On Mon, Mar 29, 2010 at 4:39 AM, Charles Matthews <charles.r.matthews(a)ntlworld.com> wrote: High value links should always be provided. Can you provide an reference to a Wikimedian arguing that links to the most useful additional resources shouldn't be provided? I'll gladly go and disagree with them. But I do believe that a list of, say, 50 links tagged onto the end of an article typically has negative value for the following reasons: * Readers will be inundated, no one is likely to follow more than a couple so the very high value links will be lost in the less valuable ones. * Wikipedia editors are unlikely periodically review links in a large collection (supported by the high density of dead links, and the malicious sites I've found in prior scans of our internals links). * Long lists provide plausible denyability for someone attempting to profit by placement, as additions to link soup doesn't look suspect. * Someone looking for a large collection of assorted links on a subject can find a larger and more current list from any of the search providers. I bothered making the argument here because I believed that Fred was likely mischaracterizing the nuanced position people have taking in trying to balance the value of additional links vs their cost as a simple "war on external links", when no one was likely carrying on any such war: Just because someone has decided on a different benefit trade-off than you doesn't make their activities a "war on all X". I wish there were a usable non-commercial search engine. But Wikipedia clearly isn't that. Wikipedia's value is in human editorial review. A search engine's value is in enormous scale automation, "machine neutrality" (not the google results are neutral, but it is resistant to many kinds of bias which wikipedia is not), and automated updates. Everyone on the internet already has access to high quality search engines. I just don't think that making Wikipedia into a poor search engine at the expensive of diluting the selectivity is a net positive for the reader.

Gregory Maxwell wrote: I have had a look around WP:EL and its Talk, and I believe it is clearly not the case (given the 20 reasons not to include a link, starting with a catchall) that the guideline is in the hands of those who have that as credo. See below for more. <snip> OK, reductio ad absurdum. But what I see around WP:EL is quite different. Basically it now stands, in relation to linkspam, as WP:N can be considered to stand in relation to cruft. But it has clearly gone further down the deletionist road, and (I presume, just as you jumped to sections of 50 extlinks) anyone who objects is supposed to love linkspam. It seems apparent that a working concept of "justifiability" has been introduced, analogous to "notability"; that the onus is on anyone adding an extlink is to show it is "justifiable", and your third point is parodied (I hope it is only a parody) as "Any site that does not provide a unique resource beyond what the article would contain if it became a featured article" (WP:ELNO). What you wrote is "I think that at its best Wikipedia should be directly including all the information available up to Wikipedia's coverage depth, linking only for citations, then it should have links to the most valuable external resources which go deeper into the subject than Wikipedia reasonably can." Obviously the word "unique" is just bad drafting - should be replaced by "distinctive" or something that doesn't mean if two web pages have the same essential content we can't have either as extlk. But "deeper into the subject than Wikipedia reasonably can" and "what the article would contain if it became a featured article" both make our criteria for "justifiability" be driven by a state of affairs that is not only hard to define, but actually in practical terms applies only to 1 out of 1000 articles, with no prospect of this proportion changing soon. In short, while no one can be for linkspam or including long lists of duplicative exlks, since "Wikipedia is not a web directory", the guideline has gone over to "necessary to inclusion" by a general criterion (so worse than WP:N) and at the same time junked good sense and "weaving the web" at the basic, nodal level. Not good at all. I don't see the trade-off. What I see is that WP:EL is now a battery of arguments for winning arguments about what is linkspam, with complete disregard for the cost on the majority of topics, which are neither likely to be spammed seriously, nor enjoy the "incorporation" cycle whereby extlk content is written into the article in a timely fashion. Charles

There are other things to do short of that. 1. try to change the interpretation of NOT DIRECTORY and the EL policy to permit a section of links with more generous standards. 2. try to get a policy for adding a subpage for links to articles 3. run a mirror of the project, with links added, which is easier & better than a true fork where the articles diverge. David Goodman, Ph.D, M.L.S. http://en.wikipedia.org/wiki/User_talk:DGG On Mon, Mar 29, 2010 at 9:17 AM, Fred Bauder &lt;fredbaud(a)fairpoint.net&gt; wrote:

On 29 March 2010 10:58, Gregory Maxwell &lt;gmaxwell(a)gmail.com&gt; wrote: Yeah. 7-10 is IMO the absolute limit for non-reference links, and I can hardly think of an article that can reasonably justify more than three or four. (I'm sure someone will weigh in with counterexamples, I'm speaking in the general case.) - d.

That probably misses the flux. How many links are added and then almost immediately removed? That won't be picked up in something like that, I don't think. Carcharoth On Tue, Mar 30, 2010 at 4:06 AM, Angela &lt;beesley(a)gmail.com&gt; wrote:

On 30 March 2010 12:49, Charles Matthews &lt;charles.r.matthews(a)ntlworld.com&gt; wrote: I'm not at all convinced there's an actual problem here. Prospective useful links and references can (and should) go on the talk page. - d.

On Tue, Mar 30, 2010 at 6:10 AM, Fred Bauder &lt;fredbaud(a)fairpoint.net&gt; wrote: I don't think there's such a consensus, site wide. I have seen articles where someone OWNs it and there is a local consensus. Keep in mind that we risk ending up with our articles web link farms which is are not maintained in any consistent manner. I support good links, and add them. But there's a downside there too. -- -george william herbert george.herbert(a)gmail.com

On Tue, Mar 30, 2010 at 2:09 PM, Fred Bauder &lt;fredbaud(a)fairpoint.net&gt; wrote: I agree that this is a similar problem. In theory, the 'external links' section of an article should grow and take shape in proportion to the article's size and maturity, not stay constant over time. We have been doing a good job of expanding footnote-style references and external links -- I spoke to a business school class yesterday where a student said "isn't excellent citation one of Wikipedia's main attractions?" -- but there is also value in links to general further reading. A feature to improve the curating and presentation of these links might be handy. We have a few places were having a "set of links" as a first class member of the wikiverse would be useful * external links or further reading * a list of images related to an article (which may not all fit neatly in the article) * interlanguage and interproject links to a set of articles about the same topic SJ

On Fri, Apr 2, 2010 at 9:05 AM, Charles Matthews &lt;charles.r.matthews(a)ntlworld.com&gt; wrote: That's an interesting point. I presume you mean wikisource here. For Commons and Wikiquote (I'm unsure about the other projects) it is fairly easy to have a corresponding page or category or both. If the Wikipedia article is a person who is an author, then a wikisource page is possible, and if the Wikipedia page is about a book or other published work that could be on wikisource, then again a single link, page or category is usually possible. But there are some articles where this system does fall down. I presume the place to put links to editorially selected wikisource pages would be in the external links, or as a courtesy link in a citation. Carcharoth

On 2 Apr 2010, at 11:21, Charles Matthews wrote: I'm really not fond of the poster boxes in their current form at all. It's far too easy for them to clutter up a page. As a suggestion, what about something like: http://en.wikipedia.org/wiki/40-foot_telescope Look at the infobox. While you're at the page, also look at the bottom - that's my preferred way of dealing with external links. ;-) That doesn't solve the issue of multiple links being needed, but IMO it does make the links look a lot better and in a more relevant place. I would expect that most multiple links to Wikisource would/ should be in the references, though - although the same probably wouldn't be true for wikinews links. I think that's just Wikisource's growing pains; over time I think it will probably end up with more disambig pages and also topic pages. But perhaps that's just my viewpoint as I'm used to Wikipedia. That sounds like a way of adding confusion to those editing a page, when they find that part of the page is stored somewhere else completely. Interwiki (as in language) links seem to be dealt with well nowadays by robots; expanding that to include wikisource links might be good. External links are best done as project-specific ones IMO, though. Mike

On Mon, Mar 29, 2010 at 2:58 AM, Gregory Maxwell &lt;gmaxwell(a)gmail.com&gt; wrote: General Thoughts: The editors who feel most strongly about these issues congregate at WP:EL, WP:SPAM, WP:COI, and the related noticeboards and wikiprojects (WP:ELN, WP:WPEL, WP:WPSPAM, WP:COIN). Some of the work that the "cleaners and spamcops" do is _immensely_ helpful, clearing out the blatant SEO/spam links, and even worse items like the malware and shocksites. However, a few take the perspective and skills of a spamcop, and apply them to "imperfect" links and user-contribs, such as when an academic archivist goes around adding links to their university's collection to multiple articles, or when someone goes around fixing urls to a site after it gets restructured. A few vocal editors would even prefer that we only ever had a single "Official site" link in the EL section, and would like all the [[:Category:External link templates]] to be deleted. eg the current (basically biannual) discussion to eradicate all links to wikis, that aren't official sisterproject wikis. http://en.wikipedia.org/wiki/Wikipedia_talk:External_links#Using_Wiki.27s eg the latest (long) discussion concerning archivists http://en.wikipedia.org/wiki/Wikipedia_talk:Conflict_of_interest#What_we_wa… eg the most recent (October) discussion concerning links to archives of official websites at archive.org (wayback machine) http://en.wikipedia.org/wiki/Wikipedia_talk:External_links/Archive_27#ELs_o… etc etc etc. There is definitely a small but active number of editors who have extreme views (as with any subjective issue). Sometimes they happen to be in the same place at the same time, which could give the impression of a "gang or cabal" engaged in a "war". The only thing that can really be done, is to provide the counter-perspective, and hope that consensus results in something sensible. Each and every time. Thankfully, most editors have moderate views on these things. Sadly, we get tired of repeating the same arguments regularly. Small specific example: http://en.wikipedia.org/w/index.php?title=Geoff_Dyer&diff=350129846&… I wrote the original stub, so I _know_ those links were used during its creation (they're 90% interviews with the author). I added them back, and left a note on the editor's talkpage, but he was more interested in removing the single link that had been apparently spammed (which as-it-happens was to a very informative video interview with the author), so he re-removed them all. However, he did take my advice, and at least left a copy on the talkpage this time). http://en.wikipedia.org/w/index.php?title=User_talk:Novaseminary&oldid=… http://en.wikipedia.org/w/index.php?title=Geoff_Dyer&diff=350145148&… I'll add them back eventually, all cited and tidy, but readers won't be likely to find them in the meantime... As has been said before: Most of these types of conflicts can be boiled down to [[m:Immediatism]] vs [[m:Eventualism]]. (imho) Immediatism is great for BLPs, and CurrentEvents, and dealing with unambiguous problems; but Eventualism is one of the core reasons behind Wikipedia's successes, a fact that is sometimes insufficiently recognized. Quiddity