Wikipedia talk:Manual of Style/Linking/Archive 20

Archive 15

←

Archive 18

Archive 19

Archive 20

Archive 21

Overlinking

I have made a change to the sub-section on Overlinking - see my diff. The sub-section looked like it had been put together by many Users, all trying to achieve something slightly different. Hopefully my change improves the readability and comprehensibility. Happy to discuss. Dolphin (t) 10:27, 3 November 2019 (UTC)

Definite readability improvement; thank you. I tweaked it.[1][2] ―Mandruss ☎ 16:01, 3 November 2019 (UTC)

It's somewhat better, but not sure I agree with "However, try to be conscious of your own demographic biases", because that isn't always common sense. What were you going for there?

There may also be the need to say that certain terms may be common in specific subject areas. For instance, when discussing a musical recording, there's no need to link to guitar, and when discussing a politician, there's no need to link to election, etc. Walter Görlitz (talk) 21:09, 3 November 2019 (UTC)

@Walter Gorlitz. The text beginning “However, try to be conscious of your own demographic biases, ...” has been there for a long time. I made no change to it so I can’t reply to your request. I see that after your edit here User:EEng relocated it substantially so that it now appears parenthetically within another sentence. I’m not convinced EEng’s change to the position of this text is an improvement.

I agree with your comment regarding no need to link to guitar in an article about a musical recording. Dolphin (t) 05:39, 4 November 2019 (UTC)

Try reading my edit summary and see if you can find a better way. EEng 11:54, 4 November 2019 (UTC)

The sentence in question (which has been in place for a long time) adds little or nothing of value to the sub-section. I think the better way is to erase it altogether. The sub-section is explaining overlinking and giving advice on how to avoid overlinking, but this sentence appears to be warning against the dangers of avoiding overlinking! Let's get rid of it. Dolphin (t) 13:34, 4 November 2019 (UTC)

We want editors to find a middle ground, so we're trying to point out the hazards in both directions. EEng 20:19, 6 November 2019 (UTC)

Thanks. I hadn't read it before. My error.

Giving additional examples on what might constitute an OVERLINK could prove helpful. Three additional examples might be enough to get the point across. Walter Görlitz (talk) 07:40, 4 November 2019 (UTC)

I made an attempt at an improvement. I think the demographic bias concept is important to highlight, but to me it makes more sense as an introduction to the entire list. Please feel free to revise/revert. CThomas³ (talk) 00:46, 5 November 2019 (UTC)

@Cthomas3: Your change is an improvement, especially as it puts the demographic bias concept in its own sentence. Previously it was attached to the tail of another sentence. Dolphin (t) 00:58, 5 November 2019 (UTC)

I'm not keen on the change, which is why I reverted. Why begin with the "Well known and understood concepts (such as those appearing on the list below) are generally not linked" piece when the "Everyday words understood by most readers in context" and "The names of subjects with which most readers will be at least somewhat familiar" bullet points exist? Why move the WP:BIAS text away from the "names of subjects with which most readers will be at least somewhat familiar" bullet point? I'm not seeing the issue with stating "Unless a term is particularly relevant to the context in the article, the following are usually not linked:" and then giving examples. Flyer22 Reborn (talk) 11:35, 5 November 2019 (UTC)

No need to ping me when you reply. Flyer22 Reborn (talk) 11:36, 5 November 2019 (UTC)

Greetings Flyer22. The reason I moved WP:BIAS to the intro to the section is because it shouldn't be only names of countries, locations, and languages that we consider our systemic bias about. What some would consider everyday words, common occupations, and common units of measurement might not be considered so by everyone. As such, I thought that the BIAS statement should apply to the entire section and not just one bullet.

Once I moved the bias statement to the top of the article, it just didn't read right without modifying the other text, which is why I took a stab at doing so. I'm perfectly happy for someone to take a better stab, but I do think that the BIAS piece should stay at the top. CThomas³ (talk) 23:22, 5 November 2019 (UTC)

I am disappointed by Flyer22 Reborn’s reversion for two reasons. The first reason relates to the art of good writing. The reversion welded two sentences together, with the second sentence in parenthesis. Prior to the reversion we had two sentences:

Be conscious of your own demographic biases – what is well known in your age group, line of work, or country may be less known in others. 26 words

The names of subjects with which most readers will be at least somewhat familiar. 14 words

After Flyer22 Reborn’s reversion we have one sentence of 43 words! It is an awkward and unwieldy sentence, struggling to do the work of two sentences, and failing. The fact that the appended sentence is in parenthesis does not mitigate the problem – it makes it worse by emphasising that the two parts are really two separate sentences. It is rare indeed that formal writing, such as an encyclopedia, can be improved by joining two good sentences into one very long sentence. Very rare indeed, and this latest creation isn’t one of them. It is much more common to find a long, clumsy sentence that is easily improved by dividing into two sentences. Reverting to Cthomas3’s version would be such an improvement.

My second reason is that the concept of demographic bias is just as applicable to common words, common occupations and common units of measurement. I see Cthomas3 has expanded on this theme much better than I could.

I don’t see any support for Flyer22 Reborn’s reversion, so I suggest the best course of action is for the reversion to be reverted. Dolphin (t) 12:53, 6 November 2019 (UTC)

Cthomas3, you moved the WP:BIAS text to after the following sentence: "A good question to ask yourself is whether reading the article you're about to link to would help someone understand the article you are linking from." I do not see how it fits better there than placed after "The names of subjects with which most readers will be at least somewhat familiar." This is because the latter piece is trying to tell us that the names of subjects with which most readers will be at least somewhat familiar may vary; it's similar to what WP:BUTIDONTKNOWABOUTIT tells us. If it's agreed that we need to expand the text beyond the names of subjects, we can do that at that bullet point. But expand in what way? The section is about linking.

Dolphin51, you are proposing that we change long-standing wording of a guideline. The top of this guideline speaks of WP:Consensus. What WP:Consensus is there to go with the version I reverted the text away from? Despite your latest comments, I don't see how the version I reverted is a better presentation. My questions above explain why. You've focused on one aspect of the revert. You are complaining about "the second sentence in parenthes[e]s." It is easy enough to put "though try to be conscious of your own demographic biases – what is well known in your age group, line of work, or country may be less known in others" in its own sentence at that bullet point. Flyer22 Reborn (talk) 13:20, 6 November 2019 (UTC)

Flyer22, thank you for your reply. I guess I am a bit confused. I absolutely agree that we should be applying BIAS to "names of subjects with which most readers will be at least somewhat familiar". But we should also be applying BIAS to all of the other bullet points as well, which is why I put it at the top. If you don't like how my initial paragraph was written, that's totally cool with me if you want to change it. However, as it is currently written, the phrase "names of subjects with which most readers will be at least somewhat familiar" is only referring to the sub-bullets underneath it (names of countries, geographic features, locations, languages, nationalities, ethnicities, and religions). If it is intended to refer to the entire list, then it shouldn't be one of the bullets. CThomas³ (talk) 19:32, 6 November 2019 (UTC)

Just in case my intent wasn't clear: I was attempting to do the following: "Here is a bunch of stuff we generally don't link. However, keep in mind (for this entire list below), you should keep your BIAS in mind. <list of 5 items below>." The way it reads at least to me, it says "Here is a bunch of stuff we generally don't link. item 1, item 2, item 3 (however keep BIAS in mind), item 4, item 5." I think BIAS should apply to all bullets, not just item 3. CThomas³ (talk) 19:37, 6 November 2019 (UTC)

Well, your Items 1, 2, 4, and 5 are Everyday words ... Common occupations ... Common units of measurement ... Dates, and while in principle there's always someone who's unfamiliar with something, I think bias is far less of an issue for those categories than it is for geographic features, languages, ethnicities and so on, and I think it's fine to target the warning specifically to that part of the overall list. EEng 20:17, 6 November 2019 (UTC)

I definitely hear where you're coming from. I suppose my final thought would be that BIAS seems to be good general advice for the entire overlinking section, and it's certainly not inaccurate to apply it to all bullets. It doesn't make the section any longer to do so, and if we put it up top as general advice, people can skip the individual examples and still get the message. But if in the end everyone wants to keep it as specific advice for a specific group of terms, I guess that's okay by me. CThomas³ (talk) 21:44, 6 November 2019 (UTC)

I think we are discussing these two edits which changed OVERLINK:

Old

New

A good question to ask yourself is whether reading the article you're about to link to would help someone understand the article you are linking from. Unless a term is particularly relevant to the context in the article, the following are usually not linked: Everyday words understood by most readers in context (e.g., education, violence, aircraft, river) Common occupations (e.g., accountant, politician, actor) The names of subjects with which most readers will be at least somewhat familiar (though try to be conscious of your own demographic biases – what is well known in your age group, line of work, or country may be less known in others). This generally includes major examples of:

A good question to ask yourself is whether reading the article you're about to link to would help someone understand the article you are linking from. Be conscious of your own demographic biases – what is well known in your age group, line of work, or country may be less well known in others. Everyday words understood by most readers in context (e.g., education, violence, aircraft, river) Common occupations (e.g., accountant, politician, actor) The names of subjects with which most readers will be at least somewhat familiar. This generally includes major examples of:

Is that the issue? And the central point is the "Be conscious..." sentence? If so, I do not see the point of the change. A guideline should provide guidance—what does "be conscious" mean? I know that Japan is a country but it's likely there are people somewhere who would not recognize that word, at least not in English. Should I link it because I might be biased? Johnuniq (talk) 00:22, 7 November 2019 (UTC)

• I agree with Johnuniq. "Be conscious of your own demographic biases – what is well known in your age group, line of work, or country may be less well known in others."—We're supposed to be aware of those matters in everything we do on WP. I don't see what the instruction-bloat achieves. Tony (talk) 06:19, 7 November 2019 (UTC)

The current issue concerns a sentence that says “try to be conscious of your own demographic biases; what is well-known in your culture, age-group, profession or country may be unknown to others, and vice versa”.

This sentence is contentious on two counts:

1. whether this sentiment is warranted in this article. See posts on this thread, dated 7 November 2019, by Johnuniq and Tony.
2. if the sentiment is warranted, where should it be located? Should it be near the end of the bulleted section beginning “The names of subjects with which most readers will be at least somewhat familiar”, and if so should it be bulleted? Or should it be near the beginning of the bulleted section? Or a third option, should it follow immediately after “A good question to ask yourself is whether reading the article you're about to link to would help someone understand the article you are linking from”, as implemented by Cthomas3 in his edit dated 5 November?

Where to from here? Flyer22 Reborn has made a good point in her post dated 6 November – she drew our attention to the direction that substantial changes to this guideline should only proceed after consensus has been achieved. This sentence is proving contentious, and no consensus has yet been achieved. So I suggest we revert the contentious sentence to its long-standing form and location. Then we can discuss the two counts I have identified above – is the sentiment warranted, and if so, where should it be located?

The contentious sentence was inserted on 1 May 2019 by Boud – see the diff. On the same day, the word “conscious” was inserted by EEng – see the diff. The sentence remained substantially unaltered until 3 November. I will restore the sentence to its long-standing status while we explore a consensus on the above two counts.

The substantial changes to this sentence since 3 November are:

1. Dolphin51 erased the bullet and replaced with an indentation – see the diff
2. Mandruss removed the indentation and restored the bullet – see the diff
3. EEng moved the sentence upwards, immediately following the sentence “The names of subjects with which most readers will be at least somewhat familiar”. EEng graciously conceded that his fix was somewhat awkward. See the diff
4. Cthomas3 moved the sentence to the top of the list of examples of OVERLINKING, so that it is equally applicable to all examples. In his edit summary he asked “if anyone can improve, please do so.” See the diff
5. Flyer22 Reborn reverted the change made by Cthomas3, thereby restoring the version implemented by EEng 34 hours earlier. See the diff

I will now restore the long-standing status of this sentence so we can explore a consensus on the two counts identified above. Dolphin (t) 12:00, 7 November 2019 (UTC)

I have restored the sentence to its long-standing status. See my diff. Dolphin (t) 12:06, 7 November 2019 (UTC)

I see that in my edit introducing the theme, I missed the word "be".

Regarding Johnuniq's point - I agree that "being conscious of bias" is not the best wording - it sounds a bit like making editors feel guilty for being biased. I put "counteract" because the point is editing, not feeling guilty. I don't see any magic guide in judging what is demographically biased or not, but quite a few people have put in effort to write WP:BIAS as an introduction for those willing to try to counteract their biases.

On Tony1's point - what we are supposed to do is not necessarily what we really do. I long ago pointed out that the en.Wikipedia front page has an objectively false phrase on it: the free encyclopedia that anyone can edit. At the time I made the statement (on a Wikipedia Talk page - sorry, I didn't link it anywhere convenient), most people could not edit the Wikipedia, since most people did not have Internet access. As of late 2015, about half of us did not have internet access - in 2019 we're unlikely to be at 100% (yet), so the statement very likely still remains false. (And having "Internet access" does not guarantee having the practical knowledge for editing - for many people "the Internet" is "Where do I click? It doesn't work but I clicked!".) There's no point trying to change the slogan, so getting back to the point here: there are many well-intentioned Wikipedians who are unaware of their biases. Quite a few of us are multi-lingual and have a broad experience of different cultures and groups, but not everybody's in that situation. And experience does not guarantee a lack of bias. Or maybe I can put it another way: effectively teaching a group of people of whom you more or less know the demographic profile (e.g. they're in front of you and you've taught similar groups in the past) is not trivial; teaching a group of people "somewhere in the world who know English" is even more difficult. Wikipedia only has one part of teaching - we don't try to verify the readers' understanding - but it's a component of teaching - disseminating knowledge.

Placement: My preference would be to include the warning immediately after "A good question ..." - to make the Wikipedian realise that "someone" is not necessarily someone from the same region/country/age-group/gender/job-profile/educational-level, with a link to explain what is meant by that. This applies to all of the bullet points. Boud (talk) 22:09, 7 November 2019 (UTC)

I should have checked to see how long the piece had been there. I didn't know that you added it, Boud. Although I felt it was fine where it was, it seems okay to have it placed here. Flyer22 Reborn (talk) 05:52, 8 November 2019 (UTC)

Oh dear. I thought this had been settled a decade ago. There seems to be a feeling among a few people that readers are incapable of typing a search target into the box at the top. Just because an item is not linked doesn't erect a wall around going there. So if we catered to everyone on earth, we'd be going back to square one and reinstating the "sea of blue". Readability would be reduced, and the linking system would be diluted. No thanks. Tony (talk) 21:50, 8 November 2019 (UTC)

Typing into a search box is a lot less convenient than clicking on a link - this is the nature of the world-wide-web served as http responses in html, as opposed to sending requests to servers to run searches in databases. In any case, nobody here is proposing to delete the Overlinking guide in total. Let's get back to the issue. The question is how to avoid demographic bias. We should indeed aim to "cater to [every-English-speaking-one] on earth" in the sense of avoiding racism, sexism, colonial-history-biases and other demographic biases, and given the objectively false (but well-intentioned) slogan we have on the Wikipedia front page, it would be hypocritical to not to try to avoid these biases. We are definitely not aiming to be the US.Wikipedia nor the UK.Wikipedia nor the SouthAsia.Wikipedia. Anyone wishing to create one of those should start a fork, using a delinking filter to generate the fork with culture-specific links "obvious" delinked. Boud (talk) 19:56, 9 November 2019 (UTC)

I am now deeply sceptical that the sentence in question (”However, try to be conscious of your own demographic biases - what is well known in your age group, line of work, or country may be less known in others.”) has any place in the sub-section titled “What generally should not be linked”. This sub-section focuses on the problem of overlinking, and what can be done to avoid it. In this sub-section, the sentence in question is contradictory becauses it focuses on one of the problems of underlinking and provides advice on how to avoid it.

If this Guideline needs the sentiment “Try to be conscious of your own demographic biases …” it should be placed in the sub-section titled “What generally should be linked” which focuses on the problem of underlinking and what can be done to avoid it. In this sub-section, the sentiment would not be contradictory, and would be correctly located where Users who want to read about underlinking will find it.

I am in favour of the sentence in question being removed from its present location. If there is an adequate case for retaining it in this Guideline it should be re-worked for insertion into the sub-section titled “What generally should be linked”. Dolphin (t) 12:18, 10 November 2019 (UTC)

Is this underlinking or overlinking?

In the lead section of the article about Keith Moon, the third paragraph reads "Moon suffered a number of setbacks [ ... ], most notably the accidental death of chauffeur Neil Boland". In my opinion, according to MOS:UNDERLINK a link "should be created for [...] proper names that are likely to be unfamiliar to readers". The person Neil Boland has no article but an own section further down on the same page. According to an administrator, linking would be a case of MOS:OVERLINK because one should "not link to pages that redirect back to the page the link is on." I am using WP:MOBILE and usually start looking up proper names or clicking on their links. What would be your opinion here please? Marcuse7 (talk) 17:53, 14 November 2019 (UTC)

I have looked at the lead of our article on Keith Moon and given it considerable thought. I see what you are advocating. On balance, I’m inclined to leave it as it stands at present, for the following reason: everything in the lead should be summarising some part of the article; therefore there should be no need to use a blue link in the lead to alert the reader that more information is available later in the article. This is most likely what Ritchie333 had in mind when he reverted your 25 April edit.

Many words in any article will be the title of another article and that is why any article has a number of blue links. If the first use of one such word is in the lead it should be blue linked in the lead. However, it appears that blue links aren’t used to take the reader to further information later in the same article. Dolphin (t) 02:05, 16 November 2019 (UTC)

Thank you for your reply. Right now here are about 22 links in the lead section of the Keith Moon article, mine would be the 23rd. Two links are for proper names, i.e. information on Ginger Baker and Peter "Dougal" Butler. Like my reverted edit, Peter Dougal has a separate section on the same page. Please apologise but I do not understand why my edit was reverted. Should I ask Ritchie333 for more information?

The blue links to Ginger Baker and Dougal Butler take the reader to comprehensive biographical articles dedicated to Baker and Butler respectively. Wikipedia doesn’t have an article dedicated to Neil Boland.

Peter Dougal is mentioned on the same page but the blue link doesn’t take the reader to this spot on the same page; it takes the reader to the separate article dedicated to Dougal.

The reversion was done by Ritchie333 so it may be useful to ask him for his reasoning. Dolphin (t) 04:28, 21 November 2019 (UTC)

WP:DATELINK and centuries

WP:DATELINK advises not to link to month-and-day or year articles, unless there is a significant connection to the subject. For example, in an article about a ruined castle, we don't write "the castle was built in 1252". But what about century articles? Is "the castle was built in the 12th century, and later fortified in the 13th century" desirable, acceptable, or should it be avoided as overlinking? Should WP:DATELINK mention it? --IamNotU (talk) 20:20, 17 November 2019 (UTC)

• I think the current advice against linking to "month and day" articles (unless etc etc etc) should be extended to "month, day, decade, and century articles" or maybe (since we may be overlooking some species of chronological article) "month, day, decade, century, and similar articles on specific chronological items" (or something like that -- articles on specific chronological items doesn't sound quite right). EEng 22:50, 17 November 2019 (UTC)

I would not link the century either, unless there is a significant connection to the century. If the particular castle was the fist of a series of castles built in that century, and there is discussion of it in the article, it makes sense to link the century. If it's just a courtesy, do not link it. The century itself can be understood without linking to it, and if a reader wants to read about the century, there's always the search tool. Walter Görlitz (talk) 02:36, 19 November 2019 (UTC)

It would very helpful if the MOS guidance on centuries, etc., could be explicit. Otherwise, editors get accused of disruptive edits and told to revert their work. Hmains (talk) 00:15, 9 December 2019 (UTC)

What to do with an editor insisting on linking dates because they aren't about opera houses?

See User talk:Kapeter77#Please don't link dates. I reverted his links at Judea (Roman province) and he's reinstated the link, and at Talk:Judea (Roman province)#Years he writes "Historical topic, years are SIGNIFICANT. But not wuith CE, rather AD". Doug Weller talk 17:26, 6 December 2019 (UTC)

They've been indefinitely blocked by another Admin. Doug Weller talk 20:52, 6 December 2019 (UTC)

I also struggle with the directives on linking dates yes or no. Wouldn't it be useful to expand on it or give some additional examples/real world problems and ruling? E.g. I'm studying the science of History. And e.g. I'm at Ban Zhao, who seems to be the first woman to have used a historical methodology to get a better understanding of her culture, etc. and the article mentions in the beginning "45 – c. 116 CE".

1. I have no idea what the CE stands for - and as it is the first time in the article this abbreviation is used for, should it be put between square brackets and piped to the right one, because for all I know CE stands for Communauté Européenne :).
2. And as she is mentioned to be the first female historian - to understand better the era she lived in, wouldn't it make sense to hotlink those years? Or pipe through to those centuries? But for me - if you link to the years a. that's easy - just put them between square brackets and b. once on the page listing all the major events of those years, clicking through to the century for a better understanding of the era/world she lived in is easy as the century will be mentioned there. What are your insights in this? The suggestion to just use the "search function" to look up the number of the year - I don't know what that is going to result into - I guess a lot of disambiguation page around the number and a lot of unnecessary reading and waste of time? Thy SvenAERTS (talk) 15:08, 30 December 2019 (UTC)

SvenAERTS, your points are not without merit. But these questions have been the subject of long debates in the past, and I don't see a compelling reason to reverse the decisions that were agreed upon. The reasons we don't link commonly-understood terms are explained fairly well in MOS:OVERLINK, for example: "An overlinked article contains an excessive number of links, making it difficult to identify links likely to aid the reader's understanding significantly ... links compete with each other for user attention." Regarding your points:

1. I have seen people link [[Common Era|CE]], but I don't think it's a good idea in general text. I think most readers know what it means, for those who don't, it's easy enough to find out. It just isn't necessary to make a distracting blue link in every article using it on Wikipedia. We don't link AM or PM, or similar abbreviations. As an example, Encyclopedia Britannica tends to link more liberally than Wikipedia, but does not link CE/BCE. It's clear from MOS:ERA that while other era systems like AH, BP, or YBP should be linked on first use, CE and BCE should not.
2. No, I don't think it's a good idea to link the years or centuries in that article, or any other article simply because they are "historical" (see also the section directly above). I agree with the consensus and the reasons given in WP:YEARLINK. It's again a form of overlinking, creating blue links to articles about common words or broad subjects, where the target article has no specific connection to the source article subject. We don't link the words historian, philosopher, politician, astronomy, mathematics or poems in that text, even though people may well want to find out more about those subjects after reading the article. The articles AD 45 and AD 116 give no additional information about Zhao. We should not link every occurence of a year in general wiki text, just in case someone wants to find out what other unrelated things were happening around the world at the same time.

--IamNotU (talk) 16:28, 30 December 2019 (UTC)

cf. "3.1 Linking month-and-day or year" - towards a script to auto-build events in the article about a specific year/century/date and allow people to show only events according to a parameter of interest e.g. "art", "politics", "notable people", "fun", "military", "priority level 1", "priority level 2", etc? Suggestion

cf. "3.1 Linking month-and-day or year"; Is it correct that when we put e.g. years, centuries or dates in whatever form, between square brackets, the article gets listed automatically on the wikipedia article of that year? I can imagine that this can lead to a page where you can't see what's relevant to you or not. Wouldn't it be interesting to ask for a script to be developed that allows people to show only those events related to e.g. a parameter "art", "politics", "notable people", "fun", "military", "priority level 1", "priority level 2", etc?

Sincerely, SvenAERTS (talk) 03:10, 31 December 2019 (UTC)

The place for an idea like this is Wikipedia:Village pump (idea lab). But I suggest you spend some time developing a much clearer description of what you're proposing -- I can only vaguely tell what you seem to have in mind. EEng 04:01, 31 December 2019 (UTC)

Not OK at all. That battle was fought and resolved back in 2009. Only link chronological items in year and month articles. Tony (talk) 05:29, 31 December 2019 (UTC)

Another opinion please

IP 165.225.81.96 made this edit, linking the Romo article to a an article-less touring member - Xavior Roide credited in the personal section of the article. I reverted it as MOS:LINKCLARITY as I felt the Romo movement had no bearing or relation in any way to the Placebo touring member credits. 165.225.81.96 disagrees and I'd value a third opinion please. Robvanvee 17:19, 5 March 2020 (UTC)

User:Robvanvee suggested I put my side of the case on here, so I'd say it's simply a matter that unless and until an article is ever written on Xavior Roide (or failing that his Romo-era band DexDexTer or some other project of his) the Romo article is the obvious place to link his name since it is the article which contains the most information on him. This fits with "consider linking to a more general article instead" on MOS:SPECIFICLINK. It is quite common to link non-notable members of notable bands to the notable band's article. Another example which I cited on Robvanvee's page was that it would be reasonable and appropriate to link a writing credit for the author Bertie Marshall to Bromley Contingent.165.225.81.96 (talk) 17:51, 5 March 2020 (UTC)

Nah, Robvanveel is right. I'd link him under "See also" at the bottom, if you really must create a red link. Tony (talk) 07:14, 7 March 2020 (UTC)

Well he was clearly an important figurehead in Romo, not only the leader of one of the core bands but also one of the "fab four" on That Melody Maker Front Cover no less (on the right in the pink top) so if there's a logical place to link him to, it's the Romo page if there's info on him there, which there is.Romomusicfan (talk) 18:35, 7 March 2020 (UTC)

Account/IP discussion

I think it's somewhat misleading to make your argument using 2 separate accounts in the same thread. You may have told me on my talk page that you edit from both but it gives the impression that 2 different people are in agreement. Robvanvee 18:42, 7 March 2020 (UTC)

Well I'm at home now (165.225.81.96 was/is my workplace) so it would have been even more confusing to write from another IP. Romomusicfan (talk) 17:34, 8 March 2020 (UTC)

Probably best not to bandy that IP about. Tony (talk) 00:48, 9 March 2020 (UTC)

Names of identifiers in citation templates

Are links to ISBN or DOI excessive? Should they link to some redirect for filtering purposes? Could using a tooltip to obscure the redirect title ever be appropriate? These are just a small handful of the questions being raised in this thread, which is actually going several directions at once. ―cobaltcigs 17:35, 11 March 2020 (UTC)

Discussion on April Fools' Day policies where MOS:EGG is being cited

 You are invited to join the discussion at Wikipedia talk:Rules for Fools#Revoke the Main Page's exemption from the disclosure requirement. {{u|Sdkb}} ^talk 01:42, 7 April 2020 (UTC)

Wikipedia:Manual_of_Style/Linking#Style

Regarding this section, it seems that numbers don't work as suffixes, either. For example LGMD2 doesn't work like LGMDr. Additionally, the linked help article [[3]] has nothing to say about this section.

Not sure if someone else has input? I'm not too experienced so I don't want to make an edit without input.

Lukelahood (talk) 15:57, 7 May 2020 (UTC)

Gracious, you're right! I'm going to add a tacky note about it, but I lack the energy to go ask whoever knows the answer. Can someone else do that at VP or meta or wherever? EEng 19:05, 7 May 2020 (UTC)

Linking to Fraternity/Sororities/Honoraries

Mu mu

-EEng

Is there any reason *not* to link to Greek Letter Organizations mentioned in an Article? For example, John Q. Smith belong to Mu Mu Mu, and Mu Mu Mu has an article, so link. Whatamatta U has a chapter of Mu Mu Mu mentioned, so link, Right The only time I tend not to link is when the Greek Letter Organization is mentioned in a reference, with something like cite Magazine and the Magazine is The Record of Mu Mu Mu.Naraht (talk) 20:04, 8 May 2020 (UTC)

(Throwing aside that trivial memberships probably aren't even worth mentioning) If the org was worth mentioning, it seems like a useful link as most readers couldn't distinguish one from the other.—Bagumba (talk) 02:38, 9 May 2020 (UTC)

MOS:OVERLINK: nationality in bio lead sentence

Per MOS:OVERLINK, a nationality which most readers will be at least somewhat familiar are generally not linked. I recently unlinked Canadian in the lead sentence of the bio Colby Cave, but there was a good-faith revert citing it as a WikiProject Ice Hockey standard. It seems that an earlier discussion in that project referenced an OVERLINK exception, Unless a term is particularly relevant to the context in the article ... That could apply to linking Canadian in Canadian American. However, that standard is not used in other sports' bios e.g. David Beckham, Serena Williams, Roger Federer or Michael Jordan. And with Canadian, English readers will be mostly familiar with the country and its people, so the debate should not be that Canadian is unfamiliar.

Should Canadian be linked in a the lead sentence of a bio?—Bagumba (talk) 09:57, 12 April 2020 (UTC)

No, definitely not. "Canadian American", maybe; but Canada/Canadian is so well-known it's dysfunctional to link to the whole article. There's a case for linking to the sport section in that article, in a "See also" section at the bottom of Colby Cave; there it can be explicitly linked rather than inaccurately piped. Better still, Sports_in_Canada#Ice_hockey, or Ice_hockey_in_Canada, which are much more specific and useful for readers—again, better not as "easter egg" links to "Canadian", but as explicit links under "See also". Our job as editors is to provide links that are as specific as possible; no one's going to click on "Canadian" if they've arrived to read about Colby Cave. Also, that opening sentence is already a sea of blue without the superfluous addition. Tony (talk) 10:14, 12 April 2020 (UTC)

Delink, is best. GoodDay (talk) 15:11, 12 April 2020 (UTC)

What or who is this "Canada" of which you speak? EEng 15:19, 12 April 2020 (UTC)

Never heard of it. Looking it up now. Tony (talk) 02:07, 13 April 2020 (UTC)

Found it .....right below the Artic circle. ..who would have guessed. I though it was just north of Floribama...my bad.--Moxy 🍁 03:01, 13 April 2020 (UTC)

Don't link names of large countries generally. There's a certain baseline of knowledge which you need in order to read our articles -- you have to be able to read English and understand most common terms which English speakers use, and "Canada" is very commonly understood term. We could cater more to people who don't know what "Canada" or "Germany" or "India" is, but if we're going to do that we would really want to link to every proper noun, and to any words that are a bit uncommon. There's a limit.

Exceptions for special cases of course, and for small countries. I would consider linking "Belizian" or whatever. Where the line is is not crystal clear, but many things aren't, so best judgement shpuld be used. Herostratus (talk) 03:20, 13 April 2020 (UTC)

No to Canada or [[Canadians|Canadian]]. Think of it this way, would someone come to an article on something and expect a link to the nation? For instance, if you were talking about the prime minister of Canada, I would expect a link to the nation. If you were dealing with a specific prime minister, maybe, but Wayne Gretzky, I wouldn't. The other question you would want answered is would a reader not know the nation itself or not know it in general? Again, if it were "United States", "Canada", "United Kingdom", "Russia", "Brazil" or any one of the major nations in the world, there's no need to link it or to its people. If it were to "Malawi", "Belize", or any one of the dozens of smaller nations, I wouldn't link it. If there's a city linked nearby, that reduces the need to link the nation as well as they can usually get to it through the city link. Those are my rules. They are somewhat based on advice I've read over the years. Not all editors would follow that prescription. Walter Görlitz (talk) 05:49, 13 April 2020 (UTC)

Can you define "large"? Last time I checked, Canada was the 38th country in the world by population, coming after Sudan, Ukraine, Uganda, Irak and Poland, slightly above Uzbekistan and Peru. I would probably not link any nationality in the lead. Place Clichy (talk) 14:19, 30 May 2020 (UTC)

Landmass, not population. Walter Görlitz (talk) 20:21, 30 May 2020 (UTC)

Frankly for a biography their nationality is very relevant in the context of the article so I would expect a link to Canadian for example which links to the people as opposed to the country. It is very reasonable to expect that a person reading about a biography about a Canadian might want to learn more about Canadians. In fact I have done it often so I know people do it, I can't imagine I am all that unusual. I am less convinced on linking to the country itself. -DJSasso (talk) 13:50, 23 April 2020 (UTC)

Nationality is usually not at all relevant. That's the point of WP:OVERLINK. Walter Görlitz (talk) 16:22, 23 April 2020 (UTC)

"It is very reasonable to expect that a person reading about a biography about a Canadian might want to learn more about Canadians." – Why? The evidence suggests that few people click links in the first place. The wikilink system is not for idle browsing. Tony (talk) 09:15, 24 April 2020 (UTC)

And to take that idea further. If someone wanted to know about Canadians, would they think, "Who are those people from the country north of the United States? Oh, I know { fill in the blank } is one of them! I'll check out that person's article because they're sure to have a ling to that people group." Or would they think, "Canadians" and drop the term into their favourite search engine, and click on the Wikipedia article because it's likely to be one of the first articles listed?

The other option. A reader comes to the article for { fill in the blank } and sees that their are Canadian, and think to themselves, "oh, I'm reading a biography about a Canadian and I want to learn more about Canadians. Wow. No link to Canadians. I guess I will never be able to learn more about them." Or will they put "Canadians" into the search box and hit enter? It's not much of an inconvenience.

In short, if there is a very close association between the subject and the nation, such as the Prime Minister of Canada, or the Governor General, then it is reasonable to include a link. For a celebrity, musician, sportsperson, inventor, writer, entrepreneur or other person, the association is much less close and so it doesn't make sense to have that link. Walter Görlitz (talk) 03:19, 25 April 2020 (UTC)

The nationality of a person in a biography is highly relevant to a person, in all walks of life, it is very influential to the development of a person. Of course they can put Canadians into the search box no one said they could not, but the entire point of wiklinks is so they don't have to do that. The wikilink system is designed for exactly this situation, for someone reading about something to then link to a page that expands on something that was particularly relevant to the subject of the previous article. -DJSasso (talk) 16:31, 28 April 2020 (UTC)

A person's nationality is relevant, which is why it's mentioned. Linking to the nation is unnecessary though. Do you understand the difference? Walter Görlitz (talk) 18:22, 29 April 2020 (UTC)

This is all very subjective. There is some sort of line that everyone has. Why do we link ice hockey on a player? Is ice hockey well-known enough or not? This could apply to all of the major sports. Seems like the same for nationalities. For Canadians born in Canada, the link to Canada in their birth_place field and in their articles was removed by me using AWB some time ago. Previously, editors had been linking Canadian to Canada. I and other editors agreed, that it was a bad wikilink. This was changed to link to Canadians instead. I do NOT believe that Canadians as a people are that well-known. I don't particularly care about linking as a basis of pride, just that we are not that well-known. As an example, I would apply this lack of knowledge to our neighbours to the south, who know little about us. And similarly for geography. So my line is to leave in the link to nationality of Canadians on that basis. I also agree with Djsasso above. I rarely do, but there you go. :-) Alaney2k (talk) 21:55, 28 April 2020 (UTC)

We shouldn't link the sport either, but the projects tend to object. Their rationale has been that they are notable for the sport, not their nationality. Walter Görlitz (talk) 18:22, 29 April 2020 (UTC)

There's more likely to be a demographic bias regarding sports that varies from country to country among English-speaking nations, making a link more relevant.—Bagumba (talk) 18:48, 29 April 2020 (UTC)

Some editors seem not to want to hear the clear consensus against linking Canada, including Alaney2k (see User_talk:Alaney2k#Overlinking). I have lived for years in each of the US, Canada and the UK, and I know that Canadians do get irritated with the lack of knowledge/interest of too many Americans have about the people north of the border. But we are not here to right great wrongs, and those readers are not going to click on the link anyway. Edwardx (talk) 12:49, 27 May 2020 (UTC)

That's not correct. Don't be misleading. I have NOT been linking Canada, I have been maintaining the links to the Canadians article. Which, I believe is NOT the subject of this discussion AND, I believe, there is no overwhelming consensus on. Some of this discussion smacks of 'I want things done this way' and that is inappropriate. I, in fact, remove links to Canada when I find them. I agree in not having Canada linked in the first sentence of articles, and as little as possible. Alaney2k (talk) 14:52, 27 May 2020 (UTC)

Linking Canadians is also an OVERLINK, and is the subject of the discussion, (first comment is "Should Canadian be linked in a the lead sentence of a bio?") and I believe you should stop now as there is consensus to not link these terms. Walter Görlitz (talk) 14:56, 27 May 2020 (UTC)

Sorry, I miswrote there. The adding Canada is not the discussion here. Alaney2k (talk) 16:16, 27 May 2020 (UTC)

@Alaney2k: What's the benefit to your changing bio leads to pipe "Canadian" from "Canadians" instead of "Canada",[4] when MOS:OVERLINK suggests that "Canadian" should not be linked at all in this context? Regards.—Bagumba (talk) 16:28, 27 May 2020 (UTC)

Canadian redirects to Canadians, that is Canadian people, not Canada as a country. That is more pertinent to the biography of a person. I am not going to go back and remove those links, as editors deliberately put them in. They are not out-right banned as some people want. I don't know why they want that. The nationality of a person is pertinent. It is not like linking 'water'. There are two opposing essays on linking/overlinking, too. Alaney2k (talk) 16:35, 27 May 2020 (UTC)

The consensus in this discussion seems quite clear. Nobody is telling you to go out of your way to remove the "Canadian" links, but if you are editing the link anyways, I don't understand why you would edit it against a known consensus.—Bagumba (talk) 16:50, 27 May 2020 (UTC)

@Bagumba: I'll try to explain. Please consider me to be in good faith. Like I've mentioned, I am not 100% in disagreement with the people who want to impose overlink on Canadians links. I do remove wlinks to Canada. That seems to be non-contentious. But editors put in those wlinks for whatever reason (project rules, etc.) and I don't intend to over-ride them. I do not think they are represented in this discussion. As mentioned in the first sentence of this discussion, the wp:hockey project wants the links to Canadian in the prose, not the infobox. The infobox is for the national team the player participated with. Secondly, I personally disagree with the designation that the Canadians wlink should be removed. The argument that it won't be clicked on is subjective. When most of the wlinks on webpages are not clicked on, then the point that certain ones are irrelevant does not follow logically. I think that the facts are simply not complete. You cannot prove conclusively that just because a link is not clicked on, that it is irrelevant. We regularly have duplicate wlinks to biographical locations in articles, wlinks to common sports, etc. On biographical articles, the person's nationality is relevant. I would relegate it to the infobox, but even that seems to be contentious. Like I've said, I think this is being a bit dictatorial and not really in the best interest of readers. Alaney2k (talk) 17:31, 27 May 2020 (UTC)

So just to be clear, you clearly do not agree with the consensus that has been put forward here, but do you at least recognize that the consensus is that there is no need to link "Canadian" to anything? Walter Görlitz (talk) 04:43, 28 May 2020 (UTC)

I would not put it in those words. That's too wide of an interpretation. Alaney2k (talk) 12:36, 28 May 2020 (UTC)

Put it in your own words then. The discussion as I have read it here is no linking to Canada, Canadian or Canadians. Walter Görlitz (talk) 04:36, 29 May 2020 (UTC)

Should links to WMF sites be plainlinks?

King of Hearts and I were just discussing 360-degree panoramas, which are displayed through a link to a WMF Labs tool, and the question came up of whether links to WMF sites from Wikipedia should be plainlinks (like this) or non-plain links (like this). The sense I have is that they're generally treated as plain. Is it okay to codify that standard here? {{u|Sdkb}} ^talk 19:25, 1 June 2020 (UTC)

We tend to prefer WMF sites in the form of interwiki links so that the HTTP scheme can be decided automatically, among other reasons. These also have the benefit of being presented as "plain links". I do not really see a need to add text here for what is general practice. --Izno (talk) 23:16, 1 June 2020 (UTC)

Closely related RFC

The RFC at Wikipedia talk:External links#RFC on how to format external links may eventually affect some of the contents of Wikipedia:Manual of Style/Linking#External links section. All editors are welcome. WhatamIdoing (talk) 18:45, 15 August 2020 (UTC)

Linking to 1,000,000,000 in articles on companies

I just did two edits relating to links to 1,000,000,000 in the form of [[1,000,000,000|billion]] in articles on companies in the Revenue field of the Infobox.

• For News Corp (2013-present) Revision as of 05:12, 6 September 2020
• For Alphabet, Inc. Revision as of 05:19, 6 September 2020

I'm not sure why this was done, and MOS:OVERLINK says, "Unless a term is particularly relevant to the context in the article, the following are usually not linked: •Everyday words understood by most readers in context (e.g., education, violence, aircraft, river)."

But perhaps there was some purpose behind it. I have not sought to go back in the article history to identify how that came about.

SuchAndSuchWork (talk) 05:33, 6 September 2020 (UTC)

This was probably done to avoid ambiguity between the long and short scales. Readers from long scale countries likely misinterpret billion as 10¹². But according to MOS:BILLION, the short scale can be assumed. —Dexxor (talk) 06:48, 6 September 2020 (UTC)

@Dexxor: Thanks very much for shedding some light on that. The only language I speak is English so I was not aware of the long and short scales. SuchAndSuchWork (talk) 22:27, 6 September 2020 (UTC)

It's like feet vs. meters only way more confusing. EEng 23:03, 6 September 2020 (UTC)

"Billion" should not be linked in normal circumstances. No one nowadays uses billion as million × million. Tony (talk) 11:58, 7 September 2020 (UTC)

No that's not true, as you'll see at short and long scales. But most English-speaking countries use short scale, so MOS provides that we assume readers will interpret things that way. EEng 15:33, 7 September 2020 (UTC)

Using two links where one will do

I am having difficulty describing it, but I thought there was a specific rule against splitting a link, by piping the first half of one link and then adding the second half as a separate link.
Rightly or wrongly, I have just changed - [[Pike County, Alabama|Pike County]], [[Alabama]] to [[Pike County, Alabama]].
Is there a rule/guideline to specifically cover this? - Arjayay (talk) 17:08, 7 June 2020 (UTC)

MOS:SEAOFBLUE.—Bagumba (talk) 17:11, 7 June 2020 (UTC)

Hi Bagumba - that doesn't cover it, because [[Pike County, Alabama|Pike County]], [[Alabama]] renders as Pike County, Alabama whilst [[Pike County, Alabama]] renders as Pike County, Alabama - they are both exactly as blue as each other, right down to the blue comma - Arjayay (talk) 17:18, 7 June 2020 (UTC)

No, Bagumba linked the exact correct section. Please read the referenced section and then explain how you come to a different conclusion. --Izno (talk) 17:22, 7 June 2020 (UTC)

My apologies Bagumba, I stand (sit actually) corrected. I thought MOS:SEAOFBLUE only referred to the density of links, but it includes "avoid placing links next to each other so that they look like a single link" - apologies again - Arjayay (talk) 07:21, 8 June 2020 (UTC)

The change Arjayay described (1 link, not 2) is entirely reasonable and covered by WP:OVERLINK. Using artificially 2 links as described can only confuse readers. -- Michael Bednarek (talk) 00:52, 8 June 2020 (UTC)

The mindless linking of [City], [State] instead of [City, State] is one of my pet peeves, especially since an editor has to go to extra trouble to do the wrong thing: [Mobile, Alabama|Mobile], [Alabama]. EEng 01:45, 8 June 2020 (UTC)

I propose adding something specific after SEAOFBLUE to address city–state like:

For geographic places within an entity (e.g. state, province, etc.), do not add an additional link to the entity. Those links are generally available in the place's article for readers seeking more information about the location.

• Use: [[Buffalo, New York]], e.g. Buffalo, New York
• Avoid: [[Buffalo, New York|Buffalo]], [[New York]]

• Use: [[Toronto, Ortario]], e.g. Toronto, Ontario
• Avoid: [[Toronto]], [[Ontario]]

—Bagumba (talk) 08:10, 8 June 2020 (UTC)

Sounds good, but can it be called something other than SEAOFBLUE, to avoid other people making the same mistake as me? - Arjayay (talk) 09:00, 8 June 2020 (UTC)

Naming new shortcuts is the easy part ;-) —Bagumba (talk) 09:08, 8 June 2020 (UTC)

• All of which raises the matter of why the state should be linked at all, when a specific item within the state is linked. Tony (talk) 09:51, 8 June 2020 (UTC)

  For the US, the common naming covention per WP:USPLACE is city, state. In real life, Americans commonly refer to city-state together a la the US postal address system. Getting back to your question, it's a case of MOS:OVERLINK to want to provide a direct link to the state also. For example, see this recent edit linking "Texas" in "Houston, Texas".—Bagumba (talk) 10:19, 8 June 2020 (UTC)

• The point is that you can get to the state/province from the first few words of the article on the city, and by creating a SEAOFBLUE -- part of which links to the city and part to the state -- you risk the user unknowingly clicking on the wrong bit. It seems to me there must be other examples of hierarchies (other than geographical) in which it's better to link [X, Y] instead of [X], [Y]. EEng 10:44, 8 June 2020 (UTC)

  There probably are similar cases, but I'm not aware of them to write a more generic proposal. In some way this is similar to MOS:SPECIFICLINK, but not exactly either.—Bagumba (talk) 11:13, 8 June 2020 (UTC)

  Not exact but good enough. Anyone have a reason the examples above shouldn't be added there? In fact, I think one or both of those examples would work nicely at SEAOFBLUE as well. EEng 13:33, 8 June 2020 (UTC)

Those are not the best examples. We do not want to encourage linking to redirects, WP:NOTBROKEN notwithstanding. Also, Toronto, Ontario may not be the best linking in general as it may be better to use Toronto, Canada in many instances. A location in the UK or Europe might be a better choice. Walter Görlitz (talk) 16:49, 8 June 2020 (UTC)

I think including both versions of Toronto would be usefully illustrative. Why do you say We do not want to encourage linking to redirects? EEng 22:43, 8 June 2020 (UTC)

I disagree. Toronto, Ontario is a prime example. For Toronto, Canada, more people would agree that linking Canada is MOS:OVERLINK, so it's not as much of an issue.—Bagumba (talk) 01:47, 9 June 2020 (UTC)

Don't change my indent style I was correct.

We do not want to encourage linking to redirects because it's wrong to do so. If someone does so accidentally, NOTBROKEN applies and we don't have to fix it, but we do not want get people to link to redirects by default. The link here is at Toronto, not Toronto, Ontario. Again, somewhere in the UK or Europe would be better. Walter Görlitz (talk) 06:21, 11 June 2020 (UTC)

[[Buffalo, New York]], is indeed better than [[Buffalo, New York|Buffalo]], [[New York]], but is equally blue. Arguably, [[Buffalo, New York|Buffalo]], New York is even better. I don't see the need to see that it links to the [[City, State]], any more than the state needs to be linked. Extra benefit is that we won't have any problem like Walter highlighted with [[Toronto, Ontario]] above.-- Ohc ^¡digame! 14:17, 11 June 2020 (UTC)

The amount of blueness isn't the issue. The (primary) issue is having two links hard up against each other so that the reader doesn't realize that clicking one bit of a single stretch of blue takes you somewhere else than does clicking a different bit of the same stretch of blue.

We do not want to encourage linking to redirects because it's wrong to do so – That just a tautology. Why is it wrong? EEng 21:41, 11 June 2020 (UTC)

I don't think it's a tautology at all. "We don't want editors to link incorrectly because it's incorrect" would be a tautology where I'm from. Perhaps you'd prefer, we don't want to encourage editors to intentionally link to redirects because it's a waste of resources. If there's a direct link, we should make every effort to link to them, however

1. if it were a possible future articles
2. if it introducing unnecessary invisible text that makes the article more difficult to read in page source form
3. if a non-piped link makes better use of the "what links here" tool
4. if it's an intentional links to a disambiguation page

In short, if editors persistently use a redirect instead of an article title, it may be that the article needs to be moved rather than the redirect changed, but we don't want to encourage such a move by encouraging editors to create redirects. Walter Görlitz (talk) 02:48, 12 June 2020 (UTC)

Thanks for the lesson in the terminology of logic, though with my degrees in mathematics and computer science I really didn't need it; I was speaking in the colloquial language of normal people. Your idea that redirects are a waste of resources is completely absurd. An actual waste of resources would the waste of human time and brainpower that would result from telling editors that they should fuss about figuring out what's a redirect and what's not, your contrarian disdain for WP:NOTBROKEN notwithstanding. Redirects are there to be used, and the clarity of the text and links we present to our readers is paramount. EEng 14:49, 12 June 2020 (UTC)

Yet your degrees didn't help you to meet MOS:INDENTGAP , no redirects are there to be used, they're there for editors who make mistakes. I can see why you would like them. Walter Görlitz (talk) 01:41, 15 June 2020 (UTC)

And your degrees don't enable you write full, grammatical sentences – as you no doubt are already painfully aware – nor to comprehend our guideline on the purpose of redirects. If you'd like to keep sputtering inarticulate blunders and irrelevancies do be our guest. EEng 03:05, 15 June 2020 (UTC)

I don't write full sentences? Gee. The guideline is simple: if a link exists that points to a redirect, there's no need to change it. The guideline does not state that we should encourage intentionally pointing to a redirect. That's not the purpose and you can't make it that way. Walter Görlitz (talk) 04:35, 15 June 2020 (UTC)

You're doing better on writing grammatically – and keep up the good work on that – but still having trouble with absorbing the guideline, which says There is usually nothing wrong with linking to redirects and goes on to list several cases in which redirects are affirmatively desirable. You even listed some of those reasons above, though in incomprehensible sentence-fragment form. A bit above that you betrayed what seems to be your fundamental confusion on this topic, which is the idea that redirects are a waste of resources – a common misconception among people who have no concept of the structure and performance of modern computer systems. Now sputter away one more time if you please, but your simple self-certainty is losing its charm so I don't think I'll be answering further. EEng 05:36, 15 June 2020 (UTC)

NOTBROKEN is about not "fixing" links. It does not state that we should fix them. Walter Görlitz (talk) 05:49, 15 June 2020 (UTC)

I think you mean "It does not state that we should use them" (or something) but we've learned to make allowances for your difficulty in expressing yourself. You're fixated on NOTBROKEN to the exclusion of everything else. For example, WP:Manual_of_Style/Linking#Link_specificity explicitly encourages linking to redirects:

Always link to the article on the most specific topic appropriate to the context from which you link ... If there is no article about the most specific topic ... consider creating the article yourself [or] create a redirect page to the article about a more general topic

and MOS:NOPIPE says

do not use a piped link where it is possible to use a redirected term that fits well within the scope of the text.

Your idea that If there's a direct link, we should make every effort to link to them is 180 degrees from guidelines and common practice, not to mention common sense. Now I really do have to get back to working with grownups who enjoy learning and improving themselves intellectually, so I'll leave you to your world of selective reading and fantasy. Sputter on. EEng 12:47, 15 June 2020 (UTC)

• EEng, it's the amount of blue, the need to focus linking on the more specific (away from the uber-well-known, like "New York"), and the need to avoid confusing readers as to whether a blue patch is one item or two bunched together. Tony (talk) 03:21, 12 June 2020 (UTC)
  • Agree, use Toronto not Toronto, Ontario. Articles for municipalities use the shortest non-ambiguous title. We use the second, province, link as a help. A province would not be expected to be a familiar topic for persons outside of Canada to know. Canadian readers are only a small fraction of the English-speaking readership. Alaney2k (talk) 04:16, 12 June 2020 (UTC)
    • @Alaney2k: There are some cases where it's useful to list "Toronto, Ontario", such as the table at Canadian Football League#Active teams, which conveys which teams are in the same province. It was not the intent to mandate inclusion of "Ontario"; it's a separate editorial decidion. When listing "Ontario" is deemed appropriate, showing "Toronto, Ontario" just invites drive-by editors to create a SEAOFBLUE and link Ontario. At any rate, we should list the Buffalo, New York case, at a minimum.—Bagumba (talk) 08:40, 12 June 2020 (UTC)
• Memo to Walter Görlitz and EEng: if people with degrees could write professional-standard English, I'd be out of a job. Tony (talk) 07:06, 15 June 2020 (UTC)

  Well some of us can, at least. But even someone with your prodigious powers of composition and editing would find WG a full-time job. EEng 12:47, 15 June 2020 (UTC)

• I'd like to see something like Bagumba's proposed text adopted, but I'm too lazy to say anything intelligent about the issue just now. Sleepyyyyyyy. EEng 15:37, 7 September 2020 (UTC)

Overlink/Repeatlink clarification needed

MOS:OVERLINK says "locations and geographical features should not be linked repeatedly" until and unless its necessary. MOS:REPEATLINK policy which says "if helpful for readers". Now England cricket team Test results (1990–2004) article have mutiple times linking London, Leeds or Manchester and many other cities. And here neither it seems necessary to link nor anyway helpful for readers since linking such cities once is enough. Can anybody clarify on this. Drat8sub (talk) 17:22, 20 May 2020 (UTC)

Also I've gone through this previous discussion, which focusses on seasons of tournament or sports which can be helpful for readers. But my particular concern is about location i.e, cities or countries. Here, two geographical features used, the stadiums and the cities. Now one can argue with table sorting, which is fine. Now to my understanding, it can be accepted that, the stadiums' links can be helpful for readers which are not so common and can be assumed totally unknown and will be helpful during table sorting, but cities which are very common and per statistics of wikipedia, as written in MOS:OVERLINK, readers hardly open them. Again, if one considers that the reader is very much interested in venues, it should be assumed that the reader would first like to know about the stadium, and once the stadium link is opened, the stadium article will again have the location i,e. city's link. So is it necessary to add links of locations (cities & countries specifically). Drat8sub (talk) 18:42, 20 May 2020 (UTC)

My feeling with these types of tables is that a reader is likely to skim the material and not read from beginning to end, so REPEATLINK would allow for multiple links. Also with tables, I'd just prefer to link all similar entries (e.g. cities) if most of them would anyways be linked. It just looks consistent for a table, and a drive-by editor almost always "corrects" links "missing" in tables anyways (less so with prose, in my experience).—Bagumba (talk) 17:12, 27 May 2020 (UTC)

Well, first, such articles are expected to be read by soemone who knows about cricket lot..one can say fans of cricket, since these list articles are subordinate records, not primary, hardcore cricket lovers will follow up such records and statistics, so will definitely go from top to bottom. And, it can be expected such reader will definitely know about these cities, like if I consider myself (I am not big fan), but sometimes watch cricket, but knows all these cities just watching cricket only, as because these are fixed cities where cricket are often played in tours. Like if India makes a tour to England, everyone knows that London, Nottingham, Birmingham, Manchester will be the venues. I mean it will be bizzare to say you need to open a cricket team's subordinate record article to know about London, Leeds, Kolkata, Mumbai, Manchester, Perth, Sydney...list goes on. I mean if we are intended to keep such cities' links, which are expected to be known to almost every reader, then is not the Overlink policy becomes redundant, and tomorrow another person will give the same argumentS that its for consistency and almost all list articles will be linked. Why there is a need of Overlink policy then for geographical location, wikipedia can remove it. Drat8sub (talk) 13:35, 28 May 2020 (UTC)

All links to cities should be omitted in that table; the link to the venue is sufficient. If a reader is interested in the city, it's 1 click away. Although WP:REPEATLINK makes allowances for tables, providing identical links to countries' teams in successive rows looks like overlinking and is unnecessary. -- Michael Bednarek (talk) 13:49, 28 May 2020 (UTC)

Bringing Slashme, Bsherr, Mathglot, EEng, Tony1, Flyer22 Frozen, Dhtwiki, Mandruss (find your names from recent page history) here, if you are interested please put your views here, it has been a regular conflict in the list articles. A productive discussion from such experienced editors and reaching a consensus would be good for the list articles. Thank you. Drat8sub (talk) 14:36, 30 May 2020 (UTC)

Responding to ping. As far as I'm concerned, your rationale regarding lists (or tables?) applies to open prose as well. Readers generally don't read articles top-down, and, regardless, they shouldn't be expected to go find a link when they want it. It would be easier to scroll to the top and use the search box, which would defeat the purpose of links. To my mind, then, it makes more sense to eliminate REPEATLINK (or dramatically modify it, and I haven't the time to think about how that could be done in a useful way) than to make lists (or tables?) an exception to it. ―Mandruss ☎ 14:52, 30 May 2020 (UTC)

Oh dear, imagine the pain of having to type Leeds into the search box. Tony (talk) 05:57, 31 May 2020 (UTC)

No need to, really. The venue next to it has a link in the first sentence. -- Michael Bednarek (talk) 11:32, 31 May 2020 (UTC)

I feel much as Mandruss does re revisiting repeat-linking philosophy for regular article prose. For tables specifically, I have little to say about whether these particular links are helpful to readers of these particular kinds of articles, but if they're helpful once, they should be repeatedly linked; lists aren't read like prose, and the argument that lists are for hardcore cognoscenti rings very hollow. The main argument against repeats is that they distract while reading, but lists aren't read. EEng 15:39, 30 May 2020 (UTC)

Clarification: I am sorry, I think I failed to convey the point clearly, my argument is not about linking in the prose but about linking cities multiple times in the table. Thank you. Drat8sub (talk) 17:07, 30 May 2020 (UTC)

I think that's understood at least by EEng and me. I can't support such a change to the guidance when I strongly feel it's the wrong solution. Apologies for complicating the issue. ―Mandruss ☎ 18:15, 30 May 2020 (UTC)

Thanks for ping. A tangentially related discussion is at Archive 132. I tend to agree with User:Beyond My Ken at the linked discussion regarding the difference between guideline and policy; in a nutshell, (paraphrasing): "it's just a guideline"; if you think it's helpful to the reader to relink the cities, go for it. Also, I have no special feeling about tables, but in my mind, user-friendliness is "improving the article", and that trumps the guideline. So if the answer to the question, "Is the reader better served by multiple links?" is "yes", then do it. Though I also understand clearly what your argument is, some articles with non-table structures may yet have something to say about your question, because they have an analogous issue; namely, like a table, there are other formats that are not read top-to-bottom like a single narrative, but are composed of many similar, discrete, pieces; instead of rows, the repeating pieces may be bullets, sections, or something else. Timeline articles, List articles, and other articles with repeating structures, like, say, this one, come to mind. They may well be consulted piecemeal and not read through as a single narrative. In these cases, failing to include "redundant" links in the repeating structure, whether it be table row, bullet item, section, or something else does not do the reader a service and may slow their understanding. Mathglot (talk) 20:08, 30 May 2020 (UTC)

I agree with the distinction Mandruss makes between lists/tables and prose, and his conclusion that repeat links in tables may be desirable since tables are not necessarily read top-down. I do however believe the guidance to only link once in prose is a good practice. In my view, the guidelines currently are consistent with these views, and no changes are warranted. As to the table in question, I find the links to cities not to be so obvious as to implicate MOS:OVERLINK (countries would be a different story), and the repeat links to be helpful. The opposition team names are more marginal, and I would defer to local consensus on the topic. I don't find the table to be inconsistent with the guideline, though. --Bsherr (talk) 20:52, 30 May 2020 (UTC)

I'm for retaining REPEATLINK and applying it with common sense. It allows for common sense usage. The reason I wouldn't want to eliminate it is because of the egregious overlinking that can and has occurred, especially among newbies. I think an overlinking concern is less of an issue for lists/tables. But in an infobox for a film, for example, I see no valid reason to link the director multiple times; once is enough. Flyer22 Frozen (talk) 23:57, 30 May 2020 (UTC)

• Why is "London" linked (unless there's some confusion with the London in Ontario, which there isn't in that cricket article). I really wonder which reader would want to click on "Birmingham" or "Leeds", too. So my inclination is to link the latter two only once, on first occurrence—or not at all. It's about cricket, not English cities. Tony (talk) 05:55, 31 May 2020 (UTC)

I a reader is interested in a town/city, a click on the venue will get them there. -- Michael Bednarek (talk) 11:32, 31 May 2020 (UTC)

Reminding people that cricket is the topic, not geography. Tony (talk) 00:57, 1 June 2020 (UTC)

• I think this is a useful discussion that should be continued. EEng 15:40, 7 September 2020 (UTC)

Semi-protected edit request on 24 September 2020

Please replace

After playing [[Germany national football team|for Germany]]

with

After [[Germany national football team|playing for Germany]]

because it would be a reasonable link and it makes the point even clearer. 64.203.186.120 (talk) 12:43, 24 September 2020 (UTC)

I think we should just switch to some other example. The current version is too much of an unexpected submarine link despite the page saying it's showcasing how to avoid surprise, your proposed version is probably more obvious to a reader but it's not commonly used on Wikipedia. – Thjarkur (talk) 12:56, 24 September 2020 (UTC)

@Þjarkur: How about:In 1845, the Republic of Texas was [[Texas annexation|annexed by the United States]]. instead of In 1845, the Republic of Texas was [[Texas annexation|annexed]] by the United States. ◢  Ganbaruby!  (Say hi!) 09:23, 25 September 2020 (UTC)

I like it, certainly much clearer – Thjarkur (talk) 09:26, 25 September 2020 (UTC)

 Done. ◢  Ganbaruby!  (Say hi!) 09:59, 25 September 2020 (UTC)

Input sought on use of Wikilinks in table

I'd appreciate input on the use of Wikilinks in the table located here. My view is that individually Wikilinking each instance of "Singles match" is excessive; another editor has contested this and pointed to the wording at MOS:REPEATLINK. The MOS states "if helpful for readers, a link may be repeated in infoboxes, tables..." I would welcome a view on whether the Wikilinking of each instance of "Singles match" in this table is or is not appropriate. McPhail (talk) 16:04, 30 September 2020 (UTC)

The user McPhail did not give much background here which makes it difficult for a reader to understand what was being asked. For example, every instance of a match is linked at WrestleMania 35#Results. Per WP:REPEATLINK: "Duplicate linking in stand-alone and embedded lists is permissible if it significantly aids the reader. This is most often the case when the list is presenting information that could just as aptly be formatted in a table, and is expected to be parsed for particular bits of data, not read from top to bottom. If the list is normal article prose that happens to be formatted as a list, treat it as normal article prose." These results are often not read top to bottom and are a perfect example of this exception, as someone may come to these results tables and will want to see what happened in the final match and not the opening matches

In addition WP:PW has already discussed this and by consensus agreed it should be repeated and created a style guide based on the interpretation of this policy. There is a hidden table behind a green bar at WP:PW/PPVG, which shows the agreed upon use based on this exception. - Galatz גאליץ^{שיחה Talk} 16:17, 30 September 2020 (UTC)

The salient question is whether the duplicate linking "significantly aids the reader". My view is that in the table in question it does not, but I would welcome views on this point. McPhail (talk) 16:21, 30 September 2020 (UTC)

If they were not duplicated, in addition to "Singles match" being linked once, 4 of the people in match #7 would not be linked. A person who was not reading the table from top to bottom, as one typically will not in results tables, is aided by this duplication. Your argument would say no one should be linked in the results table at all since they are linked in the article already. So either repeating links is helpful or its not, and clearly you think it is or else linking the article should be enough. - Galatz גאליץ^{שיחה Talk} 16:29, 30 September 2020 (UTC)

The statement in the MOS for repeating links is intended for tables where you have a count of rows > the height of a page (so, let's call it 30 rows). That table fits entirely on one page, even on most mobile displays. I would suggest removal here. --Izno (talk) 00:42, 1 October 2020 (UTC)

A question about MOS:OVERLINK

In MOS:OVERLINK, it's mentioned that only major examples of countries, geographic features, locations, languages, nationalities and ethnicities, and religions are subjects with which most readers will be at least somewhat familiar. What is meant here by "major"? What constitutes as major and what constitutes as minor?

Furthermore, it's mentioned that these are not linked unless they're particularly relevant to the context in the article. What specifically is meant here? Should Australia and Western Australia be linked in Perth? Should they be linked in Politics of Australia? What about in States and territories of Australia? Also, what cities are linked and what are left without a link? Linking only capital cities would lead to, for example, New York City not having a link and Lilongwe having one, which wouldn't make sense. _Signed, Swiftestcat ^talk 13:23, 29 September 2020 (UTC)

Major constitutes any country, geographic feature, location, language, nationality, ethnicity, or religion that a reader of the article could be expected to know. For instance, in the discussion of a recording, one does not need to link the nation or studio album in the first sentence, but the musician should be linked as there is a close association.

Highway 61 Revisited is the sixth studio album by American singer-songwriter Bob Dylan.

I'm not sure what the manual of style is for Australian subjects, but other than the United States, the state or province should not be listed unless the subject is likely to be internal to the nation. For instance, I would argue that the Rolf Harris article errs in listing Bassendean, Western Australia, Australia. If it is for internal readers (in this example, primarily Australians) then linking it is entirely unnecessary as the link will likely never be clicked. Similarly the article on Gordon Lightfoot lists Orillia, Ontario, Canada, but the province is not linked. New York, London, Paris, Munich and most other cosmopolitan cities do not need to be linked, unless they're relevant to the context in the article. For instance, I would link the city of Sydney in the article on the Sydney Airport, but not in an article about a sportsperson, or the like. However, if you're linking the city, I would definitely not link the state (if it were listed) or the country. If the article on that sportsperson had a personal life section, I would not link that the person was "Christian", or "Evangelical", or "Anglican", or "Jewish", etc. It's probably not necessary.

The way I've heard it explained is that you should ask, "will the reader click on a link to this subject because they are unfamiliar with it?" If the answer is "no", then no link is needed. I've also heard it suggested that if a reader were to ask, "what's that thing? You know, the one that <subject> is known for? Oh I know, let's go to <subject>'s Wikipedia article to read more about it." Would the link be there? If no one would ask that question, don't link it. For instance, "what the name of the mountain the Edmund Hillary scaled?" If there's a link to Everest, the reader will be happy. If no one would ever ask that question, the link to Everest is an OVERLINK.

In short, this is not a rule, it's a call to common sense. By all means, avoid a WP:SEAOFBLUE. Walter Görlitz (talk) 03:52, 2 October 2020 (UTC)

Navboxes

I inserted navboxes into the existing text, so that it reads:

...if helpful for readers, a link may be repeated in infoboxes, navboxes, tables, image captions, footnotes, hatnotes, and at the first occurrence after the lead.

because, well, doing so is custom and practice in every single case where a navbox mentions something already linked in an article.

This was reverted the edit summary "No reason for that addition per common sense, and also because the content of navboxes is not in the article text (while the rest of that section concerns those constructs which are)". The former claim is bullshit, and the latter claim is false, as neither footnotes or hatnotes are part of article text article text.

Accordingly, I restored my edit. It has again been reverted.

Can anyone make a cogent case why navboxes should not be mentioned in this context? Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 17:01, 28 November 2020 (UTC)

@Pigsonthewing: Your failure to understand the context of that sentence is clear. Those elements mentioned in that section clearly have some text appearing in a specific page that a user can choose to link. "Navbox"es do not allow the editor of a specific page that choice. 'nuff said. --Izno (talk) 20:18, 28 November 2020 (UTC)

Redirect-link in a hatnote

{{Distinguish}} says it is to be used in cases of linguistic confusion. Suppose for an article there is a similar-spelling/pronounciation unrelated topic, but that other topic is actually synonym of (and therefore redirect to) an article whose actual page-name is totally different. As a specific example, "fluorene" is similar to "fluorane" but are unrelated topics, and fluorane→hydrogen fluoride as synonymous. Is it appropriate to use the redirect in the hatnote, or should (instead, or also) the redirect's target be listed? For example, should the Fluorene article have:

• Not to be confused with Fluorane
• Not to be confused with Fluorane or Hydrogen fluoride
• Not to be confused with Hydrogen fluoride

DMacks (talk) 18:25, 3 October 2020 (UTC)

To explain to the reader why a hatnote is there, the first alternative makes the most sense. The second is confusing because the "or" is not really an "or" as both terms link to the same article, and the third leaves the reader wondering why there could be confusion. -- Michael Bednarek (talk) 00:15, 4 October 2020 (UTC)

DMacks: I'd usually opt for the redirect term, which is closer to what the reader was intending to find. Generally, I'd go for a more explicit explanation with {{for}}, unless it's really obvious what the misspelled term is. A lot of times I'm looking for a word I heard, but am not sure exactly what it is. We shouldn't make readers click just to find out if it's the right topic for them.—Bagumba (talk) 13:21, 3 December 2020 (UTC)

MOS:EGG and linking to specific pages

I've come across a similar question a few times now, and wanted to present it here for more general consideration. It arises for pages like, for a prime minister John Smith of Foobaristan:

• At Smith's page: Smith downplayed the threat of the [[COVID-19 pandemic in Foobaristan|COVID-19 pandemic]] vs. Smith downplayed the threat of the [[COVID-19 pandemic]]
• At the 2020 Foobaristan elections page: A major issue of the election was the ongoing impact of the [[COVID-19 pandemic in Foobaristan|COVID-19 pandemic]] vs. A major issue of the election was the ongoing impact of the [[COVID-19 pandemic]]

In both of these instances, plus a bunch of other similar ones, we have three options, all of which have drawbacks:

1. Link to the worldwide pandemic page, a less appropriate and less useful link than the country-specific pandemic page would have been, thus going against MOS:SPECIFICLINK.
2. Link to country-specific pandemic page, introducing a possible minor MOS:EGG issue.
3. Adjust the passages to make the link something like [[COVID-19 pandemic in Foobaristan|the country's COVID-19 pandemic]], introducing unneeded verbiage that is already clear from the page's context.

There's no perfect solution here, but my sense is that different editors have different editorial approaches to balancing these concerns, leading to slow-motion edit wars and inconsistency, so it'd be nice if we could decide on the best option for these situations. {{u|Sdkb}} ^talk 20:31, 8 November 2020 (UTC)

• Personally, in most cases I tend to go for Option 2. The more specific pages almost always link prominently to the general page in their own leads, so someone who ends up at the Foobaristan pandemic page can easily find the more general page if that's what they want. Conversely, linking to the more general page (option 1) typically means that readers who might have benefited from the more specific page won't read it. Option 3, if applied at wide scale, would lead to lots of very clunky prose that would ultimately reduce readability. {{u|Sdkb}} ^talk 20:31, 8 November 2020 (UTC)
• This is not a question of EGG actually in that it might be surprising. It is a question whether that is the most accurate inference for what that person said. Was he actually talking about the pandemic in Fooistan? Yes? Then a link there is reasonable. If not, then a link there is not reasonable, regardless of any other case. You shrug this off as unneeded verbiage but it is in fact sorely needed. Assuming the link is reasonable, then the text should always make it clear where someone is going, so that leaves us with 3. --Izno (talk) 20:53, 8 November 2020 (UTC)
• You raise interesting points that show how much skill we editors can apply to linking. Another option, too rarely used, is to link to the actual target name downs in the ==See also== section. Tony (talk) 07:48, 9 November 2020 (UTC)
• Use the specific link where it is reasonable. MOS:EGG is not necessarily a concern: "it should at least take them somewhere that makes sense". Unless the person was dealing with COVID on a larger scale (e.g. international trade impacts, general worldwide research, etc), the country specific article would generally be more applicable.—Bagumba (talk) 09:49, 9 November 2020 (UTC)
• Linking to the specific page seems better overall. As argued above, it's not a surprising target, and it's easy to find the general page from the more specific one. I'm not sure that inserting modifiers like "the country's" is all that clunky; that's perhaps down to a matter of taste. XOR'easter (talk) 21:40, 9 November 2020 (UTC)
• Option 3. The reader is much more likely to skip the link if they think it goes to a large article on a general topic. We should make it clear where the link goes so they can make a better decision if they want to read the linked article I tend to replace the EGG links if I see a way to do so with it not being overly "clunky". MB 14:59, 3 December 2020 (UTC)

Linking "song"

I want to unlink the common word "song" that is emblazoned at the top of infobox song in more than 60,000 articles. But Sdkb – a contributor to a thread above – appears to object because "song" is "absolutely particularly relevant" to an article about a song. It seems like an argument that would lead to the abolition of MOS:OVERLINK.

The reason I came across this is that I'm trying to improve music articles by unlinking a host of everyday items like "singer", "vocalist", album, and that crowd the opening sentences of many articles – not to mention "actor", "writer", "poet", etc. The "Links to song" list is currently overwhelmed by the linking of "song" in the infobox; so the list is of little use in detecting articles in which the item occurs in the main text.

Can I please have opinions on this?

Tony (talk) 04:38, 24 December 2020 (UTC)

Tony1, if you want opinions on that question, put a neutral invitation here; don't create a talk fork by asking the same question, and don't canvass by arguing your case (and misrepresenting my position) in the invitation. Anyone else who wants to weigh in, feel free to do so at Template talk:Infobox song#Recent edit, where I'll let the discussion so far speak for itself. {{u|Sdkb}} ^talk 05:09, 24 December 2020 (UTC)

It is neutral. I quoted you in context. I gave my reasoning. Tony (talk) 07:15, 24 December 2020 (UTC)

In terms of a single or other music-related article, "song" would most definitely be an over-link. I agree that actor does not need to be linked in those articles, nor does writer, poet, or even the sport for athletes. That it is done does not make it less of an over-link, it only makes it an error that is repeated across an entire class of articles. Years and dates used to be linked in this way at one point and its removal was argued against using this same logic. Walter Görlitz (talk) 23:59, 25 December 2020 (UTC)

Need for clarity on linking major American cities

Several years ago, we renamed articles on the major American cities to remove state names. Thus, Los Angeles, California became Los Angeles. Chicago, Illinois became Chicago. Houston, Texas became Houston. Philadelphia, Pennsylvania became Philadelphia. Boston Massachusetts became Boston. And so on. The rationale was that these cities are generally recognizable and do no require disambiguation.

Despite the change in naming, some users sought to bypass the new protocol by linking around the new names as follows: "Los Angeles|Los Angeles, California". User:Jweiss11 is an advocate of this work around, so I invite him to explain the basis for his position. Also, there was a 2016 discussion about this subject at Wikipedia talk:Manual of Style/Linking/Archive 18#What sort of linking here, but no clear direction was finalized.

It appears that MOS:GEO is the applicable guideline. It states: "Places should generally be referred to consistently by the same name as in the title of their article". Under this language, Los Angeles should be referred to/linked by the same name as in the title of its article, i.e., Los Angeles.

As things now stand, there remains significant variation which includes: (1) Los Angeles (which MOS:GEO seems to support); (2) "Los Angeles|Los Angeles, California" (which Jweiss11 advocates); (3) Los Angeles, California (triggering a redirect to Los Angeles); (4) Los Angeles, California (with separate blue links to city and state); and (5) Los Angeles, California (blue link to city, no link to state).

My view is that (1) is correct, though (5) might be ok if there was some legitimate reason to believe that the state ought to be included in a particular case. Wider input is sought here so that a consensus, one way or another, can be reached. Thanks for giving this a moment of your time. Cbl62 (talk) 00:01, 24 November 2020 (UTC)

• In case anyone's wondering, the origin of the no-state cities is WP:USPLACE, which provides Cities listed in the AP Stylebook as not requiring the state modifier in newspaper articles have their articles named City unless they are not the primary topic for that name. Buried in the footnotes is this list: Atlanta, Baltimore, Boston, Chicago, Cincinnati, Cleveland, Dallas, Denver, Detroit, Honolulu, Houston, Indianapolis, Las Vegas, Los Angeles, Miami, Milwaukee, Minneapolis, New Orleans, New York, Oklahoma City, Philadelphia, Phoenix, Pittsburgh, St. Louis, Salt Lake City, San Antonio, San Diego, San Francisco, Seattle, and Washington, D.C.—although Washington, D.C., does have a territorial qualifier and New York is naturally disambiguated.
  I'll just add that where GEO says cities should be referred to consistently by the same name as in the title of their article, I think that means that we write Rome and not Roma, because our article is titled that way -- I don't think it's the reason we say simply Rome and not Rome, Italy. (That's true, but I don't think the bit I just quoted is the reason.) EEng 22:58, 24 November 2020 (UTC)

  EEng, do we have any guidance regarding non-American cities? I'd imagine that if we're referring to Seattle without the state, we'd do the same for e.g. Montreal. {{u|Sdkb}} ^talk 02:38, 24 November 2020 (UTC)

  Articles certainly refer to (simply) London, Paris, Rome, Berlin, Moscow, Tokyo, etc., but I don't think there's a central list. Notice those are all national capitals, but surely not all national capitals go on that list (though having said that I went to see how the capital of Tonga is styled, and it's simply Nukuʻalofa -- not Nukuʻalofa, Tonga -- so I don't know). On the other hand, "surely" there are non-capitals outside the US that go on the list; if we feel readers don't need to be told that Cincinnati's in Iowa then surely we don't need to tell them Toronto's in Canada (much less Ontario -- or maybe we tell them it's in Ontario but not Canada? -- my head whirls). EEng 02:52, 24 November 2020 (UTC)

  Readers definitely don't need to be told that Cincinnati is in Iowa. {{u|Sdkb}} ^talk 03:56, 24 November 2020 (UTC)

  ;) EEng 22:46, 24 November 2020 (UTC)

I think there's a time and place for revisiting that list, which seems to have been drawn up around 1920. (With all due respect... Cincinnati? Really???) But for now, let's stick with Los Angeles as the canonical example. EEng 02:23, 24 November 2020 (UTC)

• I favor "1" or "3" in most cases, with "1" being more common usage. Other forms WILL be needed from time to time. I would NOT be in favor of making wholesale changes to existing pages, but if someone did do a page-wide change to one page or a small number of pages, and the results were consistent within each page, I wouldn't blindly revert them without starting a discussion. On the other hand, if someone decided to do mass-changes without a discussion first, that would be disruptive. davidwr/(talk)/(contribs) 00:19, 24 November 2020 (UTC)
• I prefer "1" though I want to recognise there will be instances in which it's appropriate to include the name of the state in the article. So, #1 is preferred in most instances, but #2 or #3 is fine when maintaining consistent prose is more important (for instance, in a list with cities that don't follow those rules and include the state), and #4 should probably be used in relevant infoboxes, for instance birth and death locations. SportingFlyer T·C 01:18, 24 November 2020 (UTC)
• Thanks for bringing this up. I've long been a little unsure just how much linking/detail is needed. I'm interested in hearing from Jweiss11 and others, but without delving too much into the prior discussions you linked, it does seem as though they are inappropriately going against MOS:GEO.

Also, at the risk of complicating things further, there are some related questions, such as whether to begin articles with Foo is a bar in Los Angeles, California, United States or just use the city/state or just the city. Other than not linking the United States, all of this seems like it could use some standardization. {{u|Sdkb}} ^talk 01:21, 24 November 2020 (UTC)

• 4 is never appropriate one next to the other per sea of blue (or somewhere in that vicinity. --Izno (talk) 01:39, 24 November 2020 (UTC)

  Well, there is the comma, so technically it's not an uninterrupted sea, but that is something to weigh. Personally, I view SEAOFBLUE as less sacrosanct than making sure useful links are present when otherwise warranted. {{u|Sdkb}} ^talk 02:16, 24 November 2020 (UTC)

  A sea with toothpick floating in it is still a sea. EEng 22:46, 24 November 2020 (UTC)

• To clarify, the disagreement between Cbl62 and me concerns usage in infoboxes and tables only, particularly place of birth/death field infoboxes, as seen at Greg Lens. For the sake of consistency across articles and between parallel elements in the same article, I advocate variation #2 above ("Los Angeles, California" or "San Antonio, Texas") because it renders a style consistent with "Marshall, Minnesota". Variation #4 above should be thrown out because it's a clear case of overlinking.

Also pinging Rikster2 and Nikkimaria, who have been involved in recent discussions about whether "U.S." should follow city and state in place of birth/death fields in infoboxes; see User talk:Nikkimaria#Adding “US” after city and state on baseball, basketball, college coach articles. Jweiss11 (talk) 01:45, 24 November 2020 (UTC)

For tables that definitely makes sense for internal WP: Consistency within articles. {{u|Sdkb}} ^talk 02:16, 24 November 2020 (UTC)

• I favor the ability to display City, State in tables and infoboxes where consistency is desired. This mirrors what most sports use in rosters and biography articles in the “real world.” I don’t think use of American (or Canadian) City ONLY should be required in all cases. As for linking, I am good with 2, 3 or 5. In sport articles that I work on the correct “unit” (usually birthplace, etc) is the city/state because the question the reader is trying to “answer” is “where specifically is this person from?” Rikster2 (talk) 03:25, 24 November 2020 (UTC)

• Option (1) makes sense to me as a default, but I can easily envision situations where (2) or (5) may be appropriate, such as in lists or tables alongside other US cities, and in situations where the state might warrant a mention anyway. I agree (4) is inappropriate (and slightly WP:EGGy given the many US cities which include the state name in their links), and (3) seems a poorer version of (2). CMD (talk) 04:46, 24 November 2020 (UTC)
• <city>, <state> is OK for consistency in tables and infoboxes. For link style, I prefer (3) to discourage drive-by editors creating a MOS:SEAOFBLUE by linking the state. (2) goes against MOS:NOPIPE, but honestly, many long-time editors don't follow it.—Bagumba (talk) 08:54, 24 November 2020 (UTC)
• Do our readers a favour: if it's clear from common knowledge or article context, don't slavishly insert the state. Use the state to identify only less-well-known cities. Tony (talk) 11:38, 24 November 2020 (UTC)

  I agree with Tony. Most of these references are blue linked, at least at first encounter, so anyone who wants to bone up on their geography is only ever one click away from enlightenment. For Badiddlyboing, for sure add in the state Odawidaho to save people the trouble of looking. But for big towns and cities, why bother? MapReader (talk) 15:05, 24 November 2020 (UTC)

  There is one school of thought that holds that we should not use Wikilinks as explanations. I'm not sure what I think of that argument, but wanted to note its existence. {{u|Sdkb}} ^talk 19:17, 24 November 2020 (UTC)

• I agree it is good to display the state also in many cases for consistency. It also avoids deciding when "it is clear from common knowledge" that the state is unnecessary - we do have a worldwide readership and shouldn't assume the reader has a typical American knowledge of geography, whatever that is. Of course, in prose you wouldn't say Los Angeles, California beyond perhaps the first occurrence. When the state is added, I would use (3), (2) is acceptable as well but I don't see a reason to "complicate" the source unnecessarily to avoid a redirect. Never (4). MB 16:22, 24 November 2020 (UTC)
• Per the list EEng has provided, we normally should not list the state for those cities unless it is in a place where the omission clearly causes consistency problems, such as in a table, list, or infobox where we have had to link cities by "city, state" names. When such it needed, it should be linked as one single piped linked eg either [[Chicago|Chicago, Illinois]] or [[Chicago, Illinois]], but definitely not making the state a separate link, that's just a waste of a link that's typically not relevant since the city is the relevant link, not the state. --Masem (t) 00:16, 25 November 2020 (UTC)
• Mostly use option 1 (Los Angeles) for major cities familiar even to non-Americans, and 3 (Los Angeles, California) when for some reason there seems to be a need to be more specific. E.g., if the article is about Ontario, use London, England, since London, Ontario, is an important place in the context. But none of this is news or a question. This is what we've been doing for something like 15 years. We do tend to include more specifics infoboxes. Dunno why. Don't really care. Tables, probably, just for consistency, unless every city in the table is a "world renowned" one like Chicago, Tokyo, and Munich. Lists, well, that just far too over-generalizing. There are a zillion sorts of lists. Generally, I would say not in lists, because a list can at any time be converted to regular prose (we actively encourage this a lot), but running text that doesn't work well in paragraph form can be list-ized. That is, the prose/list distinction is like the DVD/Blu-ray distinction; they're both still the same movie [or whatever] content. The prose-and-lists vs. tables distinction is sharper, like an original film vs. remake distinction.  — SMcCandlish ☏ ¢ 😼  11:38, 4 December 2020 (UTC)

Consensus?

@Cbl62: seems like we have wide support here for including the state in infoboxes, tables, and lists. Do you agree? Jweiss11 (talk) 02:52, 28 November 2020 (UTC)

Consensus is clear as to tables and lists where the state names are consistently recited. That makes sense. I'm still unclear whether there is a consensus for using state names in narrative text and infoboxes. I tend to agree with Tony and MapReader's view that the state should be listed only if it's a lesser-known city. As to infoboxes in particular, they are not like a lengthy chart or list where inclusion of state name is needed for consistency's sake. There's also a variety of views as to "how" the state name should be displayed. The discussion has been helpful. Let's let it run its course and see where it ends up. Cbl62 (talk) 21:26, 29 November 2020 (UTC)

Cbl62, there's support here from a number of editors (Sdkb, SportingFlyer, Bagumba, Masem, Rikster2, and me) for including the state names in infoboxes/tables/lists. It's unclear whether Tony and MapReader were offering an opinion about prose, non-prose, or both. Perhaps they can clarify? @Nikkimaria:, can you weigh in here? How would you format the place of death field in the infobox at Greg Lens? Jweiss11 (talk) 21:24, 1 December 2020 (UTC)

With regards to linking, I personally would link only San Antonio and not other components of the place. Nikkimaria (talk) 21:33, 1 December 2020 (UTC)

@Nikkimaria: But you would include Texas and U.S., i.e. "San Antonio, Texas, U.S."? Jweiss11 (talk) 21:37, 1 December 2020 (UTC)

Correct. Nikkimaria (talk) 21:38, 1 December 2020 (UTC)

@Contributor321: Please see this discussion per your reversion of my edit at California State University, Los Angeles and weigh in if you like. Thanks, Jweiss11 (talk) 19:28, 2 December 2020 (UTC)

Thanks for letting me know about this discussion. I think it's important to include the state name in the prose. Let's keep in mind the international readership who might not know that San Antonio is in Texas. As to how to display that information, one way is San Antonio, Texas, which links only to the city; an alternative is to put brackets around each of San Antonio and Texas, which links the city and links the state. The text looks the same in either case, but the latter lets a reader easily go to the state page if interested, while the former leaves out that option. I prefer the latter simply because it gives the reader more options. Just my 2 cents. Contributor321 (talk) 01:34, 3 December 2020 (UTC)

Because the latter looks the same as the former, users are not able to distinguish visually that there are two links versus just one - see WP:SEAOFBLUE. Nikkimaria (talk) 02:16, 3 December 2020 (UTC)

Putting your cursor on it readily distinguishes whether it's one link or two. Contributor321 (talk) 03:02, 3 December 2020 (UTC)

Still a problem - see also MOS:NOHOVER. Nikkimaria (talk) 04:04, 3 December 2020 (UTC)

Comment I recommend that we first resolve what to display and then we can discuss how to link it.—Bagumba (talk) 01:45, 3 December 2020 (UTC)

Bagumba, that might be needed. And we're probably going to need an RfC. I checked back in on this hoping to see some sort of clear resolution, but it appears we're not there yet. {{u|Sdkb}} ^talk 09:51, 21 December 2020 (UTC)

Sdkb: I think the general consensus is that including the state is OK for consistency in tables and infoboxes. If one is interested, a new discussion on how to link a city like Los Angeles, whose page title does not include a state, when the displayed text is "Los Angeles, California". (As an aside, I personally think MOS:SEAOFBLUE or similar should have an example that a page named <city>, <state> (e.g. Austin, Texas) should be a single link, and not a separate link to the city and state).—Bagumba (talk) 10:46, 21 December 2020 (UTC)

Yeah, we're agreed for the tables, but that's not really the main question. I want to know, when writing a lead X is a Y in Maplehurst, Wisconsin, United States, do I include United States and what do I link to where? Unless we have enough agreement to add something to the linking guideline to answer that question, we need more discussion. {{u|Sdkb}} ^talk 10:59, 21 December 2020 (UTC)

I think the country is relevant as one cannot assume the reader is famiiar with the location. I'd link to Maplehurst, Wisconsin only. A separate link to US, if not MOS:SEAOFBLUE, is MOS:OVERLINK. Depending on the domain, "X is an American Y in Maplehurst, Wisconsin" is also an option.—Bagumba (talk) 11:07, 21 December 2020 (UTC)

Bagumba, this is the "Consensus?" subsection, so I'm less interested in gathering opinions than figuring out how we should move forward based on the opinions in the survey above. {{u|Sdkb}} ^talk 11:19, 21 December 2020 (UTC)

No problem. However, your "X is Y" example was never really the focus of this discussion.—Bagumba (talk) 11:23, 21 December 2020 (UTC)

Demurrer: "seems like we have wide support here for including the state in infoboxes, tables, and lists. Do you agree?" No. There's wide support for doing it in infoboxes and tables, not lists. The support for that is weaker, and most of it takes no account of the fact that "lists" is an over-generalization; there are many kinds of lists. Nor does it account for list and paragraph content being effectively freely convertible from one form to the other, which converting from either to a table is less straightforward (aside from very bare and formulaic lists that are basically already tables misformatted as lists).  — SMcCandlish ☏ ¢ 😼  01:43, 1 January 2021 (UTC)

@Eagles247: I noticed your edit today at John Jenkins (American football coach). You want want to weigh in here, as this discussion is related to the issue there. Thanks, Jweiss11 (talk) 01:27, 3 January 2021 (UTC)

@Jweiss11: Thanks for the ping, this was an interesting discussion to read. I agree this should go to a full RfC. I recognize many of the editors who voiced an opinion here from the North American sports WikiProjects, so a wider range of viewpoints is needed, along with narrower focus and more specific examples of where and when these links are to be used. My edit was removing "U.S" from an infobox, but I don't believe that is the focal point of this discussion. (FWIW, I usually link option 3, [[Los Angeles, California]], since I can never remember which cities do not include their states in their article names.) Eagles 24/7 (C) 02:04, 3 January 2021 (UTC)

Semi-protected edit request on 7 January 2021

Please amend the MOS:DRAFTNOLINK guideline in such a way that it must permit the linking of draft articles that have been submitted for the review. You can still impose restriction on linking the sandbox articles or the draft articles that have not been submitted for the review. You can read the benefits and arguments in favor of this proposed edit here and more detailed benefits and concerns here. I was redirected to MOS:LINKING guideline/talkpage here. Please note that the similar concerns have been already documented by numerous other editors Wikipedia:Why Wikipedia is not so great, my proposed edit mitigates some of those concerns. Thank you. 58.182.176.169 (talk) 15:07, 7 January 2021 (UTC)

 Not done for now: please establish a consensus for this alteration before using the {{edit semi-protected}} template.—Bagumba (talk) 15:31, 7 January 2021 (UTC)

I was also refusing your request, but ended up in an edit conflict. If you were to make this a proposal for discussion, my response would be:-

Strongly oppose: Draft articles are of unknown quality/truthfulness/notability/neutrality and linking to them has been repeatedly used as a way of adding PoV content, trivia and blatant vandalism to Wikipedia. As a review can take 3 months, or longer. this proposal would allow unvetted "information" to be linked to directly from our articles for a very considerable period of time. Very few readers will notice the "Draft" in the title of the linked article, and of those that do, many will not understand the implications of it. - Arjayay (talk) 15:35, 7 January 2021 (UTC)

Strongly oppose: For all the reasons cogently and clearly stated by User:Arjayay. --Coolcaesar (talk) 15:42, 7 January 2021 (UTC)

Oppose Linking from articles to drafts is a bad idea, whether or not those drafts have been submitted for review. Anybody can submit their own not-even-half-baked draft for review; the fact of being submitted is not itself a marker of quality. XOR'easter (talk) 15:51, 7 January 2021 (UTC)

Strong oppose Draftspace is called "draft" for a reason - it's the place to store potential articles of unknown quality that aren't ready for prime time. This would essentially just make draftspace into a mainspace alternative, which would defeat the purpose entirely. PohranicniStraze (talk) 18:07, 7 January 2021 (UTC)

Oppose per same reasons as above. Drafts are absolutely not ready for mainspace by default and anyone can submit one for review. Readers should not be taken to these pages. A draft that can be linked to is a draft than can be published. I see no actual argument why these links help readers. —  HELLKNOWZ   ▎TALK 18:50, 7 January 2021 (UTC)

Oppose as it would be the perfect recipe for further spam/UPE storms. --MarioGom (talk) 11:00, 8 January 2021 (UTC)

• Strongly oppose Not that it's going to happen based on the consensus here, but we would never link to articles in draft space. Until an article has been accepted into main space, it should not be linked to from main space. Walter Görlitz (talk) 02:01, 11 January 2021 (UTC)

Linking to projectspace in articles

Is it acceptable to link to the Wikipedia namespace in an article? This guideline says: Do not create links to user, WikiProject, essay or draft pages in articles, except in articles about Wikipedia itself (see Wikipedia:Manual of Style/Self-references to avoid). - nothing about projectspace. Wikipedia:Linking dos and don'ts says: Don't link to user, project, draft, or talk pages in articles, but that's an info page. Does an actual PAG say that?

Context: Nathan_Larson_(politician)#Political_views_and_online_advocacy links to Wikipedia:Child protection. ProcrastinatingReader (talk) 21:15, 2 January 2021 (UTC)

• Remove the link. Project space is not part of the encyclopedia proper and should never be linked from within an article any more than an external link should be. If it's really relevant enough, put it in a reference or in the external links section. I regularly remove non-article links from pages about Wikipedia, but I feel like they get added back faster than I can remove them, so I wish we'd be a bit firmer with our rules. Maybe I'll make an editnotice for pages like Wikipedia. {{u|Sdkb}} ^talk 23:24, 2 January 2021 (UTC)
  • Well, if it is truly a violation of guidelines as WP:LINKDD would have you think, maybe we should add it to this page? I know that per Wikipedia_talk:WikiProject_Templates#Linking_to_templates_from_tables_and_hatnotes we don't accept it for templates, but I dunno which PAG explicitly prohibits this. ProcrastinatingReader (talk) 23:30, 2 January 2021 (UTC)

Moved years ago from some link guideline to WP:ADMINP as most saw it as obvious and no need to expand on.--Moxy 🍁 00:36, 3 January 2021 (UTC)

• I just created Template:Wikipedia article editnotice and added it to Wikipedia to test the waters. The language/links could use improvement, so feel free to refine those. If there are no complaints, anyone able to do so is free to add it to other pages within WP:WikiProject Wikipedia's scope. {{u|Sdkb}} ^talk 02:32, 3 January 2021 (UTC)

You just added a notice to a page that on purpose links to project namespace.--Moxy 🍁 02:41, 3 January 2021 (UTC)

Looking at Wikipedia, it has a link to WP:Autoconfirmed and a few others which should be cleaned up. It's hardly a model page; we shouldn't be looking to it for policy inspiration or anything. {{u|Sdkb}} ^talk 02:47, 3 January 2021 (UTC)

Not sure what you mean by cleanup. ..but I am sure we can all agree it happens to be one of the pages that linkage is warranted to expand on our readers knowledge on Wikipedia functionary processes.--Moxy 🍁 02:57, 3 January 2021 (UTC)

Moxy, that seems contrary to what our rules currently say. Everywhere someone might be tempted to link to a project page would be a place it'd be relevant, including at Nathan Larson as above, but they're still intrinsically inappropriate. {{u|Sdkb}} ^talk 03:33, 3 January 2021 (UTC)

Agree with most of what your saying and is why I wrote and linked WP:ADMINP. As the vast majority of articles have zero need to link to project administrative pages...but the one page you added the template to is an exception to the norm for obvious reasons. We could remove the one non informative link 'how to policies" again for obvious reasons. Not sure your usage of your new template is appropriate on a page that deliberately links to more information about the administrative structure of Wikipedia in that namespace. What is best for our readers on that page? Leave them in the dark because of a new template or continue to link to how are administrative function works, how policies work, how user levels work etc? We do have a small subset of pages that link for purely educational knowledge value Notability in the English Wikipedia ...we need to make sure people are doing right by our readers before edit notice spam and appropriate link removal.--Moxy 🍁 03:52, 3 January 2021 (UTC)

It's not worth the effort to parse Moxy's word salad, but for anyone else having doubts about the importance of this: any journalist who's ever had to report on a topic involving their publication knows how important it is to separate the business side of things (in our case, project space) from the editorial side of things (in our case, articlespace). Wikipedia has a much weaker conception of that principle, and it's at our peril. Our fundamental goals differ between articlespace and project space, and by linking to project space from an article body, we blur that boundary and let a bit of self-servingness seep into articlespace, just the same as if we put an external link in the middle of an article to an organization that promotes itself on its website (as all organizations do). {{u|Sdkb}} ^talk 04:37, 3 January 2021 (UTC)

As long as it does not lead to more negative experience for our readers your feel to insult me.--Moxy 🍁 04:56, 3 January 2021 (UTC)

• Looking closer at the guideline, MOS:DRAFTNOLINK addresses this, but I don't fully agree with it: In articles, do not link to pages outside the article namespace, except in articles about Wikipedia itself (and even in that case with care – see Wikipedia:Manual of Style/Self-references to avoid). Per my arguments above, I think this should be modified to take a stronger stance, and should modify that line, as well as WP:WAWI which gives further instructions. Cross-namespace links are one of those things that seem fine and harmless until you realize what they actually represent; I might have to create an RfC to seek broader consensus. {{u|Sdkb}} ^talk 11:25, 11 January 2021 (UTC)

Links to common terms in lists?

Hello everyone! Could you please help me out in a recent disagreement: Does the the recommendation not to link common items also apply to lists, where the problem of identifying specific links is much less an issued due to their structured nature? Should I go into detail why I think those links are helpful and should be kept in my case?
I have noticed that e.g. in the example used by WP:MOSLIST, List of vegetable oils terms like palm oil, cooking oil, plant and seed are linked. Thank you very much for your input! Daranios (talk) 08:30, 1 March 2021 (UTC)

Giving details after all: In the list in question, List of Advanced Dungeons & Dragons 2nd edition monsters, links to common terms like bear may only rarely be used for navigation, but on the other hand in my opinion they do not make reading more difficult due to the structured nature of the list. But they do provide a service: They show at a glance which of the subjects the list treats are notable as real-world-subjects (or notable as part of (A)D&D), as opposed to those which are original creations that did not achieve fame. In effect, this gives a summary of a great number of "In popular culture" sections of real-world subjects. I don't see the benefit of taking that away in this case. What does anyone else think? — Preceding unsigned comment added by Daranios (talk • contribs) 08:35, 18 March 2021 (UTC)

There are several related discussions in the recent archives you may want to read. I support somewhat more frequent linking from lists/tables than I would from prose, including more frequent WP:LINKDUP, but there's a limit. Also, redlinking is never okay if the subject is not notable. {{u|Sdkb}} ^talk 17:27, 18 March 2021 (UTC)

Thanks a lot, Sdkb! I have to confess I did not find a directly fitting discussion. Would you have specific ones in mind you could perhaps point me to? Daranios (talk) 11:37, 19 March 2021 (UTC)

Daranios, here's one. {{u|Sdkb}} ^talk 19:26, 19 March 2021 (UTC)

• Wall-to-wall linking of list items precludes one of the most important editorial contributions on linking: to funnel readers toward what we decide is a target more likely to be useful. Tony (talk) 23:40, 20 March 2021 (UTC)

Thank you both! It seems to me that in the suggested older discussion, opinions were divided, meaning there is no overall consensus for linking common terms in lists or not. So we should apply the editorial decision locally. And then I am back at my argument: Aside from navigation, which will probably only be used rarely, blue-linking the common creatures together with the less common ones in my opinion serves a purpose: To easily see which subjects within List of Advanced Dungeons & Dragons 2nd edition monsters have reached beyond the game itself, and which are original creations that did not recieve (significant) secondary coverage. So my idea would be to go back and ask that question once more locally, as was suggested by one editor in said discussion. If noone shows up to give an opinion there, things can stay where they have been put by User:Tony1 according to the non-list-specific guidance of WP:MOS. If the links are found useful in that specific case, they could be reinstated. Would this be reasonable? Daranios (talk) 11:51, 22 March 2021 (UTC)

Perhaps I am too bold.

Hello! I made a change here that I think I should probably have raised here before making. I'm sorry for that. I moved the words around because this was information I was looking for, and didn't see immediately. I feel like those are the primary examples that editors will be looking to confirm with that section, and thought it might be wise to front-load them. If anyone disagrees, then I totally understand, and they are free to revert. I did not change the words, nor (I think) the meaning; all I did was place the exceptions first, and then the general rule.

If anyone at all disagrees, for any reason, you can revert it, or ping me and I'll do so the moment I see it. I thought this was a non-substantive, quality-of-life change. — ImaginesTigers (talk) 21:09, 15 January 2021 (UTC)

@ImaginesTigers: I understand the sentiment behind this, but it's not a good idea in these documents. The MoS rules only have very limited exceptions, and it's important to indicate what the default is before allowing for enumerated variances, or people are apt to take the variances given as just non exhaustive examples, decide they can the same with whatever is on their mind, and never even get to the default rule because they've already projected the answer they wanted to hear. If you go through all the policy-and-guideline material, you'll see the same order of default-then-exceptions used consistently.  — SMcCandlish ☏ ¢ 😼  17:27, 6 April 2021 (UTC)

Linking to other language Wikipedias in articles closely related

Generally the MOS here states that links to other language Wikipedias should only appear in the External links section or the Languages sidebar, the exceptions being Wiktionary and Wikisource (in some circumstances). However, many articles clearly relate to another language, e.g. Spain has a strong relationship to the Spanish language and indeed the lead sentence gives its Spanish name "Spanish: España". I think in this context linking España to the Spanish Wikipedia would be an excellent thing for speakers or students of Spanish at all levels. Even for non-Spanish speakers, to ve able to view the images and get a flavor of the language is useful. This is easily accomplished: Spanish: España. Please note that I'm not advocating for third party languages such as a link to the Japanese WP for Spain, or for other inline links. What does everyone think? Cheers, Facts707 (talk) 03:00, 17 April 2021 (UTC)

No. That's what the language links in the side bar are for. -- Michael Bednarek (talk) 03:47, 19 April 2021 (UTC)

Linking non-major countries

At the main page talk page, I suggested that we wikilink East Timor in an item, citing MOS:OVERLINK's wording that specifies only that major countries with which most readers will be at least somewhat familiar should not be linked. My view is that, for a tiny country like East Timor, where most readers wouldn't be able to locate it on a map and the population is 10x less than some cities we'd link without second thought, I think declining to wikilink is just following rules because they're rules when the actual rules don't actually say what they're imagined to. What do others think? {{u|Sdkb}} ^talk 21:28, 5 April 2021 (UTC)

Not sure what the normal convention is on the main page....but best to ask for more input there. The guidelines on this is not stringent either way and is left to interpretation... that is normally discussed at the article level.--Moxy- 23:21, 5 April 2021 (UTC)

It's not just about the country. You should also ask if there are related links nearby, for instance a city, county, state or province. Also, is the subject closely linked to the term? Is the subject's interaction with the location vital to understanding either the country or the subject? This isn't a simple yes or no based only on the nation itself. Walter Görlitz (talk) 04:27, 8 April 2021 (UTC)

I've been having a discussion with a user about whether Iraq should be linked in the body of the Jurassic article. In my opinion, Iraq is sufficiently well-known that it probably should never be linked in the body text. Can I have a second opinion? Hemiauchenia (talk) 20:27, 8 April 2021 (UTC)

For the Main Page, I've always advocated for having the blurbs link only to the main target article. Everything else should be easily findable from there.--Khajidha (talk) 15:53, 9 April 2021 (UTC)

Oh, maybe link it, but I'm happy not to. It would help me to judge if I could locate the actual item. Tony (talk) 08:00, 10 April 2021 (UTC)

I would support linking to non-major countries. I know where East Timor is because I live practically next door to it. But I wouldn't expect the average European or North American to know where it is - just like I don't know where most of the Caribbean and smaller African countries are if I was only told their name. Think of what the average reader would know.  Stepho  talk  11:43, 10 April 2021 (UTC)

I would most certainly link East Timor - surely not "major" on any definition. Iraq is on the borderline imo, if only because of the publicity from recent history. My personal cut-off is around the level of Portugal, Argentina, & Hungary, all of which I'd link. Only Brazil and perhaps Mexico in Latin America don't need it. It's not just "the average European or North American", as we have huge numbers of readers from India & many from Africa etc. There is a case for defining "major" as a member of the G20 group and linking everybody else, though in fact I'd link Argentina and Indonesia. But it's about that sort of level. Johnbod (talk) 12:57, 10 April 2021 (UTC)

I think there should be some kind of specification in the guidelines. "The names of subjects with which most readers will be at least somewhat familiar. This generally includes major examples of: countries (e.g., Japan/Japanese, Brazil/Brazilian)" is too simple and vague. There have been some cases of absurd removal of links to Romania, such as in Bucharest. It does not make sense that the capital of the country does not have a link to it. Even Washington, D.C., the capital of probably the most famous and well-known country today, and Paris, probably the most famous capital, have links to their respective countries. And Bucharest is just one example of the many pages related to Romania that had links to Romania removed some time ago. Other examples are Romanians (absolute non-sense), History of Romania, Traian Băsescu (former president of Romania), Politics of Romania, Parliament of Romania... Removing links to Romania in these cases is obviously wrong, but since the guidelines do not give any specification (other than the word "major"), these things happen. That's why I would like to see it changed to something like "The names of subjects with which most readers will be at least somewhat familiar, unless the article in question is of important relevance to the subjects. This generally includes major examples of:" or something like that. Super Ψ Dro 11:55, 14 April 2021 (UTC)

It is intended to be vague, because a reader of an article about a subject in England might not link the city of Birmingham, but that would be linked in a topic that is primarily American. Other topics like United States or Berlin should not be linked in most articles except those that are closely linked to the topics. And if the article already had a link to Bucharest, I would not also link Romania, but in an article on geographic topic, like the city of Bucharest or one of its neighbours, it should be linked. Your example is, I believe, REPEATLINKs. Romanians should not be linked as a substitute for linking to the nation in an article about a citizen of the nation, and the nation should not be linked in the lede of subject's article. Walter Görlitz (talk) 06:10, 15 April 2021 (UTC)

Super Dromaeosaurus, MOS:CONTEXTLINK speaks to that somewhat. {{u|Sdkb}} ^talk 20:54, 21 April 2021 (UTC)

Overlinking elsewhere?

Just wanting to check and make sure this is right... MOS:OVERLINK would also apply to other places such as navigation templates, correct? For example, I don't how linking every instance of every show name here is beneficial whatsoever, which is why I then removed the duplicate linkings. Would that be correct, or is linking all of them fine for some reason? Magitroopa (talk) 08:17, 29 April 2021 (UTC)

The relevant guideline for that is actually MOS:REPEATLINK: Duplicate linking in stand-alone and embedded lists is permissible if it significantly aids the reader. This is most often the case when the list is presenting information that could just as aptly be formatted in a table, and is expected to be parsed for particular bits of data, not read from top to bottom. As a naxbox is arguably not read from beginning to end, repeating is probably helpful.—Bagumba (talk) 08:24, 29 April 2021 (UTC)

It's been changed to this now which might actually be much better. This same user created a few other KCA navigation templates in the past hour or so, but I would think what you mentioned would moreso apply to the other ones they made where there was a repeat winner for a category about 1-3 times (or so)- However, I don't think it would 'significantly aid the reader' linking the same show 18 times (most of them in a row), as was done in the original version of the Favorite Cartoon navigation template. Magitroopa (talk) 08:41, 29 April 2021 (UTC)

Sometime WP:IAR applies. It's up to consensus.—Bagumba (talk) 08:48, 29 April 2021 (UTC)

Consecutive links

Shouldn't consecutive links be avoided? e.g.: "[[wiki]] [[link]]" should be avoided as it misleads reader in to thinking that it is a single link?- Vis M (talk) 08:00, 26 May 2021 (UTC)

Vis M, the relevant guideline is MOS:SEAOFBLUE.—Bagumba (talk) 10:45, 26 May 2021 (UTC)

Ok, thanks!- Vis M (talk) 10:58, 26 May 2021 (UTC)

Languages

Per the guidance in MOS:OVERLINKING that says terms should be linked if they "help someone understand the article you are linking from", I often unlink major languages, especially English, which I would expect most readers to understand sufficiently such that they would not require further help to understand an article. Uanfala disagrees, leaving this edit summary: for "English": we avoid arbitrarily unlinking items in lists just because the reader will know what they mean . I have never seen any guidance that says if one term in a list is linked, all must be linked. Isn't English one of the most overlinked terms here? MB 18:55, 21 May 2021 (UTC)

• From my travels, I have seen precious few occasions where English warrants linking. My script sees to it that the term hets unlinked. -- Ohc ^¡digame! 19:36, 21 May 2021 (UTC)
  • What script is that, Ohconfucius? – Uanfala (talk) 21:07, 21 May 2021 (UTC)

• User:Ohconfucius/script/Common Terms. Take it for a spin! -- Ohc ^¡digame! 07:45, 22 May 2021 (UTC)

• There's nothing specific about languages that may warrant treating them differently from other topics. Yes, in most cases there's no need to link English just like there is no need to link France, but there are also a fair number of cases where linking them might be appropriate – it's all down to the context. For example, as stated in MOS:OVERLINK itself, linking will not be inappropriate if the term is particularly relevant to the context in the article. That's why Paris links in its first sentence to France, and English grammar starts off with a link to English language. Another case concerns standalone lists (as found inside infoboxes among other places). If all the other items in the list are linked, and there's one that's unlinked, then the result is unwelcome visual emphasis on the unlinked item (which leads to the sort of reader distraction that is one of the motivating factors for the guidelines against overlinking).
  It's also worth pointing out that sometimes a generic link can be replaced with something more specific. For example, if the article about the Indian territory of Ladakh mentions English as having official status there, then instead of removing the link, it may be better to pipe it to the closest thing we have to an article about the use of English in the country. – Uanfala (talk) 21:07, 21 May 2021 (UTC)

  I don't think "unwelcome visual emphasis on the unlinked item" has been commonly cited anywhere. I don't agree it's a justification to link common terms (overlinking) to avoid the distraction normally caused by overlinking. Regular text is the least distracting. I find infoboxes that are mostly blue to be highly distracting and easier to parse when there are fewer links (and flag icons, etc). Piping to the "closest thing we have to an article" is another ill-conceived justification for an unneeded blue link. MOS:EASTEREGGs are to be used sparingly. Saying English doesn't help because most readers will assume it goes to English and will never click on it since they know what English is. MB 00:51, 22 May 2021 (UTC)

• No, do not link "English" or "English language" unless there's a very specific reason—and even then it would be better to link to a more-specific offspring article or section. It's OK in Uanfala's example in English grammar, though. Tony (talk) 04:39, 22 May 2021 (UTC)
• I just unlinked English language in an the infobox of an article on American TV series. There is no need to link it there. Agree that there are some logical links such as those that Uanfala mentioned, but they are few and far between. I don't know which table was unlinked and then reverted, but I can think of none where linking would be required. Walter Görlitz (talk) 15:45, 25 May 2021 (UTC)
• Linking a piped Indian English for a list of languages for a region is not needed. WP:OVERLINK says: Unless a term is particularly relevant to the context in the article ... Indian English is slightly more relevant to Ladakh than plain English, but still not "particularly relevant" (presumably why it was piped to begin with).—Bagumba (talk) 11:03, 26 May 2021 (UTC)
  • It is relevant though. The place of English in Ladakh is similar to the rest of India, and in the absence of an actual article about the use of English in India, the section Indian English#Status will give readers some answers to the questions they may have when they see the mention of English in the infobox. – Uanfala (talk) 14:55, 26 May 2021 (UTC)
    • The disagreement is whether it is "particularly relevant", not simply "relevant".—Bagumba (talk) 01:36, 27 May 2021 (UTC)
      • I'm not going to argue about the exact phrasing we're going to use. My point is that the article is relevant enough to be linked, and as far as I can see, it's no less relevant than any other one linked on the page. – Uanfala (talk) 14:04, 27 May 2021 (UTC)
        • If the article on India wants to link to Indian English, that is particularly relevant. Doing so in a region, state, city, town or villiage in India is not, in my opinion, particularly relevant or even relevant enough. The only possible reason I could see linking an English-language variant at this level is if it were different than the levels above, and it was notable for this very reason. Walter Görlitz (talk) 05:51, 28 May 2021 (UTC)

Discussion at Talk:Donald Trump § RFC (lead sentence link)

 You are invited to join the discussion at Talk:Donald Trump § RFC (lead sentence link). {{u|Sdkb}} ^talk 06:33, 8 June 2021 (UTC)

Question regarding duplicate links

Does it fall afoul of MOS:DL (duplicate linking) if there is a link right after using the main template for subheadings? Which version is correct for MOS?:

Ludonarrative dissonance (Version 1)

Main article: Ludonarrative dissonance

In a 2007 blog post, Hocking coined the term ludonarrative dissonance as a term for ...

Ludonarrative dissonance (Version 2)

Main article: Ludonarrative dissonance

In a 2007 blog post, Hocking coined the term ludonarrative dissonance as a term for ...

Thanks.  Spy-cicle💥  ^Talk? 13:29, 29 March 2021 (UTC)

• I don't understand. So the link within an article goes to the article? Tony (talk) 13:43, 29 March 2021 (UTC)

  @Tony1: To clarify this example is dervived from a subsection of the Clint Hocking article seen here, not from the ludonarrative dissonance article itself.  Spy-cicle💥  ^Talk? 15:02, 29 March 2021 (UTC)

  No, the question is whether per SEAOFBLUE you can have both a link in main and nearby in the prose-proper, or solely in the main. --Izno (talk) 15:04, 29 March 2021 (UTC)

• I think it does, in that the inline link is unnecessary because the hatnote has already introduced the reader to the term, and in the most visible way possible. I always remove the link in such instances; it's one or the other, imo: include the hatnote, or link the term in the prose. JG66 (talk) 13:51, 29 March 2021 (UTC)
• Version 1 is preferable as it guarantees the link is present for all users. Hatnotes may be removed by display: none or by external users seeing them as unnecessary or any of a variety of other reasons. I always add it if it is not present (and I feel inclined that day).... (For example, mobile used to remove hatnotes. Since fixed of course.) --Izno (talk) 15:04, 29 March 2021 (UTC)
• Per JG66: one is quite enough. Tony (talk) 13:11, 30 March 2021 (UTC)
• The link in the actual prose should not be omitted, because hatnotes are unprintworthy self-refs. Even aside from WP:User CSS hiding them (as Izno mentioned), hatnotes will simply disappear for various WP:REUSE purposes, along with other navigational claptrap, leaving just the text of the article's per se content. This is by design. Also, what hatnote, if any, is present is entirely dependent on extraneous factors like article splits, mergers, deletions, and so on. The link in that hatnote could change or disappear at any time (and might even be cleaned up by a bot, so be less likely to be noticed by watchlisters).  — SMcCandlish ☏ ¢ 😼  17:19, 6 April 2021 (UTC)

  It seems 2 editors think 'Yes' it is duplicate linking (essentially version 2) and 2 editors think it is not duplicate linking (essentially version 1), I should probably start an RfC perhaps?  Spy-cicle💥  ^Talk? 15:11, 9 April 2021 (UTC)

An RFC is a great idea. I generally link both, but I try not to have the inline link too close to the hatnote. Generally I lean a bit towards more relinking, especially in photo captions because readers are often more intrigued by photos than a long article. But never linking twice in the text of one paragraph or section. Cheers, Facts707 (talk) 02:21, 17 April 2021 (UTC)

• I've also been a bit stymied by this question. I raised the issue on the talk page for Oscar Wilde several weeks ago but haven't got any bites (a bit surprising given it's a GA-class article on a a level-4 vital topic). In that case, I was weighing version 2 above with a version 3: in-text wikilinks and no hatnote. (Though that article is a bit of an unusual case, as it has some sections with {{main}} hatnotes that point to 3+ different articles.) Colin M (talk) 19:05, 16 June 2021 (UTC)

Linking part of a word

Is it possible to link only one part of a compound word, if that's only part of it is likely to be difficult for the reader to understand? HLHJ (talk) 23:20, 24 May 2021 (UTC)

Example, please. EEng 23:48, 24 May 2021 (UTC)

The answer is "yes", but exact​ly how best to do it might depend on your context. DMacks (talk) 17:32, 20 June 2021 (UTC)

Links with 's

I've noticed on a lot of articles, when an 's is added to a linked topic, the 's is not actually included in the link. Sorry if that doesn't make sense, see the example from Jordan Omogbehin below.

"he was presented as a surprise member of Akira Tozawa's ninja faction"

Should it be Akira Tozawa's or Akira Tozawa's? 03Heat (talk) 12:46, 1 February 2021 (UTC)

I believe it would be the former of the two, but I am not certain.  Spy-cicle💥  ^Talk? 13:19, 29 March 2021 (UTC)

Definitely "Akira Tozawa's", as far as I'm concerned. JG66 (talk) 13:55, 29 March 2021 (UTC)

Respectfully disagree with the responses above, for the following reason: Towaza's in the above example is an indivisible word, including the apostrophe-s. The apostrophe-s construction is how nouns in their possessive form are inflected in English. This is clearer when you consider how pronouns are handled in this case: possessive inflection of pronouns is highly irregular, but in all cases apostrophes are never used to indicate possession with pronouns. It would be absurdly unthinkable to have links read as, e.g., their or his or mine. The apostrophe-s used for all other non-pronoun nouns has the exact same grammatical function as the final letter(s) in the pronoun examples above, so should be handled accordingly. Firejuggler86 (talk) 08:16, 11 April 2021 (UTC)

• Agree with JG66. The subject should not include the possessive, and should be "Akira Tozawa's". Similarly, you would not link their, his or mine as they are pronouns and linking would be absurd and an WP:OVERLINK. You would not link [[band]]'s, except when discussing [[:the Band]]'s, but even then, no need to link the possessive. Walter Görlitz (talk) 21:14, 11 April 2021 (UTC)
• No surprise, this has come up before. See Wikipedia_talk:Manual_of_Style/Linking/Archive_18#What_a_waste_of_time (though that's actually a link to my withering summary which apparently ended the discussion -- scroll up to see the preceding infighting.) EEng 16:48, 6 May 2021 (UTC)

Above: (i) "Towaza's in the above example is an indivisible word" (I've spared you the boldface.) (ii) "The apostrophe-s construction is how nouns in their possessive form are inflected in English." Wrongo. I've never heard of Tozawa, but if this were a Tozawa in Beppu who needed to be distinguished from a Tozawa in Kumamoto, then we wouldn't say *"Akira Tozawa's in Beppu ninja faction"; we could say "Akira Tozawa in Beppu's ninja faction" (awkward though this admittedly is). I have divided "Tozawa" and "'s". No surprise there, as genitive ("possessive") "apostrophe s" is not an inflection on a noun, it's a clitic attaching to a noun phrase. (Yes, very often the noun phrase ends with the head noun, whereupon the clitic looks like a suffix on the noun.) -- Hoary (talk) 23:46, 26 June 2021 (UTC)

Should 's be part of the link?

Should it be linked as

• Rembrandt's paintings

or

• Rembrandt's paintings

Akeosnhaoe (talk) 17:36, 20 June 2021 (UTC)

The former. Regards  Spy-cicle💥  ^Talk? 17:38, 20 June 2021 (UTC)

• I agree. I see far too many piped links needlessly including the 's, and I usually undo the piping. -- Ohc ^{revolution of our times} 10:11, 28 June 2021 (UTC)
• Agreed, the former. Herostratus (talk) 00:51, 29 June 2021 (UTC)

Overlinking

So, I removed this redirect (New South Wales, Australia –> New South Wales) because I believe it is completely unnecessary for obvious reasons. An edit war ensued with User:Fleets, with their rationale being that it makes the location appear as one colour (they also mentioned that it's recommended at the Rugby League project page, which I can't find any mention of). After I reverted their third revert, informing them that the redirect isn't covered by any aspect of WP:RPURPOSE, they made this edit to link Australia on it's own, presumably for no other reason than to make the location appear as one colour. Is this a valid reason for linking Australia in the infobox? – ².O.^Boxing 15:10, 21 April 2021 (UTC)

Australia should be unlinked per MOS:OVERLINK. There's never been consensus for cases like "Sydney, New South Wales" whether it should be linked as 1) Sydney, New South Wales ([[Sydney]], New South Wales); 2) Sydney, New South Wales ([[Sydney, New South Wales]]; or 3) Sydney, New South Wales ([[Sydney]], [[New South Wales]]).—Bagumba (talk) 08:39, 29 April 2021 (UTC)

Why is "New South Wales" necessary at all? And I would unlink "Sydney" in that context. Tony (talk) 09:11, 29 April 2021 (UTC)

I can't speak for Australia, but in the U.S. referring to "city, state" is quite common in real life, and is even more jarring when they differ from one infobox to the next. A drive-by editor will always come in and "fix" the state, if missing. It generally just creates churn to enforce WP:USPLACE in tables and infoboxes.—Bagumba (talk) 09:23, 29 April 2021 (UTC)

I personally wouldn't have NSW there; I usually only include states for the US or Canada. I would also link Sydney, mainly because of the confusion that people seem to have about it being the capital of Australia lol. As there doesn't seem to be a consensus on those I'll leave them be as I'm sure another edit war will occur. I'll go ahead and remove the link to Australia though, "it makes the location appear as one colour" is a ludicrous reason to link something that probably 99% of English speaking people will know of. – ².O.^Boxing 11:59, 29 April 2021 (UTC)

Linking "Sydney" is not going to help confusing it as the Australian capital. Go look at the article on Sydney and see how long a foreigner would take to get that message (and of course, they wouldn't be specifically hunting for it). Tony (talk) 03:31, 2 May 2021 (UTC)

@Bagumba: I can't speak for Australia

I can.

@Bagumba: but in the U.S. referring to "city, state" is quite common in real life

As it is here.

@Squared.Circle.Boxing: I personally wouldn't have NSW there; I usually only include states for the US or Canada.

1. That's inconsistent. Biased even.
2. Canada doesn't have states; it's not a federation. Australia is.

  — 𝐆𝐮𝐚𝐫𝐚𝐩𝐢𝐫𝐚𝐧𝐠𝐚 ☎ 22:51, 26 June 2021 (UTC)

I've encountered issues with this same editor, persistently changing links to the form [[Town, New South Wales|Town]], [[New South Wales, Australia]] (example, but there are many others in their contribs). I see that Bagumba last year proposed an addition to the guideline that would clarify that this type of thing shouldn't be done, but it seemed to die out before being implemented. Pinging participants from previous discussion: @Squared.Circle.Boxing, Tony1, Arjayay, Michael Bednarek, EEng, Walter Görlitz, Ohconfucius, and Alaney2k: Are there any objections to implementing this to avoid misunderstandings of this kind? Nikkimaria (talk) 12:04, 8 June 2021 (UTC)

No objection whatsoever, please proceed - Arjayay (talk) 12:09, 8 June 2021 (UTC)

No objections from me. – ².O.^Boxing 12:14, 8 June 2021 (UTC)

(summoned, ec) Nested links like the one pointed out by Nikkimaria are confusing easter eggs. As has been discussed before, using two links where one will do is always an unnecessary sea of blue. There is no need to even mention Australian states or territories for larger cities, much less to link them. If a state is mentioned for any country, there's no need to link it, unless the place cannot be linked for lack of an article for it. -- Michael Bednarek (talk) 12:19, 8 June 2021 (UTC)

In most cases, a simple link like [[Parkes, New South Wales|Parkes]] is better than either [[Parkes, New South Wales]] or [[Parkes, New South Wales|Parkes]], [[New South Wales]] . Though there may be instances where either the second form is preferable (say, for American readers), or the third one (where the state has particular relevance in the context). I don't think it's a good idea to legislate away any two of these alternatives: this situation is well enough served by the existing guidelines (and the interaction between them, applied with common sense). Ideally, I wouldn't like to see people edit war over such matters, and as always, making large-scale changes without consensus should be an obvious no-no. – Uanfala (talk) 12:29, 8 June 2021 (UTC)

We are here because of the large-scale editing and edit warring by one user, Fleets (talk · contribs). Your optimism is touching, but apparently misplaced. The guideline should be clearer on this. -- Michael Bednarek (talk) 13:37, 8 June 2021 (UTC)

At a minimum, let's add an example for something like Parkes, New South Wales; it's a clear MOS:SEAOFBLUE to break that into two links.—Bagumba (talk) 14:30, 8 June 2021 (UTC)

Per User:Uanfala: As I have said before, Buffalo, New York, is indeed better than Buffalo, New York, but is equally blue. Arguably, Buffalo, New York is even better. I don't see the need to see that it links to the City, State, any more than the state needs to be linked. -- Ohc ^¡digame! 15:14, 8 June 2021 (UTC)

The issue is not the amount of blue in this case, it's not knowing where one link ends and another begins. But if the full page title is already Buffalo, New York, its overkill to pipe to remove the state from the link, only to add the state back unlinked. It's a different case with titles like Atlanta, which don't already have the state in its title.—Bagumba (talk) 15:38, 8 June 2021 (UTC)

The parsing of adjacent links definitely needs to be considered, but I don't think the other distinction is relevant. Whether the target article will have a disambiguated name (like Buffalo, New York), or sit at the base title (like Atlanta) will depend on the presence of other, unrelated, topics with the same name out there (WP:PRIMARYTOPIC), and I see no reason why this should ever enter the equation here. – Uanfala (talk) 15:53, 8 June 2021 (UTC)

I agree with Ohconfucius on this, and will take it further. If a link is needed, then link the city. In most contexts, the state, province or canton is not required at all, and it certainly does not need a link. A link to the nation is almost never required.

The rationale is that in most contexts, the city is the least-known element in the array and it should contain links to the next higher geographic area. It will also contain a link to the nation. As an extension to this, piping a disambiguated location, as was presented in the article, is not needed, and is essentially an Easter egg. However a link array like [[London, Ontario|London]], [[Ontario]], Canada should most definitely be avoided, but either [[London, Ontario]] or [[London, Ontario|London]], Canada is acceptable. Walter Görlitz (talk) 16:05, 8 June 2021 (UTC)

Whether the target article will have a disambiguated name (like Buffalo, New York), or sit at the base title (like Atlanta) will depend on the presence of other, unrelated, topics ... Not the case for at least the U.S. Per WP:USPLACE, virtually every U.S. location page is titled <placename>, <state>.—Bagumba (talk) 16:14, 8 June 2021 (UTC)

Yes, but the best-known cities in the USA are single-barrelled: Los Angeles, Seattle, New York City, Chicago, Miami, New Orleans, San Francisco, Las Vegas. -- Ohc ^¡digame! 20:07, 8 June 2021 (UTC)

See the list at [5]. EEng 20:28, 10 June 2021 (UTC)

• Australia has a small population that mostly resides in the state capital cities. There are only six states, and we don't have the same custom of automatically specifying state after city. But a state should be specified on first occurrence of a rural or regional town/area—definitely. And unless there's a good reason, I'd avoid linking the major capital cities. Tony (talk) 03:59, 9 June 2021 (UTC)

Aside from the US (let's not forget WP:NOTUSA), which is addressed by WP:USPLACE, locations should generally be City, Nation. The sub-national location is not usually germane to the understanding. I disagree with WP:CANPLACE that insists on including the province and nation because most non-Canadaian readers of an article about a Canadian do not care about the province, only Canadian readers might. Walter Görlitz (talk) 17:09, 9 June 2021 (UTC)

For US locations where the target article has city/state, I think it's completely unnecessary to pipe them separate. For locations like LA, I think it's completely unnecessary to link California. As for LA itself, I personally wouldn't link that either as the city is probably one of the most well-known US cities. I agree with Walter above; as somebody from England, I rarely hear/read a province mentioned after a city. If people insist on including the province, then like US locations, I think it's unnecessary to link the province sperately. I know we shouldn't make readers chase links, but if somebody is interested in finding out about a province then they can simply click the city and find the link to the province there, it's no big drama. I also agree with Tony, similar to Canadaian provinces, I rarely hear/read an Australian state mentioned after a city. However, for a small town or rural area, providing the state would be helpful.
The main thing I'd like to get some feedback on is if countries should be included in a redirect/piped link, examples are given in my opening comment as well as Nikkimaria's comment. I personally can't find any logic in New South Wales, Australia or Humberside, England. – ².O.^Boxing 19:58, 10 June 2021 (UTC)

• Thanks all for your input. I have added some wording along the lines of Bagumba's earlier proposal. I've left off the Toronto example because I don't think this page, as a linking guideline, ought to get into the question of whether/when to include subnational locations for non-US places; if anyone has strong feelings on that question they're probably best raised elsewhere. Nikkimaria (talk) 12:35, 12 June 2021 (UTC)
  • The new rule you've added is a direct corollary of other, more general, principles of the same style guidelines. I think this amounts to instruction creep. – Uanfala (talk) 13:05, 12 June 2021 (UTC)
    • I see that you've reverted the addition. I appreciate you may believe it is not needed, but the consensus here seems to be that it is. Nikkimaria (talk) 14:35, 12 June 2021 (UTC)
      • As far as I can see, the consensus is that links of this type are usually unwanted. I don't think I see consensus that the guidelines should be expanded with a dedicated rule specifically covering them. – Uanfala (talk) 14:46, 12 June 2021 (UTC)
        • I see several people above supporting the addition to the guideline, and only you explicitly opposing it. Nikkimaria (talk) 15:27, 12 June 2021 (UTC)

          The guideline should address this issue, either via explicit text or (probably sufficient) an appropriate example adjoined to the existing general text. EEng 16:08, 12 June 2021 (UTC)

          The guidelines already address the issue: at MOS:SPECIFICLINK and MOS:SEAOFBLUE. An example may be added to one of these though. – Uanfala (talk) 17:15, 12 June 2021 (UTC)

Back to Squared.Circle.Boxing's question, artificially linking the nation as you demonstrated seems odd. I would remove any such links if I found them. Walter Görlitz (talk) 22:36, 12 June 2021 (UTC)

I think Nikkimaria's reading of this discussion is correct, and her addition to the guideline at "General points on linking style" was a reasonable summary of this discussion and is non-controversial. -- Michael Bednarek (talk) 02:16, 13 June 2021 (UTC)

Yeah, I agree. Tony (talk) 02:32, 13 June 2021 (UTC)

I tried to improve the added text [6]: it was vague about when it was applicable (the way it was worded, it seemed to apply to any mention of anything that's a place), and it used strong language ("Do not...") that's ususual even for the other, much better supported, parts of these guidelines. However, I got reverted. Now guys, if you're going to introduce a big new black and white rule to the guidelines – especially one that effectively bans a very widely spread current practice – then please do the decent thing and get proper consensus for that first. – Uanfala (talk) 00:36, 17 June 2021 (UTC)

As already noted, we have consensus for the addition in this discussion. I appreciate that you don't agree, but your change (with the exception of the note, which I kept) did not seem an improvement. Nikkimaria (talk) 01:08, 17 June 2021 (UTC)

How is there no consensus? Care to prepare a count? -- Michael Bednarek (talk) 01:22, 17 June 2021 (UTC)

It often happens that a few people will find consensus among themselves, only for it to later turn out to be at odds with the views of the community at large. This concerns one of the most important pages of the manual of style, and any new rules added to it (especially such unusually strict ones) need to have the support of more than a handful of editors. – Uanfala (talk) 01:26, 17 June 2021 (UTC)

You stated above that "The new rule [I]'ve added is a direct corollary of other, more general, principles of the same style guidelines", and that "The guidelines already address the issue". By those arguments, there was already community consensus for the principle before this discussion - this is simply a clarification rather than an entirely new concept. Nikkimaria (talk) 01:31, 17 June 2021 (UTC)

From the existing guidelines: When possible, avoid placing links next to each other so that they look like a single link, and Always link to the article on the most specific topic, an editor will be able to figure out what is the most appropriate way to link in the situations at hand. This will often, though not always, result in the outcome you're proposing, which is the hard-and-fast rule For geographic places within an entity (e.g. state, province, etc.), do not add an additional link to the entity. This new rule applies to an overly specific case so effectively locks into place a rule that bypasses the general guidelines, it's worded vaguely and so is likely to be overapplied in unintended ways, it's stated unnecessarily strongly and so would discourage common-sense exceptions, and it effectively bans a practice that's very widely used (not always, but at least sometimes, with good reason). These are four massive red flags, and each one on its own would in my opinion be enough to block the addition. – Uanfala (talk) 01:56, 17 June 2021 (UTC)

We already have guidance both within this page and more broadly in MOS that uses "do not" phrasing, and Tony1's addition which you reverted had already softened that with "generally". This isn't in any way inconsistent with the tenor of MOS, and doesn't turn this page into a policy. Besides which, your argument seems contradictory: if the existing guidance is sufficient for editors to figure out the most appropriate way to link, then why is the practice of inappropriate double-linking widely used? Nikkimaria (talk) 02:48, 17 June 2021 (UTC)

Why the practice is widespread despite the existing guidelines? To the extent that these guidelines enter into the equation at all, I believe it's because either people don't care about these specific guidelines (in which case expanding them isn't going to help), or they have taken them into account but only in combination with other factors which have altogether led them to decide to link anyway (this sort of nuanced, common-sense approach will become more difficult if the guidelines are expanded with a dogmatic new rule). – Uanfala (talk) 14:45, 17 June 2021 (UTC)

I believe that one significant factor is that Wikipedia:Nobody reads the directions. That doesn't mean that we should have good directions. However, I don't think we should expect too much from them. WhatamIdoing (talk) 17:58, 17 June 2021 (UTC)

It won't, because no matter how we word our clarification - which this is, as you've pointed out, rather than a whole new rule - this remains a guideline. If there actually is a common-sense reason why some case should use [[Buffalo, New York|Buffalo]], [[New York]], then it can. But in most cases that shouldn't be done, and it seems that everyone else agrees that should be made clear here. Nikkimaria (talk) 23:32, 17 June 2021 (UTC)

The story that guidelines don't/shouldn't use language such as "Do not" is false but very widespread. The main MOS page currently contains the phrase "Do not" 85 (eighty-five!) times. The Manual of Style (which includes this page) uses this phrase and similar ones (always, never, must) more than any other set of guidelines or policies, and I explicitly include legal policies in that count. The majority of legal policies don't use this or similar phrases at all. Wikipedia:Copyright violations contains "Do not" twice, and one of those is not a command ("If the criteria for speedy deletion do not apply"...). WhatamIdoing (talk) 17:55, 17 June 2021 (UTC)

• @Uanfala: I appreciate you don't think this clarification should be made, but your count is off by at least half, you're the only person opposing, and your data may or may not mean what you're interpreting it to mean. We have strong consensus both that [[Buffalo, New York|Buffalo]], [[New York]] links should be avoided - even you seem to have agreed with that previously - and that this issue warrants being explicitly addressed. It's a minor clarification, not a radical policy change. You're welcome to propose an alternate wording if you want, but please stop stonewalling. Nikkimaria (talk) 00:39, 21 June 2021 (UTC)
  • Sorry for being such a spoilsport, but I believe there remain substantial arguments against the addition, and I don't believe these can be waived away simply because all but one of a small group of editors choose to ignore them. If indeed my position here is so far from community consensus, then there's no better way to demonstrate than and shut me up for good by querying the community with an RfC and seeing within days WP:SNOW-level support for the proposal. Yes, the data below is limited, but it very obviously contradicts the working assumptions in the preceding discussion. I understand that it may sometimes be annoying when someone continues to disagree, but I also believe it's better if arguments are into account rather than wished away. – Uanfala (talk) 14:30, 21 June 2021 (UTC)
    • There is a difference between ignoring arguments and disagreeing that they constitute valid substantive objections. Looking at the arguments you've made, you've claimed that the case is adequately addressed by the current guideline - other respondents feel the clarification is merited. You've claimed that it used "strong language" - as has been pointed out, other parts of the MOS use similar language. You've claimed it is an effective ban - as has been pointed out, the page remains a guideline no matter its wording so allows for common-sense exceptions (should there be any). You're now claiming that the data below is limited, but it very obviously contradicts the working assumptions in the preceding discussion - no, your interpretation of that very limited data contradicts the guideline, but there is insufficient information to confirm that interpretation. Other than that, about the only objection you've made that hasn't been directly addressed is that the wording is vague - I would invite you again to propose an alternative version that you feel is clearer, to see if you can get consensus for it. But at the moment, there is consensus for the version you've been repeatedly reverting. Nikkimaria (talk) 00:39, 22 June 2021 (UTC)
      • What other possible analysis is there of the data below? Whatever plausible conclusions you can draw from it, I don't see any way in which it can be made consistent with the premise of the proposal here. – Uanfala (talk) 02:04, 22 June 2021 (UTC)
        • I wouldn't presume to base any analysis on the scant data presented below. Nikkimaria (talk) 02:09, 22 June 2021 (UTC)
          • It is scant, but it is the only data that has been presented so far. If you say that all swans are black, and then you're presented with two perfectly healthy specimens of swans with naturally black feathers, then it doesn't seem very reasonable to insist that the black swans should be discounted because it's just two of them, especially not when that's all the swans there are. – Uanfala (talk) 02:43, 22 June 2021 (UTC)
            • If you are presented with an image of two swans that may be black - you can't really tell because the photo is so blurred - it in no way prevents you from writing a guideline that swans are generally white. It certainly doesn't outweigh existing guidelines that support that, nor accessibility issues for swan-users, etc. Nikkimaria (talk) 03:00, 22 June 2021 (UTC)
• In an effort to move forward, I have added a compromise wording. Nikkimaria (talk) 14:32, 26 June 2021 (UTC)
  • This is getting closer to the compromise wording that I had attempted before. However, in light of the link usage data presented below, it appears that the assumption that we all shared – links after the comma are often useless for readers – is not valid. The way to genuinely move forward from here is to look at more data, re-examining our assumptions in the process, and coming up with better, evidence-based, advice for linking (whether in these guidelines, or in an essay). Instead of doing that, you seem determined to continue trying to insert this text into the guidelines, no matter what the data is showing. I find that baffling.
    Given that we seem to be stuck in a slow edit war over this issue, I see no other way out than to start an RfC. Again, this should at any rate be done before making any major additions to such a widely used guideline. Personally, I believe it's a waste of time – we need to make use of the newly available tools to look at how links are used by readers instead of lining up to vote on a proposal that's contradicted by the data we have so far. But I see no other way out. Sorry, everyone, for the bother. – Uanfala (talk) 21:16, 26 June 2021 (UTC)
    • I find it baffling that you would insist the consensus here so far is insufficiently representative, and yet also insist that two data points out of millions of articles is sufficient to make broad generalizations. Nikkimaria (talk) 22:08, 26 June 2021 (UTC)
      • But these are not two data points out of millions, these are two data points out of two. You're welcome to expand the small dataset available here. – Uanfala (talk) 22:42, 26 June 2021 (UTC)
        • The fact that you only looked at two data points doesn't make the limited data more convincing. If you really believe we need to radically rethink our linking practices, it'll be up to you not only to demonstrate that your interpretation is correct and generalizable - as you yourself note below, there are reasons to believe these examples are not representative - but that the best/only way to address this interpretation is to support SEAOFBLUE-style linking, regardless of the issues it creates wrt accessibility etc. Nikkimaria (talk) 00:09, 27 June 2021 (UTC)
          • I don't believe you need more than one black swan to demonstrate that the statement that all swans are black is false. Even the single example of Obama shows that 1,700 people find it useful to follow that one link on that one article each month. That's a pretty significant figure. Anyway, the way the RfC is going, it appears to clearly demonstrate one thing: the community has zero interest in taking any evidence into account. I guess I've misunderstood how things things are supposed to happen here. I apologise, and I will not waste your people's time any more. – Uanfala (talk) 14:45, 27 June 2021 (UTC)

I'm not sure the RfC is ready. I don't think you're going to get much push-back on avoiding [[Buffalo, New York|Buffalo]], [[New York]] and [[Portland, Oregon|Portland]], [[Oregon]] and other examples where the sub-national area is a disambiguator, but it's not sure from this example if we should be avoiding [[San Francisco|San Francisco, California]] or if we should insist on [[San Francisco|San Francisco, California]] rather than [[San Francisco]], [[California]] or even [[San Francisco]], California. I'm afraid we will get into wikilawyering over the lack of detail. Walter Görlitz (talk) 23:56, 26 June 2021 (UTC)

Looking at usage

I'm having second thoughts on whether generally avoiding the second link is even a good idea to begin with. One solid indicator of whether a term should be linked is whether the resultant link will be useful to readers, and it will likely be useful to readers if readers end up clicking on it. So I had a look at a small sample of articles: the four last US presidents. Two of them – Barack Obama and George W. Bush – had prominent links of this form for the birthplace ([[Honolulu]], [[Hawaii]] and [[Houston]], [[Texas]] respectively). In both cases, the link to the state was more popular than the link to the city (the clickstream for March showed 1731 clicks for Hawaii vs. 1606 for Honolulu, and 16 for Texas vs. 14 for Houston). – Uanfala (talk) 20:47, 20 June 2021 (UTC)

That's because readers think it's a single link—the very reason that link-bunching is discouraged. Then they end up at a less-relevant, less-focused article. What method do you use to gain these data? Tony (talk) 23:02, 20 June 2021 (UTC)

The data is from meta:Research:Wikipedia clickstream. Some numbers on both sides are indeed likely to be down to reader errors. However, I don't think your explanation is plausible. For it to be true, then the following must obtain: 1) there were no readers interested in following the link to the state; 2) of the readers interested in the city, the majority misunderstood the two links as one; 3) the majority of those readers in turn clicked on the state rather than the city (which looks ever more unlikely if you consider that in both cases the state name is a bit shorter than that of the city, so even if readers clicked at random you'd expect fewer clicks for the state). Proposition 1) is the underlying assumption of most participants in this discussion, but it doesn't follow from these figures, and in fact, it's inconsistent with them unless both 2) and 3) and met, and of these at least the last one is extremely improbable.
Now, that's a small sample: just two articles, I don't know if the pattern will hold if we look at more data. Of course, it's possible the interest in the state rather than the city may be down the factors peculiar to those two articles (for example, they're exposed to a more general readership, where knowledge of US geography may be scant: maybe readers were interested in finding out if Hawaii, where Obama was born, is part of the US or not?) It's easy to imagine the opposite scenario: say, readers in an article about an obscure Texan politician are most likely to be residents of that state, so there'd be little reason for them to want to read more about it in exactly this context. If this is the right way of looking at the issues here, then it would appear that the proposed addition to the guidelines would be geared only towards the narrow-topic context. – Uanfala (talk) 23:49, 20 June 2021 (UTC)

Request for comment

Should the following be added to Wikipedia:Manual of Style/Linking?

For geographic places specified with the name of the larger territorial unit following a comma, generally do not link the larger unit. For example, avoid [[Buffalo, New York|Buffalo]], [[New York (state)|New York]].

The previous discussion is at Wikipedia talk:Manual of Style/Linking#Overlinking above. – Uanfala (talk) 21:19, 26 June 2021 (UTC)

• Support - appropriate clarification of the existing guidance (MOS:SEAOFBLUE and MOS:SPECIFICLINK), as per arguments above. Nikkimaria (talk) 22:08, 26 June 2021 (UTC)
  • Pinging respondents above to ensure they see this new subsection: @Squared.Circle.Boxing, Bagumba, Tony1, Michael Bednarek, Ohconfucius, Arjayay, Walter Görlitz, EEng, and WhatamIdoing:. Nikkimaria (talk) 22:09, 26 June 2021 (UTC)
• I started this RfC only because Nikkimaria insisted on inserting this text into the guidelines. At first I was inclined to agree with them that this was a specific straightforward corollary of the two existing principles linked above (and hence unnecessary per WP:CREEP). However, as the section #Looking at usage above shows, whatever little data we have looked at on actual reader usage of such links has shown that the links to be banned by this proposed guideline are in fact useful to readers, sometimes to a large degree. We really need to look at more data (now that we have a window into link usage), but any advice we eventually come up with about links to entities with comma-separated geographical qualifiers will be too specific to include in this broad guidelines page, and will instead be better placed in a supplementary info page or an essay. – Uanfala (talk) 22:22, 26 June 2021 (UTC)
• Oppose. Can a casual reader tell where either part of this is, just by reading it black on white: Jijoca de Jericoacoara, Ceará? Rather, it should be linked as Jijoca de Jericoacoara, Ceará (Brazil) as it refers to a single object in the text (much like polar bear refers to a single object, and should so be linked, not separately: polar bear). — 𝐆𝐮𝐚𝐫𝐚𝐩𝐢𝐫𝐚𝐧𝐠𝐚 ☎ 23:09, 26 June 2021 (UTC)
• Support, with modification: "avoid [[Buffalo, New York|Buffalo]], [[New York]]; instead use [[Buffalo, New York]]." That way we clarify which Buffalo is meant but also avoid a redundant extra link. --Redrose64 🌹 (talk) 23:22, 26 June 2021 (UTC)

  That's what I meant above. So, yes, Support, with that modification. — 𝐆𝐮𝐚𝐫𝐚𝐩𝐢𝐫𝐚𝐧𝐠𝐚 ☎ 23:32, 26 June 2021 (UTC)

• Comment. Definitely avoid [[Buffalo, New York|Buffalo]], [[New York]], if only because that invites the reader to a disambiguation page. (Perhaps you mean [[Buffalo, New York|Buffalo]], [[New York (state)|New York]].) -- Hoary (talk) 23:34, 26 June 2021 (UTC)

• Hoary, I've made that change. Nikkimaria (talk) 00:05, 27 June 2021 (UTC)

• Support as written, and would also prefer the addition by Redrose. MB 00:11, 27 June 2021 (UTC)
• Support + Redrose's addition, and I'd go further. Right now there's reference to "a comma", but there are use cases with multiple commas. For example, [[San Quirico, Sorano|San Quirico]], [[Sorano]], [[Italy]] --> [[San Quirico, Sorano]], Italy or maybe [[San Quirico, Sorano|San Quirico, Sorano, Italy]] . EEng 03:17, 27 June 2021 (UTC)

Italy shouldn't be linked, so I'd say [[San Quirico, Sorano]], Italy would be the better option. No point including Italy in the link at all. – ².O.^Boxing 12:07, 27 June 2021 (UTC)

Right you are. Struck part of my example. EEng 12:43, 27 June 2021 (UTC)

• Support – Tony (talk) 03:42, 27 June 2021 (UTC)
• Support and include Redrose's addition. – ².O.^Boxing 03:54, 27 June 2021 (UTC)
• Support proposal & Redrose's clarification. ~ HAL333 04:29, 27 June 2021 (UTC)
• Oh, how I have (long!) yearned for Redrose's addition to be adopted. It would simplify the wikitext, reduce confusion between separate links which appear to be the same link, and help fight WP:OVERLINK. Count me as a support in any case. — JohnFromPinckney (talk / edits) 08:53, 27 June 2021 (UTC)
• Support with Redrose's clarification per MOS:SEAOFBLUE and MOS:SPECIFICLINK.—Bagumba (talk) 11:40, 27 June 2021 (UTC)
• I like Redrose64's clarification, but I wonder if using one of the most famous regional entities in our example is leading us to being too parsimonious with links. Most people have heard of New York (and California). Most people have not heard of most states/provinces in the world. For less famous locations, maybe the advice should be City in the Province of Country or City in Country (the very smallest country names might even need to be linked), rather than "City, Province" and making people guess what part of the world that's in. WhatamIdoing (talk) 16:51, 27 June 2021 (UTC)

  I would support requiring country names, particularly for lesser-known locations, but since this is a linking guideline I don't think this is the place to address that. Nikkimaria (talk) 16:57, 27 June 2021 (UTC)

  Nothing in here addresses whether or not to give the country name, just whether/how to link it if it's present. As in my example above, if someone wants to know more about Italy, they can click on San Quirico, Sorano, and that article will undoubtedly link to Italy. This might be an argument for the link being [[San Quirico, Sorano|San Quirico, Sorano, Italy]] in order to signal the reader that the one link gives him info on all three entities. (And really, San Quirico, Sorano, Italy ought to be a redirect to San Quirico, Sorano, so then the link can be simply [[San Quirico, Sorano, Italy]].) EEng 18:14, 27 June 2021 (UTC)

  Discuss independently whether to add state/province or country, when it is not already part of the actual page title. A lot of times it's a matter of context if the added information is helpful or redundant, and there's pros and cons on whether it should be 1) mentioned and 2) piped or not.—Bagumba (talk) 02:21, 28 June 2021 (UTC)

• Support – The fewer the successive (secondary) links, the better. -- Ohc ^{revolution of our times} 05:50, 28 June 2021 (UTC)
• Oppose – I think this just subset of MOS:SEAOFBLUE, would it therefore not suffice to add "..for instance [[San Quirico, Sorano|San Quirico]], [[Sorano]], [[Italy]]' " to that guidance rather than adding a new point for one specific example? Otherwise, as there seems to be consensus so far for this change, could we instead add it to the 'Specific Examples' section? JeffUK (talk) 18:38, 23 July 2021 (UTC)