User talk:Tony1/Build your linking skills
Tutorial on linking skills
edit[Copied from Hans's talk page]
Hans, a little while ago I raised this issue at WP:LINK; you commented that it might be a good idea, with the rider that sufficient leeway should be given to the "grey" areas.
I've made a start. If you have time, would you mind providing feedback? (In addition, please let me know if the font-size is a little too small within Gary King's editing exercise templates; if so, I'll ask him to tweak it.) Tony (talk) 13:35, 26 July 2009 (UTC)
- After just a quick look I am not sure that this is the right format for this information. In the overall context of Wikipedia instructions it's a bit eccentric. It creates a teaching atmosphere that is not very wiki-like and which I think is likely to affect its efficacy. Even the most willing reader would need some patience to read through everything. Now imagine the reactions of an overlinker who doesn't like what they read.
- For illustration, here is what happened to me when I went through the first set: Read the problem text. Thought: Some pretty irrelevant links there, but also potential for a better one. Surely we have an article on the general concept of a minimum age? (We almost do, at Legal age, and also some country-specific articles such as Minimum legal ages in Belgium.) // Read the issue. Oh, it's only about the overlinking. OK. Hmm, public is stupid. Law and constitution are a bit silly. Either of vote and election is redundant. // Read the solution. Aha. Neither vote nor election? Seems a bit extremist as this is such a close topic. Basically we are in a subarticle of election. And what's the link to constitution still doing there? The explanations didn't convince me, and now you can imagine how motivated I was to continue.
- Here is an idea that may not be feasible at all, but so what. How about a screenful of text, carefully crafted to span many typical areas of Wikipedia. Perhaps someone narrating what happened in a dream. No links at all. Give this to a dozen of our best editors, and let them create the links. The are allowed minor rephrasing to prevent link clashes. When they are finished, they may look at each other's work and amend their own, if they so desire. Then produce a kind of majority version, with areas that they don't agree on clearly marked and annotated.
- Unfortunately I am not going to do anything like this any time soon because I am first flying to a conference very soon and then moving house over a distance of 1800 km. For the same reason I won't be able to give you much feedback in the near future. Hans Adler 14:24, 26 July 2009 (UTC)
- I must say that I was put off by the overwhelming negativity of this feedback. "Eccentric"? A teaching atmosphere that is "not very wiki-like"? I can't imagine what you were expecting—let's all be teacher? It's meant to be a show-and-tell resource in which the problem gradually unfolds. This has worked very well in most of my other tutorial pages.
- "Imagine the reactions of an overlinker"—it's not designed for hardened maximal linkers. No one will ever convince them to moderate and take a skilled, selective approach. They want to link as much as possible, and that is that. This tutorial is aimed at the vast majority of WPians who would appreciate exposure to the issues involved in the decision to link, and how to link.
- I take your point that it's over-wordy. I'm going to try to do something about it; it's a problem in that there's quite a lot to explain for each instance, and I'm unsure I want the samples to be smaller.
- "Legal age" might be a good link, and I'll investigate this point.
- Your comment that "Oh, it's only about the overlinking" assumes that all matters will be dealt with in all exericses; but there are too many for that. And perhaps you and I differ here, but my firm opinion is that overlinking is a much more common problem than underlinking. Yes, I will include a few underlinkings and discuss them. In addition, I want readers to ponder what is improveable in each sample before giving away the precise problems. That unfolding aspect in critical in making people think. Tony (talk) 03:58, 6 August 2009 (UTC)
A few comments by A. di M.
edit- Examples could give a little more context. For example, you could bold "voting age" in the first one to make it clearer that it's the article about it; in the penultimate one, you could include the previous sentence ("His father, Gille-Brighde of Galloway, ...") so that the reader already knows who Uhtred is and that we have already linked to the article about him. (I'm thinking of a way of showing the reader that the article about the Beatles has already been linked to in the second example, but none occurs to me right now.)
- In the first example, "capacity" could be linked to "capacity (law)". Sure, that's very similar to the ordinary "dictionary" meaning, but it is slightly more specific, and considering it'd be the only link in that sentence, that would not make a "sea of blue".
- I'm not 100% sure that "class" in the last example should link directly to the Germany section: that section starts out giving obscure details about the German classification system without even saying what a class of ship exactly is; OTOH the lead of that article is short enough that the "1.5 Germany" line of the TOC will likely be in the same screenful as the top of the article, so that a reader could get there with just one click, after reading the general definition.
- You could also show examples of terms which should be linked but aren't. I've encountered these often enough, usually in supposedly introductory articles on mathematical topics where the writer might not realize that a term they use everyday could be totally alien to common mortals. This is especially serious when such a term is a everyday English word, but used with a well-defined technical meaning which is not identical to the "ordinary" one. --A. di M. 20:35, 26 July 2009 (UTC)
Two more points:
- In the Donnachadh example you point out the awkwardness of having a nominal group in half black and half blue, and in the first example you suggest "constitutional provision" which (to my eye, at least) is considerably uglier than "King Henry II of England"? (Maybe "costitutional provision" might redirect somewhere; if not, a slight rewording such as "... set by constitution" might be OK; but that would require changing the actual prose and would then be out of the scope of this tutorial—unless it is changed in the problem text, too.)
- "[M]ost all of the links": I think you mean "almost all of the links" or "most of all the links".
One more:
- Why'd you pipe the acronym into the link in the RFID example? I'd go with "by radio-frequency identification (RFID) smart card", for the KISS principle. --A. di M. 23:26, 10 August 2009 (UTC)
- I'd link whichever term is more commonly used ("radio-frequency identification" or "RFID"), but if the latter, I wouldn't include the parentheses in the link, ie (RFID) not (RFID). Adrian J. Hunter(talk•contribs) 02:09, 13 January 2014 (UTC)
Feedback by Sebastian
editThank you for inviting my critique! I love the three tests - the names are catchy and relevant. Maybe we could work that into WP:LINK. I also like the idea of the unfolding hints exercises.
So, now to my concerns:
It takes patience to go through the tutorials; I don't know how many people have the time in our fast-paced world to really think through the problems. I must admit, I cheated, too, in the battleship example.
Without having seen the feedback you mentioned, I can imagine that negative reactions can be caused by the underlying problem : People put links in the article because they feel they're improving the article that way. Wikipedia lives from its volunteers, and some people find great satisfaction in adding links. At first glance, your tuturial seems to be only an appeal to remove links, which is a direct attack on the work many editors are proud of. I agree with you that on average we have too many links, but I think you're going too far. You seem to overlook the words "Unless they are particularly relevant to the topic of the article". For example, I think the term "voting" is particular relevant for the article on "voting age". I even would add a link to the Beatles in the second example. Not because I think anyone wouldn't know what or who "voting" and "the Beatles" are, but because it is very likely that someone who is interested in "voting age" is also interested in "voting", and because I myself often went from an article about a musical piece to the article about the artist or composer.
Details:
- Overlinking is only one issue of many. The lead talks mostly about overlinking. I would move that out in a subsection.
- The instructions don't match all exercises; in particular, it is confusing that they don't match the first (there is no hint and no numbers).
- Color coding of links (black/blue) is against WP style. (See Template:Xt#Usage for a nice explanation.) Maybe you could use strikeout (<s>) or invert the color, as in white text on red background.
- Piped link: I also used to write many piped links like "[[Ship_class#Germany|class]]", but from what I just learned on WP:LINK, this would be an ideal case for a redirect page named something like "German ship classes".
- Actually, in the battleship example, I just realized that you're not as gung-ho about removing links as I thought: I would have considered removing the link to "battleships", since it is indirectly connected with the article through "dreadnought". (But I decided it's better to leave it in, per the principle of least surprise. Some people may not know that dreadnoughts are battleships.) This brings me to the last point:
- In your uniqueness test, you speak of "indirectly" reachable. I don't know what exactly mean by that. (Almost by definition, all articles are indirectly reachable from each other. At first I assumed that you meant "1 step removed".) But I now feel that even 1-step reachabilities may have a reason to be linked, as in the "dreadnought" case above. That needs some more thought.
I hope this helps and I didn't discourage you from writing tutorials - thanks for your commitment! — Sebastian 03:42, 2 August 2009 (UTC)
- In the Beatles example, the actual article does link The Beatles in several sections before that one, but, as I told above, I can't think of a way of showing that in the example. (Perhaps, bite the bullet and pretend it isn't, linking "the Beatles" in the problem text?) As for "indirectly reachable", I think it's supposed to refer to the obvious case. (Everybody would expect the first line in Springfield, Illinois to link to Illinois, so there's no point in writing
[[Springfield, Illinois|Springfield]], [[Illinois]]
. But for the dreadnought/battleship case this doesn't hold.) --A. di M. 17:31, 4 August 2009 (UTC)
- Since this occurs only in one case, I'd bite the bullet. (You could also add a note like "The following links have already been introduced in the text before this example...", but that might distract readers, so I would only do that if you are planning to add more exercises in which the same situation occurs, and you can establish a pattern for such notes.) Regarding the Springfield example: I think the best description for that example is not "'Illinois' is indirectly reachable from 'Springfield, Illinois'", but simply: "The word 'Illinois' is already contained in 'Springfield, Illinois'." — Sebastian 22:39, 4 August 2009 (UTC)
Clearly, the exercises are too wordy. And there's always that larger context problem. It might be possible to insert a permalink to the article from which an excerpt is drawn, advising the reader to look at X number of paragraphs. But only where larger context really matters. Or one could assume that each example is at the top, or close to it. Tony (talk) 12:10, 5 August 2009 (UTC)
- No no, I don't think they're too wordy! If your intention is to teach people how to look at texts and identify appropriate links, then you have to present them with some lengths of text to look at. And I don't think it's necessary to take only text from the top. If it matters for the exercise where the text is from, you can say so explicitly and tell the student not to worry about links to X, Y, and Z as they already have been introduced before. If it doesn't matter, don't bother the student with that detail, and tweak the exercise so it remains on focus. But at some point, you should also have a whole-article exercise, and for that, a permalink may indeed be more appropriate than copying the whole article. — Sebastian 16:28, 5 August 2009 (UTC)
- I'm glad you think so. But all the same, locating the explanations (which are often long) into their own clickable section at the bottom of each exercise might do the trick. I think allowing users themselves to control the unfolding of the exercises (to use your word) is good, to avoid flummoxing them with too much at once. Tony (talk) 11:04, 11 August 2009 (UTC)
Learning from greengages
editThanks to whoever used Greengage as an example at WT:LINK, I looked up that article and found the following sentence:
Soon after, Greengages were cultivated in the American colonies, even being grown on the plantations of American presidents George Washington (1732–1799) and Thomas Jefferson (1743–1826). However, their cultivation in North America has declined significantly since the eighteenth century.
How many links would you cut out of this sentence? When asking myself this question, I realized that I would not only cut the link to "American presidents", but the whole word, too. The fact that they were presidents has nothing to do with the plant, and it is highly unlikely that someone who is reading about greengages would suddenly become interested in the article on American presidents. So you may want to include such considerations in your tutorial, too.
But thinking further, I don't think any of these links are actually particularly helpful. The one link that would really be helpful would be American colonial agriculture. Since that article doesn't exist, it would be a good candidate for creating a redirect or disambig page. It would be interesting to learn about that too, but I don't know if that exceeds the scope of your tutorial. — Sebastian 18:40, 5 August 2009 (UTC)
- Sebastian and A. di M.—see what you think of the new exercise based on this. Tony (talk) 03:41, 6 August 2009 (UTC)
- I wouldn't remove the text "American presidents": if Washington and Jefferson had been just two other random farmers, we wouldn't start that part of sentence with "even" and probably we wouldn't mention them at all. I would link George Washington and Thomas Jefferson: there will be many readers (more than you'd expect) who barely know anything about these men's lives; then the link to President of the United States is redundant as both their articles link to it in the first sentence, as they should. Colonial America, redirecting to Colonial history of the United States, is not such an awful link; it also has a #Farm life section, although it starts in a way too abrupt way for me to want to link directly to it. (I agree that I'd lose all the other links in the paragraph.) --A. di M. 17:04, 6 August 2009 (UTC)
Interesting that between the three of us, we all have different opinions on this. And it's not even that you can arrange us on a linear scale from link-inclusionist to link-deletionist. This indicates that linking is not so much a question of skill, but rather like an art, where there often is no right and wrong. It might be interesting make a little survey, maybe something like the CSD Surveys, to see what other experienced editors would keep or delete.
I will create a redirect from American colonial agriculture to [[Colonial history of the United States#Farm life]] and put that in the article. — Sebastian 18:01, 6 August 2009 (UTC)
- Just to add my thoughts here, I would also remove both the link and the text "American presidents". I would definitely leave the links to Washington and Jefferson however, in order to answer the exact question which A. di M. posed above: "who are these guys? Should I care about them?" Incidentally, I would also remove the birth and death date ranges from the text as well. The reason why is because this is exactly where links have power: the information is there for those who desire to find it. Their birth and death dates are not directly relevant to this passage. The reason it is included here very likely has to do with the fact that printed works regularly will use that style because they can't link. Print media needs to include it as a rough disambiguator ("yes, we're talking about that George Washington!").
- Basically, what I'm getting at here is that even links which may not be immediately relevant to the passage they are included in, they still may be worthwhile. Reading and writing are both different on computer screens, and it takes some adjusting to get used to (one fundamental difference is that computer monitors are "active", they push light at you, where paper simply reflects light back to you). Anyway, yea, linking is very much an art form... try not to be prescriptive about it, is my recommendation.
— V = I * R (talk) 10:44, 11 August 2009 (UTC)- I agree with some of what Ohms law says, but I am cautious about "even links which may not be immediately relevant to the passage they are included in, they still may be worthwhile." We have to draw the line somewhere, and allowing links that are not clearly relevant to the understanding of a topic is a slippery slope. Yes, there is room for editorial discretion, but every link carries a cost that must be balanced against its utility, on average, to the likely readers. On "American presidents Georde Washington ....", can there be any doubt as to which Goerge Washington is referred to? And is the article on Washington even vaguely relevant to the vegetable?
- Sebastian, the "Farm life" section-link is a good one. I'm going away for 10 days, and on return will have time to survey all of this feedback. Thank you. Tony (talk) 11:00, 11 August 2009 (UTC)
- First, be sure to emphasize "may" in the sentence "even links which may not be immediately relevant to the passage they are included in, they still may be worthwhile.". That leads nicely to rehashing the point that linking is an art form. I don't really disagree with what you're trying to get across, but the perception of "correctness" itself changes over time, and the "level of (over/under)linking" varies greatly from person to person. You're obviously firmly in the "less is more" camp, which is a perfectly reasonable and valid stance. I used to be there myself, and may have been even more "right wing" from your current position then you currently are. My view on this issue has moderated significantly in the last 5-10 years, and I'm willing to bet that when you revisit all of this in a couple of years that you'll be surprised at the level of your own dislike of linking. Finally, to answer the question about "can there be any doubt?" which you posed, the answer is a resounding yes. I know how surprising this can be to some of us (In a case like G. Washington this is really a cultural thing), but you need to kind of let go of your assumptions to see that not everyone will automatically know things. It's tough to do sometimes, but editing an article so that it makes no assumptions about existing knowledge is a goal towards reaching FA status.
— V = I * R (talk) 11:22, 11 August 2009 (UTC)- "Rightwing" meaning ... prescriptive? Controlling? If so, WP is always going to be a fluid mixture of control and freedom, rather like societies. I'll go through your suggestions (and any further ones on the exercises) when I return, Ohm. BTW, have you ever been to the French or Italian WPs? They have virtually no guidelines on wikilinks (although they do on external links—at least WP.fr does); the result is an unholy mess of blue in many articles, and a ruining of high-value links through their swamping with the trivial that almost no one would ever hit. Some people think I'm anti-linking, but I see myself as a keen supporter of wikilinking, and specifically want to ration it so that it works optimally. Eng.WP is lightyears ahead of the others, through pressure on editors to be more selective, although we have a long way to go on our less prominent articles. Tony (talk) 12:16, 11 August 2009 (UTC)
- I wouldn't say "controlling", and "prescriptive" deals with the application of views, not the views themselves (although, the two are inextricably linked, one does not automatically follow the other). I can't really sum it up in one word, but I wold associate "the right" with overlinking (conservativeness in linking) and "the left" with underlinking (liberalness of linking). As you mention in the opening here, this is all about defining the "grey area". Anyway, I haven't really looked around at the foreign language Wikipedia's, but I can imagine that some of them are a real mess. There are many of the nearly 3 million articles here on en.eikipedia that are a mess (although it's not really a systemic issue here, since there are so many participants).
— V = I * R (talk) 17:32, 11 August 2009 (UTC)
- I wouldn't say "controlling", and "prescriptive" deals with the application of views, not the views themselves (although, the two are inextricably linked, one does not automatically follow the other). I can't really sum it up in one word, but I wold associate "the right" with overlinking (conservativeness in linking) and "the left" with underlinking (liberalness of linking). As you mention in the opening here, this is all about defining the "grey area". Anyway, I haven't really looked around at the foreign language Wikipedia's, but I can imagine that some of them are a real mess. There are many of the nearly 3 million articles here on en.eikipedia that are a mess (although it's not really a systemic issue here, since there are so many participants).
- "Rightwing" meaning ... prescriptive? Controlling? If so, WP is always going to be a fluid mixture of control and freedom, rather like societies. I'll go through your suggestions (and any further ones on the exercises) when I return, Ohm. BTW, have you ever been to the French or Italian WPs? They have virtually no guidelines on wikilinks (although they do on external links—at least WP.fr does); the result is an unholy mess of blue in many articles, and a ruining of high-value links through their swamping with the trivial that almost no one would ever hit. Some people think I'm anti-linking, but I see myself as a keen supporter of wikilinking, and specifically want to ration it so that it works optimally. Eng.WP is lightyears ahead of the others, through pressure on editors to be more selective, although we have a long way to go on our less prominent articles. Tony (talk) 12:16, 11 August 2009 (UTC)
- First, be sure to emphasize "may" in the sentence "even links which may not be immediately relevant to the passage they are included in, they still may be worthwhile.". That leads nicely to rehashing the point that linking is an art form. I don't really disagree with what you're trying to get across, but the perception of "correctness" itself changes over time, and the "level of (over/under)linking" varies greatly from person to person. You're obviously firmly in the "less is more" camp, which is a perfectly reasonable and valid stance. I used to be there myself, and may have been even more "right wing" from your current position then you currently are. My view on this issue has moderated significantly in the last 5-10 years, and I'm willing to bet that when you revisit all of this in a couple of years that you'll be surprised at the level of your own dislike of linking. Finally, to answer the question about "can there be any doubt?" which you posed, the answer is a resounding yes. I know how surprising this can be to some of us (In a case like G. Washington this is really a cultural thing), but you need to kind of let go of your assumptions to see that not everyone will automatically know things. It's tough to do sometimes, but editing an article so that it makes no assumptions about existing knowledge is a goal towards reaching FA status.
Linking, delinking, changing links, and link maintenance exercises
editThis is a good set of exercises for spotting and removing unnecessary links and improving link syntax (and piping of links). What I think is missing is exercises that point out when links need to be added, and how to spot and correct incorrect links (I call the latter link maintenance). Examples are: spotting when a redlink has incorrectly turned blue (when someone creates an article without checking what links to it); spotting when a link has turned into a disambiguation page without the incoming links being disambiguated; spotting when a link has turned red because an article was deleted (OK, that is a trivial example); spotting when a blue link is incorrect (links to the wrong article), and so on. In addition, exercises where a word that should be linked, but hasn't been linked, should also be included. Rather than have separate exercises for tackling overlinking and ones for dealing with underlinking and ones dealing with link maintenance, is it not possible to combine them and edit them collaboratively, or add suggestions to this one? Tony, would it be best to set up something separate, or make suggestions here? Carcharoth (talk) 20:41, 23 August 2009 (UTC)
- Yes, it's a stub, really. Thanks for your suggestions, Carcharoth—I'd already identified the need to write exercises for the correction of underlinking (although I see it as a much smaller problem than overlinking). Also the Nickj script might be mentioned in the lead, in the context of being selective (that is, IF we can get it past the syntax errors it throws up at the start). I'm unsure what material to use for the changed-red-links idea: do you have any examples? Tony (talk) 23:28, 23 August 2009 (UTC)
- For some of the suggestions, you would need to have dummy pages set up as subpages of the test to link to, so that the 'before' example would include a redlink, and then you change the links for the 'after' example to link to the subpage and ask people what has changed and if anything needs correcting? The basic idea is to get people clicking every link and making sure there is a good reason for it to be there. You could even teach people when to delink redlinks. The ultimate exercise would be a large piece of text with 20 or so different types of changes needed, and see how many people spot them all. Or throw up a non-wikified piece of text and see what things people link and why. But that might be moving too quickly. There is a whole lot of thoughts going on in people's heads when they link and delink. Carcharoth (talk) 00:16, 24 August 2009 (UTC)
- Carcharoth, these are important issues. I'm not sure that all of them can be dealt with in this click-and-show unfolding format. And there's a limit to users' attention span in terms of the size of the exercises, given that people are not expect to do any typing—the learning process relies on memory, and if it's not an iron-clad memory, at least it needs to be sparked when the user reads the solution ("ah, I got that one, but I missed this one).
- I've already taken your advice in adding some instances of underlinking, but more are needed, yes.
- A partial solution to your concern that red links and other maintenance issues be dealt with might be to insert, intermittently between a few of the exercises, a tip (almost in the manner of a Tip of the day). Could break up the rhythm of exercises after exercises, too. I'm keen that users concentration be switched from one thing to another regularly; the turn-off factor for such concentrated work is usually a concern with these exercises. Tony (talk) 03:06, 1 September 2009 (UTC)
- For some of the suggestions, you would need to have dummy pages set up as subpages of the test to link to, so that the 'before' example would include a redlink, and then you change the links for the 'after' example to link to the subpage and ask people what has changed and if anything needs correcting? The basic idea is to get people clicking every link and making sure there is a good reason for it to be there. You could even teach people when to delink redlinks. The ultimate exercise would be a large piece of text with 20 or so different types of changes needed, and see how many people spot them all. Or throw up a non-wikified piece of text and see what things people link and why. But that might be moving too quickly. There is a whole lot of thoughts going on in people's heads when they link and delink. Carcharoth (talk) 00:16, 24 August 2009 (UTC)
Janis Joplin
editNice work on this. I removed the italics (and the recommendation for them) from Southern Comfort, as I don't think WP recommends this. Can't imagine saying that someone works on an Apple iMac. Dabomb87 (talk) 21:29, 31 August 2009 (UTC)
- BTW, I disagree with the recommendation to remove the link; it is true that ‘it is clear from the text that "Southern Comfort" is a brand of liquor’, but that alone gives no information: “her trademark beverage was a brand of liquor” would be almost a tautology. That statement only gives non-obvious information if you know what type of liquor Southern Comfort is; and some readers, on seeing that its name is not linked, might assume that WP has no article about it. --___A. di M. 18:27, 3 September 2009 (UTC)
- I don't think it needs a link (context, upper-case initials), but I will change that aspect with a note. Tony (talk) 01:41, 4 September 2009 (UTC)
- Context and upper-case initials only show that it's the proper name of an alcoholic beverage; it could still be anything from a beer to a spirit and anything in between. --___A. di M. 12:58, 4 September 2009 (UTC)
- I linked it; isn't that what you suggested? Tony (talk)
- Yes. I'm just not 100% comfortable with the sentence ‘Since it is clear from the context and the upper-case initials that "Southern Comfort" is a brand of liquor ...’ in the explanation, but after all this is your tutorial. :-) --___A. di M. 16:34, 4 September 2009 (UTC)
- I linked it; isn't that what you suggested? Tony (talk)
- Context and upper-case initials only show that it's the proper name of an alcoholic beverage; it could still be anything from a beer to a spirit and anything in between. --___A. di M. 12:58, 4 September 2009 (UTC)
- I don't think it needs a link (context, upper-case initials), but I will change that aspect with a note. Tony (talk) 01:41, 4 September 2009 (UTC)
- In my space, but if I didn't take feedback very seriously, the page wouldn't be taken seriously. Do you mean I can resolve the issue simply by replacing "liquor" by "alcoholic beverage"?
- Sounds OK to me. --___A. di M. 11:38, 5 September 2009 (UTC)
- for the record: i'd omit the sentence questioning the value of the Southern Comfort link, because i feel it undermines the credibility of the "smart-linking movement" to suggest that an item like that shouldn't be linked. i disagree that the context/capitalization make it "clear enough" what Southern Comfort is; unless readers already know (at least roughly) what it is, how would they know it's not a cult wine, or a mixed drink, or cough syrup, or '60s jargon for acid-laced kool-aid? and even readers who know vaguely that it's liquor might well find it interesting and pertinent know it's a sweet liqueuer, not 198-proof grain alcohol or something like that - and those things make a difference in understanding Joplin's image. Sssoul (talk) 07:53, 6 September 2009 (UTC)
- The semicolon also shows that it's an alcoholic beverage, since the heavy drinking occurs just before that boundary. I'll tweak the explanation again. Tony (talk) 06:15, 7 September 2009 (UTC)
- for the record: i'd omit the sentence questioning the value of the Southern Comfort link, because i feel it undermines the credibility of the "smart-linking movement" to suggest that an item like that shouldn't be linked. i disagree that the context/capitalization make it "clear enough" what Southern Comfort is; unless readers already know (at least roughly) what it is, how would they know it's not a cult wine, or a mixed drink, or cough syrup, or '60s jargon for acid-laced kool-aid? and even readers who know vaguely that it's liquor might well find it interesting and pertinent know it's a sweet liqueuer, not 198-proof grain alcohol or something like that - and those things make a difference in understanding Joplin's image. Sssoul (talk) 07:53, 6 September 2009 (UTC)
(outdent) smile: but this is the encyclopedia that anyone can edit, which means readers can't always rely on punctuation and capitalization as dependable carriers of subtleties of meaning. on top of that, whether she was chugging wine or beer or grain alcohol or a liqueur does make a difference, and using a link to clarify it seems 100% appropriate. again, my concern here is the credibility of the "smart linking movement" ... but anyway i'll shut up and leave you to your tweaking Sssoul (talk) 06:40, 7 September 2009 (UTC)
- "anyone can edit": yes, and link-target articles can change without our knowing it, too. Do you think the explanation is still unsatisfactory? You are taking a surprisingly 100% line for the linking of a commercial product (I was slightly concerned about not wanting to promote the brand, but that wasn't the clincher—the importance of the target information to understanding the sense of the statement was the motivation for not implying to users that they must link such an item; to me, it also depends a little on how densely linked the surrounding text is). Tony (talk) 09:33, 7 September 2009 (UTC)
(outdent) sorry to reply only now - i didn't see your questions earlier. i think the explanation is okay now - thanks! i still don't quite agree that context/capitalization would be very communicative for a reader who had never heard of Southern Comfort - but ... so be it! just for the record, i'm not sure what "a surprisingly 100% line" means, nor do i see why it would make any difference that something is a commercial product. records and makes of guitar are commercial products too, but i've never seen anyone suggest we shouldn't link to articles about them when they're relevant, and (as i've been saying) in my opinion it is relevant to understanding Joplin's persona/image that her "trademark" drink wasn't (for example) a sophisticated cult wine, grain spirits or psychoactive kool-aid. but anyway, the explanation looks a lot better now, so thanks Sssoul (talk) 13:50, 11 September 2009 (UTC)
- Actually, "guitar" is a bad one in popular music articles. I've been encouraging either no link, or a link to the section there on "Types of guitars". Drums, singer, vocals, keyboards—which reader of such articles doesn't know what these mean? Every kid in the world aged over 6 does. Tony (talk) 14:24, 11 September 2009 (UTC)
- Tony, i was talking about links to makes of guitar, like Fender Telecaster or Gibson Flying V, not to guitar - but never mind! swing on Sssoul (talk) 15:05, 11 September 2009 (UTC)
- Heaven. I wish the WikiProjects would encourage specialised guitar links! Tony (talk) 15:18, 11 September 2009 (UTC)
- Tony, i was talking about links to makes of guitar, like Fender Telecaster or Gibson Flying V, not to guitar - but never mind! swing on Sssoul (talk) 15:05, 11 September 2009 (UTC)
Whoopi(e) Goldberg
editGreat to see a nice guide on this important aspect of article-building. I made an edit to correct Whoopi's first name in two places. Hope this is all right. Giants2008 (17–14) 20:09, 1 September 2009 (UTC)
- Thanks! Tony (talk) 06:22, 7 September 2009 (UTC)
Question
editHi. Can you give me a sprinkling of examples of when it might be appropriate to link to, say, bicycle? --Dweller (talk) 13:18, 30 December 2009 (UTC)
- Is anyone watching this page? --Dweller (talk) 21:25, 12 January 2010 (UTC)
- I sometimes miss this on my watchlist if I arrive late in the daily cycle. To take whole articles/sections as the anchor examples, the obvious ones from which "Bicycle" might be a link-target are Cycling, Single-track vehicle, Bicycle wheel, Bicycle frame, etc, and all of the daughter articles linked from "Bicycle". From history-of-transport type articles and sections, a section-link to "History" within "Bicycle" might be appropriate. That type of thing. Now you've drawn my attention to that article, I see woeful overlinking in the lead, which I'm about to correct. Tony (talk) 22:19, 12 January 2010 (UTC)
- Dweller, there is a somewhat-related discussion about linking common terms at WT:LINKING. Dabomb87 (talk) 23:13, 12 January 2010 (UTC)
- I sometimes miss this on my watchlist if I arrive late in the daily cycle. To take whole articles/sections as the anchor examples, the obvious ones from which "Bicycle" might be a link-target are Cycling, Single-track vehicle, Bicycle wheel, Bicycle frame, etc, and all of the daughter articles linked from "Bicycle". From history-of-transport type articles and sections, a section-link to "History" within "Bicycle" might be appropriate. That type of thing. Now you've drawn my attention to that article, I see woeful overlinking in the lead, which I'm about to correct. Tony (talk) 22:19, 12 January 2010 (UTC)
So, you're saying that articles about bikes and related matters should include a link to bicycle, but say, the story about David Cameron riding his bike with a car following him should not include a link, as it's unlikely that someone reading Cameron's article will want to go read the article about bikes? --Dweller (talk) 09:44, 13 January 2010 (UTC)
- Doesn't it depend on the relevance, focus, utility of the link-target in relation to the topic? And the existing density of wikilinks in the text? It's not a simple matter, like prose, which is why I'm keen that linking gain more recognition as a skill that needs to be cultivated by WPians. I must say that I often see general links to whole articles that are simply not worth the blue, and wish they were section links or daughter-article links. I find that there's a reluctance to do just a little searching for a target that will help readers the most. This is apart from the cancer of bad piping.
- The story about David Cameron doesn't sound as though bicycle should be linked (nor road, nor helmet, nor wheel—although I haven't looked at the text you cite).
- One more point: in browsing through the quite good article on bicycle, I formed the view that the many subsidiary articles (bicycle wheel, bicycle lights, bicycle this and that) are in need of rationalisation and merging. Too itty-bitty: why not one article for bicycle parts/equipment? Tony (talk) 11:05, 13 January 2010 (UTC)
- The Cameron example was from my head, not looked for it in article. I tend to agree with you, I think, but it is a tricky and rather subjective area and would play interestingly at FAC for me and wonder how recentish FAs stack up. Take a look at a random section (not Lead) of (say) Donald Bradman - do you think it's a problem there? --Dweller (talk) 11:24, 13 January 2010 (UTC)
- The Bradman lead looks pretty well linked; there was one change I'd make (to a pipe?), but I can't remember which, now. Tony (talk) 22:58, 15 January 2010 (UTC)
Some of these examples go too far
editWhile overlinking definitely is a problem that needs to be solved (see how annoying that is?), I think some of these examples go too far in the opposite direction of excessive de-linking. In the Voting Age example, it seems very silly not to link to voting or elections, when they're directly relevant to the subject of the article. Granted, someone reading that article probably already knows what voting is, but on the off chance that they don't, they need to read about voting or elections before they can understand what a 'voting age' means. When articles are that closely related, linking of common terms is justified - it's comparable to linking alcohol from an article on drinking age.
In the Greengage example - well, this one's more questionable, but why not link George Washington? OK, he's of extremely little importance to greengages and they're of even less importance to him, and 95% of people reading the article will know who he is already; but why not help out the remaining 5% who are wondering 'who is this Washington guy, and why is he important?'. This is a more subjective case, and if a section of text was already heavily linked, I wouldn't bother linking Washington's name. But the suggested solution text had no links at all, so an extra one or two wouldn't clutter it.
Finally, the Fatboy Slim example appears to discourage creating red links, which I find a little surprising. Wikipedia has been built through red links; if a name or phrase is widely red-linked, it will encourage someone to create an article for it. As long as a name could plausibly be made into an article, I'd always link it. In this case, I'd say links like Drive thru booty and More of everything for everybody should generally be linked, as albums by notable bands are always notable; the only reason not to would be if they only existed as redirects to the Freak Power article (as perhaps they should), as the links would be unhelpful in that case.
I think this is still a very useful page - it answers a question I was wondering about recently, 'when, if ever, should terms like New York City, London or United States be linked?' (answer: rarely, and only when they're directly relevant to the context). So thanks to Tony1 for making it and putting together. I just feel that this is an area where there isn't a true 'right answer', and it's often more subjective than this page makes it appear. Robofish (talk) 14:07, 15 January 2010 (UTC)
Robofish, thanks for your note.
- "it seems very silly not to link to voting or elections, when they're directly relevant to the subject of the article". As someone pointed out above, the word—the concept—may be directly relevant to the article (as "problem" is to your first sentence here), but is diverting to and reading the link-target articles relevant to understanding the topic at hand. No. Certainly not sufficiently useful to add to the blue-clutter, diluting higher-value links in the vicinity.
- "but why not help out the remaining 5%"—(1) That is starting to be a perilously small proportion of readers who would be ignorant of who "George Washington" was. (2) Of that proportion, only a very small proportion are likely to click. (3) The George Washington point is not of major importance to understanding the topic, anyway, and the GW article provides no more information on Greengage. (4) The very rare person who wants to move from this vegetable topic to GW can type it into the search box. We are trying to select, to ration, links to flag to readers the important subsidiary or related pages to Greengage, not to provide a magic blue carpet to anywhere.
- Red links: I find it hard to come to a solid conclusion about them.
- NYC, US, etc: almost never, and if tempted, try to link to a section or a daughter article instead. Tony (talk) 22:56, 15 January 2010 (UTC)
- What "higher-value links in the vicinity" are you talking about? The only link in your solution is constitution, which is very far from being a vital one – I don't think it would have ever occurred to me to add such a link in the first place. It is more likely that someone reading the article about a particular aspect of voting becomes (or already is) interested in other aspects of voting than in constitutions. If I were allowed to have only one link in that paragraph, voting or election would be it. (I'm not quite sure of which one, because I'm not quite sure of why there are two separate articles for them in the first place.) Linking to capacity (law) wouldn't be useless, either, as it is a technical concept very close but not identical to the everyday meaning of the word; I would have it just because there are so few other links in the paragraph – less than one per line on my display, including this one – and would omit it in a more densely linked paragraph. BTW, I cannot parse your sentence containing "but is diverting to", but if a concept is relevant but the article about it has no relevant content, the solution is to expand it, not to hide it away; and the fewer links there are to it, the fewer people will visit it, and the less likely it will be that one of them will expand it. Also, I'm not that confident that less than one person in twenty doesn't know who George Washington is, and strongly suspect that that is not the case for Thomas Jefferson. ― A._di_M.2nd Dramaout (formerly Army1987) 16:19, 16 January 2010 (UTC)
- Agree some go too far, but think this is a good overall example of what we should be encouraging. As this is not a policy (see it as essays) perhaps 2 examples of each could be shown. Y because i believe at all times we should imply to our readers there our always different options that are expectable (As this is a core principle of Wikipidia - as demonstrated at Wikipedia:Verification methods) - that i believe is a great example of saying no particular method is preferred or being shoved down the throats of our editors. I am not suggestion this is the same situation here, but there is always going to be different options and/or opinions and we should be clear that this is excepted here and that we are willing to work constructively. I think that linking within the context of a paragraph and/or sentence is just as important as the context of the over all topic (article). Pipe links should reflect what the statements are trying to convey and not just what the main topic is. We should show (link) the context of the examples sentences in there paragraphs with the child links. That said i am a firm believer in that our readers can find there way around Wikipidia and we have no need to hold there hands in most cases. Our search button is widely used and very user friendly with a spell check and all. We should simply suggest what we think should be change. I do see a problem with geographical names as this is what behind lots of problems. I will try to make some second examples over the next few days and post them here and then tell me what you think. Moxy (talk) 06:40, 29 December 2010 (UTC)
passing comment
editI saw these pages recommended elsewhere and was drawn to this one, it was very interesting and thought provoking because I am already heavily opinionated about linking from and between articles :-) What a good idea, cheers for this. cygnis insignis 12:17, 29 April 2019 (UTC)
Linking tutorial UI
editHi Tony1! I was wondering if there's any reason you've kept the "solution" and "explanation" tabs separate. I think it might be better to combine them. {{u|Sdkb}} talk 20:59, 14 June 2020 (UTC)
- I'm not sure the "problem text" needs to be collapsed, either. The user should have to make as few clicks as needed. {{u|Sdkb}} talk 21:40, 14 June 2020 (UTC)
- Yeah, that's an idea worth exploring. Thanks. Tony (talk) 07:08, 15 June 2020 (UTC)
- @Sdkb: How is it now? Tony (talk) 08:06, 15 June 2020 (UTC)
John Denver example
editAlternative link for the passage: Denver became outspoken in politics in the mid-1970s. In 1976, he campaigned for the [[1976 United States presidential election|election of Jimmy Carter]], who became a close friend and ally. Denver was a supporter of the [[Democratic Party (United States)|Democratic Party]], and a number of charitable causes for the environment, the homeless, the poor, the hungry, and the [[HIV/AIDS in Africa|African AIDS crisis]]. He founded the charitable [[Windstar Foundation]] in 1976 to promote [[sustainable living]]. His dismay at the [[Chernobyl disaster]] led to precedent-setting concerts in parts of communist Asia and Europe. Note that although at first glance this may create a potential issue with the descriptive subordinate clause that follows, there really is no possible alternative way to parse the sentence. -- Ohc ¡digame! 16:10, 16 June 2020 (UTC)