Wikipedia talk:Manual of Style/Linking

Proposed merged wording

Please discuss the proposal below.--Kotniski (talk) 20:01, 9 January 2009 (UTC)

nice work, Kotniski - thanks! one suggestion for the date link section: it's rather contradictory to say multiple year-in-X links are unnecessary and then to suggest aliasing those links "in the main prose of articles in which such links are used heavily". a revision of this was discussed briefly on some talk page or other - i'll see if i can find it, but meanwhile it would make more sense to leave it at "However, piped links may be useful in places where compact presentation is important, such as tables, infoboxes and lists." thanks again for this work. Sssoul (talk) 20:13, 9 January 2009 (UTC)

ps: here's that discussion - i don't know if it adds much, but ... there it is. Sssoul (talk) 20:17, 9 January 2009 (UTC)

Thanks for the link - I see I was part of that discussion, but have only vague recollections of it... Anyway, yes, you're right, it does seem contradictory. I'll change it as you suggest.--Kotniski (talk) 20:40, 9 January 2009 (UTC)

I question whether the wording on dates reflects current consensus. It certainly does not reflect the middle ground in any of the recent debates on the matter that I have seen and it seems to me that the current opinion on the issue needs to be properly tested before that section can be part of the merge.Dejvid (talk) 22:00, 9 January 2009 (UTC)

Dejvid, that was the wording inserted by an uninvolved admin into WP:MOSNUM, you can take it up there. Dabomb87 (talk) 02:01, 10 January 2009 (UTC)

Not so. The wording on WP:MOSNUM is "Linking: Dates (years, months, day and month, full dates) should not be linked, unless there is a reason to do so." But in any case the controversy of the debate does suggest we need to check exactly what the community opinion is.Dejvid (talk) 10:51, 10 January 2009 (UTC)

I think we've already done that at extremely great length, in various RFCs and so on. In any case, this merger proposal doesn't aim to change the status quo as regards date linking guidance, so let's not discuss that here unless you think I inadvertently have changed something.--Kotniski (talk) 11:30, 10 January 2009 (UTC)

The wording that you have taken up is a result of this edit [[1]]. As the edit description was " Practicality to avoid multiple "hidden" sibling links; re-organised bullets more logically" I think there is reason to be extremely skeptical that it reflected a change of consensus. The addition of "demonstrably" is a very significant change.

You are right there has been a very diffuse and confusing debate. What is needed is however a clear vote now we are bringing everything to the same place. Without that, I doubt anyone who claims to know what the current consensus really is.Dejvid (talk) 15:21, 10 January 2009 (UTC)

The edit you refer to was four months ago; we've been through reams and reams of discussion and at least two very well advertised RfCs since then, which confirmed the status quo, so I don't see what good will come from any more voting or discussion on that issue. In any case it isn't relevant to this merger, unless you're saying that the proposed post-merge version differs in substance from the pre-merge version.--Kotniski (talk) 17:31, 10 January 2009 (UTC)

The can of worms is very much open and the RfCs have not produced any kind of conclusions. What is needed is not a RfC but proper voting on propositions.Dejvid (talk) 17:41, 10 January 2009 (UTC)

• Kotniski, we all owe you! This has been a bug-bear for so long, and I believe you've done a superb job. Tony (talk) 01:56, 10 January 2009 (UTC)
• Looks good. Great job Kotsiski, this was much-needed! Dabomb87 (talk) 02:05, 10 January 2009 (UTC)

"Deepen" --> "expand"?

While I won't immediately object to Locke Cole's change here, I find his/her continual framing of his/her views on such matters as "consensus" a little hard to take. Temporarily accepting the change is not equivalent to accepting the claim in the edit summary. I'd like discussion here as to how "expand" is different from "deepen", and why the editor is so keen to subsitute the word, which has been in the style guides for some time.

However, the claim that "demonstrably" is against consensus is harder to stomach. I don't know what is so hard about demonstrating that a year-link deepens (or expands) a reader's knowledge of a topic. Unless Locke Cole is concerned that it can't be done ....? Tony (talk) 03:14, 11 January 2009 (UTC)

Really Tony, there's only consensus to stop linking dates purely for auto formatting. There is no consensus to stop linking dates entirely or place unnecessary burdens on editors. I chose expand over deepen as a mostly semantic issue: the linked date/year may only contain links and information of events/issues/subjects with a minor relation, but a relation nonetheless, to the source subject. —Locke Cole • t • c 03:21, 11 January 2009 (UTC)

Really Locke, that appears to be spin. Tony (talk) 06:32, 11 January 2009 (UTC)

I'm sorry you see it that way Tony, but that's my opinion. It's at best a semantic change at any rate, but I think expand is a little more open ended. Again, I believe this reflects what was discussed at the second RFC. —Locke Cole • t • c 09:02, 11 January 2009 (UTC)

So in the exceptional cases where such relation exists (I'm not sure I've ever seen one, but let's assume there are some), that ought to be demonstrable, right? So can we compromise and say "demonstrably...expand"?--Kotniski (talk) 07:01, 11 January 2009 (UTC)

I'm waiting to hear from Locke as to why there's a need to change "deepen" to "expand". What exactly is the problem with "deepen"? Tony (talk) 07:26, 11 January 2009 (UTC)

Again the issue here is about presenting an unnecessary burden on editors. There's nothing wrong with simplifying it by removing "demonstrably" IMHO, and this more accurately reflects consensus from the RFC. —Locke Cole • t • c 09:02, 11 January 2009 (UTC)

We need evidence of this "consensus", a word that is being bandied about quite a lot over the past day or two. I don't accept it on the basis of what I've seen. Tony (talk) 12:23, 11 January 2009 (UTC)

I'm sorry you feel that way Tony, but I don't see consensus for this burden you're placing on editors. —Locke Cole • t • c 13:53, 11 January 2009 (UTC)

Thanks !

Thanks for the merge Kotniski ! I mostly like it, and it is definitely better than having 3 separate pages. About not linking "Plain English words", I must admit I sometimes do it when it is ambiguous. For instance, after reading "the monk was shot in the temple" I am not sure whether the author meant temple or temple (example inspired from here). This is a stereotypical example that could be fixed with rewording of the article itself, but any non-trivial article contains a number of ambiguous words, which may be misinterpreted by a newcomer. Anyway, I am happy with the current wording since it includes the word "generally". There is some junk draft wikicode at the bottom of the page, I guess someone is working on it ? Nicolas1981 (talk) 04:56, 11 January 2009 (UTC)

The meaning will almost always be clear from the context; but it it's not, it's a sure sign that the wording is ambiguous. We should not force readers to divert to another page to disambiguate a word (nor for its basic definition). Tony (talk)

I totally agree :-) How would you apply this precept to the first sentence of this paragraph ? I would say I understand English better than the average Wikipedia reader, and "temple" is an English word, but I did not know it had two definitions, so I really thought the guy had been shot while within a religious edifice. I accidentally understood a while after, when I read the same story in a newspaper that put it differently. Nicolas1981 (talk) 12:05, 12 January 2009 (UTC)

"familiar to most readers of the article"

The 3rd most accessed page of Wikipedia is Special:Random and the 4th is Special:Randompage. That is 59 times more than Barack Obama (Source:[2]) When you write an article, you can bet that most readers know nothing about the context. Those users will be helpless without wikilinks. Since we can not presume who reads an article, how about rewording to "familiar to most readers" ? Nicolas1981 (talk) 04:56, 11 January 2009 (UTC)

I've made that minor change. (I don't believe that most readers of an article come to it from Randompage though - there's a one in several million chance that anyone accessing that will come to my article, and even if they do there's no reason to suppose that they'll actually read it.)--Kotniski (talk) 08:04, 11 January 2009 (UTC)

Statistically, each article is hit by one "random page user" every month. As a member of the WP:ASE project, I have seen many article that probably had never been seen by an expert. So, for niche topics, this is probably not negligible. Nicolas1981 (talk) 09:01, 14 January 2009 (UTC)

Misleading footnoted statement?

"Academic research has shown that red links drive Wikipedia growth"—Footnote 4. This is not causally logical from the remainder of the footnote. I'm concerned about including this. Has anyone read the article? Tony (talk) 07:25, 11 January 2009 (UTC)

rumour/allegation

Kotniski, your edits substantially improved mine; thank you. On the example—it's better, but if you can think of an example that doesn't involve plain English words, all the better. Let me rack my brains (I'm thinking of a political example). Tony (talk) 08:07, 11 January 2009 (UTC)

Most dates

Locke, why do you keep changing the statement that "most dates [are not to be linked]"? This is surely not in dispute (particularly after all the RfCs), and we should be wording the guideline to make such things clear to new editors, not muddy the waters.--Kotniski (talk) 10:06, 11 January 2009 (UTC)

That's funny because my read of the results indicates a lack of consensus for delinking dates. I'm confused why we're discouraging editors to link dates when the community consensus supports their linkage under certain circumstances (and that's being charitable; realistically it's a "no consensus" which brings us back to the status quo; link all dates). —Locke Cole • t • c 11:25, 11 January 2009 (UTC)

On some pages linking of dates is not appropriate. On pages with a historical theme, however, year links are very important. For many with an interest in history, dates are important to fit individual events into a wider whole. The blanket opposition to date links seems to me to be based on the principle that "if I would never want to click on that link, no one else should be allowed to". I trust people to ignore links that do not interest them. You seem to be convinced that the RfCs have produced a clear verdict. All I have seen of the debate on this issue convinces me that "most" is extremely controversial and can't be remotely considered to be backed by consensus.Dejvid (talk) 11:38, 11 January 2009 (UTC)

We've been through this debate surely, and the consensus is very clear: we don't link run-of-the-mill dates in articles. There may be certain circumstances where dates can be linked (such as in chronological articles), but whatever those circumstances are, they constitute a small minority of cases, so "most" is perfectly legitimate. (I don't know where Locke gets the idea that the status quo is link all dates - that wasn't the case even before the decision to deprecate autoformatting.) If a new editor comes to this page wondering whether or not he should link dates in his articles, the best answer we can give is a clear "no" (with an equally visible link to a section which explains what exceptions there might possible be).--Kotniski (talk) 12:18, 11 January 2009 (UTC)

No—this "consensus" that is being spun out of all proportion needs to to be evidenced and discussed in each case. Trying to force your own views by spinning your RfC results is going to result in the failure of this merger: we'll just have to keep the existing mess. Tony (talk) 12:27, 11 January 2009 (UTC)

As the ones insisting all date linking is bad have failed to present evidence of a consensus on this, I would say the onus is on you to provide evidence of a clear consensus Tony to add that language. —Locke Cole • t • c 13:55, 11 January 2009 (UTC)

The best answer is a clear "no, not for auto formatting". The other issues (year links and month-day links) are much less clear. I sincerely wish people would stop misrepresenting the results of the RFC for their own purposes. —Locke Cole • t • c 13:55, 11 January 2009 (UTC)

I agree with Tony, a merger is a good idea, but not at the expense of agreement. Lightmouse (talk) 13:40, 11 January 2009 (UTC)

Yep, I'm afraid that Cole will go down as having wrecked this excellent move by Kotniski. It's not enough that you're dragging everyone to ArbCom, pushing your particular, personalised notion of what consensus is: you feel you need to launch the changes that you want, unilaterally, to the guidelines about to be merged. I'm sorry, but I'm not going to accept them. Why strike right now, just when the merger is being prepared? Tony (talk) 14:14, 11 January 2009 (UTC)

Because the longer this disputed language remains the harder it is to fight it. The RFCs concluded three weeks ago and for whatever reason you seem disinterested in accepting the results. I accept that dates linked purely for auto formatting must go, why can't you accept that not all date links are evil? —Locke Cole • t • c 14:17, 11 January 2009 (UTC)

Ah, I see that the merge has been done. I'm afraid that if Cole is going to engage in edit-wars to force his changes, the merge will need to be undone an we'll have to go back to the previous, messy, separate pages. Tony (talk) 14:16, 11 January 2009 (UTC)

Merge done

As you will have seen, I've made the merge, with the above couple of wording issues still to be resolved. I suggest that further discussion continue on the MOSLINK talk page.--Kotniski (talk) 13:19, 11 January 2009 (UTC)

Merged

Introducing the new merged version (incorporating material that was previously at Wikipedia:Only make links that are relevant to the context and Wikipedia:Build the web).--Kotniski (talk) 13:23, 11 January 2009 (UTC)

Wording about dates

This still needs to be resolved (see previous discussion at /merged). Any suggestions (in line with the consensus established in recent RfCs) welcome.--Kotniski (talk) 13:23, 11 January 2009 (UTC)

Rename

How about renaming this WP:Linking now? It seems to go beyond the scope of a mere style manual.--Kotniski (talk) 13:47, 11 January 2009 (UTC)

As most of the content is from the Manual of Style I think it should remain a MoS page. —Locke Cole • t • c 13:51, 11 January 2009 (UTC)

What content is from the Manual of Style? Doesn't seem to be very much to me, though I may be mistaken.--Kotniski (talk) 13:57, 11 January 2009 (UTC)

As I understand it everything is from WP:CONTEXT (a MoS page), WP:MOSLINK (a MoS page) and WP:BTW (the only page not part of the MoS, but so small that anything merged in is likely irrelevant). My concern is that if it's not part of the MoS then it needs to be vetted to gain consensus before being labeled as a {{guideline}}. If you keep it as a style guideline then of course it must be within the MoS. —Locke Cole • t • c 14:00, 11 January 2009 (UTC)

Don't understand all your arguments, but if there's no enthusiasm for this change, I'll hold off for now.--Kotniski (talk) 14:05, 11 January 2009 (UTC)

Kotniski: I'm unsure what the purpose of a rename right now would be. Tony (talk) 14:01, 11 January 2009 (UTC)

Please reverse the merger

OK, sorry Kotniski, I'm going to have to ask you to reverse the entire thing. Cole has started to edit-war, and I, for one, will not accept his unilateral demands.

Can you do this now, please? Tony (talk) 14:28, 11 January 2009 (UTC)

I'm sorry Tony but the changes are backed by the results of the recently concluded RFC I believe. Perhaps instead of constantly reverting me you should try discussing other options? Also, it would be silly to revert the merge when the only thing disputed is one sentence and two words... —Locke Cole • t • c 14:30, 11 January 2009 (UTC)

Yes, the wording dispute will remain regardless of whether we reverse the merge or not, so I don't see a need to undo what was a very popular move. But Locke, please can you say where you're coming from with this claim that the RfC supported date linking?--Kotniski (talk) 14:35, 11 January 2009 (UTC)

I'm sorry, Cole, I disagree with your spin, and if you'll look at MOSNUM talk, a lot of other people disagree with your spin. You're succeeding in wrecking the merger. Fine. Have it your way on that count. You will not be forcing your spin on this style guide.

Kotniski, can you bring back Context, please? Tony (talk) 14:38, 11 January 2009 (UTC)

Please stop with the WP:OWN behavior Tony. —Locke Cole • t • c 14:41, 11 January 2009 (UTC)

WP:MOSNUM/RFC#When_to_link_to_Month-Day_articles.3F and WP:MOSNUM/RFC#When_to_link_Year_articles, which I believe show support for some date linking (generally where it's relevant was the impression I got). At worst it's no consensus, which returns us to the prior status quo (link all dates) until consensus can be reached on a change. —Locke Cole • t • c 14:41, 11 January 2009 (UTC)

Kotniski: I will take you to ANI then. You have made a major move with the disagreement of at least two people here. I request again that you undo it, and return CONTEXT. Tony (talk) 14:51, 11 January 2009 (UTC)

Don't understand how that would help. Everyone (including you) was full of praise for the merger. We can easily restore the original wording about dates without reversing the merger (which I've just spent an hour doing).--Kotniski (talk) 14:56, 11 January 2009 (UTC)

Tony please stop threatening people. And who is this second person that's disagreeing? —Locke Cole • t • c 14:55, 11 January 2009 (UTC)

I was full of praise for the draft, before Cole walked in and changed key wordings. You have acted prematurely. Both Lightmouse and I have objected. Tony (talk) 15:01, 11 January 2009 (UTC)

I believe the merger is wonderful. Count me as part of the consensus in favor of it. Tennis expert (talk) 19:52, 11 January 2009 (UTC)

Yes. I was also full of praise for the draft. But the recent changes to meaning are worse than having split pages. Lightmouse (talk) 15:03, 11 January 2009 (UTC)

Those changes reflect the recently established consensus from WP:MOSNUM/RFC. Where is the problem with that? —Locke Cole • t • c 15:05, 11 January 2009 (UTC)

OK everyone, calm down. Let's not touch anything on the page as from now, so we at least know what version we're talking about. Now, what (of importance) does it fail to say now that any of the pages said before the merge?--Kotniski (talk) 15:13, 11 January 2009 (UTC)

Locke Cole, the prior status quo was not link all dates. The RFCs showed us that the consensus was to link dates on a very limited basis, especially in the case of month-day articles. Please cite more specific "consensus" than just the RFCs; something along the lines of this would be good. Keep in mind that there have been other places where consensus was demonstrated (FAs, FLs, User:Tony1/Survey of attitudes to DA removal, etc.). Dabomb87 (talk) 15:24, 11 January 2009 (UTC)

Cole has made it quite clear that he's going to try to force his way. He will continue to use this merger as a chance to promote his spin on the RfCs at MOSNUM. Others will simply not accept this spin. It is and will continue to be an impasse. Kotniski, I'm sorry that your work (and mine) is being capsized, but you have way-too-prematurely implemented the merge, knowing that there were disagreements. This was an error of judgement (I make them too—we all do) and reverse it. Please bring back CONTEXT; you had no right to remove it without consensus. Tony (talk) 15:27, 11 January 2009 (UTC)

Tony even if this is unmerged the issue will not go away. It'll just be spread between three pages instead of this one. —Locke Cole • t • c 15:32, 11 January 2009 (UTC)

Dabomb87, the RFC is all the needs to be linked to. You can see either by sheer number of !votes or by actual opinions expressed that there's consensus for month-day and year links to be made "sometimes". There's definitely no consensus there for "generally never link" as is being proscribed here. And yes, the prior status quo was "link all dates" because there was never a community consensus for the initial change in the first place. —Locke Cole • t • c 15:32, 11 January 2009 (UTC)

As long as I can remember (2 years-ish?), the status quo (whether actually documented or just observed in practice) was that dates are not generally linked except for autoformatting. It was recently agreed that we lose the autoformatting. So as I see it, dates are not generally linked. That is not incompatible with the RfC result that dates should be linked "sometimes", since the "current" version of the guideline also implies that dates are linked sometimes. So what is it we actually disagree about?--Kotniski (talk) 15:43, 11 January 2009 (UTC)

The disagreement is about dates being listed under "generally not linked" (or whatever the section title is). For auto formatting, I would agree (I'd even go with stronger language, "never linked for auto formatting"), but for just general linking I believe it gives our editors the wrong impression (especially given the results at the RFC). Clearly there's support for "sometimes" linking, so we need to be specific that it's just dates linked for auto formatting that aren't okay. —Locke Cole • t • c 15:47, 11 January 2009 (UTC)

Tony, "consensus" does not mean that everyone agrees, as far as I can see it is only you and Lightmouse who currently disagrees about the merging of these pages. That there is a disagreement about some of the wording of areas of this page is a separate matter. I would also suggest that Locke Cole is right about one thing, you do seem to be owning this page a bit. Can we all calm down and actually discuss the disputed area? Regards, Woody (talk) 15:56, 11 January 2009 (UTC)

That is your opinion. A third person has arrived to object (Dabomb); is this an error of fact? There was and still is no consensus for the merger as yet; only disagreement about important wording. I want to know where CONTEXT is, its talk page and its archives. Tony (talk) 16:08, 11 January 2009 (UTC)

Tony it's pretty clear that you and the MOSNUM regulars aren't interested in accepting the results of WP:MOSNUM/RFC, but I really wish you'd calm down and talk about this rationally. CONTEXT is still available via article history here: link. Click on "Talk" which hasn't been redirected yet if you'd like to see the discussion archives. —Locke Cole • t • c 16:18, 11 January 2009 (UTC)

• No, I don't accept your spun interpretation of them, which you were always going to do whatever the results. There is no consensus on what they really mean. Tony (talk) 16:29, 11 January 2009 (UTC)

• Please stop referring to my reading of the results as "spin". I'm open to discussing this, but if you continue to stonewall discussion rather than being flexible there's not a lot of choice for me, is there? I certainly don't want to degenerate this discussion down to "what the consensus of the consensus is" as you seem to want to do... —Locke Cole • t • c 16:40, 11 January 2009 (UTC)

• For all interested parties, I am writing up a detailed summary of consensus of the two RFCs here. In this case, I agree(!) with Locke Cole. It seems counter-productive to revert for such a little thing. We can always change the wording later. Dabomb87 (talk) 16:46, 11 January 2009 (UTC)

When I'm faced with what I see as large-scale distortions of the truth in your statement at the ArbCom thing, "spin" also comes to mind. It's a pattern in your contributions to the debate over the past few months, except that it has become more extreme and less compromising in the past six weeks. This has been difficult for other editors to live with, I believe. Perhaps you don't realise how you come over.

Dabomb, I now see MOSLINK as illegitimate, and will advise editors to disregard it at FAC and other forums. Tony (talk) 16:49, 11 January 2009 (UTC)

• I have no problem with that Tony, I doubt that most of those editors at FAC keep regular tabs on the MOS anyway. Dabomb87 (talk) 16:53, 11 January 2009 (UTC)
• Frankly Tony I'm getting a little tired of your constantly disparaging remarks. We disagree, and I'm sorry about that, but characterizing my opinions and remarks as you have is counter productive. I've tried, and tried, and tried again to reason with you over this. You wouldn't accept my attempts at compromise, so we held RFCs. And now you are seemingly disinterested in accepting the results of those as well. Just what does it take for you to consider alternatives other than your own preferred way? —Locke Cole • t • c 16:54, 11 January 2009 (UTC)

• Anomie (talk · contribs) did a writeup of the results (which I largely agree with) here. You might wish to consult that when doing your writeup. Though this does seem to be veering us down the road of "what the consensus of the consensus is", if it results in agreement I'm willing to tolerate it. —Locke Cole • t • c 16:54, 11 January 2009 (UTC)

• A nice analysis. However, it seems to look at the reasons for votes only rather than the consensus; it also doesn't examine the consensus on linking dates. I will definitely take that into account. Dabomb87 (talk) 16:59, 11 January 2009 (UTC)

• You're right, I missed that. For some reason I thought his analysis covered the entire RFC. Naturally I don't dispute the auto formatting linking issue. My concern is the month-day and year linking issue of the RFC. —Locke Cole • t • c 17:13, 11 January 2009 (UTC)

I have to go and do things in the real world now, but just one practical observation: 99% of all dates in Wikipedia articles are of exactly the same type - simply saying that something happened on a particular day and/or in a particular year (or maybe century etc.) It seems irrational to split these into categories of those that should and those that should not be linked - at least, all attempts I've seen to make such a split have failed. So all we can do by way of guidance is to say whether or not these regular dates are or are not to be linked. It's a decision editors have to make several times per average article, and basically they just need to know. It's yes or no - "sometimes" (if undefined, which in the light of failure to find a definition it has to be) is no help at all. So tell us, those who have been following the debates, which is it? --Kotniski (talk) 16:51, 11 January 2009 (UTC)

• All of this nonsense should STOP. You stoopid, or wot? Locke's ArbCom request is still pending, and he's digging up yet more trenches and opening another front in the battlefield? WP:CONTEXT is back where it was for now. Lets calm things down a few notches. Ohconfucius (talk) 02:43, 12 January 2009 (UTC)

• Please assume good faith, there were no objections to the merger until one sentence and two words were changed. Hardly a call to revert all the work done by Kotniski, and even the language that was changed has been largely changed back (so I seriously don't understand these objections and edit wars). Further, you've only undone part of the merge, as I mentioned at WT:CONTEXT to you, Kotniski said it took him an hour to perform the merge, so you've left a great many things undone... —Locke Cole • t • c 02:48, 12 January 2009 (UTC)

• As I said in an addendum to my statement in your ArbCom request, you are giving serious reason for me call into question your good faith. I am not saying Kotniski voluntarily involved himself in your trench warfare, but you are certainly dragging him into it whether he liked it or not. You seem to be aware that it was principally the change you introduced which got my back up, and it appears also Tony's (maybe there were others as far as he is concerned). The moment of merger is not the moment to make these sort of changes, so I suggest you remove your change, let things settle down, and then we can look calmly at whether the merger was faithfully executed, not that I have any reason to doubt it. Ohconfucius (talk) 03:05, 12 January 2009 (UTC)

• Numerous editors suggested I participate in this merger, and my proposals were made prior to the merge being performed. Kotniski rightly understood the changes to be minor compared to the overall good work being done, but for whatever reason Tony, Lightmouse and now you seem to be taking great issue with this minor minor change. What's worse here, and what gives me pause, is that my changes have been largely reverted. The meaning and wording now isn't that much different from what was "good work" and "acceptable" only a day ago. So why, exactly, are you undoing this hard work (reverting pages and redirects and making inflammatory comments here instead of partaking in discussion)? What is your specific objection? Is it something else that I didn't change that was lost in Kotniski's merge? —Locke Cole • t • c 03:15, 12 January 2009 (UTC)
• This diff compares the merged version and the version currently up on this page which demonstrates that most of my changes were undone. So why the explosion of disruption and anger again? —Locke Cole • t • c 03:59, 12 January 2009 (UTC)

• Whoever invited you to contributing to WP:MOSLINK should revisit their decision. Your record, and your narrow partisan approach speak for themselves. Ohconfucius (talk) 04:14, 12 January 2009 (UTC)

• I'm OK with the current state of the page, even though I note that the word "demonstrably", which was in WP:CONTEXT, has been removed ("demonstrably deepens readers' understanding of the topic"). While I'm not happy about this, I'm willing to accept it in the spirit of compromise. Is there an objection to removing the dispute tag from that section? And I'm OK if CONTEXT is finally removed to complete the merger. Your thoughts? Tony (talk) 04:46, 12 January 2009 (UTC)

• (ec)As you said that the changes were put through before the merger, I will need to do some combing back to be sure there are no material changes. Reserving my position. Ohconfucius (talk) 04:49, 12 January 2009 (UTC)

In any case, please don't reverse the merger... those 3 pages were really about the same thing. Nicolas1981 (talk) 05:56, 12 January 2009 (UTC)

Chronological items wording

As far as I can tell, the main outstanding dispute is over the wording of the "chronological items" section, since that's the only part that differs from the merge proposal that was accepted by all. So please make suggestions for improvements here (although I think the only actual difference is the absence of the one word demonstrably).--Kotniski (talk) 07:04, 12 January 2009 (UTC)

As I said, I can swallow the absence of "demonstrably". It's OK as is, IMO. Tony (talk) 07:39, 12 January 2009 (UTC)

i hope this isn't another can of controversial worms: the sentence "Articles about other chronological items or related topics are an exception to this guideline" isn't clear enough about what kind of exception is meant. how about something like:

In most articles, items such as days, years, decades and centuries should generally not be linked unless they are likely to deepen readers' understanding of the topic.[3] Articles that are about chronological items are exceptions: in them, links to other articles about chronological items are generally considered relevant and useful.

Sssoul (talk) 07:25, 14 January 2009 (UTC)

Well, I don't consider rafts of linked dates in those articles to be at all useful: more a hindrance to the reader, given the visual interference with the very next item to the right, which is typically linked. See the ones that bold the initial dates instead—so much better looking and easier to read. Who is going to click on a link to 3 January when they're looking at an article on "2009 in South African television"? Therefore, I suggest wording that doesn't encourage this, but merely doesn't forbid it, in these articles. Tony (talk) 09:58, 14 January 2009 (UTC)

how about:

In most articles, items such as days, years, decades and centuries should generally not be linked unless they are likely to deepen readers' understanding of the topic.[3] Articles that are about chronological items are exceptions: in them, links to other articles about chronological items are more likely to be relevant/appropriate.

i personally think the formatting used in chronological-item articles is something for the editors of those articles to settle "locally", but ... well, any other suggestions for how to word this? Sssoul (talk) 13:21, 14 January 2009 (UTC)

List of major historical events

This manual currently says "It is generally not appropriate to link items that would be familiar to most readers, such as the names of major [...] historical events [...]".

The last really big event that happened since Hiroshima/Nagasaki is probably the German reunification, so that would mean editors should generally not link to this article ? I am not too sure what "major events" encompasses. There is probably not so many historical events that are familiar to most readers, so it would be helpful to list them (at least here under), so that we know what are talking about. Please list what you consider are the major historical events, thanks ! Nicolas1981 (talk) 11:52, 12 January 2009 (UTC)

I don't think there are many: the two World Wars, probably. I think a lot depends on the context, and this concept probably still needs to be made clearer - while it might be appropriate to link WWII in the introduction to an article about a major episode of that war (where it provides immediate context, an immediately next-higher node in the tree), linking probably wouldn't be necessary in the vast majority of incidental references to that war throughout articles.--Kotniski (talk) 12:09, 12 January 2009 (UTC)

OK, so for you that would be World War I and World War II, depending on context. Anyone else ? It is better to make it clear now rather than face different interpretations later. Nicolas1981 (talk) 15:01, 12 January 2009 (UTC)

I agree with Kotniski; it's context-dependent. "Reunification" looks like a reasonable link. Tony (talk) 15:42, 12 January 2009 (UTC)

So, Attack on Pearl Harbor would need a link to WWII or not ? Bernadette Chirac (WWII impacted her life a bit, but that is clearly a detail in the article) would need a link to WWII or not ? Thanks ! Nicolas1981 (talk) 02:57, 13 January 2009 (UTC)

List of major religions

This manual currently says "It is generally not appropriate to link items that would be familiar to most readers, such as the names of major [...] religions [...]".

There will probably be very few, so we'd better cite them explicitly. According to the Religions article, the 5 major religions are Christianism, Islam, Hinduism, Chinese folk religion and Buddhism. I have never heard of the sixth one. I somehow thought that Judaism would rank higher, but it is probably because of my cultural background. So, what religions do you think are familiar to most readers and generally should not be wikilinked ? Nicolas1981 (talk) 15:20, 12 January 2009 (UTC)

In country articles, christianity and possibly the others appear to be overlinking, given the readership of the eng.WP. It depends on what kind of information about the religion in the context is going to deepen the reader's understanding of the topic at hand. You need to visit the Islam article, for example, to see whether this is a reasonable diversion by most readers of the original article. Does the original article provide the basic information required there? Is any further information required? If you can come up with a few examples, we might have a better idea. Tony (talk) 15:47, 12 January 2009 (UTC)

You're right, we'd better try with a few examples. Sandalwood has a pretty good article, and among other things, a short part of the article mentions how this product is used in Hinduism. So Hinduism is quite secondary in this article, nearly anecdotal. Should Hinduism be linked or not ? Nicolas1981 (talk) 02:10, 13 January 2009 (UTC)

Why no wikilinks in section headings?

Just came across the first bullet point under General Principles, and this was news to me: while I think links in headings should be rare, I can see some cases where they would make sense. Just for my edification, what is the rationale that they should not be used? UnitedStatesian (talk) 02:55, 14 January 2009 (UTC)

I guess it's just to make headings look nicer. Links in headers are authorized and widely used on the French Wikipédia, so it is just a local consensus of the English-language Wikipedia community. Personally, I am fine with it as long as appropriate terms are wikilinked soon after the heading. For instance, this paragraph about Eurofor is the first occurrence of "Eurofor" in the WEU article, so it has an Eurofor wikilink in the beginning. Nicolas1981 (talk) 04:50, 14 January 2009 (UTC)

Can you please provide an example of paragraph where it would make sense ? Thanks Nicolas1981 (talk) 08:14, 14 January 2009 (UTC)

I'll go back through my contributions; I know I have added them one or two times in the past, and would appreciate the specific feedback. UnitedStatesian (talk) 13:59, 14 January 2009 (UTC)

Bring back WP:CONTEXT and WP:BUILD and mark them as historical

For the record, I support merging the CONTEXT and BUILD guidelines into MOS:LINK, so that there is only one normative guideline instead of three. However, in retrospect, I don't feel that they should redirect here. Both of them had a comparatively long history on Wikipedia, and were familiar to many editors. Both of them date from April 2002, while MOSLINK wasn't started until November 2004, over 2.5 years later. I think that both Wikipedia:Only make links that are relevant to the context and Wikipedia:Build the web should be restored to their former states and marked with the {{historical}} tag. A hatnote can be placed at the top of each noting that "The scope of this former guideline is now ruled by WP:MOSLINK" or something to the same effect. Both guidelines were highly influential over the past few years, and the dynamic tension between them is part of our history. They should be preserved as such for posterity, and because we will still want to refer to them as points of view and recommendations to be followed, even though we now hold MOSLINK to be the controlling guideline in this matter. All opinions are welcome as to the wisdom of this suggestion. If consensus is overwhelmingly against this option, then I will just move them to my userspace for posterity's sake and leave the redirects pointing to MOSLINK.--Aervanath talks like a mover, but not a shaker 16:01, 14 January 2009 (UTC)

I see that the WT:CONTEXT archives have already been linked to above. I am adding a link to WT:Build the web as well, since that was also merged here.--Aervanath talks like a mover, but not a shaker 16:04, 14 January 2009 (UTC)

No objection from me if you want to bring them back as historical pages.--Kotniski (talk) 16:16, 14 January 2009 (UTC)

I think it would be unwise: pages linking to the historical pages should jump here for current guidance regarding links (and any historical links on talk pages, etc. should likewise come here). You could, perhaps, move the pages to subpages of their former location (restoring the redirect post-move), then marking those subpages as history. For example, move Wikipedia:Only make links that are relevant to the context to Wikipedia:Only make links that are relevant to the context/Historical. Thoughts? —Locke Cole • t • c 16:20, 14 January 2009 (UTC)

Locke, could you elaborate more on why you feel they should automatically redirect here? Even though they're not current guidelines, I feel that a big ol' historical tag at the top, with a hatnote (make it as prominent as you like) telling people that the page is explicitly overruled by MOSLINK, would be enough so that people could find the right rules quite quickly. I don't think it would take very long before everyone figured out that they had to cite MOSLINK as an authority, instead of CONTEXT or BUILD.--Aervanath talks like a mover, but not a shaker 16:33, 14 January 2009 (UTC)

Because we should use the redirection system to make it easier to get to what's relevant quickly. Historical pages aren't relevant when you're following links to guidelines/policies. Plus the pages will still be categorized by tagging them with {{historical}}, so it shouldn't be hard to find them that way. I'm just more interested in making sure readers are directed with as little effort as possible to what's current. —Locke Cole • t • c 16:47, 14 January 2009 (UTC)

We'll agree to disagree on that, then, as I still think they should continue to be at their previous names. Let's wait and see what other users think before doing anything.--Aervanath talks like a mover, but not a shaker 18:37, 14 January 2009 (UTC)

We can keep them with a very clear warning that it is not the current guideline, and modify incoming links to point to WP:MOSLINKS (on a case-by-case basis). I started doing the latter yesterday, there is not that many if you filter out talk pages and archives Nicolas1981 (talk) 02:04, 15 January 2009 (UTC)

I would leave the shortcuts WP:CONTEXT and WP:BUILD as pointing to the new merged page, if that's an issue.--Kotniski (talk) 12:08, 15 January 2009 (UTC)

That would be acceptable to me if and only if there was a prominent hatnote at the top saying something to the effect of "WP:CONTEXT and WP:BUILD now redirect to this page, as their function has been subsumed by this guideline. To see those historical guidelines, see Wikipedia:Only make links relevant to the context and Wikipedia:Build the web."--Aervanath talks like a mover, but not a shaker 06:15, 17 January 2009 (UTC)

I don't think that's necessary. The historical pages are just for academic interest and relatively unimportant. They should be listed under "See also" at the end of the merged page, with (historical) after them, but no need to give them any more publicity than that. The shortcuts are used with the intention of linking to current guidelines, and the current relevant guideline on those subjects is this one. --Kotniski (talk) 11:13, 17 January 2009 (UTC)

Kotniski has a point, the important existing links are meant to show the current guideline, not some history. Anyway, not a big issue, and updating the links themselves would make it a non-problem. Cheers Nicolas1981 (talk) 15:59, 17 January 2009 (UTC)

Articles containing links to the User space

Recently, I came across this report, and set about removing some of these links, which were mostly people signing when they shouldn't, or plain vandalism. One of my edits was reverted with the question "Why". It was a good question - if that editor is, indeed, the 'Director of Media Relations', of course - but I could not find any policy or guideline that says this such links should not be present. Other articles, such as San Diego Trolley, link to the user space via a template for the authorship of a photo. So, the question is: is there a guideline or policy? Cheers, Stephenb (Talk) 20:49, 21 January 2009 (UTC)

• Elonka Dunin does not link to User:Elonka; Jimbo's article is linked to his userpage, but that is in the form of 'Official Website' link. But Jimbo's case is demonstrably different to this one. There is a problem with the San Fernando Valley Quakes article, the most obvious one being conflict of interest. The article's premier editor is the team's PR manager. The article is unsourced and I feel it has a 'partisan commentator' tone to it. As for the templates, it may be an idea to leave message on the creators' talk pages suggesting these templates be moved to mainspace. Ohconfucius (talk) 03:52, 4 February 2009 (UTC)
• They most definitely don't belong. {{Notable Wikipedian}} should suffice. I did just notice, though, that the user who added back the link was the user who the link went to. If anything it's WP:SELF in a way and his readdition is something of a WP:COI. §hep^Talk 05:31, 22 February 2009 (UTC)

Wikilinking in list tables

A number of us have been discussing whether or not links should be repeated in list tables, and if so what rules should be followed. This conversation is here. The discussion started with an edit to the List of operas by Mozart, see [3] with three instances of the Teatro Regio Ducal being linked on lines 88, 100 and 123.

MOS:LINK says '"Link only the first occurrence of an item. A link that had last appeared much earlier in the article may be repeated, but generally not in the same section. (Table entries are an exception to this; each row of a table should be able to stand on its own.)" but how should this be applied to, say List of operettas by Offenbach? Does the list turn into a sea of blue? There are about 100 entries in this particular table with Paris in almost every one. Any thoughts? --Kleinzach 03:10, 28 January 2009 (UTC)

Interesting problem. I've reviewed quite a bit at Featured List Candidates, where this is an issue. Frankly, bright-blue lists can be a bit garish, but what can be worse is speckled blue and black down a column. IMO, horizontal inconsistency doesn't matter so much (they're typically quite different categories, and a vertical unity looks neater/more logical in terms of formatting). Year-in-foo links in tables, I believe, are most unlikely to be clicked on, since they deceive the reader into thinking they lead to a plain, sea of irrelevant information on a year-page.

If I were involved in an article, I'd suggest that the most important few year-in-foo links (maybe even plain year-links, if someone insists), be highlighted in the See also section at the bottom of the main text, where they can be piped more flexibly and addition information added where useful. That way, tables are more likely to be viable as plain black text without worrying about speckle or overall garishness. Just my opinion. Tony (talk) 08:56, 29 January 2009 (UTC)

In this case, we were looking at names rather than dates. I don't think anyone has very strong opinions on this, we are really looking for guidance - that might be usefully recycled into the MoS. --Kleinzach 09:53, 29 January 2009 (UTC)

Got some examples? Tony (talk) 11:06, 29 January 2009 (UTC)

Almost every list of Category:Lists of operas by composer can serve as example; take List of operas by Hasse: should every occurrence of every genre, librettist, theatre for which Wikipedia has an article be linked or only the first? This is a sort-of trick question because the tables are sortable and the notion of "first occurrence" doesn't mean much. I wouldn't mind if every occurrence gets linked because of the sortability issue, but I understand the dislike of a sea of blue. The question was brought here because of this dilemma. Michael Bednarek (talk) 13:04, 29 January 2009 (UTC)

I checked a few from that category, as you suggested—they look fine (not heavily blued at all, nor speckledy). In fact, when the linking is selective, I find myself much more likely to hit a link. On another issue I probably shouldn't raise on this page, why are the operas of these composers in separate articles to their other works? Tony (talk) 13:51, 29 January 2009 (UTC)

That's the point - they were edited on the 'first instance linked' rule - which was then challenged. (I've answered your other question elsewhere). --Kleinzach 01:12, 30 January 2009 (UTC)

The problem there is that the table is sortable; if the table is sorted a certain way, links that were once near the top are now near the bottom. That is why there should be an exception for sortable tables. Dabomb87 (talk) 03:32, 2 February 2009 (UTC)

Historical events versus recent events

OK, I'm confused. First of all: what historical events are most of our readers actually familiar with? And why is linking to recent and current events OK, when readers are far more likely to know about those events? The way this is written now, in an article about a General who fought in the War of 1812 (although I doubt most readers are familiar with it, per se) we wouldn't link to it, but in an article about a General who fought in the Iraq War we would link to it, even though many many more readers are familiar with that war. That seems completely backwards. -- Kendrick7^talk 16:21, 1 February 2009 (UTC)

I actually think that linking to important historical events is far better than linking to lone years or year-in-x articles. Dabomb87 (talk) 03:27, 2 February 2009 (UTC)

When this was raised before, the only historical events that were suggested as generally not linkworthy were the two World Wars. If there are no other suggestions, maybe we could say something like "very well known historical events such as World War I and World War II"?--Kotniski (talk) 10:38, 2 February 2009 (UTC)

Articles can span multiple topics, so linking to multiple "Year in X" articles should be OK

This should go without saying, I would think, but my attempt to change Wikipedia:Manual_of_Style_(links)#Chronological_items was reverted. An article like Ben Franklin could easily link to "X in invention", "X in politics", "X in diplomacy", "X in finance" etc. I don't understand the need to pick just one. -- Kendrick7^talk 16:27, 1 February 2009 (UTC)

Probably because your wording was ambiguous. Dabomb87 (talk) 03:28, 2 February 2009 (UTC)

Temporal links should be for context, not simply topical

The current wording here is disingenuous (Edit: for lack of a better word). None of our year and era articles are specifically topical in a way that would ever deepen a readers knowledge on any topic. It would be better to come straight out and forbid the links, or say that providing historical context (per WP:CONTEXT) is OK. -- Kendrick7^talk 16:33, 1 February 2009 (UTC)

The thing is, most the year articles are filled with random trivial facts and don't even provide context. Now, if you want a helpful year link, see 1345. Dabomb87 (talk) 03:29, 2 February 2009 (UTC)

I'd be happy if what was done with that article was done a few thousand more times. (Removing year links makes such a project more difficult, since an armchair historian can't bootstrap a better article by using "What links here", as I've tried to suggest elsewhere.) But under the current wording, we shouldn't even link to 1345 or any other "nearly GA" article because no year article is inherently topical. -- Kendrick7^talk 04:33, 2 February 2009 (UTC)

That's been normal practice for a long time, and recent RfCs seemed to confirm it. Of course all guidelines have occasional exceptions.--Kotniski (talk) 10:40, 2 February 2009 (UTC)

What has been normal practice? -- Kendrick7^talk 19:44, 2 February 2009 (UTC)

Being a member of the WikiProejct Years, the gathering of information for year-articles is, of course, a concern to me. What is wrong with the "search" box? This yields 2157 results. Granted that some are simply to "1345 in [topic]" articles, and some may be false postives, it does present a rather large database. Tony (talk) 13:22, 2 February 2009 (UTC)

You'll actually get tons of false positives. For example, off the top of my head, searching for 1897 will list Marie Curie in there somewhere, when the only notable event in her life in that year was her giving birth to a daughter, which I would never include in the 1897 article. 3 digit and 2 digit years are also problematic -- searching on those numbers will yield articles that have nothing to due with those numbers as years. Anyway, this was just an aside and not my main point. -- Kendrick7^talk 19:44, 2 February 2009 (UTC)

So we certainly don't want "1897" linked in Curie's article when it talks of the birth of her daughter. Tony (talk) 23:33, 2 February 2009 (UTC)

I would say not, among Mrs. Curie's many claims to fame, being a mother isn't generally one of them. -- Kendrick7^talk 02:58, 3 February 2009 (UTC)

OK ... but that's what we used to do! Tony (talk) 03:39, 3 February 2009 (UTC)

I agree. I agree wholeheartedly that there has been chronological overlinking, encouraged especially by autoformating. But as I still see shades of gray here such that I oppose underlinking, I disagree with any scorched earth solution. I certainly disagree with the misdirection of the wording of this section of the this part of the MoS. It's just a sugar coating over saying, more directly, that years should never be linked, because, AFAICT, that's what it really amounts to. -- Kendrick7^talk 06:55, 3 February 2009 (UTC)

• Well, I believe the onus should be on the article editor who wants to retain a special-case link to make a demonstrable case. Otherwise, after years of an ingrained, unconscious culture of linking every year (even centuries, decades, days of the week if you please), one fears blue-creep. I agree that it's not easy to make a hard-and-fast rule, but there are still people who believe that every year should be linked, and every date autoformatted (although their number has dwindled significantly over the past year or two). There's also the issue of the central role played by automated (which spare editors much grunt-work and have been given bad press by a few people who are offended by their role in this particular issue). One could always pipe the year-link, I suppose, but I believe we shouldn't encourage this as general medicine. The "See also" section, IMO, is an ideal solution. Tony (talk) 10:28, 3 February 2009 (UTC)
  • I believe that we are not in the business of constructing a legal system, complete with waivers for breaches of the linking code. If Tony1 wants to do that kind of thing, he should go into local politics, and volunteer to oversee statutory planning. Septentrionalis PMAnderson 18:02, 27 February 2009 (UTC)

Non-style aspects of this guideline

I went to review some wording in WP:BUILD and I was somewhat dismayed to find that it had been merged with a style guideline. I certainly think that all of the style-related aspects of linking should be in one place but my understanding of WP:BUILD was that it was more fundamental than a simple matter of style: it enjoined the editor to specifically create Wikipedia as a richly interlinked and interconnected encyclopedic work.

In contrast this page seems to take a sort of ho-hum attitude to it and relegates it to a simple matter of styling rather than a core element of the essence of Wikipedia. Whereas "Build the web." was before a directive of the project, this page seems more like "Build the web, y'know, whenever it's convenient, you can make an argument of context, or it looks nice."

I'm against overlinking and promiscuous linking of things like dates, I don't think date-linking fulfilled WP:BUILD as it was written. But if it's really just a matter of style now it seems to me that Tagishsimon's comments in VP about orphan and wikify tagging is kind of appropriate; if "Build the web" is no longer a fundamental principle of WP and we're only talking about how you're going to style it when you do happen to build the web, it seems that in many cases {{orphan}} and {{wikify}} may not be appropriate, because it's basically okay if an article has a context that doesn't make internal linking needed or appropriate.

(Whereas before it appeared to me that we were saying articles need to be formulated and written in a manner that makes linking appropriate - and if it wasn't the case that an article had such a context, the article needed to be reformulated / rewritten / re-envisioned under the preceding guideline in such a way that it would be appropriate for it to include linking.)

If I'm correct about this, it seems a sentence or two in the header about internally-linked content being a primal part of WP's essential purpose could fix things. But Wikipedia is a dynamic project, so maybe it has actually changed and "Build the web" is no longer a motivating part of the essence of WP. Or maybe I misinterpreted that old guideline and it never meant what I thought it did. What do people think? --❨Ṩtruthious ℬandersnatch❩ 06:15, 5 February 2009 (UTC)

The lead starts with this: "Linking is one of the most important features of Wikipedia. It binds the project together into an interconnected whole, and provides instant pathways to locations both within and outside the project that are likely to increase our readers' understanding of the topic at hand." Is that what you are referring to? Tony (talk) 09:17, 5 February 2009 (UTC)

I did see that but it seems more descriptive to me than WP:BUILD was - it's like saying something equivalent to "The PHP scripting language and interpreter is a powerful piece of software that is essential to Wikipedia." That's true but the Wikipedia Project does not specifically endorse PHP or advocate its use. Conversely when we had a guideline telling people to "build the web" we were being much more prescriptive than simply stating "hypertext is a very powerful medium and its use is a significant aspect of Wikipedia" - we were enjoining WP editors to do something in the course of improving the encyclopedia.

If I'm correct in my various interpretations of things I would want to add something like, "An important guideline of Wikipedia is to build the web: articles should be written in a manner that promotes interlinking with related topics and subjects that provide context to the reader." And maybe even something like "If an article cannot be linked to any other Wikipedia article at all or if links cannot be established to it from other articles, this may indicate that something is wrong: either the subject of the article may not fulfill the criteria for notability or it may lack a thorough enough description of the subject to give the reader the context necessary to understand it."

(I could understand if people objected to the second one there; I couldn't quite remember what WP:BUILD conveyed to me and I'm not sure if that's a good summary.)

But I'd also note that I'd wonder if it would be appropriate for a style guideline to say something like that. (I really don't know, I haven't spent much time examining style guidelines.) If it's not, is there maybe some other guideline somewhere it could go in? --❨Ṩtruthious ℬandersnatch❩ 10:14, 5 February 2009 (UTC)

BTW wasn't written like a normal style guide; more like an essay pushing a view. That, I think, was its original status, which was then changed to some halfway-house kind of policy, then a style guide, somehow (I could never find the consensus for either). I think you're right, that WP, and perhaps wikis in general, have become pickier about linking. This is reflected in the changes to CONTEXT and MOSLINK over the past two years, and in the dispensing with the old date-autoformatting practice. I have no issue with the insertion of the point that there is usually at least one link in an article. This is probably more appropriate as a point in the main text rather than in the lead. However, some stubs lack a link; this is only natural, and should be mentioned, don't you think? Tony (talk) 14:17, 5 February 2009 (UTC)

Just to make it a little easier to follow this, here is a link to the build the web article just before the recent merge into this article. Build the web Zodon (talk) 07:23, 6 February 2009 (UTC)

So Tony, your wording there assumes that BTW was never anything other than a style guideline - am I interpreting that correctly? (I'm having difficulty being certain of whether you're relating your personal impression like I was or if you're indicating a convention or broader usage.) Was there any discussion in the past which indicated this? (edit) Urk, that was a monumental misreading on my part, sorry. It just seems to me that as the name of the guideline itself is an imperative, I would think that it can't have been the original author's intent to merely conditionally talk about styling.

And thanks for the link, Zodon. I note that BTW was not categorized as a style guideline and had the general guideline infobox and navboxes rather than style ones.

Also, going back to the very first version, the following line would seem to agree with my interpretation of the guideline's intent: "Don't just write the article, but also consider its place in the link web." And from a quick survey of some history entries I don't see anywhere that the page was identified as an essay; in the history entries I looked at it was either identified as a guideline or "semi-policy" at one point, or not classified at all. (Though I don't know when the policy-guideline-essay distinction began except that I know the classification "info page" is recent.)

So with that evidence in hand I'm going to add the first sentence I mentioned above, though we might need some BRD cycles or rewriting to tighten it up. (Or, if as I said before it belongs better in another essay, once we secure consensus to transfer the language there we can.) --❨Ṩtruthious ℬandersnatch❩ 09:53, 6 February 2009 (UTC)

I may be missing something, but all this talk of web-building just seems to be saying the same as what we already say in the lede, except in different (and rather vaguer) language. I'm not sure why anything needs to be added.--Kotniski (talk) 11:02, 6 February 2009 (UTC)

Having said that, I've just added something to emphasize what I understand you want to emphasize.--Kotniski (talk) 11:14, 6 February 2009 (UTC)

• I can't see that these changes add anything useful to the lead. What was wrong with it before? Tony (talk) 12:45, 6 February 2009 (UTC)

Wikilinking educational background

There's a mini-biography of a shooting victim, and one editor is arguing that, though it's important enough to include details of the victim's educational background, the wikilinks to the decedent's high schools should be deleted. What's the proper style? Discussion at Talk:BART_Police_shooting_of_Oscar_Grant#Wikilinking_high_schools. THF (talk) 19:33, 7 February 2009 (UTC)

script to remove wikilinks from section headings

Copy and paste the article's source into "orig.txt" and run the following script. The results will be output in "new.txt". Any wikilinks in headings will be removed and placed into a {{Main}} directly below the heading. Make sure to double check the results for any errors before submitting.

perl -pne 'if (m/^(=+).*\[\[([^]|]+)(?:]]|\|([^]]+))/) {$a = $3?$3:$2; $_= "$1$a$1\n{{main|${2}}}\n"}' orig.txt > new.txt

-- Intractable (talk) 11:15, 13 February 2009 (UTC)

BTW merge

Tony twice attempted to move discussion here, it's been moved back to Wikipedia talk:Build the web, where it belongs.

Resurrect this guideline?

Namely WP:Build the web. Note: for the time being, I have restored the text of the guideline, as it is unfair to expect that people can argue for the life of someone when then have already been executed. This is for discussion purposes, not edit warring, and I will adhere to the eventual result of the discussion. -- Earle Martin [^t/_c] 12:48, 17 February 2009 (UTC)

Let's try to get the discussion back to the issues. It is proposed to restore, as a separate page, the text of WP:Build the web that existed before it was merged with WP:MOSLINK and WP:Only make links that are relevant to the context a few months ago. Arguments:

This "poll" is an excellent example of why we say WP:Polling is not a substitute for discussion. Polling is only useful to the extent that it is used to gauge support for a given idea (and even in that capacity, it rarely gives definitive answers). Many who are responding here have already hashed out many of these opinions elsewhere, and are at this point just screaming at each other from across the aisle. Further, the way this poll is framed -- "resurrect or kill WP:BTW" -- is guaranteed to further polarize the issue and drag us further away from any hope of resolution.

While I cannot compel anyone to follow my lead and leave this poll closed, I urge everyone to consider how little it is accomplishing: no "votes" will be counted at the end, and no action will be sanctioned by it, as this shouting match cannot in any fashion be interpreted as a consensus-building process. If you do seek a peaceful resolution here (the only kind that is ever actually upheld on Wikipedia), engage in discussion: express thoughts that you actually expect the other side to consider -- dogmatism will accomplish nothing -- and consider the views of others as well.--Father Goose (talk) 04:51, 19 February 2009 (UTC)

Reopening as I think you misunderstood. It is not a poll, and I would like to continue discussing. (I'll make the headings clearer.)--Kotniski (talk) 07:47, 19 February 2009 (UTC)

Arguments in favour of unmerging BTW as a separate guideline page

• It is absolutely shocking that one of the oldest philosophical tenets of this project has been swept aside in such a hasty and unadvertised fashion. Father Goose has said it above well: the very phrase "build the web" was evocative and compelling. It is a rich, subtle, and important part of Wikipedia's soul. Additionally, we should be publicizing this discussion. -- Earle Martin [^t/_c] 12:37, 17 February 2009 (UTC)

Nothing's been swept aside, it's just been put in a more appropriate place and in hopefully more helpful wording. Please don't overdramatize the issue - BTW was/is just a few paragraphs of vague rhetoric; there's nothing even remotely rich, subtle or important that hasn't been preserved in the merged page. Or if there is, let us know and we can work it back in. The phrase "build the web" is certainly still there - I've just bolded it so that it stands out for those with an emotional attachement to it.--Kotniski (talk) 13:19, 17 February 2009 (UTC)

Three words (not even the original phrasing - it says "build a web") buried in a giant morass of Thou Shalt Not. Your replacement certainly has swept a lot of things aside: charm, feeling, and subtlety at the very least. -- Earle Martin [^t/_c] 13:45, 17 February 2009 (UTC)

I would suggest that charm and feeling are not particularly valuable as attributes of guidelines, but subtlety certainly isn't - we want people to understand the things with as little effort as possible. I still don't see anything charming in the text of BTW anyway.--Kotniski (talk) 13:59, 17 February 2009 (UTC)

I've said my piece here. This recent trend of replacing anything that encourages our editors to think with iron-fist rules of mindless obedience is a sad one indeed. -- Earle Martin [^t/_c] 14:04, 17 February 2009 (UTC)

• I also agree with FG, this text describes one of the fundamentals of Wikipedia. Merging it into a style guideline which deals with the details of linking waters down the philosophical aspect of BTW. —Locke Cole • t • c 15:32, 17 February 2009 (UTC)
• In favor of resurrecting BTW, as well as CONTEXT in an abbreviated form; BTW described a philosophical point of view, and wasn't a style guideline. CONTEXT should be resurrected as a philosophical counterpoint to BTW. The style elements of CONTEXT have certainly been superseded by MOSLINK, but I think that the injunction against overlinking deserves a separate page. It doesn't need to go into detail, but BTW and CONTEXT always went (in my view) hand-in-hand, delimiting the extremes. BTW is especially relevant with the discovery that more than a quarter of our articles are orphaned, but CONTEXT is important to prevent the "sea of blue" that in the past proliferated on some articles.--Aervanath (talk) 16:26, 17 February 2009 (UTC)
  • tl;dr summary: CONTEXT and BTW should be resurrected as generalized injunctions, with MOSLINK providing the specifics.--Aervanath (talk) 16:26, 17 February 2009 (UTC)
    • Very well said. I endorse these points wholeheartedly. -- Earle Martin [^t/_c] 16:41, 17 February 2009 (UTC)
      • • Do any of you actually care about the people who are going to navigate and read these things, or are you just interested in your little philosophical disputes? If you want separate philosophical tracts, then write essays, and put them in userspace if you want to keep them pure of any opposing sentiment. Meanwhile, let the guidelines provide people with accurate guidance. Splitting a topic between three separate pages, each kept deliberately incomplete, where readers of one will probably not realize the existence or significance of any of the others, and where they are already part of a messed-up jungle of hundreds of pages purporting to offer guidance of one sort or another, is just a recipe for misleading people. But in some cases I think that's what you may actually want. --Kotniski (talk) 18:56, 17 February 2009 (UTC)
          • Tes, actually; there has been relatively little outside comment on this proceedural dispute on date linking, but one recurrent thread is, "I find it useful to see what else is going on in the world in the same year as [this important event]." This is not all readers; it may not be a majority; but the large minority should be served. Septentrionalis PMAnderson 17:24, 27 February 2009 (UTC)
• Restore it. The MOS should work with these odd and interesting pages and not try to fit them into broader style guidelines that most of us will never read anyways. I understand the problems of competing texts and the promise of standardization. I understand that "Resurrect" and "kill" are not the appropriate phrases to use WRT to these guidelines. However I liked build the web. It described what we as long term editors did (Surprise! We don't write most of the content). It should be restored. Protonk (talk) 02:17, 18 February 2009 (UTC)
• Restore. When did "consensus" suddenly decide that a fundamental underpinning of what made Wikipedia better than every other encyclopedia was no more than a stylistic consideration? There's no reason this can't be treated on two different pages. Joshdboz (talk) 06:20, 18 February 2009 (UTC)
• This should be restored, it's one of the central tenets of Wikipedia, and although we've grown, we still need it and it describes exactly how a wiki is built. The MOS needs to work with this guidance. Perhaps it is time to consider pruning the MOS, there are likely a number of editors who no doubt remember the time when Wikipedia didn't have a style guide. At the very best, the style guide should reflect the consensus of Wikipedians, per policy. Hiding T 09:59, 18 February 2009 (UTC)

The style guide and related pages (there isn't a clear boundary, as this debate shows) certainly need a lot of tidying up. This merger was part of that effort. But as you see, try to rationalize anything in the way these guidelines are organized and we get jumped on by people like you who see any such change as a threat to our very soul. Getting WP properly documented is a big task, but it's one that could realistically be carried out if the "oh-my-God-you-can't-change-this-it's-always-been-here" brigade could be kept at bay.--Kotniski (talk) 15:24, 18 February 2009 (UTC)

I'm gratified you've summed my objections up in such a neutral manner. My objection is not "oh-my-God-you-can't-change-this-it's-always-been-here", it's, I do not believe this should be changed since it describes what we do and what we should aspire to do. I do not believe we should discard that simply because that doesn't fit in with something somewhere else. I hope that clarifies, and perhaps allows a base fromn which discussion in a good faith manner might continue. I'm not really one to tar with the brushes you have dripping so heavily there, but thanks. Hiding T 18:03, 18 February 2009 (UTC)

Sorry, that "brigade" wasn't supposed to refer to any specific individuals. But question (genuine, I really do want to know): in what way do you think BTW describes anything we do/aspire to to better than the current text of MOSLINK does? (As Greg points out below, BTW can easily be interpreted to describe something we quite decidedly don't do or aspire to do, namely overlinking.)--Kotniski (talk) 20:54, 18 February 2009 (UTC)

My problem with WP:MOSLINK is outlined at WP:TLDR, or less snarkily at WP:KISS. And now I'm babbling in wiki-speak, which is another no-no. Basically, if BTW and WP:MOSLINK say the same thing, then I'd rather kill WP:MOSLINK, or have both. My humble opinion is that BTW is better because it is shorter, easier on the ear and the eye, and is therefore more likely to be read and to be understood. If BTW can be interpreted to mean something other than consensus would like, how can we fix that and keep it brief, to the point and simple? Can we not find a way to have our cake and eat it? At some point we lost the advice that this page was in dynamic tension with Wikipedia:Make only links relevant to the context, but I'd argue that part of this policy is that it is in tension with other parts of policy. I'm one of those people that has no issue with policies being in tension with each other, though. I appreciate that seems to cause problems with people who don't have the ability to hold two conflicting ideas in their head at the same time. So I don't have a solution as yet. But I hope I've outlined my thoughts a little better. I guess my best solution is that those people who can't hold conflicting ideas in their head are bashed repeatedly with a clue stick. Hiding T 22:02, 18 February 2009 (UTC)

Your motivation seems to be that "conflicting" ideas can't be on the same page, or that any page more than a few paragraphs long is too frightening for Wiki editors to read. But most of our key policy and guideline pages are quite long, and we don't necessarily expect people to read them from top to bottom - they scan them for the information they're looking for (or just read the summary at the top). And as for the conflicting ideas - if there really are conflicting ideas (which there aren't in this case - everyone more or less agrees where the balance between not underlinking and not overlinking should lie), then clearly no page that calls itself a guideline on a subject should confine itself to presenting only one of the conflicting ideas, since that misleads readers very badly. It's like giving parents two leaflets (mixed up with a whole lot of other leaflets so we have no idea which if any they'll read), one exclusively about the dangers of underfeeding your child and one about the dangers of overfeeding. Result: some kids starve; some get obese; and the fact that they're just right on average is no consolation at all.--Kotniski (talk) 07:58, 19 February 2009 (UTC)

Arguments against unmerging BTW as a separate guideline page

1. There was little concrete useful guidance on this page, apart from exhortations to create lots of links, and these can now be found at WP:MOSLINK (in the lede and elsewhere; of course changes to the wording can be proposed at that page).
2. The fact that there were no objections for months after this page was merged implies that the community doesn't value it highly as a piece of guidance (and is possibly largely unaware of its existence).
3. WP already has far too many guidelines for anyone to find their way around them properly or keep track of what changes are being made to them. We should be working hard to reduce that number, not increase it.
4. Having unnecessary separate guidelines on the same subject makes it harder for readers to get a complete picture, and makes it possible to mislead people in discussions by referring to the particular guideline that seems to support one's own arguments.--Kotniski (talk) 11:04, 17 February 2009 (UTC)

• Most policies and guidelines have a neutral title which describes what they regulate, not how they regulate it. When consensus changes their content is updated. This guideline is an example for what can happen if the message of a guideline is part of its title: Once people stop believing in it they simply stop using it, rather than correcting it. When I joined Wikipedia this guideline was already obsolete, and I only learned about it when it was cited by editors who tried to defend what general consensus called overlinking. --Hans Adler (talk) 11:21, 17 February 2009 (UTC)

  You're right, and I also think that these redirects should not be advertised: WP:OVERLINK WP:UNDERLINK. Let's find a better name for this. Nicolas1981 (talk) 12:23, 17 February 2009 (UTC)

• Against. Although Kotniski and I had severe exchanges over the manner of the merger, it turned out OK, aside from what I regarded as a few serious compromises to accommodate the wishes of BTW people. Kotniski did a sterling job in merging the text, and was by and large very diplomatic in forging a solution. WP's MoSs are a dog's breakfast, a plethora of mostly poorly coordinated pages. It is going to take some time to rationalise them all. This merger, some time ago now, was an important move in that direction, and the least you'd expect from a professional outfit that aims to help, not hinder, editors who are seeking advice on linking. Tony (talk) 11:35, 17 February 2009 (UTC)
  • Decreasing the area of MOS is a start; BTW was never a MOS page, and should not be one; neither should this. Septentrionalis PMAnderson 17:24, 27 February 2009 (UTC)

• Weak against. We'll never reach a consensus if there are two pages. BTW contains general ideas that would better be transformed in clear rules, and integrated into MOSLINK. By the way, the merge had been discussed: Wikipedia_talk:Manual_of_Style_(links)#Proposed_merged_wording Nicolas1981 (talk) 12:15, 17 February 2009 (UTC)
• Against. Better to have less of these and just integrate these sub-guidelines into MOSLINK, as we did with OVERLINK and BUILD. Dabomb87 (talk) 13:21, 17 February 2009 (UTC)

• Do not resurrect. I believe the purpose of this campaign is to have a page to link to when restoring a link that an editor wants to keep that expresses a pro-linking point of view rather than a page that balances the merits of linking with the distracton of overlinking. --Gerry Ashton (talk) 19:09, 17 February 2009 (UTC)
• Against. It would be wrong to reverse the recent improvements.--Goodmorningworld (talk) 19:36, 17 February 2009 (UTC)

• Against. the merging of these guidelines was a great improvement and "resurrecting" separate pages would be a step backward. Sssoul (talk) 20:15, 17 February 2009 (UTC)

• Against I've been active on Wikipedia since late 2005 and was unaware of this 'central' guideline. I agree that it's long-been supplanted by other guidelines and as it's a simple statement of the obvious there's no real need for it. Nick-D (talk) 07:23, 18 February 2009 (UTC)

• (more) So far no-one has even attempted to point to anything in the old BTW that isn't included in MOSLINK now (or in some other guideline, since BTW jumps around a bit, suddenly going into WP:Categorization, for example). If there isn't any such thing (or if there is but it can be worked into MOSLINK) then I simply don't understand the alleged need for a separate page. All the fundamental stuff has been retained, so nothing's been lost, and we now have a page where people get the full story - the fundamentals and the details. All that needs to be done is to rename it so it isn't a style guideline, and everyone should be happy. But if we do want a separate "philosophical" guideline (which rather misstates what BTW actually was) then it should certainly include both sides to the story - on one hand saying why linking is important, but also cautioning against overlinking (which BTW previously failed to do). Of course MOSLINK currently does all this and more, and (unlike this comment) is not too long, so separation is totally pointless except as a sop to people's sentiment - but if we must do the wrong thing, then let's at least do it in a reasonable way.--Kotniski (talk) 10:36, 18 February 2009 (UTC)

• (It should be pointed out as well, for those who have been at WP so long that they sometimes forget how real people understand language, that "build the web" is a pretty poor name for a page on this subject. People will understand "the web" to mean "the Web", and assume that this is about external linking. Or just won't understand it at all.)--Kotniski (talk) 11:40, 18 February 2009 (UTC)
  • One last teeny little comment: that's deliberate. Wikipedia was always meant to be part of the global Web (that Tim Berners-Lee envisaged; "Enquire Within upon Everything"). That's why I was very unhappy to see the wording changed to "build a web". -- Earle Martin [^t/_c] 15:45, 18 February 2009 (UTC)
    • Well, you must make your mind up. If that's the intended meaning, then it should be called "Build the Web" and the emphasis of the page should be completely different, concentrating on external links at least as much as internal ones.--Kotniski (talk) 16:00, 18 February 2009 (UTC)

• Against The opinions expressed on Build the web do not represent the community consensus that…

  Per Wikipedia:Why dates should not be linked, it should be a rare date indeed that is linked in regular body text. All links should be particularly topical and germane to the subject matter. Links to lists of historical events that have little to nothing to do with the subject matter at hand should generally not be made.

  Clearly, ‘Build the web’ is an essay, not a guideline or policy of any sort, and must properly be marked with an {{essay}} tag so the disclaimer shown below this post appears at the top of the article. And, since the essay ‘Build the web’ (‘Overlink articles’) is diametrically opposed to the clear community consensus that “All links should be particularly topical and germane to the subject matter” and effectively advocates that editors be bold in overlinking articles to turn them into Treasure hunt games that look like Sewer cover in front of Greg L’s house, Build the web should be in user space, not article space. I find this proposal to be forum-shopping in an effort to circumvent well established community consensus that has been recently reaffirmed. Greg L (talk) 16:43, 18 February 2009 (UTC)
  

• Against. I've been monitoring this process and it's been going in the right direction; reinstating BTW is a step entirely in the wrong direction. --Laser brain (talk) 17:50, 18 February 2009 (UTC)
• I am against, for I do not believe that the two opposite guidelines should be on different pages, not only because, as has been said, it is harder for one to see the whole picture, but because the element of date links upsets any balance that may have existed between these opposites, and generally makes things more complex—too much, I believe, for us and the readers to afford to engage in semantic inter-page acrobatics. Also, it is easier, trying to justify overlinking, to borrow authority that does not really exist by citing a guideline page encouraging linking without mentioning many restrictions (the reverse also applies); linking to a page presenting both sides of the issue in equal depth is not as effective for these purposes. Essays are created by users and are thus exempt from this, but the principle of neutrality does, I think, apply to guidance at least partially. After all, it is very often stressed that it is not obligatory to follow guidelines, and people are supposed to be persuaded to follow a guideline instead of ignore it; being neutral in presenting both sides of the case on links (to the extent that editorial discretion is encouraged) is only fair to the editors. Waltham, The Duke of 03:01, 20 February 2009 (UTC)

Comments and discussion

A vexatious, frivolous and timewasting proposal by Earle Martin. This issue is settled already. Dozens of editors commented; eventually discussion culminated in an RfC which was duly closed by an admin in October 2008. See here. In January 2009 implementation was discussed and agreed: here and here. I hereby request that an admin close this discussion per WP:DEADHORSE and WP:SPIDER.--Goodmorningworld (talk) 20:54, 17 February 2009 (UTC)

Not my proposal. Get your facts straight. And what is "by Earle Martin" except a personal attack, especially in the light of your getting it wrong?

A number of editors have become aware of a very poorly-publicised change to our guidelines, and raised objections; just because you happen to disagree doesn't mean that you can arbitrarily cut off a discussion before it's run its course. -- Earle Martin [^t/_c] 21:01, 17 February 2009 (UTC)

Amnesia, Earle? You started this section. You penned the introduction at the top. You wrote the words, "It is proposed to restore…". So don't whine if I attribute the proposal to you!--Goodmorningworld (talk) 21:19, 17 February 2009 (UTC)

No, I didn't. Have you always been this poor at reporting facts, or is the opposition to your opinion dizzying you? -- Earle Martin [^t/_c] 21:24, 17 February 2009 (UTC)

Earle I based my comment on this diff here. If I'm wrong then obviously it was an honest mistake. Next time, show some common courtesy to readers and make it clear who is proposing what!--Goodmorningworld (talk) 21:39, 17 February 2009 (UTC)

How about first you try the common courtesy of checking your facts before naming other editors vindictively? And then thinking twice about personalizing issues in the first place. -- Earle Martin [^t/_c] 22:05, 17 February 2009 (UTC)

That poll from October 2008 was held over a span of 36 hours... hardly time enough for something like this, IMO. —Locke Cole • t • c 21:03, 17 February 2009 (UTC)

Also, there weren't "dozens of editors", that's a flat out lie. Once again we see how those in MOS operate: starting little advertised polls to push their POV and allowing them to be closed prematurely, ending debate. —Locke Cole • t • c 21:10, 17 February 2009 (UTC)

LC, there is no use reporting you to WQA, I would be spamming that noticeboard every day with your uncivil posts. The community is fast catching on to you, that is punishment enough.--Goodmorningworld (talk) 21:19, 17 February 2009 (UTC)

Report me for what? You said dozens of editors. A visit to your link to the October discussion showed, at best, a dozen. If you wish to correct yourself, do so, but don't blame me. —Locke Cole • t • c 21:24, 17 February 2009 (UTC)

You might be good enough to admit that you were one of those who supported the merger when it happened. You never said then anything about its not having been widely enough advertised.--Kotniski (talk) 21:26, 17 February 2009 (UTC)

Because I assumed good faith that you already had consensus for the merge... my mistake. It won't happen again. —Locke Cole • t • c 21:50, 17 February 2009 (UTC)

Remedial reading is indicated for Locke Cole. I wrote, "Dozens of editors commented" (over time, as indicated by the word "eventually") SEMICOLON NEW THOUGHT "eventually discussion culminated in an RfC". The two thoughts are RELATED BUT NOT THE SAME.--Goodmorningworld (talk) 21:30, 17 February 2009 (UTC)

That's nice, and still a misrepresentation. —Locke Cole • t • c 21:50, 17 February 2009 (UTC)

• WOW!. We're basing all this in part on a poll that was open, what, less than 28 hours? Wow, that's, you know what, that's shameful. And regardless of prior discussion, let's not forget that consensus can change. Now if people want to get their own way so badly they're prepared to ignore a major behavioural policy, I would perhaps suggest that might also be shameful behaviour. Hiding T 10:05, 18 February 2009 (UTC)

What are you taking about? I don't remember any poll; there was plenty of discussion that lasted weeks, and there was no need for a poll since people were unanimous that the merge was a good idea. And I have no idea who is supposed to be ignoring what major behavioural policy. Establishing whether consensus has changed is the reason we're having this discussion (and we're only having it because I initiated it - as usual, the noisy unilateral consensus-overrulers did nothing to set a proper reasoned discussion in motion).--Kotniski (talk) 10:27, 18 February 2009 (UTC)

• I'm referring to both this and this comment: "This issue is settled already", both found in the first post in this section. I hope that better contextualises my comments for you. Hiding T 18:08, 18 February 2009 (UTC)

Where were people unanimous about the merge? I took a brief look at the talk page of Wikipedia:Build the web and the only message related to this merge I saw was the one posted just prior to you implementing it. No discussion or poll seems to have taken place there, and certainly no opportunity to object for those concerned with that guideline. —Locke Cole • t • c 10:34, 18 February 2009 (UTC)

What? You were involved in the discussion and you know there was unanimous consensus. Of course it didn't take place there - the purpose of the note was to inform people where the discussion was taking place. Those "concerned" with that guideline could have objected then or anytime since - since none did, we must assume that there were no objections, or (more probably) that there was no-one concerned with that guideline at all, and the sudden voices of support are just expressions of sentiment about something some people liked to think was always there, like a much-loved toy gathering dust in a cupboard.--Kotniski (talk) 10:45, 18 February 2009 (UTC)

I assumed good faith that you and the others involved weren't misrepresenting things. As has been clearly demonstrated, there no reasonable amount of time for those concerned with BTW to object (28-36 hours for a straw poll in October which was not advertised on the BTW talk page; 48 hours or less from the time you placed the notification on BTWs talk page until you performed the merge). As far as BTW was concerned, this was totally mishandled. —Locke Cole • t • c 11:10, 18 February 2009 (UTC)

Nonsense, even if they didn't have enough time then, they've had plenty of time since.--Kotniski (talk) 11:24, 18 February 2009 (UTC)

Yes, five weeks and we're here with objections now. I don't see the problem (again, other than the problems I noted above about lack of notification). —Locke Cole • t • c 11:35, 18 February 2009 (UTC)

Use discussion instead of polling, please.

In response to Kotniski's arguments, above:

The fact that the merger was not actively disputed for a few months five weeks is meaningless; at the time that you performed the merger, it didn't come to the attention of interested parties. You can't claim that such parties don't have the right to oppose the merger now because it's only come to their attention now. (And as I understand it, the merger opposed by some editors back when it was performed; it is opposed by more now.)

It was only opposed then because of a silly squabble about one or two words, which had nothing to do with BTW and was quickly settled. And who said anyone doesn't have the right to oppose it?--Kotniski (talk) 21:36, 17 February 2009 (UTC)

BTW and MOSLINK are not guidelines on the same subject: MOSLINK is a page about how to format and use links; BTW is a page on why to use links. Though you touch upon the "why" in this guideline, "whys" aren't the role of a style guideline, and as a result the "why" gets short shrift. If BTW were still around (and not protected), I would add material from this recent Signpost story which underscores the importance of the exact advice given by BTW. If anything, BTW should have been merged not with an MoS page, but with WP:REDLINK, another editing guideline on a related topic. Editing guidelines and style guidelines do not serve the same role, so it's no surprise the merger with MOSLINK is being seen as a poor choice at this time.

What you should do in MOSLINK is touch upon the importance of linking (as you do), and link to BTW for more detailed thinking on the subject. Just because both pages are "about links" doesn't mean it was a good idea to shoehorn BTW into MOSLINK.--Father Goose (talk) 21:12, 17 February 2009 (UTC)

Detailed thinking? Have you read it? It's a couple of paragraphs basically telling people to make links, which is what they already do. (Orphans come about because people don't create links TO the articles they create, and that was never stated clearly in BTW - it's stated a bit more clearly in MOSLINK now, and only because I added it.) The kind of "reasons" BTW gives for making links are of the sort "because articles are nodes in a hypertext system". Current MOSLINK has far more in the way of "whys" than BTW ever did. I keep saying that MOSLINK should be renamed WP:Linking to show that it's not really a style guideline, but I can only assume that this suggestion is too eminently sensible for anyone to respond to it.--Kotniski (talk) 21:23, 17 February 2009 (UTC)

So far that's the only part of this issue I agree fully with you on. -- Earle Martin [^t/_c] 21:29, 17 February 2009 (UTC)

I am not sure why you think this shouldn't be part of MOS. Most of the advice given here is pure style advice. Much of it wouldn't make sense for a print medium, but we must expect MOS to adapt to the medium it's being used for. There is a small amount of technical, non-style advice such as WP:MOSLINK#Link maintenance, but IMO not enough to justify removing this page from MOS. --Hans Adler (talk) 23:29, 17 February 2009 (UTC)

Different people have different definitions of "style". The main reason people seem to want to unmerge BTW seems to be (apart from knee-jerk sentiment) that they don't see its message as forming part of a style guide. Simply renaming the page ought to address that concern, and avoid the need for harmful reseparation.--Kotniski (talk) 10:53, 18 February 2009 (UTC)

There is nothing harmful about "reseparation". The harm was in merging an editing guideline (which describes one of the central philosophies of Wikipedia) with a style guideline (which spells out as rules when to do certain things). MOSLINK/CONTEXT are style guides, BTW is an editing guide, they should not be merged. —Locke Cole • t • c 11:12, 18 February 2009 (UTC)

Your edit summary at Template:Cent indicates that you think style guidelines are in a separate and lower category than other guidelines. Where do you get this idea from? And why is "philosophy" (which never existed in BTW in the first place) any less appropriate in a "style guideline" than in an "editing guideline"? In fact, is there any value in this distinction at all? The borders between style and other aspects of editing are often so blurred that imposing an artificial separation does make the guidance less usable.--Kotniski (talk) 11:28, 18 February 2009 (UTC)

Style guidelines are a separate and lower category than editing guidelines: ArbCom made this clear (to me anyways) in the jguk 2 case from 2005. —Locke Cole • t • c 11:36, 18 February 2009 (UTC)

Wrong link, probably - or it's hidden somewhere other than under "Final decisions", or I'm blind.--Kotniski (talk) 11:45, 18 February 2009 (UTC)

Sorry, from Jguk: Wikipedia:Requests_for_arbitration/Jguk#Style_guide. —Locke Cole • t • c 11:58, 18 February 2009 (UTC)

You mean where it says "[WP:MOS] is not binding"? Or something else? I don't see any implication that something called "editing guidelines" (not mentioned by ArbCom) are any more binding than style guidelines.--Kotniski (talk) 12:05, 18 February 2009 (UTC)

Yes, that part. Editing guidelines (and behavioral guidelines) are, to an extent, binding (with few exceptions). Policies are of course almost always binding. ArbCom has stated that MOS is not binding. Do you see the difference now? —Locke Cole • t • c 12:10, 18 February 2009 (UTC)

I don't see that ArbCom has said anything about "editing guidelines", so we have no idea whether it considers them more or less binding than style guidelines. WP:Policies and guidelines certainly makes no distinction between the bindingness of different types of guideline.--Kotniski (talk) 12:20, 18 February 2009 (UTC)

They didn't need to say anything about editing guidelines. The community hasn't had nearly the same kind of trouble with those as they've had with the MOS. MOS editors don't help their case when they constantly change the MOS (and the various subpages), often without any large amount of consensus, further diluting the intended nature of the pages. BTW should definitely not be associated with watered down style guides. —Locke Cole • t • c 12:48, 18 February 2009 (UTC)

I'm not sure this has anything to do with anything. If "watering down" means including various aspects of a complex issue instead of just one in the way that BTW attempted to do, then I don't think any page deserves to remain as a guideline if those maintaining it refuse to accept that kind of watering down.--Kotniski (talk) 13:04, 18 February 2009 (UTC)

I've decided to dramatically curtail my involvement in all this. My last thoughts on the matter: I could live with seeing BTW preserved and marked as historical; and I would like in that case for WP:BUILD and WP:BTW to link to it rather than just redirecting to MOSLINK, in order to preserve links in ancient discussions. There could always be a notice on top of BTW saying "superseded by MOSLINK" (or indeed WP:Linking as suggested above). That's it, I'm out. Thanks. -- Earle Martin [^t/_c] 15:09, 18 February 2009 (UTC)

Take your war elsewhere

I'm really astonished by the hostility of some respondents here toward Wikipedia:Build the web. And the poll, above -- "resurrect/leave dead" -- is doing a great job to further polarize the issue and foster misunderstandings and hard feelings.

Digging a little deeper, I see now that this is just an extension of a multi-party edit war that has spilled over from Wikipedia:Requests for arbitration/Date delinking and elsewhere.

Well, I don't care. Build the web is not a part of that debate, and should not have been made a victim of it. To the very minor extent that it could be used by either side of the "should dates be linked" debate to push a POV, it should be rewritten to state things in a more broadly accepted way, and otherwise left in place as a vital, still very much relevant part of our encyclopedia-building philosophy.

Is anyone here prepared to discuss what parts of WP:BTW are felt to be wrong, if any? I suggest not starting with a blanket insistence that all guidelines that have anything to do with links should be folded into MOSLINK. I do not dispute the wisdom of consolidating all style guidance regarding links into one MoS, which is why I support the merger of CONTEXT and MOSLINK. But the MoS, being a catalog of formatting dos and don'ts, is not the place to discuss more practical aspects of linking -- such as found in WP:REDLINK -- or philosophical aspects, found in WP:BTW. To the extent that BTW expresses any view that does not have consensus, it should simply be rewritten. Not eliminated, and not crowbarred into the MoS, where its message has been buried, perhaps deliberately, beneath a mound of proscriptions.--Father Goose (talk) 04:00, 19 February 2009 (UTC)

• Is there "hostility" to building the web? Let's not forget that this is just like any article, whose fate can be decided on by the community. Some editors apparently feel that it should be merged somewhere - note there is no suggestion to lose or delete anything. I say great, let's have that discussion. Would you not be in favour of it having its status somewhat elevated, into a style guideline? Right now, no one pays any notice to it exactly because it's an essay. That is the default position. Ohconfucius (talk) 04:31, 19 February 2009 (UTC)

It's not an essay; up until it was turned into a redirect to MOSLINK, it was an editing guideline. If you're suggesting that it somehow gained potency by being merged, in very diluted form, into MOSLINK -- well, I don't see it. But never mind the issue of "potency"; as a nugget of philosophy on the importance and value of links on Wikipedia, the whole thing disappeared in the process of the "merger" into MOSLINK.--Father Goose (talk) 05:06, 19 February 2009 (UTC)

Well, not exactly my fault. That's the tag on the top of that page right now. Anyway, I still thkn a discussion is warranted right now. I'm not dealing with the nugget issue here. Ohconfucius (talk) 05:19, 19 February 2009 (UTC)

Oh, sure, I want discussion. Discussion, mind you; a lot of people are treating the overwriting of BTW as fait accompli -- that will get us nowhere.--Father Goose (talk) 05:33, 19 February 2009 (UTC)

Perfectly civil discussion is taking place above. Please join in.--Kotniski (talk) 07:43, 19 February 2009 (UTC)

Proposed actions

Based on the arguments advanced above (mostly under #Resurrect this guideline?) I would suggest the following steps:

1. Restore WP:Build the web as a historical page (or essay, but historical seems more appropriate since that's what it is)
2. Place a prominent message on that page directing people to the current guideline on that subject, which is this page
3. Rename this page Wikipedia:Linking

Given that we've gone through all the arguments already, are there any objections to any of this?--Kotniski (talk) 08:49, 21 February 2009 (UTC)

Ugh.. it's fascinating to me that a merge was used to effectively kill a guideline that's enjoyed wide consensus for years. Everyone commenting above insists it must be an essay, historical or moved to userspace. Yet the original !vote was only to merge what was (and IMHO, still is) a guideline together with other related topics. Or maybe the definition of "merge" has changed drastically since I started here nearly four years ago. —Locke Cole • t • c 09:47, 21 February 2009 (UTC)

No, it was a ridiculous situation that three pages were giving advice on linking. It was unfair to our editors, and such cavalier fragmentation was bringing the MoS into disrepute. Be as fascinated as you like, but we do not want to turn back the clock. Nor do we want to revisit internecine bickering that resulted in the fragmentation in the first place.

The proposal to rename is fine by me, as are the rest of Kotniski's suggestions. Tony (talk) 10:02, 21 February 2009 (UTC)

Don't know what you're getting at. Nothing's been killed - despite many invitations, no-one's pointed to anything substantial that's been left out of the merged guideline. That would seem to make it satisfy any reasonable definition of merge.--Kotniski (talk) 10:58, 21 February 2009 (UTC)

If nothings been killed then why the resistance to undo the merge? Why the calls for marking it an essay? If, as has been alleged, everything at BTW is in MOSLINK then it should maintain guideline status. But that's not what those opposing it are acting like at all; they're acting like they accomplished something other than a merge. —Locke Cole • t • c 11:21, 21 February 2009 (UTC)

The reasons why it should not maintain guideline status have been pretty well set out (principally that it only tells half the story). --Kotniski (talk) 11:31, 21 February 2009 (UTC)

You're not getting it. It was a guideline before. People (very few, incidentally, in a span of 24-36 hours) !voted to merge it with MOSLINK and CONTEXT. Merge doesn't mean the material suddenly lost guideline status. And certainly if the merge is undone within a short time (as is the case here) there's no reason to insist on labeling it historical or an essay (or userfying it; as some have suggested). —Locke Cole • t • c 11:38, 21 February 2009 (UTC)

The result of the discussion above, then (both in terms of numbers and - more importantly - strength of arguments) indicates that it is no longer supportable as a guideline.--Kotniski (talk) 11:42, 21 February 2009 (UTC)

Unsurprising that seven MOSNUM regulars show up to !vote against this. The arguments are flimsy and poorly considered. And I see now you're trying to kill attempts at wider discussion just days after this was opened (see Template:Cent). —Locke Cole • t • c 11:47, 21 February 2009 (UTC)

The arguments haven't even been responded to - if you're suddenly claiming they're flimsy, then let's hear why. (It's the arguments on the other side that have proved flimsy in discussion so far.) --Kotniski (talk) 11:53, 21 February 2009 (UTC)

They're generally of a baitish nature, attempting to entangle this dispute into the larger date linking/delinking dispute. In other words, they appear to ignore the genuine pleas for this guideline in favor of assuming bad faith. —Locke Cole • t • c 11:56, 21 February 2009 (UTC)

You've lost me completely now - this is nothing to do with date linking, and the "genuine pleas" have been answered, unlike the far stronger reasons for not marking BTW as a guideline. Please be specific if you disagree with any of the arguments advanced.--Kotniski (talk) 12:02, 21 February 2009 (UTC)

WP:BTW (see, isn't that a handy shortcut?) has the support of some users therefore I'm fine with marking it as an essay. I'm also fine with linking to this guideline on the top of the essay, as has long time been the case if I recall correctly. Hopefully, down the road, after the ArbCom case and post Ryan's RfC, we can come back and discuss this all more cordially. This may just be a pendulum swing of sorts, and I doubt the sentiment about hyperlinks expressed by BTW will remain some outlaw view forever.

Although I haven't reviewed the renaming debate for this page. I would imagine if this page will be drastically re-scoped then it would need to go thru process again as a proposed guideline at some point along the way, but I'll leave that up to you folks to figure out. -- Kendrick7^talk 03:22, 22 February 2009 (UTC)

• I think that Kotniski's suggestions above are a fair reading of the debate though I would prefer essay status for BTW rather than historical. I do think, however, that an overview of the issues linking this discussion with the date linking debate (where I was only peripherally involved) is useful. They are related because the primary argument against date linking is that they constitute overlinking and that many wikipedia articles have too many hyperlinks rather than too few which is, IMO, the underlying assumption of BTW. While orphans and dead-end articles remain than should be integrated into the web of Wikipedia links, consensus seems to be that the danger of overlinking is the greater at the current stage of Wikipedia's development. In that case, a page that advocates the creation of links without explaining when they are inappropriate cannot reflect a broad consensus and thus be a guideline. Eluchil404 (talk) 21:46, 22 February 2009 (UTC)
  • No-one said that overlinking was a "danger", or more or less of a problem than underlinking. The fact is that there are some links that we want editors to make and some that we don't, and putting all the relevant advice on one page makes it less likely that they will be misled as to what they are being encouraged to do. --Kotniski (talk) 07:11, 23 February 2009 (UTC)

Maybe I'm slow, but... (Build the web edition)

Is this just an extension of the date delinking feud? I'm seeing a lot of names here that I saw in those numerous and competing RfCs and that I saw in the RFAR on the subject. Is that a mistake of mine (in that I don't know who the general MOS regulars are), or does that seem to be the case. If is isn't a mistake, then perhaps both sides on this date delinking business could step aside and let the folks who aren't going to use this guideline simply as a means to an end discuss it. Protonk (talk) 04:52, 23 February 2009 (UTC)

Not really. The initial 3-page merger was carried out with the explicit support of people on both sides of the date-linking fight. I understand that the decision to try to partially undo the merger may have come out of something going on at the ArbCom case, but anyone was welcome to join in the discussion, and anyway we seem to have pretty much reached conclusions now.--Kotniski (talk) 07:04, 23 February 2009 (UTC)

I'm perfectly agreeable to that suggestion, Protonk. I'd certainly like to continue discussing the issue to attempt to reach an agreeable consensus. Hiding T 11:37, 24 February 2009 (UTC)

We were doing that, but you kind of stopped. Anyway, we seem to have reached the stage where there are two separate issues - the name of this page and the labelling of BTW - so I'm taking the BTW issue back to that talk page with a suggestion, and making a rename proposal for this page, which I shall announce at WP:RM.

I didn't kind of stop, I think the above post quite clearly indicates the opposite. Also, we were not as I recall discussing the issue at hand. I believe your last comment was an attempt to engage me in my motivations rather than work out what was best for Wikipedia, so I hadn't as yet worked out how to reply. I hadn't realised I was on a time limit. Hiding T 14:05, 24 February 2009 (UTC)

Requested move

The following discussion is an archived discussion of the proposal. Please do not modify it. Subsequent comments should be made in a new section on the talk page. No further edits should be made to this section.

The result of the proposal was PAGE MOVED per discussion below. -GTBacchus^(talk) 18:11, 1 March 2009 (UTC)

---

I propose renaming this page Wikipedia:Linking (which currently redirects to it). Reason: since its merge with WP:Only make links that are relevant to the context and WP:Build the web, this page deals with the whole subject of linking, addressing issues which are viewed as more than just style issues (see discussions above).--Kotniski (talk) 13:16, 24 February 2009 (UTC)

Makes sense to me.--Goodmorningworld (talk) 16:56, 24 February 2009 (UTC)

Yes, I agree. -- Earle Martin [^t/_c] 18:42, 24 February 2009 (UTC)

That merger was unwise, done to make points at a current ArbCom case. But if {{styleguideline}} is removed, it may be worth renaming. Septentrionalis PMAnderson 20:19, 24 February 2009 (UTC)

That wasn't the reason for the merger. It might have been the reason for the recent attempt to undo the merger.--Kotniski (talk) 08:29, 25 February 2009 (UTC)

The name "Linking" might suggest this page covers external links as well as internal links. --Gerry Ashton (talk) 18:45, 24 February 2009 (UTC)

• Agree with Gerry. Does Wikipedia:Internal links seem any better? Currently that redirects to Help:Contents/Links but has few incoming links in that guise, so has potential. [4] Hiding T 22:04, 24 February 2009 (UTC)

Look again. There's a whole section on external links, with an onward link to the main guideline that it summarizes.--Kotniski (talk) 08:29, 25 February 2009 (UTC)

support Kotniski's proposal, for the reasons he stated which I've also voiced above in this talk page: that WP:Build the web dealt with more than just styling. --❨Ṩtruthious ℬandersnatch❩ 09:34, 25 February 2009 (UTC)

Support the rename: the proposed title is more accurate. UnitedStatesian (talk) 15:17, 25 February 2009 (UTC)

I'm not sure you'll find many administrators willing to close this discussion before the end of the ArbCom case. I would stay away from it until then, anyway. Not to discourage you or dispute the merits of a move in any particular way. Dekimasuよ! 07:08, 26 February 2009 (UTC)

What ArbCom case? What are you talking about?--Kotniski (talk) 07:12, 26 February 2009 (UTC)

Wikipedia:Requests for arbitration/Date delinking. Dekimasuよ! 07:16, 26 February 2009 (UTC)

The rationalisation of MoS in this merger a while ago has nothing to do with the ArbCom case. Tony (talk) 07:45, 26 February 2009 (UTC)

Merely coincidence, then, that this extensive restructuring took place the day before ArbCom accepted the case. Septentrionalis PMAnderson 05:49, 27 February 2009 (UTC)

Yes. I did the restructuring, and was totally unaware of any impending ArbCom case. The case certainly has nothing to do with the name of this page.--Kotniski (talk) 09:40, 27 February 2009 (UTC)

That will do for Kotniski; nonetheless, the "approval" consists of one side of that case altering Wikipedia space to suit their position. Septentrionalis PMAnderson 17:56, 27 February 2009 (UTC)

I have no knowledge of that ArbCom case, and I'm otherwise willing to close this request. Would it be very disruptive if I were to do that? Should I ask over there? -GTBacchus^(talk) 17:40, 1 March 2009 (UTC)

Since everyone seems to be agreed on the rename, and ArbCom is hopefully not that stupid as to be influenced by the name of a page, I see no reason why it would be disruptive to make the move or any need to ask anywhere else.--Kotniski (talk) 18:03, 1 March 2009 (UTC)

• Weak support. The new name makes sense, but I guess that the word style in the name "Manual of Style" is intended to have a broader meaning than just "punctuation and formatting": see e.g. the sections "Which units to use", "Unnecessary vagueness", "Identity", "Gender-neutral language", ... in WP:MOS; therefore, the sky isn't going to fall if this page stays here. --A. di M. (talk) 11:04, 26 February 2009 (UTC)
• Weak support; the fewer pages that can be abused by the MOScrufters, the better. Septentrionalis PMAnderson 05:51, 27 February 2009 (UTC)

The above discussion is preserved as an archive of the proposal. Please do not modify it. Subsequent comments should be made in a new section on this talk page. No further edits should be made to this section.

So it's moved. If that upsets anyone, I'm sure we'll find out soon.

Now there's a fair amount of cleanup to do, which anyone is welcome to help with. I'm fixing double redirects first. There's also a lot of red tape related to classifying this page as part of MoS. I don't know how much of that, or how urgently that needs to change. -GTBacchus^(talk) 18:11, 1 March 2009 (UTC)

Should this page be part of MOS? Really, linking is content, not style. Septentrionalis PMAnderson 23:32, 1 March 2009 (UTC)

Perhaps I should have said, "red tape related to declassifying this page as part of MoS." I think it's on a lot of templates and stuff now. -GTBacchus^(talk) 02:42, 2 March 2009 (UTC)

What generally should not be linked

I'm not that happy with the advice given by this section of the guidance, and given the priority the FA and GA processes give to the MOS, I think that's a little worrying because the MOS is becoming a de facto policy rather than merely guidance, which does not always have to be heeded. Now my ideal compromise would be to ask if we could consider exempting the lede of an article from this section? My compromise stems from both philosophical and practical issues. The philosophical ones I am prepared to waive, we all differ ideologically, and no-one should attempt to argue out of philosophical necessity. But in practise, what concerns me is the impact this guidance will have on readers and editors. Readers are losing a navigational tool that is part and parcel of an internet based project. And editors are losing a valuable tool that will see them resort to other methods to achieve the same results. We're either going to see a growth in nav-box templates or a growth in see also sections, or both. And I'm worried that those may lead to issues in the future. I throw my weight whole-heartedly behind the Link density and Example sections of Overlinking and underlinking, and What generally should be linked is okay, but I'm concerned that this section is too far removed from what makes Wikipedia better. It's the second and third clauses that concern me. The first one is simple enough, and I'm not touching the fourth one with a barge pole for fear of explosion. Like I say, my best compromise is to ask that the lede be considered an area of exemption. Hiding T 10:29, 27 February 2009 (UTC)

I agree that there should probably be items linked in the lede that would not be linked elsewhere. But I don't think it's as simple as just "exempting" the lede from certain principles. We could say something like (and I'm not suggesting this as a final wording): "in the lead section of an article (and possibly elsewhere?), common terms should be linked if their articles are essentially related to the topic of the present article." No, that doesn't make much sense, but what I'm trying to express is that (for example) there doesn't have to be a link to "Australia" from "XY is an Australian actor", but there should be a link to Australia from an article on one of the states of Australia.--Kotniski (talk) 10:54, 27 February 2009 (UTC)

I don't want to get too bogged down in arguing examples, because I'm undecided on birthplaces, and we end up arguing about articles like Martina Navratilova, where I'd say it probably is informative to link to Czechoslovakia. But then again, maybe the info-box is the right place to wiki-link these terms. Thinking about it, maybe we should merge the points for and against into a section titled What to consider when linking. We could then discuss the merits of given reasons, and hammer home the general point that every link needs a stronger reason for being than because you can. Hiding T 12:01, 27 February 2009 (UTC)

Thinking about it, I removed the following:

• terms whose meaning (as relevant to the context of the article) would be understood by almost all readers.

I couldn't find it in Wikipedia:Only make links that are relevant to the context and I think it oversteps the mark between a perfect marriage between WP:BTW and that page. Hiding T 12:57, 27 February 2009 (UTC)

I've reverted this removal. It's important that links must add value by helping the reader, not just be there because they can be linked. For example, in a typical introduction such as "Jennifer Lopez is an American singer and film actress", none of the links have any value. Colonies Chris (talk) 13:16, 27 February 2009 (UTC)

If that's what you want to preclude, then you don't need the context part in brackets. But let's be clear; this was never in guidance before, so it is a new addition. That seems wrong, if this page is intended to be a merger. Hiding T 13:21, 27 February 2009 (UTC)

It's not new. That statement (or at least an early version of it) was added in August last year. Colonies Chris (talk) 13:31, 27 February 2009 (UTC)

It's new to me. I'm coming to this rewrite late and under the impression that it was intended to merge Wikipedia:Only make links that are relevant to the context and Wikipedia:Build the web. I don't happen to have every single policy and guidance page on my watchlist. But my main point stands, this never should have been added to guidance as it is overly prescriptive, and if it was added in August it conflicted with other guidance. By the way, do you have a diff handy so I can see what was added and how it has altered. Hiding T 14:08, 27 February 2009 (UTC)

This could certainly be cleared up (perhaps by combining this point and the point preceding it into one, and allowing for possible exceptions). I think many people would link singer and actress in Chris's example, since these terms are so essential to what the article subject is, but certainly links on American and film would be over the top. --Kotniski (talk) 13:37, 27 February 2009 (UTC)

I'm not sure why American and film are over the top and singer and actress aren't. This seems to be straying into areas of personal preference. I mean, you can argue there's a use in linking to Lopez, because I learnt more from that link than I think I'm likely to from the other four links which exist. It's possible that this is such a gray area that our guidance needs to be as loose as possible while retaining the overall message that every link needs a stronger reason for being than because you can. Hiding T 14:08, 27 February 2009 (UTC)

Yes, it was a deliberate move away from the previous guidance, which over-encouraged indiscriminate linking, and towards a more selective approach. Although the facts that she is a singer and actress are a vital part of the article, a typical reader is unlikely to find a link to 'singer' or 'actress' useful. Any reader with the intelligence, curiosity and understanding of the language to use WP at all must already have a grasp of such basic concepts, so the links have no value. The beginning of the changes was round about here, but there were a lot of changes after that. Colonies Chris (talk) 14:35, 27 February 2009 (UTC)

Those changes seem to be mirroring Wikipedia:Only make links that are relevant to the context and were guidance at the time. The text I altered is substantially different to the text I can see back there. Also, text which mirrors the text I can see back there was still in the guidance after I edited it just now. I agree we should be advising against indiscriminate linking, and towards a more selective approach. The rub seems to be in how selective. Nothing too obvious, but certainly stuff that could be informative. Hiding T 15:00, 27 February 2009 (UTC)

But links also have a navigational purpose, not just the purpose of allowing readers to look up words they had never heard before. No-one would remove the links in "In mathematics, the complex numbers are ..." on the ground that everyone knows what mathematics is. A link to Croatia (or better to Croats, especially in an article about someone who lived before the Republic of Croatia was established) in an article about a Croat isn't any worse than that. (But I agree that "film actress would be excessive; film actress would be better.) --A. di M. (talk) 16:26, 27 February 2009 (UTC)

A. di M, your point was also made by Kotniski above. In an article about theatre, you might link to actor as the reader would probably be interested in exploring the subject in more depth, but in a bio, you wouldn't. Similarly, in your example, it would be legitimate to link to mathematics, but in a bio you wouldn't link "he got his first degree in mathematics". It does depend on context, which is why I think the whole statement, including the parenthetical part, should remain, though we might try to reword it to make it clearer. Colonies Chris (talk) 16:53, 27 February 2009 (UTC)

But that's a very narrow exception, and may be left to IAR. Jean-Robert Argand links to both mathematician and complex numbers; the first may be excessive (although some readers will follow it); the second is necessary for much of our audience. Septentrionalis PMAnderson 17:31, 27 February 2009 (UTC)

Neither would I link "mathematics" in your example, but I would keep the link currently present in "Amalie Emmy Noether, German pronunciation: [ˈnøːtɐ], (23 March 1882 – 14 April 1935) was a German mathematician known for her". --A. di M. (talk) 17:40, 27 February 2009 (UTC)

• The problem I have with the text in the parenthesis is that it means that you wouldn't link words which people familiar with the subject of the article already know. That seems counter-productive to me. I'm also starting to wonder if our practise of linking the first occurrence of a word is a bad idea that causes problems. I'm thinking of something like, say Marlon Brando, I wouldn't exactly link method acting the first time, but given his importance in establishing method acting, I'd link it in a section on his acting style. But I think there has to be a cognitive plan behind how we build the web. We don't want to get people to America from Jennifer Lopez in one link, but I think you should be able to get to America from Jennifer Lopez in a natural progression. Probably through either the awards or more pertinently the South Bronx, because I'm thinking that perhaps the specific location in a birth place or child-hood might be the best thing to link, because it does impart some understanding to know what growing up in the South Bronx means compared to growing up in Beverly Hills, as it were. Hiding T 19:48, 27 February 2009 (UTC)

Much like Hiding, I became aware of this change after the fact; in my case, it was when I saw mention of it in the Signpost late last summer. This is not surprising given that the change was never properly discussed, nor was consensus sought prior to implementation. The unilateral July edits abruptly reversed the spirit of the linking guidance; they were first applied to the MoS, and then used a few hours later as the rationale for changing the "CONTEXT" page. Since then, the language has been used to justify stripping out links to countries, languages, and other articles - well beyond what many would consider "common" terms. (Not wanting to be accused yet again of seeking to "blue" the entire encyclopedia, I'll state that I'm not opposed to delinking everyday terms. However, there is no reason to mass-delete links to countries such as the US or Canada just because someone feels we don't need them.) --Ckatz^chat_spy 00:36, 28 February 2009 (UTC)

← Maybe, whoever added that point about countries was concerned about links such as "The quaternions were first described by the Irish mathematician Sir William Rowan Hamilton in 1843 ...", where the fact that Hamilton was an Irishman and not a Spaniard or an Icelander is totally irrelevant to the mathematics of quaternions. But he/she had over-reacted, in my opinion: before I (under my former account User:Army1987) added the "except if they are particularly relevant to the topic of the article", the wording could be taken to imply that "Italy" shouldn't have a link to "Europe". (Also "major" and "familiar to most readers" aren't necessarily correlated; I guess Vatican City is more familiar to Uttar Pradesh to most readers.) --A. di M. (talk) 01:24, 28 February 2009 (UTC)

In many cases, linking to articles about well-known geographical locations is unhelpful. Most if not all readers what the United States or Africa is. Moreover, if I am reading an article about an actress born in the United States, the United States article would tell me little that is relevant to the actress. However, in response to the above, I believe that the article on Italy should link to Europe. Dabomb87 (talk) 01:59, 28 February 2009 (UTC)

God I love systemic bias. How else would we get people working on a project for world knowledge making blind assertions not only about what all our readers know, but what they should not be allowed to easily learn about if they do not? -- Earle Martin [^t/_c] 02:05, 28 February 2009 (UTC)

Links should be to articles that aid a reader's understanding of the topic. One can learn new things from just about any article on Wikipedia, just hit Special:Random and one can learn to their heart's content. There is also a search box if one wants to specify the area of learning. Dabomb87 (talk) 02:13, 28 February 2009 (UTC)

So if I'm reading about a film director from Azerbaijan, a country which I know nothing about, my understanding of him will somehow not be improved by clicking Azerbaijan. That makes a whole lot of no sense at all. -- Earle Martin [^t/_c] 02:23, 28 February 2009 (UTC)

Well yes, but that is different from the United States. To amend the above comment on our hypothetical actress article, I meant "the United States article would tell me little that is relevant to the actress info that it is not already known for the reader of the actress article." (underlined text is new) Now, to compromise, I might not link United States outright but would instead pipe link it to the more germane Cinema of the United States article. 02:30, 28 February 2009 (UTC)

Comment by Tony. What is missing entirely from this discussion is the inescapable fact that the greater the density of links, the less prominent each link, and the greater the dilution of each link. Our valuable wikilinking system has such potential to enrich the readers' experience of WP, and it has been only over the past three or four years that the penny has dropped: our early enthusiasm for undisciplined linking (and blueing) significantly weakened the utility of wikilinking. There has since been a clear trend towards what some people have called disciplined linking, or smart linking. The community now takes a more conservative line on linking practice, since most WPians know—either intuitively or through overt awareness—that linking needs to be rationed to be effective. This underlines the pure folly of bright-blue date-autoformatting (apart from the other reasons that it's a bad idea in any form), and the linking of common country names, geographical names, professions such as actor (given Colonies Chris's possible exception above), and the rest.

User:Holcombea is a prominent US researcher in certain aspects of visual perception and processing, currently at the University of Sydney. He wrote to me on this issue:

This is a fairly basic principle of attention and perception research. Salience of an odd-colored object (eg a blue link) will be higher if there are fewer other blue things around. This is so ubiquitous that I don't know a basic reference for it, but it is incorporated in standard models of attention like that used in the attached. Also, there is an emerging subfield on "crowding", which is what happens when things get way too dense.

He attached a pdf file of Einhäuser et al., (2007) "A bottom–up model of spatial attention predicts human error patterns in rapid scene recognition", Journal of Vision 7(10):6:1–13. I would be only too pleased to email an attachment or to explain more about the contents of the article to anyone who is interested. It has an an extensive list of references.

My own basic knowledge of information theory confirms this conclusion, simply in the notion of signal-to-noise ratio. Aside from that, it is obvious.

We can define in general terms, and in some cases specifically, what is best not linked or not linked in what context. To assist editors further, and to minimise the risk of disputes, it may help to specify the issues that need to be balanced in each case. Here, I refer to the original article as OA and the linked article as LA.

The issues to balance include:

• how likely it is that a linked item is familiar to most readers;
• how useful it would be in the context to divert to the LA, including:
  • the extent to which the LA adds useful knowledge to the OA topic;
  • the ease with which relevant information in the LA can be located (again, signal-to-noise ratio);
  • whether the relevant information in the LA is either duplicated should be duplicated in the OA;
  • the extent to which the LA more generally would deepen the reader's understanding of the OA.
• the additional dilution of other links in the vicinity in the OA;
• the extent to which following a link will disrupt the reader's comprehension of the OA;
• the risk that readers will soon become inured to high densities of links—whether on WP generally or in a particular article they are reading—and will tend to follow links less often in the light of previous unsatisfactory experiences in linking;
• the additional key/mouse actions required to key the item into the search box instead;
• the alternative of listing the link in the "See also" section in the OA, where it can be supported by adjacent explanatory information and has the benefit of a critical mass of links from which to choose. Tony (talk) 14:40, 28 February 2009 (UTC)

Yes, but one should not only consider the present status of the LA, but also its potential status. An article could be a stub or in a very bad status, but that would not be a good reason to avoid linking it from other articles, provided it is reasonable to expect that it will become more informative later. (If it's not reasonable to expect that, the stub should be merged into some other article anyway...) Otherwise, we would never add red links. The more readers "land" on underdeveloped articles, the more of them are likely to eventually improve them. --A. di M. (talk) 15:24, 28 February 2009 (UTC)

Possibly, but I'm reluctant to rest judgement on many of the issues above in trust or expectation of future improvements of LAs. What is there at the time is what really matters in weighing up the pluses and minuses. A link can always be added later if a potential LA improves. It is more important that a high standard of wikilinks be maintained to maximise its utility (and reputation) at any given time: that flexibility, dynamic characteristic is essential for a wiki—if there's no periodic (hopefully, regular) quality control of articles, as well as links, we bely that characteristic. It is a disadvantage as well, of course, since articles can be degraded, too; but it is also WP's great trumping of static sources such as The Encyclopedia Brittanica. Tony (talk) 15:37, 28 February 2009 (UTC)

But if I'm writing articles "Foo", "Bar", and "Baz", to which the article "Quux" could be relevant, although it doesn't contain much as of today, then when someone expands "Quux", the articles "Foo", "Bar" and "Baz" will already have a relevant link, without the need that someone expanding "Quux" brainstorm all the possible articles which could link to it. That is more or less the same reason why it is suggested not to fix links to redirect with possibilities, or to remove red links to titles which clearly deserve an article (which provide zero benefit to the readers who never ever intend to edit, but positive benefit to the encyclopedia as a whole, as other readers might be encouraged to create the article). Replace "zero" with "very small" and "create" with "expand" in the sentence before, and the same thing will apply to links to stubs. WP:WIP, WP:TIND, WP:DEMOLISH, and all that. The various cleanup banners clutter the visual appearance of a page much more seriously than a couple of links to articles which aren't perfect right now, but all the proposals of hiding them away were rejected. --A. di M. (talk) 16:08, 28 February 2009 (UTC)

What is missing entirely from Tony is the inescapable fact that nobody is arguing for "indiscriminate" linking. (By the way, "undisciplined" is a nice piece of subtly insulting those you disagree with. I see from WT:Build the web you've been throwing that word around since at least last August.) Therefore pulling in some "expert" - who appears to be neither a librarian nor an encyclopedist - and throwing in references to some random scientific paper (how very erudite) gives us an end result that is a remarkable combination of both straw man and argumentum ad verecundiam. Well done. -- Earle Martin [^t/_c] 15:47, 28 February 2009 (UTC)

• That nobody is arguing for indiscrimination does not mean it does not occur. I have seen plenty, and I would be surprised if you have not. Your attitude is a great shame, as the above discussion appeared to be taking place with well-reasoned argument until that last bucket of water from the British isles. There has been much claimed uncomprehension, and unsubstantiated debate about the benefits of over-linking, and now Tony cites a scientific journal, presumably peer reviewed, and you brush it aside with a simple dismissive "references to some random scientific paper". Why did you not just say "I don't like it" and be done. You could have saved yourself a fair bit of typing. ;-) Ohconfucius (talk) 16:08, 28 February 2009 (UTC)

Tony didn't "cite" anything.

"That nobody is arguing for indiscrimination does not mean it does not occur" - and you, Tony, et al, are using it to argue for throwing out the baby with the bathwater. -- Earle Martin [^t/_c] 16:40, 28 February 2009 (UTC)

• The baby's already dried and changed. The poopy bathwater needs to be poured away. ;-) Ohconfucius (talk) 04:53, 1 March 2009 (UTC)

Yes, but the solution to the problem that people would add links like this one (partially undone by me here) isn't discouraging all links to "major" places altogether (whatever "major" means, see the example above about Vatican City vs Uttar Pradesh). --A. di M. (talk) 16:25, 28 February 2009 (UTC)

Reply to Tony I indicated in my first post in this thread that I throw my full weight behind the guidance on link density. That's a done deal for me. And I'm perfectly happy with the page as it stands right now, because I think without the parenthesis it works very well. My main fear was that we'd see the removal of links someone with knowledge of the topic would be expected to understand. I couldn't agree more on link density. In fact, I'd like to see your pointers introduced somewhere, because I think they should be the factors considered when disputes occur regarding a given link. Perhaps we could introduce them to a rewritten WP:BTW, but if that's impossible, I'd like them somewhere. What I do want you to consider is the potential impact on nav-boxes. In some areas of Wikipedia these are growing exponentially to a point they can fill a monitor screen on less capable browsers. (See the bottom of Iron Man with all three expanded for an example) I don't really know what impact they have with regards accessibility and screen readers, but given this page discusses links, maybe there should be something regarding such navigational templates. It's a worry that there isn't a mos for them. Infoboxes doesn't seem to acknowledge them from my skim of it. But that's my real area of concern. Hiding T 16:41, 28 February 2009 (UTC)

Problems with navboxes should be addressed at WP:NAVBOX (where, as it happens, they disagree: every item in a navbox should be linked; if it's not important enough to be linked, it shouldn't be wasting space in the navbox). As for Tony's pointers, make them an essay, and see how many people actually agree with them. Septentrionalis PMAnderson 23:42, 1 March 2009 (UTC)

Build the web is a rule

To further elucidate my recent revert of Tony1's change I should explain that the course of events that has occurred related to WP:Build the web during the past month or so, although I'm sure it's all in good faith, has appeared to me dangerously close to an accidental out-of-process deletion of that policy / guideline. So I feel that I must oppose anything other than a prominent and unambiguous mention of the "build the web" WP:RULE in the header of this new, merged guideline. (At least, without some consensus for a policy change to remove "build the web" as a Wikipedia rule.)

So we really need to avoid gradually softening and minimizing the language that specifies it as a rule. If because of any reason along the lines that it would fit poorly as part of this project page "build the web" can't be prominently and unambiguously stated in the lead here I really think it needs to be re-created as its own project page. (Though the current state of affairs with it now represented as a single paragraph within the lead of this merged guideline is agreeable to me also.) We definitely must avoid setting any precedents for it to be acceptable that policies or guidelines be accidentally changed in the course of a merge. --❨Ṩtruthious ℬandersnatch❩ 06:48, 2 March 2009 (UTC)

• I am a bit surprised with the language in the article which relates to this point (which you reverted back to). Principally, the wording appears not be be consistent with the status of this page, nor with policy matters in general. That a guideline should refer to itself as "a fundamental rule" is curious in that if something was fundamental, it is likely to be stated as part of WP:5P. Furthermore, I have heard argue that guidelines, not being policy, probably do not enjoy the status of 'rule'. Ohconfucius (talk) 07:10, 2 March 2009 (UTC)