2008/12/6 Eugene van der Pijll <eugene(a)vanderpijll.nl> Sorely tempted to add a reference to Ren Höek to [[Hoek]]. :D Michel

On Sat, Dec 06, 2008 at 03:22:55PM +0100, Eugene van der Pijll wrote: The assumption that if A-B and linked and B-C is linked then A-C are linked is perfectly reasonable. If page A on wiki A is not on the same topic as page B on wiki B, there really shouldn't be an interlanguage link. For example, if wiki A has separate articles on "Beer" and "Wine" and wiki B has only one article on "Alcoholic beverage", no interlanguage link is appropriate. Otherwise, we run into the sorts of problems described by L. Bolikowski above. I realize that, historically, this principle was violated, and some interlanguage links were set up on the principle of "find the closest article that does exist, even if it is not on the same topic". I don't think that's sustainable given the current sizes of the wikimedia projects, because we will need to rely more on automated tools to manage interwiki links. We can resolve the issue of different divisions of material, to some extent, by pointing the interlanguage link to a redirect that ''would be'' the exact same topic if the article were located there. In the example above, that means creating redirects on wiki B from "Beer" to "Alcoholic beverage" and from "Wine" to "Alcoholic beverage". Then we can set up the interwiki links from A to B as desired. - Carl

On Sat, Dec 6, 2008 at 9:22 AM, Eugene van der Pijll <eugene(a)vanderpijll.nl> wrote: "en:December" to "en:City", or "en:Alpine Ibex" to "en:Western culture" Correct? really? I think he's demonstrated that there exist problems, but not yet demonstrated that there exist many. So long as we support interlinking wikipedias of dramatically different size there will be some drift because the larger projects will make finer subdivisions. Lukasz analysis depends on linking being communicative, but this can only be true when there is only one kind of link (x is the same subject no more, no less as y). If we limited ourselves to that it would preclude the "x is covered by broader article y" link which is absolutely necessary if we want to produce useful interwiki links from bigger projects to smaller ones. (And I'd argue that the interwiki links from big projects such as en,es,fr,de to smaller projects like hi are important to getting native speakers of these smaller projects to discover the existence of those projects on the predominantly English speaking internet) It would be interesting to re-run the analysis including only linkages which among the largest few Wikipedia and resolve those first: Those really should be much closer to the ideal "x==y" behavior.

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 First of all, thanks a lot for your replies. Let me clarify a couple of assumptions that I've made: i) there should be at most one article on any given topic in a language edition, which is not true in sh:, az:, ku: and possibly others. ii) the sum of interwiki links should form an equivalence relation iii) an interwiki link to a redirect is treated as an interwiki link to the target of the redirect My rationale: from a user's perspective, the links form a dictionary, so a user might expect that a visit through an interlanguage link will not carry them to a page on a (slightly) different subject. Also, a number of scientific projects seem to be based on this assumption, check for instance the articles "A Bilingual Dictionary Extracted from the Wikipedia Link Structure" (Springer) or "Analyzing Interlanguage Links of Wikipedias" (Wikimania 2008). Some of you may say that the interpretation of interwiki links as a dictionary is incorrect, but the problem exists nevertheless, as I will try to show below. Let me list a couple of popular patterns, to show that some of the problems with interlanguage links are independent of the questioned assumptions (the list is not exhaustive). 1) absurd links, incorrect by any conceivable definition, resulting either from vandalism, like ro:Nicolae Steinhardt -> de:Penis, or complete ignorance regarding the target language: fr:Rick Ankiel -> ja: 日本語 2) systematic errors, like this off-by-ten: http://wikitools.icm.edu.pl/show/en:36898/en:12681205/, or editor's laziness during copy&paste edits, like wuu:5月26号 -> bn:মে ১, wuu:5月27 号 -> bn:মে ১. Generally, automatically generated articles on days of year and years of current era tend to introduce conflicts, but fortunately these are easy to detect and fix. 3a) links to disambiguation pages la:Benedictus (nomen) -> en:Benedict 3b) links to incorrect meanings of a homonym: it:Rubinetto -> es:Grifo 4) combination of redirects and interwiki en:Mother-in-law -> ru:Тёща + redirect ru:Тёща -> ru:Родство + interwiki ru:Родство -> en:Kinship 5) a series of links widening and narrowing a meaning (excluding disambigs): pl:Województwo krakowskie (I Rzeczpospolita) -> en:Kraków Voivodeship (14th century-1795) -> pt:Voivodia da Cracóvia -> pl:Województwo krakowskie. The first two cover the period 14th century-1795, the third: 14th century-1998, the fourth: 1945-1998. 6) Problems stemming from the cultural and linguistic limits of the translation process, for example regarding food: http://wikitools.icm.edu.pl/show/en:57572/ or meals: http://wikitools.icm.edu.pl/show/en:71691/ Note that types 1, 2 and 3b are clearly incorrect, while the other ones are disputable. From my experience, all the types occur quite often. I encourage you to explore the incoherences yourself: take a random path and find the source(s) of semantic drift. As you can see, most of the examples contain links between the major language editions. I would hypothesize that only the second category is "generated" by the small language editions, the rest is dominated by larger editions, simply because there are more opportunities to make an incoherent edit. I'll run the statistics for the top 10 editions to test Gregory's hypothesis. Finally, let me write a few words about the possible large-scale solutions to the problem. I'm afraid that "centralization" of interlanguage links, ie. a separate service where all the interlanguage links would be stored and manipulated, would inadvertently impose an Anglocentric ontology, which is (presumably) not desired. The "decentralized interwikis + lots of bots" model, with all its flaws (for example: it's not feasible to find the incoming interlanguage links), will probably reflect the ontologies of smaller editions better. Introducing two flavors of interwikis, "exact" and "approximate", would help both retain both the valuable interlanguage links that are incorrect under my narrow definition of correctness, and express the equivalence where it occurs. Cf. the concepts of defined meaning and relations in OmegaWiki: http://www.omegawiki.org/DefinedMeaning As a side note, adding semantics to interlanguage links very nicely fits the model used in the SemanticMediawiki extension, which is unfortunately not integrated into Wikipedia (yet). Regards, Lukasz

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Carl Beckhorn wrote: That's true. That's why I opt for more than one type of interlanguage links. Apart from "identical meaning" type, OmegaWiki also has "narrower term", "broader term", "related term" and "part of the theme". I'm not sure if we need such fine-grained relations, in fact introducing more than one type could be quite confusing. Thus, I thought about "exact" and "approximate" only. Your idea, if I understood correctly, is that interlanguage link + redirect should represent an "approximate" relation. It is quite tempting, esp. since it does not require any change in the engine. However, IMHO, it might still be confusing to a user, and it would be very difficult to convince all the 250+ language editions to accept this interpretation. Regards, Lukasz

2008/12/12 Thomas Dalton <thomas.dalton(a)gmail.com>om>: Ok, I used the wrong phrase, but it's still very much the case that the interlanguage relationship definitely isn't transitive, and he's assuming it is. -- -Ian Woollard We live in an imperfectly imperfect world. Life in a perfectly imperfect world would be much better.

2008/12/12 David Gerard <dgerard(a)gmail.com>om>: The problem here is... well, I want to read the German version of an article, say. And down in the sidebar... * Deutsch * Deutsch * Español * Français * Nederlands * Português * Русский * Svenska ...how do I know which of the German articles is which? The system's really set up to expect only one link... -- - Andrew Gray andrew.gray(a)dunelm.org.uk

David Gerard wrote: C'mon, even mathematicians have common sense, sometimes :) A mathematician would simply say that the relation of being a close friend is not transitive. Thus, if the interlanguage links were to mean "roughly equivalent", then it wouldn't be transitive and it would be unsound to perform a transitive closure. In other words, if the links are interpreted as "roughly equivalent" then you're absolutely right: it doesn't make sense to do the analysis that I've done. The point is, according to the policies (at least the way I interpret them) and according to some bot authors and probably according to a large share of users, the interlanguage links do represent equivalence. See: http://en.wikipedia.org/wiki/Help:Interlanguage_links http://meta.wikimedia.org/wiki/Help:Interwiki_linking#Interlanguage_link Besides, note that multiple interlanguage links from a given page to a given language edition are discouraged (although partially supported by the MediaWiki engine). If the ILLs were to mean "similarity", then multiple links to the same language wouldn't have been discouraged (an article can be "similar" or "roughly equivalent" to more than one article in another language edition). Summing up, IMHO the intended meaning of ILLs is "equivalence", and it's supported by four arguments (two links, "discouraged multiple ILLs" argument, and bots). Another thing is that some of the ILLs are clearly incorrect even using the relaxed interpretation (like June 26 -> June 27, Rick Ankiel -> Japanese language, Tap (valve) -> Griffin, listed in the other post). Regards, Łukasz

http://meta.wikimedia.org/wiki/Interlanguage_use_case#Use_case It's artificial, but possible, and quite amusing. Regards, =C5=81ukasz

On Sat, Dec 13, 2008 at 12:16:47AM +0000, Ian Woollard wrote: Indeed - so the enwiki article should not link to the french one, since they are not about the same topic. Could you explain what you mean by "intended"? I have long thought that the intention of ill links is for articles that cover exactly the same subject. In each of those cases, it seems to me that no interlanguage link is appropriate. In theory, there is plenty reason for the links to be transitive. Since that is the best way to automatically extend the links to other wikis via ill bots. - Carl

Carl Beckhorn wrote: As a human user of interwiki links, I don't necessarily use them as a way of saying "here is an article on the same topic", but "here is an article that also covers this topic". I do agree with you in this particular case---maybe I'm wrong on this, but I don't generally expect links to *narrower* topics. But links to *wider* topics are fine, if the linked article does cover the specific topic. So if, for example, en chose to cover some subjects A1 and A2 as part of one big unified article on A, but fr chose to make two different articles, A1 and A2, then I would expect both fr:A1 and fr:A2 to interwiki link to en:A (the English article where the equivalent content can be found), but no en: interwiki links to fr:, since there's no article with equivalent content. Actually I might link en:A to link to *both* fr:A1 and fr:A2, to indicate to me that the equivalent content can be found spread across those two articles, but I'm not sure you can link to multiple pages of the same language in one page. I realize this is problematic for automatic linking, but automatic linking really is problematic unless we were to keep our editions much more in sync than we actually do. It's not just shifts in meaning as you translate between languages that are a problem, but different editorial decisions made in organizing articles, that would result in many articles just having no interwiki links at all, if we were to insist on only on exact equivalents as links. This is especially true of conceptual subjects, as opposed to proper nouns, which do tend to be more directly equivalent---so for example, there is more gray area and lack of transitivity when linking articles about law in general, than when linking articles about specific legislative acts or court cases. -Mark

On Tue, Dec 16, 2008 at 8:33 AM, Ian Woollard <ian.woollard(a)gmail.com> wrote: I wholeheartedly agree. IMO, this is another instance where the human usability of something on Wikipedia is being damaged by the drive to make something more compatible with bots. -Matt

On Wed, Dec 17, 2008 at 2:47 AM, Lukasz Bolikowski <L.Bolikowski(a)icm.edu.pl> wrote: Since policy describes practice and not the other way round, I suspect that the policies may need fixing or clarifying. By this do you mean the en.wp community, or the community of all projects in all languages? If the latter case, I suspect consistency is not practically possible. This feels to me essentially the same issue as with the category tree: while some might prefer a strict IS-A relationship for categorization, in practice we have to accept that all that category membership means is some kind of hopefully human-understandable relationship. -Matt

From: "Lukasz Bolikowski" <L.Bolikowski(a)icm.edu.pl> (...) (...) A bug or two is in the sub-category tool (+) that is on cat pages -- as compared to catscan, which was written into those cat pages with enhancements. I think that's what the author means by "snafu" on catscan. Catscan can make a tree out of [[:category:neurology]] that includes more information.

Could you explain the nature of this debate? While there is certainly a grey area between "descriptive" and "interpretive" I think the basic plot elements of a novel aren't generally open to interpretation (there will be exceptions for certain parts of certain novels, and those can be dealt with on a case-by-case basis). Also, it's important to note that novels were just an example - in most cases, there isn't even a significant grey area, eg. you can use someone's birth certificate as a reference for their date of birth, you cannot use it as a reference for them having been born during the Great Depression and thus having had a tough childhood. The former is a simple fact, the latter is a (somewhat speculative) interpretation of that fact.

Please don't! I try and avoid reading about lit crit (I completely avoid reading actual lit crit), far too many long words used purely to sound clever.

2008/12/17 Phil Sandifer <snowspinner(a)gmail.com>om>: What? WP:NOR barely mentions lit crit or anything related to it. It's a general policy that applies to literature as much as it does to science or history. I think you may be a little paranoid...

2008/12/17 Phil Sandifer <snowspinner(a)gmail.com>om>: I disagree. A scientist learns how to read and write scientific papers in their science classes. A historian learns how to read and write historical papers in their history classes, etc. An English professor isn't going to know the first thing about judging the credibility of a paper on Quantum Electrodynamics.

Perhaps in universities where students study a wide range of subjects. In UK universities, most people never go near the English department but almost all of them will learn how to do academic research and writing (often as you go along, rather than in a specific class on it).

I didn't write off the subjects, I just wrote off reading about them when they are written using a bunch of made up words. If someone would like to summarise the debate in English, I'd be very interested to read it. There are appropriate uses of jargon, but the way lit crit uses it is just designed to make the author look clever. When it becomes impossible to tell the difference between the real thing and satire, you know something has gone horribly wrong.

2008/12/17 Phil Sandifer <snowspinner(a)gmail.com>om>: And this is one of the reasons why wikipedia policy is what it is. This fight has been done by others and they have done it better ( eg http://www.info.ucl.ac.be/~pvr/decon.html ). Wikipedia policy is a pragmatic approach to the situation. -- geni

2008/12/17 Phil Sandifer <snowspinner(a)gmail.com>om>: It's very pragmatic. People in general seem to want plot summaries. It can also be rather hard to talk about a book/film/legend without one. They also appear to be expected of encyclopedia articles. So pragmatically we need to produce something the majority of people will accept as plot summaries. Unfortunately for less prominent works there tends to be a lack of secondary sources for such summaries to be based on. There is a further lack of accessible sources but that is a fairly universal problem. However experience shows us that wikipedians are for the most part able to write things that both the majority of wikipedians and our general readership are prepared to accept as reasonable summaries of the plots of the work in question. This being the case it is perfectly acceptable to allow them to continue doing so. -- geni

2008/12/17 Phil Sandifer <snowspinner(a)gmail.com>om>: Depends if you are considering a general encyclopedia or a specialist encyclopedia. Strawman. Blanket denial without at least a line of reasoning behind it is not a helpful approach to debate. -- geni

Mathematics: Can't get enough original research or validation or exceptions, and please state your assumptions. Physics: No perpetual motion machines, please. Original research in nuclear science is covered by treaties and your local American army base. Biology: Please restrict your original research to photographs. Chemistry: This department actually has a policy regarding notability and what you can say about compounds that hav not been synthesized. Biochemistry: No man's land. There's big pharma, there's governmental collusion, cover-ups, smear campaigns, drug promotions and oodles of tobacco funding. Wanna meet spooks and Britain's future king? Then make a name for yourself in this field. Arts: um...anything goes...sorta...but not here...we just write about arts like policy making and other methods of screwing you, here. We try to make it artful, anyway, just so you do not forget. If I bored you to tears, then I am sorry. Stay away from [[vagina]]. No man will ever be happy until his wife is there. ----- Original Message ----- From: "Phil Sandifer" <snowspinner(a)gmail.com> To: "English Wikipedia" <wikien-l(a)lists.wikimedia.org> Sent: Wednesday, December 17, 2008 10:30 AM Subject: [WikiEN-l] The Community vs. Scholarly Consensus

On Wed, Dec 17, 2008 at 9:30 AM, Phil Sandifer <snowspinner(a)gmail.com> wrote: Like most things on Wikipedia, attempts to define boundary cases rigidly is impossible. We seem to labor under the (intentionally blinkered?) delusion that it should be possible, if we try hard enough, to set unambiguous policy that removes any need for editorial judgment. -Matt

From: "Ian Woollard" <ian.woollard(a)gmail.com> (...) (...) I don't know. I tend to think that my redirection of "Zeigarnik Effect" to "Suspense" in English was a hearty recommendation to all languages. Heck, you can almost feel it in wikipedia with weight of all those plans you hav for articles that gets SUSPENDED for research you would like to do in old books that you can't find in your library anymore. My library has the option of e-mailing receipts for books I've taken out...YAY...now I can archive them, just in case it was more than ten years ago that I took the thing out, ay. There was, for instance, this absolutely dandy book on nutrients, and formatted in a way that might let me exclude Tin and Arsenic if I could just get estimated figures on probable and incidental exposure, then rule out benefits from other nutrients. I saw an absolutely crappy list of essential elements that included Uranium (Holy $#!+ of a Bull!). _______ [http://ecn.ab.ca/~brewhaha/ BrewJay's Babble Bin]