Wikipedia

Wikipedia talk:Manual of Style/Disambiguation pages/Archive 5

< Wikipedia talk:Manual of Style‎ | Disambiguation pages
Latest comment: 18 years ago by Wahoofive in topic Wikilinking
Archive 1 ← Archive 3 Archive 4 Archive 5 Archive 6 Archive 7 → Archive 10

Wikilinking

Latest comment: 18 years ago39 comments10 people in discussion

I disagree with the instruction not to link "every other word". If I am not an expert, it is useful for me to click at these immediately, in order to make a selection. Also, people are not robots. They click links out of curiosity, and this is good. And depriving them of this possibility for purely technical pretexts is bad. mikka (t) 16:12, 28 July 2005 (UTC)

  • I disagree. A dab page is not an encyclopedia entry; it is a navigation aid. If I am looking to learn about mathematical fields, and I type field in the search box, then I am taken to the Field dab page. And I see the following entry:
  • Field (mathematics), an algebraic structure in which one can add, subtract, multiply and divide
I can easily find the thing I was looking for. On the other hand,
gives me too many blue links for my eye to parse. If I want to know what add means in this context, the field page will explain it and give me all the links I could ever I want. These pages are navigation aids to get the user to their destination quickly and painlessly. They should be "as simple as possible, but no simpler". - grubber 17:34, 2005 July 28 (UTC)
But notice that you've missed the obvious approach, used on most dab pages as of 2005: to wikify only what you'd wikify in a good article, such that the user is helped, not hindered. Thus:
If you use common sense, you don't need to worry about having too many or too few links. Use links where the reader might need them, that's all. --Quuxplusone 04:46, 2 August 2005 (UTC)
  • I'm with grubber. Highly-interlinked articles are good, because they let people wander around paths through the encyclopedia that they would never have gone down otherwise. I view dab pages differently; they're a way to get to a specific place. If I get on a tour bus, I expect to be shown all sorts of interesting stuff and take random detours along the way. But, when I get in a taxi and say, "Take me to The Met", about all I expect from the driver by way of feedback might be, "Did you mean the opera house or the museum?". I'm not expecting to get an extensive list of alternate places I might want to go instead. RoySmith 18:08, 28 July 2005 (UTC)
How is a disambiguation page analogically related to a taxi? First, readers do get there by entering a search, which can be an open-ended "random" quest for information. To relate it to your analogy, the searcher may be starting his trip and wants to chooose which tour bus he gets on. Second, you assume that a reader clicks on a link in an article specifically and only to get what the link says (and that too assumes that the article links to precisely what the word would indicate). A reader who clicks on a link in an article and gets to a disambiguation page may very well be on that tour bus looking for interesting stuff and with random detours. To relate it to your analogy: The reader tells the tour bus to go somewhere, but the tour bus only vaguely knows how to get there, or the reader only makes a vague request, because he only vaguely knows where he wants to go. Under the one-link-only system, the tour bus, depending on the sitation, will only go to the place it thinks is right, not allowing any deviation from its wrong path, or the reader makes an uncertain request, and is then locked into that for the trip--the trip he thought was going to be interesting with opportunity for random detours. Note also that the one-link-only system disallows highly related links. So, you tell the tour bus to go to a concert, but you can only listen to the one song you mentioned when you made the request. You tell him to go to Monticello, but you can't learn about Thomas Jefferson. You say you can just "type it in the search box", but that's a justification for any nonlinking. It's not a good reason for a wiki, when there's no good reason against having more relevant links. After all, by this argument, you don't "need" a tour bus at all: you can just walk everywhere. - Centrx 20:11, 1 August 2005 (UTC)
  • Your (User:mikkalai) tone suggests that you're angry about our somehow hurting readers, denying them their rights or in some way damaging them, denying them, in effect, the advantages of an online encyclopedia by keeping our heads stuck between the musty pages of a (useless?) paper tome. I'm with both User:Grubber and User:RoySmith in trying to keep things to one link per entry on a dab-page. The key is to provide sufficient information for a person to move through the "station" (using a subway analogy) so that their movement from point A to point B is done as easily as possible; extending that analogy, people tend not to linger in subway stations reading all the interesting signs unless they are stuck there waiting for a train .. or are distracted by the advertisements of vendors. What we're getting at is trying to sweep away the distractions so that the navigational aid can function as well as possible. Ideally, people would not need to pass through the dab-page but go directly from point A to point B; this is made possible by going through the pages that link to a dab-page and revising the links so they lead to the specific targets rather than to the dab-page, a time consuming but worthwhile process (p.s. I just finished this for 1992 Olympics where all references could be changed to point at 1992 Summer Olympics Courtland 03:08, July 29, 2005 (UTC)). Lot's more to be said (particularly about the search route to landing on a dab-page), but I'll stop there and look forward to your follow-up comments. Courtland 22:22, July 28, 2005 (UTC)
As for your subway analogy, the difference is that the subway advertisements are entirely unrelated to your destination. A more accurate analogy for having more relevant disambiguation links would be having a sign that pointed out a faster or alternate subway route to the destination you want, or a sign that tells you that even though the name of your destination station is "Symphony", you would actually get to Symphony Hall faster by using a different subway. - Centrx 21:49, 1 August 2005 (UTC)
  • The articles linked to contain all of the wikilinks that could be placed on the dab page; it's inappropriate for disambiguation to make things harder to find. One link per term is appropriate. Josh Parris ✉
Adding additional relevant links does not "make things harder to find"; in certain cases, it even makes it easier for the user to find what they want. - Centrx 21:49, 1 August 2005 (UTC)
  • If you want to explore, then click on one of the main articles listed in the dab page; usually there is a wealth of wikilinks there. Ease of navigation on a dab page far trumps useless links that maybe 1/100 people will click on. -- jiy 01:55, July 29, 2005 (UTC)
First of all, I don't think any one is advocating useless links; the links advocated would be highly relevant to the subject of the primary link. Second, though you are pulling the number 1/100 out of a hat, and I do argue that more relevant links would be helpful to many more people, even one out of a hundred adds up to millions of people who would find the links useful. There are many visitors to Wikipedia in a day, and hopefully Wikipedia will be here for many years. So, the question is, how do additional links impede navigation? - Centrx 21:49, 1 August 2005 (UTC)
Yes, relevent links would be helpful to many more people, in the article that's being linked to. Additional links impede navigation because they often create a sea of blue, as grubber described it, that the eyes must sort through. -- jiyTalk 03:44, August 3, 2005 (UTC)
Just as relevant links are helpful for a user clicking through an article they don't want, they are more helpful if the user does not have to click through that article. No one is arguing for a "sea of blue" links; this, as a matter of course because it is a negative characterization, would be undesirable. The proposal is for the allowance of more than one link; two or three is hardly a sea. Further, a user would have to do no wading at all if the primary eponymous link be bolded, with such formatting there would be no more sorting for the eyes needed than is currently needed. - Centrx 15:27, 3 August 2005 (UTC)
I have no idea what the first sentence in your reply means. Anyways, I don't agree with you about bolding, for many reasons (1) bolded links are ugly and clumsy in my opinion (2) it's inconsistent with the subject of the article being bold in an article (no, links to articles are not the subject of a disambiguation article, the subject is the word that is being disambiguated, and that is bolded already in the "meaningless" leading line) (3) People may end up using the bold to denote links to articles they think are most important while leaving the lesser ones with a normal link, which would be inconsistent and ugly, and besides I've never seen a dab page that used bolding consistently (4) in my opinion, bolding used in conjunction with wikilinks on the line will never be as elegent or efficient for the user as non-bold links with no other links on the line.
Moving on, I believe dab pages are meant to be scanned vertically, not read horizontally as one does in a normal article. The "sea" from two or three links may not be as apparent in horizontal-reading terms, but in scanning-vertical terms it is.
I've constructed an example page which illustrates my concern in this issue: User:Jiy/DabTest. The example assumes bolding is not used for article links, but I don't think bolding would be any remedy -- jiyTalk 11:53, August 4, 2005 (UTC)
The first sentence of my reply means: In the same way that links are useful in the article being linked to by the disambiguation page, relevant links are more useful in the disambiguation page because passing through the intermediary "article being linked to" is not necessary to get to the page wanted.
(1) Having words that are bold, underlined, and colored is not an uncommon use of formatting on the web, and I do not see how it is ugly. Regardless, using italics for instance would be better than not having any primary formatting of the type described. Less subjectively, how are bold links at all clumsy? Editing a page with them is no more clumsy than other apostrophe, bracket, colon, etc. formatting of Wikipedia, and navigating through a page with bolded links is no more difficult than navigating through any page with links--in fact, such bolded links actually make such navigation easier.
(2) The subject of a disambiguation page is not the word (by which I assume you mean the vague object of confusion that is the reason for the page); in fact, a disambiguation cannot properly be said to have a subject at all in the same way an article does. A subject is the singular purpose and end of an article, in itself; a subject is the object of study. A page with a subject directs the reader to further study that subject, and the opportunity for that study is maximized by treating the subject as comprehensively as possible.
If the subject of a disambiguation page were the word, the page would have the etymology, history, and meaning of that word, more like a dictionary. Further, as a subject is singular, words with unrelated history and meaning, despite having the same spelling, would go in different articles, as they do in a dictionary.
A page without a subject, such as a disambiguation page (or a category, for instance), instead solely aims to redirect the reader elsewhere to an entirely different topic. It aims to minimize the time spect on that intermediary page. Such pages are structures which organize into a form the content of articles. So, a disambiguation page, from an external perspective, is an index of articles with similar names. From an internal perspective, that is for the user within the structure passing from some content to the next, a disambiguation page is a mapping (more in the mathematical sense) or a redirect that provides information for a user's required decision between possible meanings of an ambiguous set of characters (word). At the extreme, if a disambiguation page has only one possible entry, so only one choice, it would be properly changed to a redirect page to that one choice.
(3) The suggested change is that bolding would be used to mark only the term which corresponds with the term being disambiguated (the "vague of object of misunderstanding" I mention above), whether that term is linked or not. Links that do not correspond to the disambiguated term would not be bolded (this might alleviate your concern about the bolding being "ugly" and "clumsy"). The benefit of this is that it allows for two flexible levels of relevant terms in an item, one that directly corresponds with the disambiguated term (bolded) and one that that links to highly relevant and useful, but not directly correspondent, terms. (Re: flexible combination: Terms that directly correspond to the disambiguated term but which properly have no Wikipedia article, or where the title is sufficiently confused that a redlink might not be appropriate, can be bolded but not linked; this is currently not possible and causes confusion for the reader when the linked article for an item is not clearly related to the disambiguated term.) I use bolding consistently on all disambiguation pages I maintain; it is no more difficult than consistently handling other formatting. Part of the inconsistency you may have noticed is probably because the new policies of this styleguide are not propogated through the wiki so elements of the style that was fairly common before this guide remain. (I surmise that, because the new leading line bolding style was much less common so any instances you find of that bolding are consistent because it was recently changed to conform.) That you notice inconsistent bolding is no more a reason against the suggested bolding style than any inconsistent formatting would be a reason against any style in the guide. Any editor who follows the styleguide would use bolding consistently, and any editor who does not carefully follow the styleguide might not use bolding consistently, which is the same situation as for all policies all the time.
(4) The current recommended style is less efficient at directing the user to the relevant article he seeks, for it does not indicate highly relevant related articles or allow for their easy link-following, and it confusingly treats links that directly correspond to the disambiguated term in the same way it treats links that are only there because there was no other relevant linkable term. That the style might look a bit more gracefully simple is not a reason that it must be retained at the expense of the usefulness and purpose of the disambiguation page.
((5)) Regarding eye-scanning, the suggestion for bolding is unrelated to the location of the primary link, that link could still be at the beginning of each line, which is efficient for vertical scanning. Regardless, bolding is effective at distinguishing the primary term so that it is easily apparent when scanning. Your example in no way corresponds with the reality of the proposal, and is irrelevant to this discussion. No one is proposing having a black background behind blue links where the user has to squint to even read what the link says. Also, this example seems like it might have the same the same misunderstanding that bold would be used for all links rather than just one. If your argument on the example page be valid, which I disagree with and which is not determinable from that example, the bolding would still "anchor" the eyes to left. - Centrx 17:29, 4 August 2005 (UTC)
(0) I think you exaggerate the ignorance of Wikipedia users. People usually have enough information to know which article they want to choose in a disambiguation page. If "relevent" links were encouraged, I am confident they would serve little to no purpose to any significant amount of users in deciding which article they want, except in special circumstances (which is why the "break rules" guideline exists). Most users know what they want.
(1) Bolding links is uncommon on Wikipedia. It's ugly because it's ugly - I don't feel obligated to explain why I am aesthetically displeased with a bolded link. Italics are not a solution because they conflict with titles of major works being in italics. A link already has primary formatting - it is blue and underlined. When working with items in a list that are aiming to be as brief as possible, any extraneous piece of punctuation or formatting is clumsy. Bolded links used in conjunction with extra wikilinks on the line does not make navigation easier than non-bold links with no other wikilinks on the line.
(2) A page is supposed to put the subject in bold. In a page like Rose (disambiguation), the subject would be the ambiguous set of characters "Rose". This ambiguous set of characters, the subject of the disambiguation page, is bolded in the leading fragment. "Rose Bowl" is not the subject of Rose (disambiguation), so therefore should not be bolded.
Pages are encouraged to use {{wiktionarypar}} for etymologies and dictionary definitions.
(3) No, your suggestions do not allieviate my concerns of ugliness and clumsiness or potential misuse.
(4) I don't think any of your suggestions bring a disambiguation page any closer to achieving its purpose than the current guidelines. The grace of the present guideline in no way expends the usefulness of the page.
(5) I did not use the example as a primary illustration of why bolding should not be used. It is a primary illustration of why more than one wikilink on the line should not be used. It is not meant to correspond to reality, it is meant to illustrate how I feel the eyes react when vertically scanning a page containing a sea of wikilinks. I do not misunderstand your proposal for having only the primary term bolded on the line, and in fact I expressedly stated the example assumes bolding is not used at all. I repeat: the example's primary function is to illustrate why more than one wikilink should not be used on the line. If the example takes bolding into account, then this is my stance: bolding is less effective in achoring the eyes to the primary term than the current guideline. The example is not irrelevent. -- jiyTalk 22:18, August 4, 2005 (UTC)
(0) This is not the main reason for having additional relevant links, which is to indicate highly relevant articles that many users are likely to be seeking. This is not directly related to helping a user choose which item they want. Regardless, who are these all-typical Wikipedia users? Wikipedia is used by millions of people from all regions of the globe and all walks of life. Wikipedia is and will be used by persons who barely know English, by persons who know nothing about vast swaths of subjects, by persons who actually are mentally retarded, and of course by everyday native English speakers of above average intelligence who simply do not know something. Suggesting that a person is ignorant of some knowledge is not an insult; no one knows everything and no one has all the information they need to find the information they seek. I hope you are not as confident of the purposelessness of additional relevant links as you are confident of the nature of "most users" of Wikipedia, who never even post a single edit, and have little in common with you or any other randomly selected individual.
(1) Bolded links in conjunction with extra wikilinks on the line do make navigation easier than non-bold links with no other wikilinks on the line; they make it easier to navigate to highly related information and they do not diminish the ease of navigating to the primary link.
(2) Why is the subject of the disambiguation page the ambiguous set of characters? I demonstrated wherefore this is not the case, so you must interrogate further or your position is unreasonable and invalid in an encyclopedia. If the subject of a disambiguation page is the set of characters, ought disambiguation pages which include similar spellings, which have different meanings but are easily confused, be split up? It is extremely common to have these together, and it is useful for disambiguation. Or are you saying that, in the page meter (disambiguation) for example, "meter" ought to be bolded by "metre" ought not be, even though many meanings for both spellings are identical? If the subject of a disambiguation page is the word, why are users directed elsewhere for the etymology and the definition?
(3) Why? Your objection has no meaning without reason. Your concerns about misuse are thoroughly unfounded, which I shown to be the case; you must demonstrate wherefore, or your position is unreasonable and invalid in an encyclopedia.
(4) Why? I have demonstrated how these suggestions are better at satisfying the purpose of a disambiguation page than the current guidelines and how the present guideline is less useful. You must demonstrate the reason that is not the case, why my reasoning is flawed, etc.
(5) How does an example that does not "correspond to reality" have any bearing on the matter? If the proposed format actually diminished ease of scanning, why not use the actual format? If the proposed format actually made scanning more difficult, then that difficulty would be shown when using the actual proposed format and would actually demonstrate a point that had anything to do with links and bolding. - Centrx 02:18, 5 August 2005 (UTC)
I am done sifting through your ridiculously verbose posts. I've tried stating my case several times now, and I'm sick and tired of every word I write getting picked apart. The "discussion" is becoming increasingly non-constructive. This is not a scientific matter where everything must be founded on infallable logic and sheets of evidence. Style is largely a matter of preference, opinion. My personal preference, in favor of the current guidelines, remains unchanged.
I see you've reverted the New England (disambiguation) that I recently cleaned up, back to your own style. Is piping links also now a part of your superior schema as well? You're going to have to argue for that change to the manual seperately, in another post. -- jiyTalk 03:39, August 5, 2005 (UTC)
Where I analyze your words and statements, it is because they are unfounded or logically inconsistent, which has a substantial effect on their validity and application to the issue. I do not pick your words apart, I demonstrate that they were apart in the first place. This is, in fact, a matter where the proper promulgated format can be concluded by logic founded on the principles of a disambiguation page, and can be shown by example. It is not merely personal preference or opinion that the essential purpose of disambiguation pages is to resolve the ambiguity of name conflicts, and that this is indicated in the Wikipedia:Disambiguation policy, where a discussion regarding that purpose belongs. It is not personal preference that additional highly relevant links connect the user to information regarding the meaning of the name, where the linked article with the exact name might be truly secondary in meaning to the additional link(s) or might be incomplete. It is not personal preference that formatting can be used to indicate which link is primary. It is your personal opinion that boldfaced links are "ugly", an opinion which I do not recall anyone else in this discussion making, but the aesthetic opinions of the general readership or population could be tested, we just have not done so or may not have the capability to do so. So, even the aesthetic quality as it pertains to the proper format of pages in the appeal of most readers, can be decided, we just have insufficient information for that decision. An individual personal preference, insofar as it comes from sentiment of a vaguely unknown cause rather than a reasonable consideration, has no bearing on what ought to be established and promulgated on Wikipedia. - Centrx 15:11, 9 August 2005 (UTC)
  • It seems that both sides of this argument are advocating an extreme. While it would be unhelpful to have a surfeit of links, it is also unhelpful to say that there must only be a single link every item line. An other link in an item line might not exactly be what the title of the page says, but that does not mean that it would not be particularly helpful to the reader, and might actually be what he was looking for in the first place, having reached that disambiguation page by entering a related search term or following a link on a page. Having additional links can actually make what the user is looking for easier to find, and the directly related link can be bolded in order to distinguish that, as it should be. - Centrx 16:45, 29 July 2005 (UTC)
  • I recently struggled with this when adding two related entries to Cary. I could have put:
Howard Cary (1908-1991), founder of Applied Physics Corporation (Cary Instruments).
but I think the way I did it (two distinct entries, with a single link per entry) is easier to read. RoySmith 17:41, 29 July 2005 (UTC)
  • Sometimes I think there should be two wikilinks. For example, on the MC page:
the second wikilink is necessary because going to Monaco is not likely to explain anything about country codes or how mc relates to it. Extra wikilinks can be useful, but in general the article will explain subtle nuances and related topics much better than a dab page can do in one line. - grubber 11:31, 2005 July 30 (UTC)
I think your example would have been better as:
That would have explained what the mysterious connection was. Anybody who was really interested in exploring ISO country codes could have typed "ISO country code" into a search box. Even better, you could have just had:
and had someplace in the Monaco article that said The ISO country code for Monaco is MC. RoySmith 12:50, 30 July 2005 (UTC)
  • But when you search for "ISO country code" you get another dab page. If you dont know what a country code is, this is going to just confuse you further. At least with the second wikilink, it is very obvious what the reference is. - grubber 15:23, 2005 July 30 (UTC)

  • The disambig page for MC isn't really organized very well because there are abbreviations, alternate meanings, and codes intermingled. As an example, the second item listed should be "Member of Congress, see Congressperson", or if it is allowed, simply a link "Member of Congress" which redirects to Congressperson.

    As for the Monaco link, it should fall under the "as part of another page" guideline in the MoS. Use "the ISO 3166-1 alpha-2 country code for Monaco" instead. —Mike 08:17, July 31, 2005 (UTC)

  • Summing up the above, the MoS has a rule that says, One link per dab entry. We should keep that. The MoS also has a rule that says, You can break the rules if you have a good reason to. We should keep that too. RoySmith 23:10, 30 July 2005 (UTC)
First, while it makes sense to split the Cary's up as you did in the example, that is a special case where both names are the same as the disambiguation title. There are many other cases where it would not work this way. Second, what is in the styleguide has no bearing on what ought to be in the styleguide. The argument here is what the styleguide ought to say, not what is actually in the styleguide. Third, the reason that it is good to not follow the rules when they are not appropriate is because it is recognized that the general rules are not always good at covering special cases. An ideal formulation of general rules would encompass all special cases, but this may not be possible. Regardless, the argument here is that current styleguide rules do not properly cover general cases, let alone special cases, and that we can implement a better formulation of the rules. A "good reason" to break the rules need not be so common that they invalidate the rules. Note also that, while in theory it might be nice to say that it's fine to break the rules if there is a good reason, the fact remains that when those rules are broken for good reason, there will be others who will revert the changes so that the page conforms to the general rules of the styleguide, and will point to the styleguide as justification when doing so. If this is disputed, the end result will be that the page will conform to the styleguide, despite the sentiment that breaking the rules may be good, or there will be another discussion in this same discussion page where it will be argued what the styleguide ought to be, which is just what this discussion should be about. - Centrx 21:49, 1 August 2005 (UTC)

Is it appropriate to set up a survey for this topic? As a member of the WikiProject Disambiguation, I would like to have this issue firmly established before we begin massive edits. - grubber 11:46, 2005 July 30 (UTC)

  • It's not a bad idea, but I doubt it will reveal something we do not already know, that there are folks at both extremes (1-wikilink vs. as-many-wikilinks-as-possible) and an underrepresented group of moderates (1-wikilink usually, but add more if it adds to the understanding of where I'm going). I've been aggressive in dropping all but one wikilink, but perhaps too aggressive in doing so. I think that a position of moderation where the position is that 1-wikilink is desirable but additional wikilinks on special terms that assist the reader in providing an alternative but related path to information. A guideline could be established that suggested any secondary wikilink associated with an entry should be to either a closely related concept or a (slightly) more general concept. For instance, take the case of AAA, a TLAdab-page, and consider the first abbreviation entry, pointing at AAA proteins. I would suggest that it is not appropriate to wikilink protein but it might be appropriate to wikilink protein superfamily ... which is a red-link, which leads to a corollary of my suggestion. If we encourage a 2-link guideline of "specific target"/"parent target", this will require more thought to be put into the crafting of the sentence fragments. The result will be a much finer product but at a significant cost in effort. As an example, the AAA proteins line now stands as "AAA proteins (ATPases Associated with various cellular Activities), a protein superfamily"; perhaps it would be as helpful or more so to write it as "AAA proteins, a protein superfamily involved in the transduction of energy from ATP hydrolysis into mechanical force", where the chief function of the superfamily is wikilinked, assuming it is possible that the reader was searching for "AAA" as a surrogate for "AAA proteins" as an instance of the general class of players in intracellular energy transduction. This type of entry enhancement can be very helpful as it adds an intentional dimension to the page; however, as can be seen from this example, it requires an understanding of the target and the context in which the target sits ... i.e. it requires domain knowledge. Therefore, I would encourage us as dab-page editors to try our best to whittle down the number of wikilinks on a line/entry but if there is a term linked that appears to be domain-specific, leave that link in place (for now). That was a long winded way of putting everything down, but I hope I've made myself clear in advocating a middle-ground established on utility to the reader that goes beyond a mind set of "the more links the better" to "a related link is ok". Regards, Courtland 02:06, July 31, 2005 (UTC)
    • p.s. a more mundane example from the AAA article ... for the entry on AAA battery, the original text read
[[AAA battery]], a standard size of dry-cell [[Battery (electricity)|battery]]

According to the guideline I suggested (which is just a more formal version of what User:Centrx and User:Grubber suggested) a suitable replacement could be 

[[AAA battery]], a standard size of [[Dry cell|dry-cell battery]].

Courtland 02:22, July 31, 2005 (UTC)

  • Although I like the style of including (mini) definitions for assisting users, the recent drive has been to scale down the dab pages to minimize the amount of text and links needed. I think your definition for AAA proteins is longer than needed, and I don't see a problem with the way it is currently written. Anyone familiar with the topic will know to choose that AAA proteins link, and anyone not familiar with it will click the link if they want to find out more about it. What I find to be more disturbing than the dab page is the articles for AAA proteins and AAA battery. The former has a link to ATP, but not ATP hydrolysis; the latter doesn't even mention "dry cell". Maybe this would be a good opportunity to improve the lead paragraphs of the articles instead of creating additional links within the dab page.—Mike 08:54, July 31, 2005 (UTC)
That "recent drive" is detrimental to the purpose of disambiguation pages. It cannot be assumed, as this drive would assume, that a person knows exactly what article they are looking for when they get to a disambiguation page and that they merely need to be shuttled along. If they searched for some random term they found while reading something in a magazine or on a webpage, or if they clicked on a link an article because they didn't know what a term meant, that person needs as much context as possible to find the information they are looking for. A person who does know what they are looking for is not going to be impeded by a longer line for some items. - Centrx 21:49, 1 August 2005 (UTC)
If we look at the middle ground, then perhaps it is worth keeping in mind that a person who arrives at a disambiguation page probably already has some idea of what it is they're trying to find. If they know it has something to do with maths, they will scan the page for maths. Of course, occasionally, they might just be trying to find out what something is — for those times, either way, they will need to go through the pages in turn, or seek help from another person; so there is nothing we can do about that. We should therefore design these on the assumption that the user already has some idea of what they're after.
Also, remember that even if you do include second wikilink, either way the user has to click a link and go to another page to figure out what it's about. Ideally, the descriptions on dab pages shoudl be done using simple enough words, or at least a field that everyone knows, so we shouldn't need to link any complicated words.
I'm still convinced that no links apart from dab ones are needed, but if we must include some, I think we should have at least some faith in the reader; if they really know nothing, then they should be expecting a bit of a search, and those links probably wouldn't help them anyway. Neonumbers 09:31, 1 August 2005 (UTC)
See my above comment. Also: math is not a good, generally applicable example because it is a highly specialized field, so there are unlikely to be multiple math-related terms with the same name (still, though, I can think of many) and readers will not confuse something they know with some abstruse concept in math. With more common subjects, however, the context which comes from having some information about the term is more useful to more people. If someone is reading a history book, or Wikipedia article, and searches or links for a term, they will not necessarily know what field the term on such general terms as "mathematics", "politics", or "war" (for it is a history book which might cover many subjects) but they are sufficiently familiar with common knowledge about diplomacy or armies that more context is helpful. Further, it is a huge, and wildly incorrect, assumption that all readers either have so much information that they need no help getting to the right article or have so little information that context in the disambiguation page would not help them get to the right article. - Centrx 21:49, 1 August 2005 (UTC)
  • Reading through responses I see that there might be an underlying reason that I feel that explanatory text is useful: the imperfect nature of article titles. If we assume that an article title has sufficient information for an informed reader to understand the content of the article to the extent that it facilitates decision making around whether or not to read/consult it under a particular circumstance, then additional explanatory text (including secondary wikilinks) is superfluous on a dab-page. There are many cases where this is true ... for instance, User:Smack recently made revisions to a page I revised, both in the name of Wikipedia:WikiProject Disambiguation; lookinat at the diff (see [1]) it is apparent that the main differences are trimming explanatory text down to a minimum, which is justified assuming the titles of articles play the role they should (this an opinion .. I looked briefly for a specific guideline/policy that recorded this best practice and did not find it). What I'm getting at is that the notion of whether to add explantory text and/or wikilinks might be best based on the quality of the article title to which the dab-page entry refers. What are thoughts on this being a practical, workable guideline? Courtland 10:59, August 1, 2005 (UTC)
* I agree, that page does not have enough information. It seems useful for every entry to have 5-20 words describing it, depending on the complexity of the subject. - grubber
We've wandered somewhat off topic in this thread, but I agree with Courtland that there's some judgment involved in determining how much explanatory text (if any) is needed in addition to the article title. I don't see a need for any additional rules, though. At most we could suggest a minimum of explanatory text.
I'm going to boldly assert that we've reached a consensus to keep the "don't wikilink any other words" statement and remove the "disputed" tag from the article.—Wahoofive (talk) 15:54, 1 August 2005 (UTC)
This is far too bold an assertion and really doesn't reflect the discussion. I'm putting it back. - Centrx 21:49, 1 August 2005 (UTC)
The discussion or your discussion? It seems that, aside from your lengthy objections, most people lean towards the current guideline (while also observing the "break rules" guideline) -- jiyTalk 04:03, August 3, 2005 (UTC)
No, I do mean the discussion. At the time of the tag removal there were two comments advocating more relevant links that had not been replied to at all. While this may seem minor in contrast to the five or six others who advocated having only a single link for each item, the major reason why 'concluding' the discussion on 'consensus' was invalid is that the many of the comments had been made in the last day or two, and not a single comment had been made more than three days ago. This is insufficient time for discussion; in that time a regular participant in this page might not even visit Wikipedia. If there had been no objection to current policy at all, maybe closing it would have been more appropriate, but the fact remains that there were four users who questioned current style or argued for a change. That's hardly consensus for the opposite position. Further, that users largely might not respond to my comments, which would be the only reason this discussion could ever be considered just my discussion, is more reason that the discussion is not concluded. My comments are eminently reasonably--there may be disagreement with them, but that does not entail that they are empty or inflammatory and so worthy of being ignored. That they are not responded to means that they represent an outstanding, unresolved issue, not that they are void of relevance to the issue. - Centrx 15:27, 3 August 2005 (UTC)

Summary

I think the guideline ought to remain as supporting one wikilink per line

  1. grubber
  2. RoySmith
  3. Courtland
    -- How is this "vote" here when the last comment by this user indicates support of more relevant linking? - Centrx 16:26, 2 August 2005 (UTC)
    comment in response to User:Centrx: the current guideline suggests one wikilink per line ... I support that, but it is a guideline and not a policy and can justifiably be broken. We do not straightjacket ourselves with rules, merely straighten ourselves.
    But the problem is that I have seen people systematically go through Wikipedia removing links on the basis on this guideline and citing it as law, so clearly some people do see this guideline as a "straitjacket," as you put it. —Lowellian (reply) 01:38, August 16, 2005 (UTC)
    I suppose I'm one of the people referred to by Lowellian, as I've been doing a lot of mono-link-ification of existing dab pages. I see the guideline as neither a straightjacket nor a law, I see it as a style guide, and I generally use an edit summary of something like "Conform to Manual of Style". Had I felt differently, I would use something like, "One link per line: it's not just a good idea, it's the law!". Style guides are just that, guides. They can be ignored when appropriate (and, in fact, this one even explicitly states that you can do that). That being said, I suspect I have a higher standard for when it's appropriate than many others. RoySmith 02:07, 16 August 2005 (UTC)
  4. Josh Parris
  5. jiy
    Whether being placed into this category is considered a "vote" or a "summary", I consider it a valid reflection of my opinions -- jiyTalk 04:11, August 3, 2005 (UTC)
  6. Neonumbers
  7. Wahoofive
  8. Mike
  9. Smack (talk) (Disclaimer: I have read virtually none of the discussion here)
  10. Eugene van der Pijll
  11. Jerzy·t The existing guideline is right to put the burden of making an arguement for an add'l lk on the editor who wants to include one. And no, no vote is needed, but this lopsided straw poll should be valuable in identifying this fringe idea as such.
  12. Reinyday -- I fully agree with only one wikilink per line, the one that goes to the article being disambiguated.

I think the guideline ought not support a norm of one wikilink per line

comment: the guideline does allow more than one link because it is a guideline and can be justifiably violated, but it indicates that the norm should be one link per line. I have changed the section name to reflect this Courtland 00:03, August 3, 2005 (UTC)
  1. mikka
  2. Centrx
  3. Quuxplusone (notice that the current practice is to use wikilinks where appropriate anyhow, MoS or not)
  4. Lowellian
  5. Hyacinth 08:23, 16 August 2005 (UTC)

I think a vote now is ridiculous

  1. Centrx

I think that this wasn't a vote, it was a summary for such a stupidly long discussion - and votes ought to be signed and dated by the voters at the least rather than tallied by some third party

  1. Josh Parris
A summary, by definition, contains the chief points or the sum and substance of a matter or the general idea briefly. The above list, however, was a tally of advocates on two sides, as though they were teams in a sporting event. Unlike a summary, such a tally holds no information on the process of the game, or its outcome. While in a sporting event looking at the familiar players on the team rosters might give someone an idea of the match up and who will prevail, here instead is a game in which the players may be unknown, and in which the team size is unlimited. Any one may join a team without any test of their fitness, for the only test in this game is the reasonable discussion that is not included in this "summary". A true summary provides a reader with an understanding of the substance of the matter without having to wade through the prologue of argument and counterargument. This "summary" holds no such information; what would be the disadvantage in deleting it? It is either a vote, which is preliminary, or it is absolutely useless. - Centrx 03:52, 3 August 2005 (UTC)

Objection to such superficial voting

A vote listing of this sort is far too preliminary and not useful at this stage of discussion, where there are scarcely responses to main points. If the "voters" do not provide reason for their objection to proposals in the discussion, and one even admits not having read the discussion, the votes are meaningless. This is not a forum where a hasty decision is necessary, which would be the purpose of having a majority fiat when discussion is no where near resolved. There is plenty of time for actually resolving the issue, for this is a pretend vote and does not actually summate anything. This so-called "Summary" is rather a generalization of more subtle, and even unrelated areas of argument into a "do not change" and "change" camp, disregarding that there are many possible changes indicated in this discussion. It furthermore purports to categorize commenters into these general groups which, for their inane generality, make it seem that a person has so crystallized his thinking as to have reached a decision and classifies people without their consent; it even appears that one person is classified incorrectly. - Centrx 16:26, 2 August 2005 (UTC)

  • Your comments are becoming unhelpful (in my likely minority opinion); rather than trying to break the organization of the comments, please try to improve it. Rather than disparaging, try to support some action .. unless you are opposed to any action, which itself is a valid suggestion. Courtland 16:33, August 2, 2005 (UTC)
I'm not sure what you mean by breaking the organization of the comments. If you mean that this list of users on sides is to serve as a contents of which users support and oppose which arguments, that is different from a vote, but it still ignores all substantive discussion, where a binary categorization is inaccurate. How have I become unhelpful? I don't think I've attacked anyone, only various arguments, and dispassionately so. Further comments by me may be less useful without an overarching, discriminated structure to guide the discussion, but I don't see how the above "Summary" section has any use either. Creating such a propositional structure is something I will soon do, but I do admit it is a smaller and easier task to respond to individual comments than to synthesize all relevant discussion on the page and generate a formal proposal. I am clearly not opposed to any action; I have proposed positive changes and almost all rejoinders I have argued have been in opposition to keeping everything precisely as it is, that is, taking no action. - Centrx 00:07, 3 August 2005 (UTC)

New page with extracted comments

Wikipedia talk:Manual of Style (disambiguation pages)/Wikilinking

  • this page contains quotations in support of several options related to wikilinking. This is meant to help organize thinking on the matter.

Courtland 10:47, August 5, 2005 (UTC)

Underlying philosophical issue

The question of multiple linking on dab pages stems from a philosophical question: is the disambiguation page:

  1. a special-purpose page with the singular purpose of guiding a user who may have typed an ambiguous term in the search field (or clicked on a link with an ambiguous name) to the page s/he was expecting
  2. an article page like any other in Wikipedia, allowing users to explore the wide universe of human knowledge

Both have valid arguments, but the first view has been Wikipedia policy for a long, long time (far predating this MOS). So the burden is on those who wish to change it to generate a consensus to do so.—Wahoofive (talk) 16:24, 3 August 2005 (UTC)

This is not the issue at all. First of all, the special purpose of "guiding" a user is subordinate to or included in the greater purpose of "allowing users to explore" knowledge. The reason for getting the user "to the page s/he was expecting" is to expedite that reader's exploration. So, the ends of the above two "philosophical questions" are the same. Similarly, and this seems to be the core of the misunderstanding, the argument proposed here is that the special guiding purpose of disambiguation pages is best served by the inclusion of more, but not excessive, relevant links. Users who types "an ambiguous term" in the search field, or clicks on a link with an ambiguous name, are, in disambiguation pages with link, led more efficiently to the pages they were expecting. Corrollary to this argument is the assertion that a user who searched or was linked to a disambiguation page is not necessarily and exclusively looking for an article with the same title as the term searched or linked.
Contrary to your above statement, it has in fact been Wikipedia policy for a long time, far predating this styleguide, to have more than one link in disambiguation pages. It remains common, maybe predominant, practice. In looking through the archive of this discussion page, there was no substantial discussion and there was no consensus formed for allowing only a single link. The only argument against more than one link is that it would produce some ill-defined "clutter", as though a Wikipedia page could be similar to a room randomly filled with nondescript junk that has to be stepped over and searched through to find the one wanted item, rather than a webpage where a whole screenful is immediately apparent to the eye, where there are there are mere few additional links, and where it is possible to distinguish those links from the primary links. Thus, a Wikipedia disambiguation page with more relevant links is more like a map with a few items, some distinguished as being more important, where a mere press of a button will retrieve the item selected. So, if you agree with what you said above, it is the burden on those who wish to change longstanding practice to generate a consensus to do so, and the styleguide ought to be changed to reflect the common, reasonable practice supported by many. - Centrx 00:18, 4 August 2005 (UTC)
Your argument is a straw-man argument. What I said above was that the view that dab pages are for quickly and easily guiding users to the page they were looking for has been longtime policy. When we created this style guide we made decisions guided by that principle. It's true that many preexisting dab pages have multiple wikilinks; the reason this style guide was created was because of the large variation in page style for dab pages. Why is "clutter" an invalid argument, when it's obviously a shorthand for principle #1 above? We solicited input on WP:Surveys, WP:Policy Proposals, Village Pump, and the Dab policy page, and managed to reach consensus without the use of {{disputed}} tags. So far, I'm only hearing one dissenting voice from the consensus we came to then: User:Centrx. How long must a {{disputed}} tag remain there to satisfy one disaffected user? I'm open to change if Centrx can generate some more support. —Wahoofive (talk) 16:32, 4 August 2005 (UTC)

I have been asked to come and voice an opinion. My opinion is: this is all a matter for case by case editorial judgement, and anyone who desperately needs a written rule on the subject (a) considers others to have no judgement (b) probably lacks it themselves and has problems with a world with shades of grey (c) anyone who thinks this is a matter for a vote is very possibly on crack (disambiguation) - David Gerard 23:14, 4 August 2005 (UTC)

The fact is that there is a written rule on the subject, and that written rule is enforced by some users by reason of it being the written rule. I, for one, would be happy if that written rule were removed - Centrx 02:18, 5 August 2005 (UTC)
"One" is the most important word in the above statement. There's a huge amount of text on this talk page from one user. As I've said before, I'll be happy to reconsider any of these guidelines if there are numerous people dissatisfied with them, but so far you've convinced no one and gotten little support. —Wahoofive (talk) 00:06, 6 August 2005 (UTC)
I responded to a comment that disparaged a desperate need for a written rule, so it would seem that the commenter would also be happy or at least not displeased if there were no written rule. Note also that there has been no discussion here on whether or not there ought to be a written rule, so the level of agreement with that is unknown.
Rather than reconsidering these guidelines if there is support, you ought to consider the proposal on its merits alone. If the proposal is reasonably compelling then you ought to support these changes regardless of the number of users expressing agreement (of course, compromise in the actual implementation is reasonable, but that is unrelated to support of the merits of an argument), and if the proposal is unreasonable then you ought to demonstrate why it is unreasonable and oppose it regardless of the number of users expressing support. This argument is not about me and you, and this and that person, and the guy who said something 6 months ago; it is about the proper format of a disambiguation page in light of the purpose of Wikipedia and of disambiguation pages. - Centrx 15:11, 9 August 2005 (UTC)

More on multiple wikilinks

Just adding my 2¢: I wholeheartedly agree with the sentiment that extra wikilinks on disambiguation pages are a bad thing. The policy is to change all wikilinks to disambiguation pages to link directly to the article; therefore, any time a user arrives at a disambiguation page, it is because she has searched for a term with multiple meanings, she clicked on a link that was not properly disambiguated, or she clicked on a link to a _(disambiguation) page on a main page because the main page was not what she was looking for. In all three cases, the user is looking for a particular thing and has not found it. It is therefore our duty to help her get to where she wants to go and not lead her astray with distracting extraneous links. Therefore, any wikilinks on the page that have zero probability of being the page the user is looking for should not exist, because they will just distract her in her purpose of finding the page she seeks.

Extraneous wikilinks are especially pernicious when the disambiguation links are piped. This happens because when the disambiguation link is piped the piece of information that the reader sees that indicates which item they want is not the piped link at the left but the extraneous wikilink that performs the disambiguation. For example, I recently fixed Sprite (disambiguation). It had a line that looked like this:

I bet that a large fraction of users that stumbled onto that page clicked on preternatural or legendary creatures and not sprite. People tend to click on the thing that gives them the information they want, and in this case, the thing that gives them the information they want is the part that says "preternatural legendary creature". The current handling is much better because users will always find the page they were looking for:

  • Sprite (creature), a class of preternatural legendary creatures commonly associated with elves, fairies, pixies, and spirits

Not only does the disambiguation link itself contain enough information to indicate it's what the user is looking for, but there are no extraneous links that might lead the user away from the page she seeks. If the user really wants to find out what preternatural legendary creatures, there are links on Sprite (creature). The current disambiguation page policy makes for infinitely more useful disambiguation pages and I definitely recommend against extraneous wikilinks on disambiguation pages because they are almost always traps that lead users away from what they're looking for, not towards it. Nohat 09:20, 16 August 2005 (UTC)

I moved the dab page to Sprite since that was only a redirect to it. —Mike 18:03, August 16, 2005 (UTC)

HMS Grafton

  • Just to throw in another couple of cents... The disambiguation pages I most frequently deal with involve Royal Navy ships (see, say, HMS Grafton for a random example). There's many hundreds of these articles yet to be written, probably, and the brief descriptions often don't make much sense. If we don't have an article for a ship, we may have an article for her class, or for the particular battle she was notable for serving in (mentioned in some entries, eg/ "HMS Foo, launched in 1907 and sunk at the Battle of Jutland") - and it's often very useful to have that linked as well, to allow somewhere for a reader wanting to know about that ship to go; these pages are often quite useful. In addition, a lot of ships are simply not notable enough for their own page ("served for twenty years, never saw combat, sold 1872") and so they get a brief one or two line entry on the disambig page instead - which is effectively their normal entry, so it's wikilinked as usual. This sort of thing may be an unusual use of disambiguation pages, and a special case, perhaps - but it's certainly a use where multiple wikilinks are useful. But, in general, I do agree with David Gerard above... Shimgray 00:33, 17 August 2005 (UTC)
There's already an exception for redlinks. —Wahoofive (talk) 02:37, 17 August 2005 (UTC)
I notice that HMS Grafton doesn't even conform to the index template shown at Wikipedia:WikiProject_Ships#Index_page_template. I would think that a better way to put it together would be as follows:

HMS Grafton is the name of nine ships of the Royal Navy:

Also, a destroyer ordered in 1944 was to be called Grafton but was cancelled in 1945.

This would do a better job of conforming to the current guidelines and still provides enough context for a viewer to figure out which ship he is looking for. —Mike 07:02, August 17, 2005 (UTC)
Properly speaking, this isn't really a disambiguation page, since so many of the entries don't lead to other Wikipedia pages. —Wahoofive (talk) 23:41, 17 August 2005 (UTC)
Retrieved from "https://en.wikipedia.org/w/index.php?title=Wikipedia_talk:Manual_of_Style/Disambiguation_pages/Archive_5&oldid=617460782"