviation from that whether it is best served there or not" so there is no allowance for even placing the allowed 'one time only link' in the most useful place in the text where is has most relevance - instead it must potentially be buried in a non-relevant section if it happens to have been mentioned there. This guideline imposes a a negative user experience and is prejudiced against anyone using adaptive technologies, anyone using the inbuilt page navigation aids or anyone accessing a mobile version and should be changed.Bladeboy1889 (talk) 07:26, 15 June 2014 (UTC)
Tony: thanks for your comment. I am, indeed, "not as oriented ... to the negatives", but that is the criticism I am making: that the negatives have not been demonstrated. Bladeboy has recapitulated the main points against the absolute "once only" rule. I have not seen any similar explanation of why, or even how, allowing some "duplication" is going to result in "large scale overlinking". ~ J. Johnson (JJ) (talk) 21:13, 15 June 2014 (UTC)


Bladeboy1889 and J. Johnson, your behaviour is coming very close to WP:IDHT. J. Johnson said (14 June 2014 (UTC), emphasis added):

If you all cannot, or will not, explain why this proposal is unwise, or would lead to large scale over linking, etc., then such mere statements of opinion, in failing to contribute to understanding, are properly ignored

.

According to WP:IDHT:

In some cases, editors have perpetuated disputes by sticking to an allegation or viewpoint long after the consensus of the community has decided that moving on to other topics would be more productive. Such behavior is disruptive to Wikipedia. Believing that you have a valid point does not confer upon you the right to act as though your point must be accepted by the community when you have been told that it is not accepted. [...] Sometimes, even when editors act in good faith, their contributions may continue to be disruptive and time wasting, for example, by continuing to say they don't understand what the problem is.

Time to let it drop. Jayaguru-Shishya (talk) 09:33, 15 June 2014 (UTC)

JS: Indeed, that is one of my points: as stated above: I have yet to hear (or see) any explanation or argument as to why this proposal is "very unwise", etc. But is that because I am deaf and/or blind? or that there has been nothing shown? Yes, there have been statements of opinion and belief, but where are the explanations or examples? We are supposed to reach consensus through reasons and persuasion, but on this matter we do not yet have consensus. ~ J. Johnson (JJ) (talk) 21:15, 15 June 2014 (UTC)
  • Bunkum. The "once only" rule that you cite (conveniently only in part) actually says: "Generally, a link should appear only once in an article, but if helpful for readers, links may be repeated in infoboxes, tables, image captions, footnotes, and at the first occurrence after the lead." Emphasis is mine. -- Ohc ¡digame! 04:44, 16 June 2014 (UTC)
  • However, I would agree that no consensus has been reached for a more permissive style of linking. -- Ohc ¡digame! 04:49, 16 June 2014 (UTC)
  • You say you tend to agree with me, but then by saying zero is better than one is classic case of "Reductio ad absurdum". I think this discussion is at an end. -- Ohc ¡digame! 04:52, 16 June 2014 (UTC)
  • I have already covered this. Of course "generally" can be interpreted as relaxing the "only once"; no where have I claimed otherwise. But whether that is a valid interpretation is not the point. My main point, as I said at the very beginning, is: 'Some editors have ignored the "generally" and interpret this strictly....'. And again on the 26 May: 'Yes, [WP:OVERLINK] is supposedly qualified by the leading "generally", but in practice it is strictly interpreted by some editors as sufficient basis for removing all "duplicates".' Do we need to go over this yet again?
  • Reaching consensus is not facilitated by having to play whack-a-mole because someone is not paying attention.
  • Do you actually understand the application of reductio ad absurdum? In the present case you stated "less is more" (meaning better), without limit or qualification. Such a limit is extremely important, because having accepted the implied progression that 4 is better than 5 (because it is less), and 2 better than 3, and 1 better than 2 — this last bit being the essence of your argument, right? — there is no reason to stop there, and therefore zero is just as validly "better" than one. Which is, indeed, absurd, because we do allow an occasional need for "one", but absurdity is a logical consequent of the unqualified "less is more". If you want to allow "one" but not "two" (i.e., "duplicates") then a specific limit or qualification is needed for that exception. ~ J. Johnson (JJ) (talk) 00:41, 17 June 2014 (UTC)
  • You are deliberately twisting things to suit your linking proclivities, and the examples you chose to prove your case are plain simply wrong. You will notice I said "less is more". I would have given you the benefit of the doubt, thinking you were choosing to be the devil's advocate. This has been going on fruitlessly for too long, and I'm inclined to say you are playing your absurd games and "re-interpreting" what I said into "none is more". There really is no point for me to continue this discussion because you are not listening. I am not playing at sillybuggers any more and become a troll. I wish you a good and contented life. -- Ohc ¡digame! 01:21, 17 June 2014 (UTC)
That "less" leads to "none" is not my "re-interpreting", it is the straight-forward logical result that follows from your assertion. If "none" is absurd, and the process needs to stop at "one", you need to specify that. ~ J. Johnson (JJ) (talk) 00:37, 18 June 2014 (UTC)
For anyone wondering if expansion leads to a similar absurdity: the problem with the simple, unqualified "less is more/better" criterion is (as demonstrated) that there is no basis for treating "one" as special. Yet there is an excellent argument for treating "one" as better than "zero", which suggests an alternate criterion: that more is better. And possibly this illustrates the essence of the issue here: if one is better than zero, why shouldn't two be better than one, and so forth? The position of the opponents is that relaxing this entirely arbitrary limit of "only once" thusly leads to "excessive" overlinking, which raises the previous question (but inverted) of: what is the limit or qualification that keeps two from progressing to unlimited ? As a matter of fact "unlimited" does not apply here at all, as the proposal has a very definite upper limit: once per section. (In the text, which is to say, exclusive of the existing exceptions for infoboxes, etc. And assuming that sections are not unlimited.) Furthermore, the other criteria (relevancy, usefulness, etc.) still apply, so (depending the context) the number of instances of a given link is quite likely constrained to less than this upper limit, even to zero. ~ J. Johnson (JJ) (talk) 21:36, 20 June 2014 (UTC)

I'm taking the discussion to User talk:J. Johnson#Misinterpretation of WP guidelines since it's no longer related to WP:MOS. Possible individual misbehaviour doesn't belong to WP:MOS, and should be discussed with either directly with the user or at forums like WP:ANI. Cheers! Jayaguru-Shishya (talk) 12:49, 17 June 2014 (UTC)

This discussion is not about "misbehaviour" of editors violating some clear and unambiguous interpretation of the guideline, but that some editors do (and I would add, in good-faith) follow a certain interpretation of the guideline with results here questioned, and a proposal made for changing the guideline. There is no element of misbehaviour, because the problems I allege are rooted not in editorial application, but in the guideline itself. ~ J. Johnson (JJ) (talk) 00:40, 18 June 2014 (UTC)
We should allow each other with flexibility and kindness, even when there's an adversarial environment. A quick look at JJ's talk page shows appreciation in barnstars by other editors. Nice. While I find it hard to go along with his views here, this page should not become personal in tone. Tony (talk) 12:34, 21 June 2014 (UTC)

Ohc: You should decide whether you are in or out of this discussion. And if in, then please keep in mind that your last comment characterising my argument as "absurd games", and your recent edit summary referring to "absurd logic", do not advance the discussion. You have not provided any basis why any of this is "absurd"; it appears that you have only called out a name you have heard but don't understand. Please note: the actual absurdity here is not the logic, but the result logically reached by the unqualified and unlimited application of your simple dictum "less is more/better".

When you first mentioned reductio ad absurdum I had a brief hope you might understand something about how this form of argument is applied. Apparently not, you truly "don't get it", so I suggest you leave off flinging that term around. Perhaps someone else can make a better argument, but I have yet to see it. ~ J. Johnson (JJ) (talk) 21:00, 24 June 2014 (UTC)

JJ, you do valuable work for WP, and you're not absurd (even though I don't agree with your line of argument). Please pass over my friend OC's use of the word, which was ill-judged, IMO. We all become passionate about WP, which is a good thing I guess. Tony (talk) 15:40, 26 June 2014 (UTC)
My takeaway from this little storm in a teacup is that JJ thinks his argument is perfect and everyone else's is lacking. If I ever said JJ was absurd, and I don't think I did, I would sincerely apologise. It was the absurdity of his "taken to its logical conclusion" logic that I believe is faulty, and I still maintain that to be the case. -- Ohc ¡digame! 15:46, 26 June 2014 (UTC)
My thanks again to Tony for trying to keep things mellow, and I shall try to follow suit. But (!) as Ohc can't decide whether to grab hold or let go, and has been the biggest opponent to my proposal, I think some resolution is needed. I propose we discuss the proper application of "absurd" at his talk page. ~ J. Johnson (JJ) (talk) 23:49, 28 June 2014 (UTC)


In the three weeks since I requested an explanation of the opposition to this proposal nothing has been established (except that Ohc doesn't understand the application of reductio ad absurdum). Can't any you nine opposers support your opinions that the proposal is "unwise", etc.? ~ J. Johnson (JJ) (talk) 22:18, 7 July 2014 (UTC)

  • I don't think the best way to elicit discussion is to badger folks who registered an opinion (Which is fine and all, discussions aren't a vote but neither are they a mock debate, not every comment needs to be a self contained affirmative case for a position). Further, by singling out opposing comments which didn't leave a rationale, you run the risk of convincing yourself that the opposition as a whole doesn't frame an argument, rather than specifically selected comments not comprising one. Someone closing a discussion like this (and many discussions are like this) should look at the preponderance of opinion and the arguments raised by each side. Each editor needn't invent a new argument to oppose or explicitly endorse particular arguments, so long as an actual argument exists on that side. Protonk (talk) 17:14, 29 July 2014 (UTC)
  • "Let's not bicker and argue about 'oo kill 'oo." I don't care what JJ and OhC want to hash out between themselves personally. I think it's clear that the proposal as a whole has failed to carry, yet there seems to be a notable multi-party view that the extant wording could use some minor clarification, to suggest that in particularly long/complex articles a re-link is okay, when it's especially pertinent and helpful.  — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  02:27, 2 August 2014 (UTC)
I'm not certain it hasn't carried as the bulk of those opposing have either given no reason or raised reasons that have nothing to do with the proposal. Part of the problem is attempting to make two changes in one proposal which means opposition to one part allows another part to be vetoed by association. Including discussion about 'see also' sections muddied the waters. Bladeboy1889 (talk) 18:47, 2 August 2014 (UTC)
In this sub-section I was hoping for someone to offer an explanation of or argument for (and by that I do not mean "heated bickering", but syllogistical reasoning) some of the opinions cited in opposition. Which is not meant as picking on anyone, only that I would like to see that someone else's opinion has at least as much support as mine. Additionally, perhaps the proposal should have been simpler, though that would have other problems. At any rate, I'm tired of this, it's not going anywhere; I'm going to close it. ~ J. Johnson (JJ) (talk) 21:47, 3 August 2014 (UTC)

Closure?

J. Johnson's proposed close, refactored to here from the {{Archive top}} he hatted the discussion with:

  • No consensus, and trying to work out a consensus not worth the aggravation. ~ J. Johnson (JJ) (talk) 21:48, 3 August 2014 (UTC)}}
Procedurally, you can't close it because you're not just deeply involved in the discussion, but the proponent to boot, and other editors have well-expressed stakes in the outcome(s). Your proposed close was also snarky, with a sour-grapes tone to it, addressing nothing at all about the merits or basis of any arguments, nor suggesting anything constructive toward resolution. Ask WP:ANRFC to close it. You and Bladeboy1889 have a view that no opposing positions have been well explained or defended, while those holding those views feel they've done so adequately and that you're engaging in WP:IDHT tactics. I'd call that a dispute that you can't settle, including by dismissively shutting down the discussion simply because, and in a way that lets you vent about, how frustrating it has been for you personally. Given that there are two, not one, well-argued-out proposals on the table, it's quite possible that a neutral closer will come to one conclusion or another about at least one of them. That would be better than a blanket "no consensus" stamp, rendering the entire discussion a total waste of time. For one thing, there may be a clear consensus against any expanded usage of "See also" sections, and as I suggested above, there may be a consensus to improve the wording a bit to permit re-linking in a later section, if an article is especially long or complex. The fact that the J. Johnson proposal was for something more specific doesn't mean that only "yes" or "no" on that idea are the only possible outcomes of this discussion.  — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  12:57, 4 August 2014 (UTC)
I filed a WP:ANRFC request for closure.  — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  13:02, 4 August 2014 (UTC)
Sorry; I didn't realize that closing it myself would improper. Nor did I mean to be snarky. I would like to comment on that in a little bit, perhaps the closure could be put off for a day or two (or indefinitely?) so this issue can be better settled. ~ J. Johnson (JJ) (talk) 21:51, 4 August 2014 (UTC)
  SMcCandlish has chided me for closing this discussion on my own. I had thought a formal closure by someone else was not necessary, as 1) I made the proposal, 2) there was enough opposition to show a lack of consensus to adopt, 3) it appears that discussion is not getting anywhere, and is unlikely to result in sufficient change of opposition to generate consensus for the proposal. At this point it seems clear: no consensus likely, drop the proposal (at least in this form).
  Perhaps where I could be chided is in not providing a neutral summary of how the discussion ran. Ideally this would include a summary of the arguments pro and con. But here is the rub: I don't understand the arguments against because I have not seen them. What I see ("hear" in the sense of WP:IDHT) are opinions against (e.g.: "a terrible idea", "take Wikipedia back to the Stone Age of linking", "far too loose", etc.), but (as I stated above at #Request for explanation of opposition) I do not "hear" (or see) the arguments on which these statements of opinion are presumably based. Please note: by argument I do not mean "heated rhetoric"; I mean such statements as logically connect from positions or viewpoints we share to the point of discussion. E.g., Tony1 said the proposal would be "very unwise". Oh? How is that? Has this specific proposal been tried before and found wanting? For all that Tony might have some good reason for his view, even arising from some consideration I would find compelling, there is no argument explaining that. And as I am not a mind reader, I cannot assess what is not communicated. In that respect I find the opposing opinions not well explained, and despite my request for explanation.
  To recapitulate: the opposers don't seem interested in my arguments, and I can't get them provide to arguments for their views, so discussion is moribund. We might wait a few days in case anyone wants to chime in, but I don't feel like dragging the horse to the water trough. ~ J. Johnson (JJ) (talk) 00:15, 5 August 2014 (UTC)
Oh Stanton, why couldn't you let sleeping dogs lie? It was much better when it was closed, improper though it may have been. Now it's going around the houses again... ;-) -- Ohc ¡digame! 01:43, 5 August 2014 (UTC)
Because improperly closed discussions are dogs that wake up; it's grounds for anyone later to claim there's really no consensus after all. It's not a mtater of chiding, it's matter of gauging neutrally what this discussion concluded if anything and not leaving it open to later WP:GAME stuff (I don't mean by J. Johnson, whose feeling that this discussion has been a dead end I take at face value, but by anyone, generally). The ANRFC horse will come to the water by itself. If the discussion is moribund this will be apparent, and will consist of people like us not talking about it any more.  :-)  — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  06:32, 5 August 2014 (UTC)
I see where u are coming from. However, bearing in mind that nobody is claiming a definitive position of consensus, I fail to the the risk that you are talking about will later turn back into a conflict situation. Anyhoow, the request has been lodged, so we let our dutiful admins do their work. Regards, -- Ohc ¡digame! 08:00, 5 August 2014 (UTC)
I'm apparently too simple to see how this might be gamed. But I think we all agree that not only are there no claims of consensus here (either way), there is in fact no consensus (either way), and, in this context, little chance of such. To the extent there may be potential consensus for some kind of changes (perhaps as a simpler proposal), that will have have to start afresh. I think we're done here. ~ J. Johnson (JJ) (talk) 22:27, 6 August 2014 (UTC)
The discussion above is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.

Wikiwand's link innovation

Anyone familiar with Wikiwand's innovation by which hovering the mouse over any wikilink produces a small pane of the opening paragraph or two of the target article? I rather like it. Example Tony (talk) 06:31, 7 September 2014 (UTC)

Try Preferences → Beta features → Hovercards. --  Gadget850 talk 09:31, 7 September 2014 (UTC)
Oh, thanks Gadget. I never knew. It's not as nice-looking as Wikiwand's, but I guess it will develop. Is it only for logged-in users, or can our readers use it? Tony (talk) 09:46, 7 September 2014 (UTC)
As with all preferences, only logged in users can select this.--  Gadget850 talk 11:13, 7 September 2014 (UTC)

HTML Lunacy: Annoyance or Asset?

I have from time to time run across "New York Times" in blue when an entry has mentioned a report in the Good Grey Lady. I have now learned, after a couple of tries, that the click-through will inform me that the New York Times is a newspaper. It will not give me the text of the relevant article.

Moving right along now... In a discussion of David Foster Wallace's landmark novel "Infinite Jest," "endnote" is blued as a click-through, because the novel has 388 endnotes. Massive end-noting is an interesting and uncommon literary device, so it's perfectly reasonable for a further discussion be to linked through the miracle of html. But no! This link tells us, in great length, a lot of stuff interesting to typographers, about endnotes in books.

I doubt that typographers are more common among those interested in this novel than they are in the common population.

There is no discussion of the novel's interesting use of the device.

In the spirit of positivity which always infuses me when I run across idiocy loose in the world, I wonder whether Wikipedia might be able to use these hilarious gaffes as an asset for improving the great encyclopedia over the long run.

Are moronic html a reliable sign of inappropriate people being loose among Wikipedia's editors? What needs to be done to hunt these rats down and exterminate them? Uh, excuse me. Is there a gracious but still effective way one can suggest that editors be perhaps encouraged to improve and be suitably informed, trained, upgraded -- if not stood out in the cold for 48 hours of Maoist self-criticism? Yet.

DavidLJ (talk) 00:55, 23 June 2014 (UTC)

What you expected to be linked should have been inline reference cites following the sentences that mentioned then; we use hyperlinks to point to relevant other articles on WP, and reference citations for "more information" on the actual fact. --MASEM (t) 01:51, 23 June 2014 (UTC)
  • Most of our editors are ordinary people who may have an amateur interest in the subjects they treat. That's why a lot of our articles are in a sorry state. I would agree that a more professional approach is desirable.

    Linking is one problem area, as some editors seem to have difficulty in applying the relevance test when it comes to linking. Both your examples are good illustrations of why our work fails. And as you said, the latter 'endnote' link would be made more meaningful if there was a discussion in the article as to the author's uncommon use of this literary device. Instead, we have this meaningless short cut and no discussion. So it's both a content and contextual issue. I don't know if there is a way to find such issues and address them in a systematic way, or if there is any answer. In the absence, we need to just accept Wikipedia is a work in progress, and do what we can to fix problems as and when we come across them. It would be great if you could help. -- Ohc ¡digame! 02:09, 23 June 2014 (UTC)

The example you cite of the article Infinite Jest is exactly how inline links should be used - the link should be to provide more information on what endnotes are, not for any further discussion of their relevant in this instance. If explanation of it's specific relevance to the topic of the original page is required it should be included in the artcile itself, or if externally to wikipedia, placed in a reference. So it appears it is you DavidLJ, that doesn't understand how inline linking should work so maybe you ought to consider if you should stand out in the cold for 48 hours? Bladeboy1889 (talk) 08:02, 23 June 2014 (UTC)
In my opinion, the example you brought up certainly should not be linked. Totally useless with respect to the article. HTML lunacy indeed. Jayaguru-Shishya (talk) 15:51, 23 June 2014 (UTC)
Not really. Proper names (save for major geographic features) should be linked on first use even if we're talking a work like the NYTimes that everyone should know. And "endnote" is not a common enough literary device that it is a germain term. What is the issue is expecting these links to lead to the "raw source", to speak, instead of another WP article, which all other links of this style go to other WP articles. --MASEM (t) 15:55, 23 June 2014 (UTC)
I think we should always ask ourselves, "is this link relevant to the article? Does this link help the reader to gain a better understanding on the subject?" A hypothetical example, if one is writing an article about "Koala-bear incident" and he/she is citing "Defleppardian Times" but still doubts that the users might not know that it is a newspaper, one could also rephrase it like "The newspaper "Defleppardian Times" reported that..."
If the only purpose of the link is to make the readers know that "Defleppardian Times" indeed is a newspaper, but has nothing else to do with the article, then it is not so good idea to link that one. It's so so tangentially related to the subject and makes a great disservice for the author (and distracts the reader to a lot of irrelevant text). If this happens from one article to another, it'd lead into a quadratic growth. Intelligent linking, that's I'd like to see!
I think User:Tony1 has a pretty good article about smart linking here. Jayaguru-Shishya (talk) 16:09, 23 June 2014 (UTC)
But for many less familiar publications it will not be clear: if, for example, the title of the publication is in Defleppardian, few readers will recognize it. Or consider something published in New York -- living there, I will immediately recognize the publication, and know it's a magazine not a newspaper, but will people outside the US? DGG ( talk ) 17:17, 10 September 2014 (UTC)

"major geographic locations"

I would like to propose that this includes country names and names of a few world-famous cities, at least when they are used in a list. The reason for not linking "Europe" is obvious; there's a similar reason not to list "France" or "Paris". Everyone will know what is meant. DGG ( talk ) 17:21, 10 September 2014 (UTC)

Application of MOS:NOPIPE and WP:NOTBROKEN

(I have posted a link to this discussion on Wikipedia talk:Redirect as to keep this all in one discussion.) On Timeline of the 2014 Atlantic hurricane season and Timeline of the 2014 Pacific hurricane season, there has recently been back-and-forth reverting over issues regarding redirects. An IP editor keeps changing wikilinks to place names to pipe the direct target, such as [[Manzanillo, Mexico]] (how it is used in the source) to [[Manzanillo, Colima|Manzanillo, Mexico]] (here is one diff, although the same edit has been pushed multiple times), even though both work. At WP:GUIDES, it states that occasional exceptions may apply, but I see no reason for which this could be an exception. Is there a specific way in which it would be preferable to apply this guideline? I cannot determine what to do, but I know that if I don't do something, the IP will just go on with it's no-discussion edit summary-only reverts. However, this thread shouldn't be about the IP, but it should be about the application of MOS:NOPIPE. Does anyone know the best way to apply this guideline? I know this may not have been the best place to bring it up, but I think it is necessary that I mention this somewhere... I am not the best when it comes to certain aspects of Wikipedia, and I think I am better off avoiding any mistakes by discussing this. To anyone who is willing to help, thank you. Dustin (talk) 22:16, 18 September 2014 (UTC)

Improving link specificity

Hi! There are around many wikilinks with a low level of specificity and lack of clarity/intuitiveness. Nevertheless I noticed that some of them can be improved safely with a bot. Some real examples:

  • canton of [[Canton of Bern|Bern]] --> [[canton of Bern]]
  • attack on [[Attack on Pearl Harbor|Pearl Harbor]] --> [[attack on Pearl Harbor]]
  • the [[The Exodus|Exodus]] --> [[The Exodus]] (in this kind of cases I will capitalize the article)
  • [[Boeing]] [[Boeing 737|737]] --> [[Boeing 737]]
  • [[Bible]] [[Bible translation|translation]]s --> [[Bible translation]]s
  • [[golf]] [[Golf club (equipment)|club]] --> [[golf club (equipment)|golf club]]
  • [[Atlas Mountains|Atlas]] mountains --> [[Atlas Mountains|Atlas mountains]] (I prefer to avoid unnecessary changes in the visible capitalization)
  • [[brown trout|brown]] [[trout]] --> [[brown trout]]
  • [[Hesychius of Alexandria|Hesychius]] of [[Alexandria]] --> [[Hesychius of Alexandria]]
  • [[open architecture|open]] [[computer architecture|architecture]] --> [[open architecture]]
  • [[Central Intelligence Agency|CIA]] [[CIA World Factbook|World Factbook]] --> [[CIA World Factbook]]
  • [[weak base|weak]] [[basic (chemistry)|base]] --> [[weak base]]
  • [[Bristol Aeroplane Company|Bristol]] [[Bristol Blenheim|Blenheim]] --> [[Bristol Blenheim]]
  • [[Lake County, Ohio|Lake County]] [[Lake County Courthouse (Ohio)|Courthouse]] --> [[Lake County Courthouse (Ohio)|Lake County Courthouse]]
  • other similar substitutions

As you can see the link quality can be greatly improved. I run a flagged bot and before to ask the approval for this additional task I would like to hear your opinion. -- Basilicofresco (msg) 21:33, 20 September 2014 (UTC)

I have indeed manually done this type of thing occasionally—even moving wording from outside to inside a link or pipe. Most of these are good, but if automated editing is employed, do take it slowly as the false positives are minimised. Ohconfucius is very experienced at that. Tony (talk) 23:04, 20 September 2014 (UTC)
Thank you for the hint, of course I will start slowly. I have some experience too. ;) -- Basilicofresco (msg) 23:44, 20 September 2014 (UTC)
OK, can you keep us posted here? Editors on this page might be able to provide a little feedback if you wish. Also, would it be a script or a bot (second of which I guess requires wp:bag approval and would run without much human supervision. Tony (talk) 01:41, 21 September 2014 (UTC)
Usually I receive the feedback on my talkpage, but any feedback is precious and I will have this page in my watchlist. It is a scipt for a pywikipedia bot. You will find the details here: Wikipedia:Bots/Requests for approval/FrescoBot 12. -- Basilicofresco (msg) 07:43, 21 September 2014 (UTC)
That's great work, Basilicofresco! I've ran to this problem numerous times, and I've been fighting it manually as well. I look forward to see how your work is progressing. :-) Cheers! Jayaguru-Shishya (talk) 14:13, 21 September 2014 (UTC)
  • 👍 Like I try to cleanup with my script. It's added to from time to time when I work out a common pattern with chain linking. The encyclopaedia is full of links created by people being too clever (and creating needless chain or juxtaposed links) that could do with rationalising, so any extra help in that direction would be most useful. -- Ohc ¡digame! 04:58, 23 September 2014 (UTC)
Oooh, you have a script of your own! I must look into that one! :-) Cheers! Jayaguru-Shishya (talk) 18:48, 24 September 2014 (UTC)

Since a wider discussion was needed, I opened a new discussion here: Wikipedia:Village pump (proposals)/Archive 114#Improving the link specificity. There you will find a more detailed description of the idea. -- Basilicofresco (msg) 01:18, 24 September 2014 (UTC)

Why is it that congress does not want Puerto Rico a state.

I heard that Puerto Rico, on its last elections, voted for Puerto Rico to become a State, rather than a territory. Why is it that Congress has refused to grant them statehood, if it is true that they did vote to become a State. — Preceding unsigned comment added by 96.249.28.71 (talk) 23:06, 7 October 2014 (UTC)

Unrelated to the topic, see WP:REFDESK. ‑‑Mandruss ☎ 00:05, 28 October 2014 (UTC)

What should this say?

"Do not link to pages that redirect back to the page the link is on (unless the link is to a that links to an appropriate section of the current article)." I get the drift, but it's not good English. What is it meant to say? 86.44.196.15 (talk) 23:37, 28 October 2014 (UTC)

Never mind, I figured it out. It was missing a colon that would have made it "(unless the link is to a redirect with possibilities that links to an appropriate section of the current article)." Fixed now. 86.44.195.47 (talk) 18:29, 30 October 2014 (UTC)

Specificity

Sorry if this has been covered before, but when linking to articles such as [[Santa Barbara, California]], is that how the link should be formatted, or is [[Santa Barbara, California|Santa Barbara]], [[California]] preferred? The second seems better, 1) to give the reader the choice of where to go, and 2) to give the option of Santa Barbara, California, when an overlink to California would result. Thanks! Spicemix (talk) 15:06, 16 November 2014 (UTC)

Aren't we going around in circles? See #Improving link specificity above. --Redrose64 (talk) 22:05, 16 November 2014 (UTC)
Thanks. I note the example [[Lake County, Ohio|Lake County]] above and I'm pleased to see that's the consensus. My problem though is that several times recently I've been reverted for piping links to be more specific, e.g. here, and I don't see a wording or example at WP:SPECIFICLINK that establishes [[Lake County, Ohio|Lake County]] as best practice. As there is a countless number of confusing non-specific links of the [[Lake County, Ohio]] type across WP, can something be added to the MOS please, so that policy can easily be quoted in the edit summary? Spicemix (talk) 13:40, 17 November 2014 (UTC)
Actually, you're wrong as to the consensus; although not incorporated into WP:SPECIFICLINK, it seems, to be, in order of preference
  1. Santa Barbara, California
  2. Santa Barbara, California
  3. Santa Barbara, California
— Arthur Rubin (talk) 15:53, 24 November 2014 (UTC)
I think WP:LINKSTYLE gives a pretty good picture about the issue, and it also does supports WP:SPECIFICLINK. Anyway, if one takes a look at the former, there are similar examples: [[Riverside, California]] and [[Riverside, California|Riverside]]. The first one is a direct link, the second one a pipelink. Like Arthur Rubin listed above, the first would be what WP:SPECIFICLINK calls "specific"; the second one "related, but less specific". The third one, however, would be similar to what WP:LINKSTYLE describes: [[Ireland|Irish]] [[Chess]] [[Championship]]. Hmm, I think user Tony1 has a pretty goor article about it here[6]. Cheers! Jayaguru-Shishya (talk) 17:52, 27 November 2014 (UTC)

Why not link several times in the case of long Wikipedia articles?

I have a comment to make about this: "Generally, a link should appear only once in an article, but if helpful for readers, links may be repeated in infoboxes, tables, image captions, footnotes, hatnotes, and at the first occurrence after the lead." I would propose that for long articles, a link could be repeated again if the reader had to scrol down several pages from the lead to get to it. As long articles are unlikely to be read from start to finish, a reader my jump into the section that interests him/her the most. And there, if a link is not repeated it could leave the reader wondering if a Wikipedia article already exists for that particular item. Example is the long Wikipedia article on malnutrition where I would propose to link the term "stunted growth" more than just in the lead and in the first occurrence after the lead, namely also in the section that deals with children (which is quite far down).

Therefore, I would propose to modify this to: "... at the first occurrence after the lead or in the first sentence of a distint section if that section is a long way down from the lead. As a rule of thumb if the reader has to scrol down by two pages, then it is reasonable to repeat a link, as readers are not always reading an article from start to finish but may jump in at the middle into a section heading that interests them." Thoughts? EvM-Susana (talk) 20:43, 9 November 2014 (UTC)

Agreed for the most part, but be advised that you have unwittingly entered an ideological, almost religious war zone. ‑‑Mandruss ☎ 05:15, 10 November 2014 (UTC)
Yes. Strongly suggest you do some searching through the archives to get a sense of what you are up against before proceeding. Pay particular attention to the issue of "overlinking". ~ J. Johnson (JJ) (talk) 22:51, 11 November 2014 (UTC)
I think Susana's made a practical suggestion which would help readers. Spicemix (talk) 15:12, 16 November 2014 (UTC)
This is not the first time that this or similar suggestions have been made. It would be wise to consider why they have not been adopted. ~ J. Johnson (JJ) (talk) 22:39, 24 November 2014 (UTC)
I also agree with Susana's reasoning. It exactly describes my sentiments when I add a link which later gets reverted. Why spend time scrolling back through an article looking for a link, when it could also appear in another section. If it doesn't appear twice within one or two sections, it is harmless and could only be helpful. And the statement "Generally, a link should appear only once in an article..." sounds more like a guideline than a hard-and-fast rule. WP:COMMON encourages common sense, which when applied to this discussion should allow multiple links for the situations described here. CuriousEric 18:43, 9 December 2014 (UTC)

Prefer anchor linking to section linking

A proposed reworking of Wikipedia:Manual of Style/Linking#Linking to sections of articles, based on an orphan comment by @Altenmann: from January 2014 (see Wikipedia talk:Manual of Style/Linking/Archive 16#Section links)

If I want to link to [[Foo#Bar]] then I don't think <!-- the article WP:LINK links here --> is ever optimal. If it's likely a once-off link, just linking [[Foo#Bar]] may suffice, with no edit to Foo. Otherwise I suggest best practice should be:

  1. Add Bar of foo as #REDIRECT [[Foo#Bar]]{{R to section}}{{R with possibilities}}
  2. Edit Foo to change ==Bar== to ==Bar=={{anchor|Bar}}<!-- [[Bar of foo]] redirects here -->
  3. Link to [[Bar of foo]]

Advantages:

  • Linking to an anchor is better than linking to a heading, because the heading text may be changed, breaking all the incoming links. The current MOS is too diffident about this point.
  • Specifying the redirect once you have added it is useful
    • in case later another editor also wants to link to [[Foo#Bar]]; they can also link to [[bar of foo]] rather than, e.g. creating Fooian bar as another redirect.

jnestorius(talk) 16:24, 18 December 2014 (UTC)

Post-nominals

I have boldly removed the reference to post-nominals in WP:OVERLINK. The usual practice is (now) certainly to link them, per Template:Post-nominals. In any case, we should not expect the average reader to know what "PC" or "CC" might stand for. StAnselm (talk) 10:56, 25 November 2014 (UTC)

Unless a post-nominal is arcane or known only to a few, the practice is not to link them. Please restore the guideline. Tony (talk) 01:27, 28 November 2014 (UTC)
I can't imagine how anyone could know how much of the Wikipedia reading population knows each post-nominal. That being the case, what does one do with "known only to a few"? I'll confess to being ignorant about anything more obscure than PhD or OBE (and I didn't learn what OBE stood for until I was probably well over 30). ‑‑Mandruss ☎ 01:47, 28 November 2014 (UTC)
Well, sometimes we just have to make decisions as to what the readers would know, and thus what would be most helpful. Tony, if this is the practice, how do you explain, for example, the Ian Smith article? It is a featured article, and yet the postnominals are linked in both the lead and the infobox. This is also the case with David Lewis (politician), Neville Chamberlain, and Stanley Bruce, which are all featured articles. StAnselm (talk) 01:59, 28 November 2014 (UTC)
OBE is one of those that should be linked. A lot of people - even in Britain - think that it's Order of the British Empire. In fact, the O stands for Officer, and the full award is Officer of the Most Excellent Order of the British Empire. --Redrose64 (talk) 11:47, 28 November 2014 (UTC)
OBE, probably yes. StAnselml, GCLM, ID, CC, QC—they're on the specialist/technical side. I'd link them, once. Tony (talk) 04:58, 29 November 2014 (UTC)
Well, what would be an example of a non-technical one, given that we don't use PhD as a postnominal anyway? StAnselm (talk) 05:06, 29 November 2014 (UTC)
In fact, now that I look at WP:POSTNOM, it explicitly encourages editors to wikilink. I really don't understand why this was included among the things not to link. StAnselm (talk) 05:16, 29 November 2014 (UTC)
  • St, that question was gurgling around in my mind, yes. Tony (talk) 13:53, 29 November 2014 (UTC)
Sorry for the late reply. I've been dead busy the past few days. Anyway, I oppose StAnselm's proposal. Pre- and post-nominals have been included ever since November 2012. Changing the prevailing consensus would definitely need discussion and achievement of new consensus before changing the guideline.
I also think that the process has started rolling on the wrong track here: the proposal is about "post-nominals", but the change included both pre- and post-nominals.
In my humble opinion, linking pre- and post-nominals is quite redundant. When it comes to pre-nominals, I don't think it's a good idea to link, e.g. [[Ph. D.]] [[Paul Krugman]] or [[Professor]] [Paul Krugman]]. For the sake of link specificity, linking directly to the subject would be a better idea. And post-nominals, I simply think that when it comes to more "bizarre" ones, it would be more advisable to write those open; that'd be better for the flow of the text, it would make the text more self-supported, and it for the reader it would save a lot of trouble. For example, instead of including a post-nominal such as "KBE" and wikilinking it, one could simply write it open as a "Knight Commander of the Order of the British Empire". Jayaguru-Shishya (talk) 19:37, 6 December 2014 (UTC)
Ph. D. isn't a pre-nom, it's a post-nom. That aside, I get the impression that what you would like at Winston Churchill is |honorific-suffix=Knight of the Order of the Garter, Member of the Order of Merit, Companion of Honour, Territorial Decoration, Deputy Lieutenant, Fellow of the Royal Society, Royal Academician written out in full and without any links at all. If so, Oppose as way too bloated and unhelpful. --Redrose64 (talk) 13:22, 8 December 2014 (UTC)
Aaah, it seems to be a Finnish speciality. According to pre-nominal letters: "In Finland, abbreviated academic titles can appear before or after the name (for example, FM Matti Meikäläinen or Matti Meikäläinen, FM)."
Oh boy... The British sure fancy their honorifics, don't they? I agree Redrose64, writing it out makes it quite bloated, but I don't think the current '''Sir Winston Leonard Spencer-Churchill''' {{postnominals|country=GBR|size=100%|sep=,|KG|OM|CH|TD|DL|FRS|RA}} is any better either. I mean, if the reader bumps into 7 different post-nominals in the very first sentence of the article, and he/she probably doesn't even have any clue what do they mean, I think it's like throwing a bucket of cold water onto him/her in the first instance. Besides, I don't know if those honorifics really are that central to the actual article.
Anyway, I'd suppose that Winston Churchill is kind of an extreme case? If the others have significantly smaller number of honorifics, the text of course won't be so bloated. And when it comes to writing it out, I think the upside is that the reader immediately gets the meaning and can decide for himself/herself whether he/she wants to know more or not. Or what would you like to suggest, Redrose64? =P Jayaguru-Shishya (talk) 20:31, 8 December 2014 (UTC)
"The reader... probably doesn't even have any clue what do they mean" - and that is precisely why we wikilink. "Knight Commander of the Order of the British Empire" is never written in full after a person's name in the British system. StAnselm (talk) 21:23, 8 December 2014 (UTC)
Anyway, I think pre-nominals are much the same issue as academic post-noms - WP:HONORIFIC says that "styles and honorifics should not be included in front of the name" so the issue of wikilinking is pretty much redundant. StAnselm (talk) 21:28, 8 December 2014 (UTC)
I understand your point StAnselm, but forcing the reader to go through seven different articles on post-nominals isn't a solution either. If we wouldn't "encrypt" these honorifics, the reader wouldn't have to go through all of those articles. After all, they are not really central to the subject at hand. Anyway, as said above, pre- and post-nominals have been included ever since November 2012, and changing the guideline would need a consensus. I'd like to suggest that the proposal is first discussed before any changes. =P Cheers! Jayaguru-Shishya (talk) 19:11, 9 December 2014 (UTC)
I think linking unexpanded pre- and postnominals violates WP:EASTEREGG. Relegate them from the lede to a footnote, or part of a "Titles and styles" or "Awards" section. Margaret Thatcher#Styles and titles could do the job better than the opening does (Currently "Margaret Hilda Thatcher, Baroness Thatcher, LG, OM, PC, FRS"). jnestorius(talk) 23:41, 18 December 2014 (UTC)
It's just like country-names: don't link the obvious ones (Dr, PhD, etc), but all of Thatcher's little baubles could do with a link. Tony (talk) 04:07, 19 December 2014 (UTC)
Good point Jnestorius, you just expressed what I couldn't put into words clear enough earlier. Placing them into footnotes instead of linking is much more appropriate, and the readers can more easily follow what's going on. That's just one reason why the previous change is problematic. I restored the previous version that's been around ever since November 2012. Cheers! Jayaguru-Shishya (talk) 19:48, 28 December 2014 (UTC)

Editor not seeing the benefit of WP:NOTBROKEN

At the Suicide among LGBT youth article, I reverted EChastain and cited the WP:NOTBROKEN guideline. Despite this, EChastain reverted me, stating, "avoiding clunky redirect is better (using Pipe trick)." I reverted again, adding. "No, it isn't, and WP:Edit warring over a clear guideline (WP:NOTBROKEN) is silly. Follow the guidelines, with few exceptions. And this is not one of those few exceptions; not a WP:Ignore all rules case in the least. Redirects like this help readers. If you don't know why, then ask at WP:Manual of Style. Or I will so that you are educated on it." EChastain reverted yet again, stating, "how does it help readers? just your presumption - readers don't tell you what they experience - please provide proof." Apparently, EChastain thinks that the WP:NOTBROKEN guideline exists just to exist and has no valid reason for existing. Either that, either the guideline is not fully clear on why it's beneficial, or EChastain was reverting simply because of the heated words that I recently left on the EChastain talk page. Will anyone else watching the Manual of Style/Linking talk page explain to EChastain why the WP:NOTBROKEN guideline helps readers? To briefly answer EChastain, it helps readers because not bypassing the redirect automatically lets them know that they are most likely at the right article; when they see the tell-tale sign, at the top of the article, that the term was redirected to the article in question, they know that they are most likely at the correct article, and that there is no article specifically for that term. Flyer22 (talk) 22:54, 27 December 2014 (UTC)

I also alerted editors of WP:Manual of Style to this discussion. Flyer22 (talk) 23:01, 27 December 2014 (UTC)

Correction: EChastain didn't revert back to the WP:NOTBROKEN violation a second time; that edit was rather a revert of my WP:Dummy edit; therefore, I struck through that part above. Flyer22 (talk) 23:07, 27 December 2014 (UTC)

As a reader I didn't see the benefit at all. In fact, it was confusing. As an editor I can at least figure out what's going on. Don't know how Flyer22 knows what the anonymous reader experiences. She's been here too long to know what's like to have wiki jump around for unnecessary redirects. EChastain (talk) 23:21, 27 December 2014 (UTC)
Is there a particular reason why the text should speak about LGBTQQ instead of LGBT? I mean, since the link is taking one to LGBT article, is it possible to use that abbreviation in the text as well? =P Jayaguru-Shishya (talk) 19:55, 28 December 2014 (UTC)
The particular pipe mentioned first, [[LGBT|LGBTQQ]], confuses both readers (at least with popups[note 1] or link display on mouseover[note 2]) and editors. If you're going to avoid the redirect, use [[LGBT]]QQ, although I believe [[LGBTQQ]] is more understandable yet.
Of course, it would be nice to know how readers generally see it. But, considering the mistakes the Foundation has made interpreting their surveys, I don't see any hope of a reasonably-well-sampled survey occuring. — Arthur Rubin (talk) 05:26, 28 December 2014 (UTC)
EChastain, these redirects aren't "unnecessary" as you put it, and this guideline exists for a solid reason. Dustin (talk) 06:18, 28 December 2014 (UTC)
  1. ^ Wikipedia:Tools/Navigation popups, the gadget or script
  2. ^ Many browsers have an option to display the URL on mouseover

Nitpick: this isn't the pipe trick. --NE2 13:32, 30 December 2014 (UTC)

Lists of shipwrecks

There is a discussion taking place re overlinking in various lists of shipwrecks. Please feel free to contribute at WT:SHIPWRECK. Mjroots (talk) 08:01, 14 January 2015 (UTC)

Comment requested on link related template

Please comment at Template talk:interlanguage link#A significant flaw on how a possible problem with this template should be addressed. Oiyarbepsy (talk) 15:35, 15 January 2015 (UTC)

Should Wikiquote have its users place bolding and wikilinking within quotations?

canvassing
The following discussion has been closed. Please do not modify it.


I note that WP:LINKSTYLE on the Manual of Style page Wikipedia:Manual of Style/Linking says:

  1. Question: Should our sister website Wikiquote allow users to place their own chosen emphasis on quotations by adding bolding or wikilinking within quotations from other sources?
  2. Please see discussion there, at q:Wikiquote:Village_pump#Wikilinking_within_a_quote.

Thank you,

— Cirt (talk) 15:08, 9 February 2015 (UTC)

Nutshell?

I have personally thought this article a little too long, and that some editors might just want a snappy summary of the page. Because of this, I have tried to make a "nutshell" summary several times, but rejecting each attempt. I was wondering if some editors could come over here and we could help each other figure out how to make an adequate summarization? If one shouldn't be made, please give me a good reason why. Thanks. Thatguytestw (talk) 14:49, 30 January 2015 (UTC)

What do you mean exactly? The current introduction seems to be as good a summary as any. You mean some kind of bulleted list of highlights? W. P. Uzer (talk) 20:47, 30 January 2015 (UTC)
You can’t really sum up something like this in a {{nutshell}}, for the same reason WP:MOS doesn’t have one. It’s a collection of best practices, not an overall piece. —174.141.182.82 (talk) 10:19, 11 February 2015 (UTC)

Comments please on avoidable links

The section "General points on linking style" seems to me to include one or two questionable points, and to urge at least one pernicious principle. Observe the following points in particular:

  • Do not unnecessarily make a reader chase links: if a highly technical term can be simply explained with very few words, do so.
  • Don't assume that readers will be able to access a link at all; remember that it is not always possible. For example, a reader might be working from a printed copy of an article without access to facilities for following links.

And the following one is closely related:

  • Do use a link wherever appropriate, but as far as possible do not force a reader to use that link to understand the sentence, especially if this requires going into nested links

Those recommendations have merit and sound very good in general, but in the wrong context or comprehension they bear Danaan gifts. When "... a highly technical term can be simply explained with very few words..." out of context, or by an author or editor who knows everything except for how little he knows, or actually well-informed, but who fails to realise his own inability to convey the necessary point to the readers who need it most, and how dismally his shortcomings in the name of WP will depress informed readers, then really, it would have been better to rely on more linkage and less generous explanation to the reader.

More importantly, there is one principle that should be observed closely, even obsessively, in any encyclopaedic structure, and which these guideline points violate, sometimes venially, more often mortally:

  • Thou shalt avoid duplication of data in independent places and contexts
  • Thou shalt particularly avoid duplication of explication in independent places and contexts

Every time an editor helpfully avoids a link, either for fancied user convenience, or because he had neglected to check whether a relevant linkable article might be available, he creates items that, if condensed:

  • Might mislead innocent readers into thinking that no other guidance on the topic might be available
  • Might mislead innocent readers into thinking that no other, more advanced material on the topic might be available or necessary (the principle of "a little knowledge")

and in any case demand:

  • consistent checking of accuracy and authoritativeness wherever the same point is casually included; many an author or editor blithely includes throw-away wisdom of the type: "member of the Hemimetabola (insects that do not have a complete metabolism)" where he mercifully omits links to "Hemimetabola", "metabolism", or "incomplete metamorphosis". Any competent biologist encountering that sort of rubbish is at hazard of an aneurism and any reasonable editor might want to know why the parenthetic material was included at all, if it shows no sign of relevance to the topic, accurate or not. Unfortunately, in a work of millions of items it could be years before anyone notices the faux pas, and in the mean time it winds up in published books and google searches (the latter commonly within seconds, and I mean that LITERALLY!). What does that sort of thing do for our name, our respect, or (I hope) our pride in our work and our role?
  • continued maintenance of consistency of such items with each other; consider where one main article (or a few on the same topic sometimes) might contain (let us hope) flawless information and exposition, but the basic facts or received wisdom might change. For example, "The Mantophasmatodea are the first new order of insects to be discovered since..." might need updating to perhaps: "Originally hailed as the first new order of insects to be discovered in recent decades, the Mantophasmatodea now are classed as the family Mantophasmatidae..." That might not sound terribly shocking, given the obscurity of the topic and accordingly the few articles that refer to it, but already there are rumblings about the status of the order Coleoptera as well. Changes there could affect thousands of articles that otherwise would have remained unaffected, if only their authors had used linking effectively. In case that sounds too academic to matter (in which case shame on you!) consider the likes of "...which never exhibits antibiotic resistance..." or "...is easily treated by..."
  • continued updating in all the places where the items might occur. Consequent on the previous point, it is too great an unprofitable burden to be justified for the sake of reducing properly conceived links. There are other ways of assisting readers of hard copy, and if there were not, WP has more realistic objectives to attend to than merely being all things to all readers.
  • that any gratuitous remarks on tangential material should be carefully chosen and phrased to minimise the scope for future required changes, and tolerated only when it would impoverish the theme to omit them; in such cases there should be links to relevant main articles so that readers need not rely on second-hand versions. Less avoidable subject matter that is of immediate significance to the article but has its own main article, obviously couldn't reasonably be omitted but as far as possible should refer to the material in the main article by suitable links, and if necessary by "see main article" links. JonRichfield (talk) 13:52, 24 February 2015 (UTC)
Retrieved from "https://en.wikipedia.org/w/index.php?title=Wikipedia_talk:Manual_of_Style/Linking/Archive_16&oldid=1144428190"