Wikipedia talk:Manual of Style/Linking

Address hyperlink-era usage patterns in "Repeated links" section

Suggested addition for below the bulleted list:

In addition to the traditional top-to-bottom reading pattern of the print medium, Web-based content (including Wikipedia articles) is also frequently read nonlinearly, as hyperlinks take the reader to anchors within a larger article and the reader reads (or skims) that section only. This is why repeating links across sections—especially when they are a long way apart, but also even when only somewhat far—can be valuable to some readers, who can pursue a line of thought across (only snippets of) hyperlinked pages, then backtrack to the jumping-off point of their digression and resume their main reading. Meanwhile, the second link is harmless to linear-mode readers, who ignore it. However, this principle must be applied sparingly—if applied to common terms, it would result in pointless overlinking; but it has value for more obscure or technical terms.

That this nonlinear mode of reading is common nowadays will be recognized by most users. It is the healthy, reasonable end of a spectrum whose maladaptive extreme is humorously lamented at xkcd.com/214. The question is whether we duly incorporate the reasonable part of the spectrum into WP:LINKS, or whether we try to pretend that it doesn't exist, that is, try to pretend that the entire spectrum is equally bad. Of course, we'd have to codify it properly so that idiots can't cite the codification to justify completely mindless overlinking of even common terms. — ¾-10 16:42, 11 February 2010 (UTC)

Ah, the assumption that people who read an article in the intended sequence can "ignore" repeated links is flawed. Every link has the potential to dilute the others in the vicinity. Our readers, in all likelihood, follow links rarely anyway. It is our job to ration wikilinking so that readers are offered the most valuable. If a reader section-links from somewhere else, they can either have a skim through the rest of the article or type into the search box any item they see in that section. Turning everything blue in the fear that one in a thousand readers might want to click on it is the old justification for blue-carpeting. If you want to see that, please go to the Fr, Ge and It WPs, which apply no notion of selectivity in their internal linking. The result is a degraded linking system, which is a huge pity.

The slippery slope here is that allowing people to justify the linking of an item once in every section will not only hugely increase the density of bright blue, but will throw the notion of skilled linking out the window. Tony (talk) 02:34, 12 February 2010 (UTC)

PS Love that cartoon you linked to. It should be widely promulgated on WP. Please see User:Tony1/Silliest wikilink of the month award. Tony (talk) 02:38, 12 February 2010 (UTC)

Extremely good counterargument. I can see all sides of it. You know, one thing that provides a handy bypass or workaround to my argument is that, within the past year or so, and led (TTBOMK) by the Chrome browser team, a feature has cropped up in the major browsers whereby you can select any word or phrase at all, right-click it, and choose "Google this" from the shortcut menu. That feature is absolutely brilliant, absolutely basic, and it's really embarrassing for the makers of IE that they never made it happen (AFAICT) until Chrome shamed them into it. I believe that, as of this writing, Firefox users (and possibly users of Chrome, IE, et al) can customize that shortcut via add-ons to add "Wikipedia this" to it. The one snag is that, as of this writing, TTBOMK, only geeks do this; average Joe web surfer doesn't have any habit of shortcut-searching—yet. I imagine that within 10 years (if not 5), even grandma will be doing it regularly. That will be really great, IMO, and would make my purple argument above 90% moot. Shortcut-searching beats "traditional" searching (i.e., copy, then paste into a search field elsewhere in the window or in another window), because in many low-need-to-know instances, people wouldn't quite bother to do the latter but *would* bother to do the former. In other words, it is not just more convenient but is the difference between the search getting done at all, or not. ¶ Well, to finish up, I will say that links will never become needless or moot in all instances (even *with* the advent of widespread shortcut-searching by readers), because (a) target≠anchor text [i.e., piping] will always have a place in life; (b) the WhatLinksHere concept will always have a place in life; and (c) even when target=anchor text, links will always be a way for the message sender to convey a metamessage to the message receiver, that is, to say, "We both know that you could digress from this message by jumping off from any single word within it; but we also both know that you don't want to do that, and that it would be totally impractical. So I am using my linking choices to *suggest* to you, 'If you only jump off into digression from *one* word in this sentence, make it *this* one!'" (cf snowclone "If you only see *one* [insert_genre] movie this year, make it *this* one!"). Thanks for the interesting and thought-provoking discussion. I've decided that my purple suggested addition to WP:LINKS#Repeated_links can safely be left on the cutting-room floor. Best regards, — ¾-10 00:19, 13 February 2010 (UTC)

You raise future issues that I'd never thought of. It would have been possible for the WMF to have introduced a right-click system that treated any clicked word as a link; it could still be done. But I think you see the disadvantage in relying entirely on this: part of the benefit of flagging items with blue is to tell the readers that "this item is focused, relevant and useful to this context, and we have a decent article or section on it". If every (black) word were a potential link, it would be the ultimate dilution and, IMO, would discourage clicking. The blue highlighting, for the very same reason, needs to be constrained.

Tim Starling, WMF developer, told me that the choice of that ugly bright blue colour for wikilinking was probably done without much thought right at the beginning. Now we're stuck with it. I use a monobook code to take the garish edge off it, and I still find it an unprofessional (and discouraging) look to have more than a certain density of items linked. Tony (talk) 01:09, 13 February 2010 (UTC)

I think you hit the nail right there ("this item is focused, relevant and useful to this context, and we have a decent article or section on it"). The act of linking, which is to say, the choice of what to link or not to link, is ideally a form of metainformation that should be conveyed with pedagogic skill that respects the reader's cognitive bandwidth limits. — ¾-10 02:29, 13 February 2010 (UTC)

As for link colours, wouldn't it be most logical to just leave the browser default alone? What is the "right" colour for a reader will be too bright for another and too dark for another– and that's not just a matter of personal tastes but a gamma correction issue as well. (Ditto with many other things such as the font size (currently "xx-small" times "127%") and so on.) ― A._di_M. (formerly Army1987) 11:02, 13 February 2010 (UTC)

• You mean my personal choice of colour, or the global default, which I strongly believe was foolishly and blithely chosen without regard for utility and appearance. If I had my way, in the settings page you'd be given a range of colour choices for your wikilinks, ranging from the current garish bright blue down to a more mellow darker blue. If you want to see a range of basic choices (of which I chose the second-darkest), see my user page (scroll down). You may wish to personalise it, you may not. However, this is quite separate from my point above, which was that even though I use a less in-your-face colour, I still find high-density linking to be a problem. When I go to other WMF sites, it's hideous by comparison, I'm afraid. Tony (talk) 11:49, 13 February 2010 (UTC)

  All I meant was that if I had my way, I'd not specify a link colour at all, so that anyone (not just logged-in users) can set the colour in their browser's settings, and not just from a range but from the entire system palette. (I wasn't commenting on the main point at all.) Did you misunderstand me? ― A._di_M. (formerly Army1987) 12:02, 13 February 2010 (UTC)

Most overlinked article?

Gratuitous links galore. This recent version of John Quinlan (wrestler) deserves to be considered a serious candidate, courtesy of Dabomb87. Ohconfucius ^¡digame! 01:54, 18 February 2010 (UTC)

It's got my vote. Anyone who says seas of blue do not interfere with readability should be shown that oldid. Some of the more useless links include "descent" (piped to "ancestor"), "Fall", "Spring", "Annual" (piped to "year"), "competed", and "placed 1st" (pipe linked to "champion"). Dabomb87 (talk) 03:00, 18 February 2010 (UTC)

Don't forget these other beauties: "December", "income", "past decade" (piped to "decade"), "event", "feud", "table" (piped to "Table_(furniture)"), "months" (piped to "month"), and "concrete".  HWV258.  03:13, 18 February 2010 (UTC)

I wonder what could have brought about that rather serious case of 'linkitis'? It just seems like somebody linked to all those terms because the were articles there to be linked to. The mind boggles... Ohconfucius ^¡digame! 04:09, 18 February 2010 (UTC)

Probably stems from boredom and ownership issues—e.g. someone just wanted to keep tinkering with their pet article. Of course it's doubtful that just one wrestler received this treatment...  HWV258.  04:28, 18 February 2010 (UTC)

• have at it ;-) Ohconfucius ^¡digame! 04:51, 18 February 2010 (UTC)
• Found the linking edits easily enough. See: this and this. Ohconfucius ^¡digame! 05:01, 18 February 2010 (UTC)

Oh, this is worthy of being made a museum piece. Perhaps the "Silliest wikilink of the month" should be expanded to include the most embarrassingly overlinked article each quarter. This one was a doozy. 06:53, 18 February 2010 (UTC)

"Concrete"—in an article about a wrestler?!?!?! (Shakes head in disbelief.)  HWV258.  08:03, 18 February 2010 (UTC)

It's not surprising, as the main editor of the article has no experience in editing any other WP articles - probably has not even visited any policy page (I suspect it is the wife or business associate of the subject). Ohconfucius ^¡digame! 08:40, 18 February 2010 (UTC)

I have to admit, that one was truly impressive.--Father Goose (talk) 09:39, 18 February 2010 (UTC)

A good laugh, but re noobness, we should remember not to bite. I can well imagine what it's like to enthusiastically make some first-time contributions to Wikipedia and then have people bite your hand off for being hyperlink-happy. You go from not even knowing that you did anything wrong to being laughed at and shamed. It's no good for developing the encyclopedia, because 2 outcomes are likely: either they leave in disgust and never contribute again, or they overlearn and go around hypercorrecting (read: no link is OK) until someone else bites them for being a delinkification nazi. Actually I think such a process is the general principle of pedant creation. Pedants begin life as insecure students getting slapped with rulers for trivial offenses. So remember to guide without slapping! — ¾-10 02:23, 20 February 2010 (UTC)

Yeah, I agree that I would be more comfortable if the stance against "overlinking" were not quite so, um, passionate. More than newbies have been chewed in the name of anti-overlinking, as the datelinking case demonstrated.

That's not to say I'm for overlinking, but this is one of those issues on Wikipedia that I wish could be discussed on a more rational and less emotional level. Just saying.--Father Goose (talk) 06:20, 20 February 2010 (UTC)

• I quite see what you're saying, Father. (I'm feeling devotional.) The linking debate runs deeply because it concerns nothing less than an evolutionary change in this early part of the history of wikis. We began with this great new distinctive functionality and people understandably used it as they wished, without much skill or guidance, for at least the first few years. Then after three or four years, there was a realisation by many editors that the project was covered with blue that wasn't useful and had the unintended effect of diluting valuable links and reducing the readability and professional look of the text. There is still a divide between those who value these things differently. The encouraging aspect is that once someone understands that wikilinking is a skill, like writing good prose, and that it should be used strategically, they usually fall completely into the "ration it or it's wasted" side. But that underpins why it's sometimes a politicised matter rather than just a technical one. Tony (talk) 07:11, 20 February 2010 (UTC)

  Tony, our differences on the concept aside, the comments above are an excellent example of why the execution of this matter needs to be approached with more care than has been done to date. There is no value at all to mocking people as above, nor is there any benefit to making this out to be a "critical matter" rather than the stylistic difference it actually is. The proponents behind the "delinking" campaign do need to accept their share of responsibility for the way that this has been a) imposed upon the community, and b) presented in a rather confrontational manner. One would be hard-pressed to not mistake the delinking campaign for some sort of new Wiki-religion, what with the strident comments, the over-the-top characterizations ("sea of blue" for one), and the way in which those who don't meet the "test of faith" sufficiently have been misrepresented and vilified. Even the method of delinking is somewhat suspect; "cleanups" are done by scripts that mix the delinking with various other less controversial tasks, which makes it harder to revert individual contested changes. More to the point, this particular script-based approach does not incorporate any form of consensus-driven agreements as to what exactly should be delinked. The lists of terms are buried in the script's code, which itself is hosted on a user page, not a public page, and the specific choices have never been open for debate or discussion. There needs to be a greater emphasis on respecting the middle ground in this stylistic difference of opinion, rather than on pushing through one particular point of view at all cost. --Ckatz^chat_spy 08:55, 20 February 2010 (UTC)

Ckatz, I do wish you'd not start with an accusation. Where did I "mock" someone? What is the difference between "critical" and "stylistic"? Is your language itself confrontational? ("imposed", "campaign"?). I've yet to see a single word from you in substantive debate; just personal stuff. As far as scripts go, I usually unlink much more than what is listed on any common-term script, manually, and I check through what has been unlinked already. It's important to do so in terms of context. Tony (talk) 12:47, 20 February 2010 (UTC)

Tony, please review the message I left on your page, where I specifically said that I wanted to keep the back-and-forth personal bickering between us off of this page. My comment is directed at this entire thread, not your note in particular; you weren't even a participant in the top half of the section. Both 3/4 10 and Father Goose's comments above suggest that they are uncomfortable with the way in which the editor who made those links was described; just think how that person would feel if they came here themselves and read that. As for the scripts, many terms that end up being unlinked are embedded in the code, with little or no chance for the average Wikipedian to discuss or change them. For example, there is no easy way to identify all of the cities and countries involved without the expertise needed to wade through the script's code. --Ckatz^chat_spy 23:11, 20 February 2010 (UTC)

Ckatz, I don't think anyone's means to mock the other editors, who, as Three-quarter-ten said, were contributing in good faith and may not have been aware of Wikipedia's linking guidelines. It was just surprising to find an article where the principle of building the web was taken to a rather extreme level. I do take issue with your comment that a "sea of blue" is an over-the-top characterization. Dabomb87 (talk) 15:31, 20 February 2010 (UTC)

"Sea of blue" gets tossed around far too often, in situations where it really doesn't apply. It is also a personal opinion, based on individual preferences, as to whether or not the number of links is excessive. That is central to the problem here; it really is a difference of opinion as to what and when to link, yet it is being treated as an "I'm right, you're wrong" issue with little room for seeking common ground. Again, if we are performing wide-scale script-based operations to strip out links, there has to be some way for the average editor to easily have input into the process. There needs to be a simple, readable list in an easily accessible location, one that can be reviewed and discussed. There also needs to be a better method of assessing context; I've seen some script-based changes that make no sense whatsoever in the context of the article they are in. --Ckatz^chat_spy 23:22, 20 February 2010 (UTC)

• I'd like to take responsibility for the article I delinked largely. In view of the comment above, as there has been little comment [from CKatz] on where linking/delinking balance should lie, I'd like to be enlightened as to what extent Ckatz thought I went overboard - firstly if (s)he agrees the article was overlinked - if there were any terms/words I delinked which (s)he felt were inappropriately done, and if so, which ones they were? At least then, we can have a concrete rather than abstract discussion about the whole subject, and no more beating about the bush and accusations of religious zealotry. Ohconfucius ^¡digame! 17:36, 21 February 2010 (UTC)

• Don't you mean a concrete discussion?--Father Goose (talk) 23:13, 21 February 2010 (UTC)

• You took the words right out of my mouth. ;-) Ohconfucius ^¡digame! 01:36, 22 February 2010 (UTC)
• Yes. This discussion has come about simply because: in the past decade—during the Fall to Spring months (especially December)—an annual feud regarding the descent in income from the sale of a concrete table competed to be placed 1st as an event.  HWV258.  02:22, 22 February 2010 (UTC)

• That unfortunately would be an example of the mocking attitude mentioned earlier. Or at least a 'way to kill the joke' moment.
  I keep meaning to follow up in this thread on a serious level. Ckatz has a lot of good points, but this conversation almost immediately drifted back into the old tit-for-tat.--Father Goose (talk) 03:01, 22 February 2010 (UTC)

• I'll make a suggestion that everyone needs to lighten-up over this issue. There are horrendously over-linked articles, and we just have to accept that and move forward (as those articles get encountered and fixed). I guess that some of us don't mind a little (light) mockery—to encourage other editors to realise that over-linking isn't desirable.  HWV258.  03:32, 22 February 2010 (UTC)

• You should see that I don't mind a little light mockery, considering I cracked the 'concrete' joke. But mockery in general is not a good strategy for attracting sympathizers to your cause. If you really want the reform to move forward well, treat your opponents -- and even those overlinkin' fools -- with the utmost of respect, and spend your time on winning them over instead of beating them down.--Father Goose (talk) 04:02, 22 February 2010 (UTC)

• OK, lets separate out the issues of defining to what extent any link is not desirable, from the mocking, which I would tend to agree may get people the wrong way and thus become rather unproductive. Ohconfucius ^¡digame! 04:21, 22 February 2010 (UTC)

Three classes of desired links

Thinking to what's wrong with the above article and what we're trying to do as a hyperlinked work, I would say there are 3 classes of desirable links that we want to encourage:

1. Proper names - Names of people, companies, products, and places should always be linked if used. This however excludes obviously common locations in non-geographic articles - continents, major countries, major cities like New York City or Paris.
2. "Parallel" terms - These would be terms that are of the same type as the article's topic that are brought up in the article - for example, in an article about a chemical element, we would link any other chemical element, compound, or term that is mentioned. This is for the purpose of someone researching that topic to make it easy to jump to related articles.
3. "Distant" terms - Given a article in a field, a uncommon term from a completely unrelated field that one would not reasonably expect to be connected to the article's field should be linked. The most common thing here would be, for example, describing the medical condition that caused a person to die. There's likely a better way to describe it. But here, the idea is to promote understanding of unexpected terms to the person researching the main topic.

This would not preclude any other types of preferred links, but this type of advice - covering near terms and distant terms - would see to help give value to a lot of our links. --MASEM (t) 04:58, 22 February 2010 (UTC)

This is a good starting-point, Masem. Even so, the categories need to be examined carefully. I have seen "Jesus Christ" linked many times in contexts where, IMO, it does not qualify. It would qualify for a link in the article on the RCC and any other religion, but not in almost every other context. Tony (talk) 05:56, 22 February 2010 (UTC)

This is a good starting point. I'd like to add one more class, or pseudo-class: links in tables and other non-prose sections such as infoboxes. It's my sense that in non-prose sections, links are valuable as a compact way of explaining the meaning or context of a term (table headings, single-word entries, etc.). Not to say that a table or infobox is a free pass for linking, but I'd say they have more value and less intrusiveness there than in prose.--Father Goose (talk) 08:59, 22 February 2010 (UTC)

I would definitely consider those, but keep that as a separate concept (eg that is more how tables and infoboxes should be presented, with standalone links), and keep the issue at hand for what is present in prose, where meanings of words can otherwise be inferred from surrounding text. --MASEM (t) 15:05, 22 February 2010 (UTC)

Separate concept, fine, but we do need to develop it alongside our "links in prose" standard so that the differences between what we expect in each are formalized.--Father Goose (talk) 00:29, 23 February 2010 (UTC)

Links in tables are already treated as a different matter, aren't they? I'd be inclined to focus on running prose. In addition, you may wish to examine the "Four key tests" for whether an item should be linked, here. Tony (talk) 07:14, 23 February 2010 (UTC)

Except when they arent: [1]. I can't really figure out why you de-linked terms like Norway and West Germany (in some places but not others), and didn't de-link Yugoslavia or Lithuania, and you left an entire table full of United Stateses alone. Some of the de-linking you did made sense, such as delinking country-after-city, but given that none of it was prose (except the lead), I'm not sure how critical it was to de-link any of the country names appearing alone, regardless of their familiarity.

I'm not intending this as a criticism of that specific edit, but as a demonstration of how de-linking in tables and templates is an entirely different art from de-linking prose; consistency of formatting, among other things, comes into play. So it's important that the differences we recognize between tables and prose are spelled out clearly in the guideline, so that they may be consistently observed in practice.--Father Goose (talk) 09:22, 23 February 2010 (UTC)

It's probably because templates such as {{USA}} and {{UK}} etc. link to the country article at the same time as generate the flags. Ohconfucius ^¡digame! 09:34, 23 February 2010 (UTC)

Ah yes, that does explain the inconsistencies.--Father Goose (talk) 10:10, 23 February 2010 (UTC)

Speaking of country flags - it has been brought to attention that flag icons may not be a free as some users expected (eg the EU flag), in addition to being questionable means of identification at the size used, and we may need to depreciate their use within tables or infoboxes. But that's not directly to linking in prose, just a different issue. --MASEM (t) 15:12, 23 February 2010 (UTC)

Overlinking At Steamroller

We should keep internal links (bluelinks or wikilinks) to a mininum:

• if it is not a piped link
• unless it is to a different article

road roller is linked 3 times:

• "A steamroller (or steam roller) is a form of road roller"
• "This is another example of how the use of "steam roller", to describe a modern road roller, still persists in the English language."
• "* Road roller – internal-combustion-powered road rollers"

We should reduce the number to 1. These 3 instances seem to have different definitions. We should write these occurrences so that it is readably more clear, but I have a dispute with some editors. Please help resolve this dispute.174.3.99.176 (talk) 00:13, 24 February 2010 (UTC)

Extraordinary abuse of our wikilinking facility

Here. It has just been entered at the Silliwili awards for March. Tony (talk) 02:36, 4 March 2010 (UTC)

Looks like vandalism: [2]. Dabomb87 (talk) 03:19, 4 March 2010 (UTC)

Announcement: Silliest wikilink of the month awards (January–February 2010)

Evidence has recently surfaced confirming that the Moon really is made of cheese.

Greetings, my fellow Wikipedians, and welcome to the second SILLIWILI awards ceremony for the year two-thousand and ten of the Common Era. This time I was supposed to make the announcement earlier, but have delayed it as a courtesy to the Oscars. Let it never be said that I wish to steal anyone's thunder.

Before we begin, an announcement. As some of you may have noticed, there have been a couple of changes to the rules. Or, rather, we have created a couple of rules. These will apply from this month onwards, and are the following:

Firstly, the scope of the contest has been restricted to articles. It is to the mainspace that the Manual of Style applies, so it makes sense not to include project pages and templates, which often follow different (or no) guidelines. Therefore, last December's winner will probably be the only project page to win the award, at least as long as I am judge. But that's not necessarily a bad thing; I suppose such anomalies and exceptions are at the core of any event's history and cherished traditions...

Secondly, a limit has been placed on the number of links that can be nominated for each article. I cannot put enough emphasis on the fact that in this contest we reward links, not articles; each submitted link is judged individually on its own merits, and therefore it is counter-productive to flood the nomination tables with them in the hopes of increasing an article's chances. The new limit of four links is meant to encourage nominators to think harder about the specific links they enter. Note that, as only one link can be silliest wikilink of the month, groups of links are necessarily represented by one of their number if selected (see October 2009). Among the elements evaluated in all links are their surroundings, so nominators can submit one link from a group and make a note about this. It will be noticed anyway, but it is best to draw attention to the fact that the link is part of a group, just to be sure.

And now, ladies and gentlemen, on to the awards for the last two months. Drum roll please...

• With 82 entries, January 2010 broke all records of participation, and I cannot begin to say how delighted I am. This was, of course, what made me realise that a nominations limit was necessary, but I am certainly very happy to see so many new faces. Unfortunately, this also means that my choice is made harder, and that I'll have to pass over many excellent candidates. Now, the winner... Enough drum roll, thank you. Ahem. From the article Civil confinement, an unexpected bit of linguistic meta-commentary in the form of the link [[Litotes|not without]]. Nominated by Robofish, it is a type of Easter egg that we should never, ever have to see in articles, and for this reason it wins this busy month's award.
  • Two honourable mentions this time. First, SuperFlash101's "3", from Phineas Flynn. Not only was it entirely unnecessary but, as most Wikipedians know, it leads to the article for a year—specifically, the year when the rule of emperor Augustus was renewed for a ten-year period, and Wang Mang foiled a plot by his son, his brother-in-law and the Wei clan to oust him. Interesting stuff. Second, Majorly's "linked" from Rubber duck. I actually found "spoke" funnier, but one has to appreciate the self-referential nature of redundantly linking the word "link".
• In February there were many bizarre links (although definitely fewer than in January), and I liked many of them, but the award goes to one with a great educational value, which illustrates one of the greatest mistakes an editor can make when linking. The article in question is Marvin Gaye, and in it could be found the following gem: "Martin Lawrence". See how hard it is to tell that they are two links? After some dithering, I decided that the first link—"Martin"—was more ridiculous (the disambiguation page was longer), and so this shall be the silliest link of February 2010. A warm round of applause for its nominator, Belovedfreak! All right, enough applause; I should also like to tell him that his diff was, rather oddly, for the current version of the article, and I had to search for the version with the link. If I cannot find the silly link, I cannot select it, so please be careful.
  • My second choice for this month was another confusing link: Tony1's[User:A. di M. (Edited by Ohconfucius ^¡digame! 15:16, 11 March 2010 (UTC):)] "LHC" from Large Hadron Collider was a delightful false tautology with a great potential for throwing readers off track.

    That entry was mine, not Tony1's... ― A._di_M. (formerly Army1987) 15:00, 11 March 2010 (UTC)

    • So it was!

And so ends the second awards ceremony of our competition. I do enjoy what I do, but make no mistake: entertainment value aside, if I could wish out of existence all links like the ones seen above, I'd do so. However, in the knowledge that there will be such links for a long time to come, I can only hope that they will be found, eliminated, and placed here for all to see and learn from. And on this note, I leave you for another two months of judicious de-linking. From Waltham Hall, I bid you a good day! Waltham, The Duke of 06:04, 11 March 2010 (UTC)

What? My goodness, the card was misprinted. Please, Mr A. di M., accept our sincerest apologies for this terrible error. Here, have some money, too. Please don't sue us.

Now, let's have a talk with my loyal and effective employees... They deserve a bonus for their hard work, don't you think, Cartwright? A severance package, perhaps... Waltham, The Duke of 17:07, 11 March 2010 (UTC)

I shan't sue you; I acknowledge how hard such a work must be, and errare humanum est. ― A._di_M. (formerly Army1987) 19:52, 11 March 2010 (UTC)

Delighted to make history as the first and last anomalous winner in defiance of these very reasonable new rules :-) Commendable work, your Grace. Adrian J. Hunter^{(talk•contribs)} 09:36, 15 March 2010 (UTC)

'Germane' again

I have noticed during my recent travels across Wikiland that articles carry a large number of links which I believe to be unhelpful although they may satisfy the 'relevance criteria' because they are not sufficiently germane. We all know now that whilst linking in an intelligent fashion helps the readers' understand the topic in question, linking can also be value destroying. In much the same way that linking to dates should only be done when the date is germane - and it now seems to be accepted that dates in the vast majority of instances are not - I have amended the text accordingly, to state that other links should also be germane. Ohconfucius ^¡digame! 09:44, 6 April 2010 (UTC)

Looks good to me. CKatz, would you mind going into the issues that you feel make it an undesirable edit ... substantive issues? Tony (talk) 02:01, 7 April 2010 (UTC)

Reverted, because the existing version is simpler, clearer, and doesn't merge the "technical terms" language. (Not sure why this didn't post previously.) As an aside, I'd ask per WP:BRD that you please refrain from reverting your change back in until we can get a good range of input. This being a guideline, it is important that we establish consensus for revisions first. --Ckatz^chat_spy 03:10, 7 April 2010 (UTC)

Simplify and clarify if you wish, but there is every good reason to provide editors with advice on the linking or non-linking of technical terms. Tony (talk) 03:14, 7 April 2010 (UTC)

I merged simply because I felt that these are similar ideas which are more logically grouped. In essence, I made two changes: one was the merger of 'technical terms and jargon', the second one was the insertion of the word 'germane'. Your argument above targets the former, though it's good that you do not appear opposed to linking what is not germane. Ohconfucius ^¡digame! 03:30, 7 April 2010 (UTC)

CKatz, your recent edit summary is on the boundary of WP:CIVIL: "Please have the decency to revert yourself when you realize I did post a discussion". Whatever you may feel emotionally, please do not allow it to spill over into statements that accuse another editor—in this case, Ohconfucius—of indecency. As an admin, we look to you to exemplify the community's expectations of calm and even-handed behaviour towards other editors. Tony (talk) 07:17, 7 April 2010 (UTC)

Tony, please stop the incessant overuse of WP:CIVIL; it is really quite tiresome, and unbecoming of a long-time editor such as yourself. For some reason, you've chosen to tag me alone for this while completely ignoring the fact that OC used near-identical phrasing ("Please at least have the decency to discuss") in his post immediately prior to mine. You've also chosen to ignore the fact that OC and I have already discussed and resolved this between ourselves in user space, a more appropriate venue, rather than on a more public page. Now, I'm virtually certain that you'll try to censure me for this post too, but before you do so I simply ask you to review your posts (and related edit summaries) to various talk pages over the past few months and note how many times you have brought up WP:CIVIL against other editors. --Ckatz^chat_spy 01:49, 8 April 2010 (UTC)

I'm not offended by the edit summary. In fact, I discovered that the edits crossed cyberspace in a short period of time, so we were slightly cross-purposes. I left a note on Ckatz's talk page to the effect. Ohconfucius ^¡digame! 07:38, 7 April 2010 (UTC)

Ckatz, you're right, I did not examine previous edit summaries (they're now not on my watchlist). All the same, I think you are meant to be the example of moderation in dealing with other editors. Tony (talk) 03:37, 8 April 2010 (UTC)

Could you point out some examples of the links you found to be "to be unhelpful although they may satisfy the 'relevance criteria'"? Maybe there's a clearer way to make your point. ― ___A._di_M. (formerly Army1987) 13:26, 7 April 2010 (UTC)

Offhand, I can think of a number of them which fall into that category. I note in particular the tendency to systematically link to 'President George W. Bush', 'Prime Minister Gordon Brown' or 'President Hu Jintao'. In most cases, because it relates to somethink the individual did or said (usually related but not central to the subject), the title - and more often than not, the biography - of the individual has little bearing on the article in which the link appears. The title is what Tony would call a 'chain link', because the title is always linked in the biography, and thus wholly unnecessary. These are 'relevant', but fail germane, IMHO. A somewhat better test might be to ask oneself, before placing such a link, whether there may be a reciprocal backlink. Ohconfucius ^¡digame! 02:01, 10 April 2010 (UTC)

Aren't these already covered by the no-chain-links "rule"? ― ___A._di_M. (formerly Army1987) 10:00, 10 April 2010 (UTC)

Even if they are, the rule is not observed; also, links to George W. Bush and Hu Jintao under those conditions are not covered.

If a rule already here is not observed, why do you expect that another rule against links which are relevant but not germane could change the situation, especially when the difference between relevant and germane will not be obvious to many readers? Also, I'm not sure I get your point about "links to George W. Bush and Hu Jintao". If they are relevant (enough for a reader to be likely to want to read them) link them, if not don't link them, as with anything else. ― ___A._di_M. (formerly Army1987) 12:30, 10 April 2010 (UTC)

If it's of interest to anyone, I'm having just that sort of argument with someone at Talk:Hong Kong by-election, 2010#Chain links. Ohconfucius ^¡digame! 04:41, 13 April 2010 (UTC)

To someone who has never heard of Zhuhai, "the annual meeting of the Chinese People's Political Consultative Conference in Zhuhai" conveys no more information than "the annual meeting of the Chinese People's Political Consultative Conference", so either the place where the meeting was held is relevant (in which case the reader needs to be able to easily find out where the heck it is), or it is irrelevant (in which case, why would it be mentioned in the first place?). (I'd guess it's the latter.) The chain-link rule doesn't apply, as neither of the other links in that sentence even mentions Zhuhai. ― ___A._di_M. (formerly Army1987) 15:56, 14 April 2010 (UTC)

When did we all of a sudden decide Zhuhai, Liaison Office of the Central People's Government in the Hong Kong Special Administrative Region and Secretary for Home Affairs were such easy links? Benjwong (talk) 05:12, 13 April 2010 (UTC)