-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1
Hi,
I've done some research on the network of interlanguage links as a whole, you can see the results here: http://wikitools.icm.edu.pl/
I wrote to this list earlier this year about incoherences in the interlanguage links, but two things have changed since then: the problem has got more serious, and I've developed a more usable tool to correct it.
A short introduction: let's say that two articles are connected if there is an interlanguage link from one to the other in at least one direction. Next, let's say that if A-B and B-C are connected, then A-C are too. Next, for each group of connected articles, let's check if it is coherent, ie. if there is at most one article from each language.
It turns out that about 5% of articles belong to incoherent groups. The largest such group is growing quite fast: it had 48'000 articles in March 2008, now it has over 76'000! With over 3'000'000 links to check, it has to be corrected semi-automatically. There are tens of thousands of other incoherent groups to fix, too.
Right now, you can find some really absurd connections using the interlanguage links alone, like "en:December" to "en:City", or "en:Alpine Ibex" to "en:Western culture". The site I've created let's you see a path connecting given two articles, and suggests a course of action. The suggestions are a result of a heuristic and should be taken with a grain of salt, but maybe you'll find them useful.
Regards, Lukasz Bolikowski
PS. Last time my replies were coming several days after I'd post them. If I don't respond it's probably because my response is still moderated. Anyway, I guess http://meta.wikimedia.org/wiki/Interwiki_synchronization is the best place to discuss this matter.
Lukasz Bolikowski schreef:
A short introduction: let's say that two articles are connected if there is an interlanguage link from one to the other in at least one direction. Next, let's say that if A-B and B-C are connected, then A-C are too.
Hold it right there. That assumes that every wikipedia divides its content into pages in the same way. Since content policy is made at the level of individual projects, this assumption is incorrect.
With over 3'000'000 links to check, it has to be corrected semi-automatically.
It doesn't have to be corrected at all, since you've not actually demonstrated that it is incorrect now.
Eugene
(who is a bit ticked off about this subject because of all the work he's done to keep incorrect wikilinks off the [[Hoek]] dab page...)
2008/12/6 Eugene van der Pijll eugene@vanderpijll.nl
(who is a bit ticked off about this subject because of all the work he's done to keep incorrect wikilinks off the [[Hoek]] dab page...)
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:
Lukasz Bolikowski schreef:
A short introduction: let's say that two articles are connected if there is an interlanguage link from one to the other in at least one direction. Next, let's say that if A-B and B-C are connected, then A-C are too.
Hold it right there. That assumes that every wikipedia divides its content into pages in the same way. Since content policy is made at the level of individual projects, this assumption is incorrect.
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
Carl Beckhorn schreef:
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.
Consider the example I mentioned: [[Hoek]]. It is a dab page of all things referred to as "Hoek" in English. It is linked to [[nl:Hoek]], which is a page of all things called "Hoek" in Dutch, which is covers much more (places called "Hoek", and things that are called "hoek" in Dutch).
Different coverage, so invalid wikilink?
The Dutch dab page for "angle" and "corner", which are synonyms in Dutch, was linked to a Polish dab page, since "angle" and "corner" are synonyms there too (I assume). Is that also an incorrect link?
I would say both links are valid.
Unfortunately, then bots started linking the English article on [[Hoek]] to the Polish article on angles, and I had to revert them at least 15 times.
I would be interested in Lukasz' conclusions if he had left out dab pages.
Eugene
On Sat, Dec 6, 2008 at 9:22 AM, Eugene van der Pijll eugene@vanderpijll.nl wrote:
Lukasz Bolikowski schreef:
A short introduction: let's say that two articles are connected if there is an interlanguage link from one to the other in at least one direction. Next, let's say that if A-B and B-C are connected, then A-C are too.
Hold it right there. That assumes that every wikipedia divides its content into pages in the same way. Since content policy is made at the level of individual projects, this assumption is incorrect.
With over 3'000'000 links to check, it has to be corrected semi-automatically.
It doesn't have to be corrected at all, since you've not actually demonstrated that it is incorrect now.
"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.
On Sat, Dec 06, 2008 at 11:40:13AM -0500, Gregory Maxwell wrote:
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.
I was saying in another message that this seems like a case where a redirect would be desirable. Instead of linking from [[en:Specialized topic]] to [[tlh:General topic]], we should created a redirect at [[tlh:Specialized topic]] and point the interlanguage link at that. Then human navigation will work smoothly, through the redirect. But automatic analysis of the interlanguage link system will not follow the redirect and will not think that [[en:Specialized topic]] is the same as [[tlh:Generalized topic]]. Especially with the advent of unified login there is little difficulty in creating redirects on other-language wikis, if you know the other language.
- Carl
-----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
On Mon, Dec 08, 2008 at 01:53:49PM +0100, Lukasz Bolikowski wrote:
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
(ii) and (iii) look to be inconsistent with the idea that every article should have interwiki links.
For example, enwiki has a single article on [[eigenvector, eigenvalue, and eigenspace]]. If another wiki were to have three separate articles on these topics, and we wanted to keep the interlanguage links as an equivalence relation, then we cannot link those three to the main enwiki article.
How would you propose handling a situation like that?
- Carl
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1
Carl Beckhorn wrote:
For example, enwiki has a single article on [[eigenvector, eigenvalue, and eigenspace]]. If another wiki were to have three separate articles on these topics, and we wanted to keep the interlanguage links as an equivalence relation, then we cannot link those three to the main enwiki article.
That's true.
How would you propose handling a situation like that?
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
Gregory Maxwell wrote:
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.
Hi, I've rerun the analysis as you proposed. I've taken all the articles from the 10 largest editions (de, en, es, fr, it, ja, nl, pl, pt, ru) and the interlanguage links between them. I was looking for incoherent components, as defined in my previous posts.
In this setting, there are 44245 incoherent components, containing, in total, 436529 articles from the 10 editions. Which shows that this is not only a problem of linking to/from small wikis. Also, trying the engine at: http://wikitools.icm.edu.pl/ you'll see that the differences between the ontologies of the large and the small wikis are not the only issue (as Eugene suggested).
Let me rephrase my concerns: on one hand, the policies state that interlanguage links represent equivalence: Meta says they connect "corresponding" articles, the English edition says they connect articles "on the same subject". There are third-party projects which assume this (I've given two examples before), not to mention an army of bots.
On the other hand, editors don't respect that strict interpretation since they want to show (valuable) relations between non-equivalent articles. Without seeing the "big picture", any such inexact link seems OK: what could possibly go wrong? And the global view does not seem to be commonly known...
I don't have a ready solution, although, as I've written before, we could take a closer look at the way OmegaWiki is dealing with the issue (in my perception, the project's existence is motivated solely by the existence of the issue in question), and the potential offered by the SemanticMediaWiki extension.
My main goal is to convince the community (or be convinced otherwise) that this is a serious, growing problem, which requires attention, and stimulate a discussion which might lead to a reasonable solution.
Regards, Łukasz
PS. An example: the following English articles are mutually accessible using only the interlanguage links between the top 10 editions (assuming that a link A -> B makes A accessible from B, which doesn't necessarily match users' experience, but bots and harvesters "see" it): Administration Administration (business) Administrator Aktiebolag Aktiengesellschaft Aktieselskab Apostolic Administrator Besloten Vennootschap Brother (disambiguation) Brotherhood Brotherhood (album) Business Businessperson Compagnons du Tour de France Companies law Company Company (disambiguation) Contract Corporate law Corporation Corporation (university) Entrepreneur Entrepreneurship Fraternities and sororities Fraternity Fraternity (disambiguation) General partnership German Student Corps Gesellschaft mit beschränkter Haftung Government-owned corporation Guild Hermano Hermano (band) Incorporation (business) Joint stock company Journeyman Junior Chamber International Kabushiki kaisha Legal name (business) Limited company Limited liability company List of general fraternities Management Management science Maszoperia Naamloze Vennootschap Public company Public limited company S.A. (corporation) Sibling Sister (disambiguation) Société à responsabilité limitée Société par actions simplifiée Society Society (disambiguation) Sole proprietorship Studentenverbindung Student society Trade name Types of business entity Yugen kaisha
On 05/12/2008, Lukasz Bolikowski L.Bolikowski@icm.edu.pl wrote:
A short introduction: let's say that two articles are connected if there is an interlanguage link from one to the other in at least one direction.
Yes.
Next, let's say that if A-B and B-C are connected, then A-C are too.
No. That's only generally true for equality relations.
Equivalence relations are not equality relations.
For example in maths, if you need a value to be 'equal' within 0.1, then 9.95 is equivalent to 10.04, but not to 10.06, even though 9.95 is equivalent to 10.04 and 10.04 is equivalent to 10.06. That's because 'equal' within 0.1 is an equivalency, not an equality.
Interlanguage links are equivalence relations (at best.)
Really, there is no solution to this problem. 1:1 translation of concepts/topics between different languages does not in general occur and hence these relations are at best equivalencies NOT equalities.
Your analysis is assuming that they are equalities, and that's why you're getting nonsensical results.
Fun though.
-- -Ian Woollard
We live in an imperfectly imperfect world. Life in a perfectly imperfect world would be much better.
Next, let's say that if A-B and B-C are connected, then A-C are too.
No. That's only generally true for equality relations.
Equivalence relations are not equality relations.
Actually, transitivity is a requirement for an equivalence relation. (In addition to symmetry (A~B => B~A) and reflexivity (A~A for all A).)
2008/12/12 Thomas Dalton thomas.dalton@gmail.com:
Next, let's say that if A-B and B-C are connected, then A-C are too.
No. That's only generally true for equality relations.
Equivalence relations are not equality relations.
Actually, transitivity is a requirement for an equivalence relation. (In addition to symmetry (A~B => B~A) and reflexivity (A~A for all A).)
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.
2008/12/12 Ian Woollard ian.woollard@gmail.com:
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.
Yeah. The common usage of interwiki links on en:wp is "that article over there is roughly equivalent to this one." I have seen en:wp articles pointing to two de:wp articles, quite reasonably. They are designed to be useful, but are only useful to the reader going from one place to the next one. The next link is in no way guaranteed to be useful in the same way.
If A and B are close friends and A and C are close friends, only a mathematician would think that *should* mean A and C are close friends. They might well be, but there's no reason at all to presume so.
- d.
2008/12/12 David Gerard dgerard@gmail.com:
Yeah. The common usage of interwiki links on en:wp is "that article over there is roughly equivalent to this one." I have seen en:wp articles pointing to two de:wp articles, quite reasonably.
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...
On 12/12/2008, Andrew Gray shimgray@gmail.com wrote:
...how do I know which of the German articles is which? The system's really set up to expect only one link...
Yeah, it could probably do with a decent hover-over.
--
- Andrew Gray andrew.gray@dunelm.org.uk
David Gerard wrote:
Yeah. The common usage of interwiki links on en:wp is "that article over there is roughly equivalent to this one." I have seen en:wp articles pointing to two de:wp articles, quite reasonably. They are designed to be useful, but are only useful to the reader going from one place to the next one. The next link is in no way guaranteed to be useful in the same way.
If A and B are close friends and A and C are close friends, only a mathematician would think that *should* mean A and C are close friends. They might well be, but there's no reason at all to presume so.
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
2008/12/12 Lukasz Bolikowski bolo@icm.edu.pl:
C'mon, even mathematicians have common sense, sometimes :) A mathematician would simply say that the relation of being a close friend is not transitive.
Yeah, but a physicist would say: 'Surely you're joking Mr. Feynman' and everyone would laugh.
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.
Yup. Fraid so. Cool research though.
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.
Well, let's take an example, like:
http://en.wikipedia.org/wiki/Rocket
Down the side are a huge number of links including the French one:
http://fr.wikipedia.org/wiki/Fus%C3%A9e_spatiale
This title translates as 'Space Rocket'.
Now straight away we are in trouble. The English wikipedia's Rocket article is about the general case of rockets- any vehicle that is propelled by a rocket engine, including a rather awesome Russian torpedo, some drag racers, aircraft, and the worlds fastest train (Mach 8.5!!!), whereas the French article is about only space rockets.
But there's nowhere else to go. And this feature is working exactly as intended.
Now the English wikipedia pretty much has an article on that too 'Launch vehicle', so really the return link from the French article could go there instead... not back to rocket: and we've moved already. (As it happens the actual link from fr goes back to Rocket, but there's no reason that the wikipedia doesn't have a precise article on space rocket in which case my example would be even clearer, it's just a fluke.) It really doesn't take many hops and we would be somewhere completely different.
And at no stage is the linkage strictly wrong. The underlying problem is that you're assuming exact correspondence, whereas it's more like a thesaurus; these are *synonymous links*. The phrase for people in the know is: 'There's no such thing as a true synonym.' And that's what blows it up.
The problems are many fold. Linked articles can have a definition that makes them a subset, partial overlap or superset. If you go through a few rounds of going to subset and then partial overlap and back up to superset you can end up practically anywhere, as you've shown rather admirably. There is absolutely no reason to think that these links are transitive in practice or theory.
Regards, Łukasz
http://meta.wikimedia.org/wiki/Interlanguage_use_case#Use_case It's artificial, but possible, and quite amusing.
Regards, =C5=81ukasz
Ian Woollard wrote:
Well, let's take an example, like: [snip]
We're talking about a bit different things. You're trying to convince me that in practice ILLs often represent "similarity" (BTW, it's a good example). But I already know this.
What I'm saying is that there's a big discrepancy between the policies and the reality.
And at no stage is the linkage strictly wrong. The underlying problem is that you're assuming exact correspondence, whereas it's more like a thesaurus; these are *synonymous links*. The phrase for people in the know is: 'There's no such thing as a true synonym.' And that's what blows it up.
It's Wikipedia: cities, countries, famous people and dates are examples of articles that either have the exact correspondence in another editions, or no correspondence at all.
Thus, if you start from [[:en:Griffin, Georgia]] and end up in [[:en:Classic RISC pipeline]], it makes more sense to correct some of the *incorrect* ILLs that caused it, rather than conclude: well, that's what ILLs are, let's get used to it.
The problems are many fold. Linked articles can have a definition that makes them a subset, partial overlap or superset. If you go through a few rounds of going to subset and then partial overlap and back up to superset you can end up practically anywhere, as you've shown rather admirably. There is absolutely no reason to think that these links are transitive in practice or theory.
Some ILLs represent subset/superset, others represent loose similarity, others are simply wrong and yet another represent exact correspondence (yes, there are such ILLs!) Wouldn't it be nice to know which are which? Right now part of the community interprets all the ILLs as loose similarity, while another part interprets all of them as exact correspondence.
Regards, Łukasz
PS. OpenPGP somehow ate all this, leaving only the last paragraph which already came in the previous post. Sorry for that.
On Sat, Dec 13, 2008 at 12:16:47AM +0000, Ian Woollard wrote:
Well, let's take an example, like:
http://en.wikipedia.org/wiki/Rocket
Down the side are a huge number of links including the French one:
http://fr.wikipedia.org/wiki/Fus%C3%A9e_spatiale
This title translates as 'Space Rocket'.
Now straight away we are in trouble. The English wikipedia's Rocket article is about the general case of rockets- any vehicle that is propelled by a rocket engine, including a rather awesome Russian torpedo, some drag racers, aircraft, and the worlds fastest train (Mach 8.5!!!), whereas the French article is about only space rockets.
Indeed - so the enwiki article should not link to the french one, since they are not about the same topic.
But there's nowhere else to go. And this feature is working exactly as intended.
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.
The problems are many fold. Linked articles can have a definition that makes them a subset, partial overlap or superset.
In each of those cases, it seems to me that no interlanguage link is appropriate.
There is absolutely no reason to think that these links are transitive in practice or theory.
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
2008/12/15 Carl Beckhorn cbeckhorn@fastmail.fm:
On Sat, Dec 13, 2008 at 12:16:47AM +0000, Ian Woollard wrote:
Now straight away we are in trouble. The English wikipedia's Rocket article is about the general case of rockets- any vehicle that is propelled by a rocket engine, including a rather awesome Russian torpedo, some drag racers, aircraft, and the worlds fastest train (Mach 8.5!!!), whereas the French article is about only space rockets.
Indeed - so the enwiki article should not link to the french one, since they are not about the same topic.
Gee, there was I think this UI feature was something intended to be useful for the users. How naive was I?
But there's nowhere else to go. And this feature is working exactly as intended.
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.
Actually the articles in the wikipedia are on whatever the definition in the lead says. So, your argument is that if the definitions in the articles are *exactly* the same (even though they are written in different languages) then they can be linked otherwise not? And I presume you have teams of bilingual lawyers standing by to check this for each and every link in the wikipedia?
I think, in that case you might as well delete all the links, the chances of the definition being exactly the same in any two cases must be so near to zero as to be not worth worrying about.
The problems are many fold. Linked articles can have a definition that makes them a subset, partial overlap or superset.
In each of those cases, it seems to me that no interlanguage link is appropriate.
It seems to me that the users of the wikipedia might prefer synonymous links over no links at all. It also seems to me that the users of the wikipedia are the ones creating the links not you, and somehow I doubt that they are that worried about it. The links are for them, not the bots.
There is absolutely no reason to think that these links are transitive in practice or theory.
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.
In theory there is no difference between theory and practice, but in practice there is.
And "ill bots". Freudian.
I hope these bots are marking the links they're making so they can be removed if necessary. The number of problems that bots/users can cause if they are working at cross purposes, which they clearly are here, must be almost astronomical.
- Carl
Carl Beckhorn wrote:
On Sat, Dec 13, 2008 at 12:16:47AM +0000, Ian Woollard wrote:
Well, let's take an example, like:
http://en.wikipedia.org/wiki/Rocket
Down the side are a huge number of links including the French one:
http://fr.wikipedia.org/wiki/Fus%C3%A9e_spatiale
This title translates as 'Space Rocket'.
Now straight away we are in trouble. The English wikipedia's Rocket article is about the general case of rockets- any vehicle that is propelled by a rocket engine, including a rather awesome Russian torpedo, some drag racers, aircraft, and the worlds fastest train (Mach 8.5!!!), whereas the French article is about only space rockets.
Indeed - so the enwiki article should not link to the french one, since they are not about the same topic.
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 16/12/2008, Delirium delirium@hackish.org 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 English wikipedia 'Rocket', which is intended to be an encyclopedic article on rocket *vehicles* including rocket planes like the X-1 ( :-) ) wouldn't link anywhere? Is that right?
Having checked the other articles, neither the French nor German articles are anything like as broad. So you seem to be arguing that that's right and all the links should be removed. I'm really not sure about this.
One big problem is that the article names frequently affect the article scope. Any two languages use the term "rocket" in different ways and this lends a different scope. In one language it may be a vehicle. In another a missile only. In a third it might be the engine.
I think using this link for anything other than synonymity reduces the usefullness and gets nothing back.
So, unless you have coordination of article scope between languages, then these links cannot be exact. And trying to get coordination between languages just to impose exact cross-linking for a bunch of bots- don't even think about going there, the problems would be *massive*.
So far as I can tell, these links are used by humans to denote synonimity. Using bots to assert synonimity based on transitivity might probabilistically work for at most a few hops. Beyond that, the bots will get it wrong too often to be much use; they just don't know what's going on well enough.
-Mark
On Tue, Dec 16, 2008 at 8:33 AM, Ian Woollard ian.woollard@gmail.com wrote:
So far as I can tell, these links are used by humans to denote synonimity. Using bots to assert synonimity based on transitivity might probabilistically work for at most a few hops. Beyond that, the bots will get it wrong too often to be much use; they just don't know what's going on well enough.
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
Matthew Brown wrote:
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.
Not in this case. Personally, I'd simply like to make the ILLs more compatible with the policies that describe what they're for.
Also, I'd be nice if the community used the ILLs in more or less consistent manner (which is apparently not the case right now).
Regards, Łukasz
On Wed, Dec 17, 2008 at 2:47 AM, Lukasz Bolikowski L.Bolikowski@icm.edu.pl wrote:
Matthew Brown wrote:
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.
Not in this case. Personally, I'd simply like to make the ILLs more compatible with the policies that describe what they're for.
Since policy describes practice and not the other way round, I suspect that the policies may need fixing or clarifying.
Also, I'd be nice if the community used the ILLs in more or less consistent manner (which is apparently not the case right now).
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
Matthew Brown wrote:
Since policy describes practice and not the other way round, I suspect that the policies may need fixing or clarifying.
Fixing or clarifying the policies is definitely one of the possible options to solve the problem, although I suspect that it will not be easy to reach a consensus in this case, for at least two reasons:
1) the engine was not designed with the proposed interpretation in mind (at least in one point it is assumed that there is at most one ILL from a page to a wiki);
2) my guess, partially based on the statistics that I've gathered, is that over 95% of the ILLs are "exact", while the remaining 5% or less are either "approximate" or plain wrong -- the vast majority of edits is done in line with the current policies.
In the perfect world, we'd have both the "exact" and "approximate" ILLs, with tags specifying which are which, and the coherence of the "exact" ILLs would be guaranteed by the engine.
Also, I'd be nice if the community used the ILLs in more or less consistent manner (which is apparently not the case right now).
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.
I meant the whole community. Yes, it might be tough, but not impossible. If we had two types of ILLs ("exact" and "approximate"), people would be forced to know their intended meanings, and hopefully they'd respect them.
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.
That's a very sober observation. I do try not to be a technocrat too, apparently with mixed results.
Anyway, the discussion here has focused on whether the "approximate" ILLs are OK or not, while the most funny/surprising results of my analysis are not a result of the "appropriate" ILLs, but the plain wrong ones. And *these can be fixed without waiting for a new functionality in the engine, or a consensus among the editors*!
Sticking with the example that I've given before: all the paths from the plumbing-related topics to the vulture-related ones are leading through three clearly incorrect links to: [[:es:Grifo]] (instead of [[:es:Llave de paso]]) from the Farsi, Italian and Swedish editions (I've already corrected all the three links). You can find thousands of clearly incorrect links by analyzing the incoherent paths at http://wikitools.icm.edu.pl/
Regards, Łukasz
From: "Lukasz Bolikowski" L.Bolikowski@icm.edu.pl (...)
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.
(...)
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.
Picking up on a thread from the anti-intellectualism thread, WP:NOR currently reads " Any interpretation of primary source material requires a reliable secondary source for that interpretation. Without a secondary source, a primary source may be used only to make descriptive claims, the accuracy of which is easily verifiable by any reasonable, educated person without specialist knowledge. For example, an article about a novel may cite passages from the novel to describe the plot, but any interpretation of those passages needs a secondary source."
At a number of points, this steps squarely into a decades-long debate in literary studies about the nature of reading and of interpretation. This is a debate that is still - perhaps permanently -unsettled. However the view Wikipedia is taking - that there is some core of knowledge that is "descriptive" as opposed to "interpretive" - is decades out of the realm of accepted. It's a discredited view.
The counter, which I've seen discussing this on the NOR page, is that the language represents community consensus, and that's good enough.
This is troubling to me - most significantly because it suggests that the community is empowered to set an official Wikipedia position on the nature of language and meaning, and to settle a decades long dispute. On the one hand, this seems to me a problem of NPOV - but it's also an intractible one, as we can't avoid having some policy on the nature of reading.
But on the other hand, surely we do not actually intend to empower the community (by which we really mean the people who wrote NOR) to rule on issues like this and ignore a huge scholarly debate.
I don't have a good answer here, so I figured I'd ask the list - to what extent is the community empowered to set a Wikipedia policy on a scholarly debate? How do we square this with NPOV? Is there a clever out to the underlying problem in NOR that sidesteps the debate?
Thoughts?
-Phil
At a number of points, this steps squarely into a decades-long debate in literary studies about the nature of reading and of interpretation. This is a debate that is still - perhaps permanently -unsettled. However the view Wikipedia is taking - that there is some core of knowledge that is "descriptive" as opposed to "interpretive" - is decades out of the realm of accepted. It's a discredited view.
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.
On Wed, Dec 17, 2008 at 5:46 PM, Thomas Dalton thomas.dalton@gmail.com wrote:
At a number of points, this steps squarely into a decades-long debate in literary studies about the nature of reading and of interpretation. This is a debate that is still - perhaps permanently -unsettled. However the view Wikipedia is taking - that there is some core of knowledge that is "descriptive" as opposed to "interpretive" - is decades out of the realm of accepted. It's a discredited view.
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).
<snip>
I think Phil means the debate over authorial intentionality (and related topics):
http://en.wikipedia.org/wiki/Authorial_intentionality
Though he might have meant other aspects of that debate.
Other articles that might help:
http://en.wikipedia.org/wiki/Literary_criticism http://en.wikipedia.org/wiki/Deconstruction http://en.wikipedia.org/wiki/Post-structuralism http://en.wikipedia.org/wiki/New_Criticism http://en.wikipedia.org/wiki/Literary_theory http://en.wikipedia.org/wiki/Reader-response_criticism
I could go on...
Carcharoth
I think Phil means the debate over authorial intentionality (and related topics):
http://en.wikipedia.org/wiki/Authorial_intentionality
Though he might have meant other aspects of that debate.
Other articles that might help:
http://en.wikipedia.org/wiki/Literary_criticism http://en.wikipedia.org/wiki/Deconstruction http://en.wikipedia.org/wiki/Post-structuralism http://en.wikipedia.org/wiki/New_Criticism http://en.wikipedia.org/wiki/Literary_theory http://en.wikipedia.org/wiki/Reader-response_criticism
I could go on...
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.
On Dec 17, 2008, at 1:37 PM, Thomas Dalton wrote:
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.
Not to call Thomas out particularly here, but this attitude is much of the problem - WP:NOR reads like it was written by a bunch of people who were really hostile to anything to happen in literary criticism since World War II, and who thought it best to just ignore it all.
I suspect that, in fact, it was.
-Phil
2008/12/17 Phil Sandifer snowspinner@gmail.com:
On Dec 17, 2008, at 1:37 PM, Thomas Dalton wrote:
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.
Not to call Thomas out particularly here, but this attitude is much of the problem - WP:NOR reads like it was written by a bunch of people who were really hostile to anything to happen in literary criticism since World War II, and who thought it best to just ignore it all.
I suspect that, in fact, it was.
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...
On Dec 17, 2008, at 1:45 PM, Thomas Dalton wrote:
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...
Lit crit, however (or more broadly, English departments), is the general discipline under which writing and research skills are taught these days. It is the discipline that generally teaches courses on research, the nature of sources, etc. It is also the discipline out of which the question of what reading is, what one interprets from reading, and what a given passage of text "means." The PSTS section, and particularly the portion in the primary sources section that I am complaining about does make clear declarations about these things. And it's clear that they were written by people who haven't cracked open a book of scholarship on the underlying issues in recent memory.
-Phil
2008/12/17 Phil Sandifer snowspinner@gmail.com:
On Dec 17, 2008, at 1:45 PM, Thomas Dalton wrote:
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...
Lit crit, however (or more broadly, English departments), is the general discipline under which writing and research skills are taught these days.
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.
On Dec 17, 2008, at 2:05 PM, Thomas Dalton wrote:
2008/12/17 Phil Sandifer snowspinner@gmail.com:
On Dec 17, 2008, at 1:45 PM, Thomas Dalton wrote:
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...
Lit crit, however (or more broadly, English departments), is the general discipline under which writing and research skills are taught these days.
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.
Sure. And if we want to break NOR into individual field-specific guidelines, it changes. But a general principles of research and writing class? That's an English class in almost every University. I know, because I've taught it. And NOR is a general principles of research and writing policy.
I'd also say that an English professor knows a great deal about judging the credibility of a paper on Quantum Electrodynamics. They know the basics of the system of peer review and of academic credentials. They know what an academic journal is. They even know enough grammar and vocabulary to verify, in many cases, whether a given statement matches the one given in a reference.
Now, could I peer review an article on Quantum Electrodynamics? No. But I can judge the credibility of a published article on it. I'm even capable of preparing a bibliography on the subject - find articles that mention Quantum Electrodynamics, then classify them based on reputation of journal (not that hard to figure out as an outsider), frequency with which the article is cited, and degree to which the term appears in the article, and I can create a pretty good bibliography of essential sources on the subject.
The skills that allow that are taught in English departments.
-Phil
Sure. And if we want to break NOR into individual field-specific guidelines, it changes. But a general principles of research and writing class? That's an English class in almost every University. I know, because I've taught it. And NOR is a general principles of research and writing policy.
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).
On Dec 17, 2008, at 2:22 PM, Thomas Dalton wrote:
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 am not an expert on the UK university system, but if my hazy memory is correct, the UK system is generally not as invested in the notion of general education as the US system. Thus you don't get a general class on the subject period.
We get to another fundamental bias here, but I do think that there is a regard in which Wikipedia, by being invested in being a general resource on everything, is a bit more American in flavor than British.
In any case, my point remains - inasmuch as there is a general, multi- field approach to and belief in principles of research and scholarship, as it stands such approaches are more easily located in English departments or in fields that share a large amount of theoretical figures with English departments than elsewhere.
If nothing else, the list of stuff Carcaroth provided that you wrote off is a pretty good list of fundamental debates in the question of how to read sources and what they mean - debates that have ramifications in all fields. To declare them irrelevant to our process is... problematic.
-Phil
If nothing else, the list of stuff Carcaroth provided that you wrote off is a pretty good list of fundamental debates in the question of how to read sources and what they mean - debates that have ramifications in all fields. To declare them irrelevant to our process is... problematic.
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.
On Dec 17, 2008, at 3:29 PM, Thomas Dalton wrote:
If nothing else, the list of stuff Carcaroth provided that you wrote off is a pretty good list of fundamental debates in the question of how to read sources and what they mean - debates that have ramifications in all fields. To declare them irrelevant to our process is... problematic.
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.
The debate is in English.
There are appropriate uses of jargon, but the way lit crit uses it is just designed to make the author look clever.
That's flatly untrue. The jargon is there for a reason. Jargon free summaries lose vast amounts of the content.
When it becomes impossible to tell the difference between the real thing and satire, you know something has gone horribly wrong.
I don't know about you, but I have no problems along this line.
-Phil
2008/12/17 Phil Sandifer snowspinner@gmail.com:
On Dec 17, 2008, at 1:37 PM, Thomas Dalton wrote:
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.
Not to call Thomas out particularly here, but this attitude is much of the problem - WP:NOR reads like it was written by a bunch of people who were really hostile to anything to happen in literary criticism since World War II, and who thought it best to just ignore it all.
I suspect that, in fact, it was.
-Phil
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.
On Dec 17, 2008, at 1:48 PM, geni wrote:
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.
Usually, yes. The passage in question, however, is not pragmatic in the least.
-Phil
2008/12/17 Phil Sandifer snowspinner@gmail.com:
On Dec 17, 2008, at 1:48 PM, geni wrote:
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.
Usually, yes. The passage in question, however, is not pragmatic in the least.
-Phil
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.
On Dec 17, 2008, at 2:09 PM, geni wrote:
It's very pragmatic. People in general seem to want plot summaries
Pandering to popular opinion is a dodgy proposition. We regularly ignore what people in general seem to want on fiction articles.
It can also be rather hard to talk about a book/film/legend without one. They also appear to be expected of encyclopedia articles.
This is largely our invention - Britannica has relatively few plot summaries. Now, mind you, I agree we should include them, but we should recognize that they are not the traditional focus of encyclopedic coverage of fiction.
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.
Sure. The issue is, there are other things that are as obvious and easy to do as plot summaries, that are vital parts of the traditional conception of how to encyclopedically describe literature, that we do not allow, and in fact explicitly forbid for reasons that are not the pragmatic "the customer demands it" argument you present, but rather because they are "interpretation" and not "description."
I'd prefer a pragmatic approach comparable to what we do to solve the "we don't want citations for obvious statements" problem on WP:V - a hedge like "material challenged or likely to be challenged." That's a pragmatic approach.
What we have now is not.
-Phil
2008/12/17 Phil Sandifer snowspinner@gmail.com:
On Dec 17, 2008, at 2:09 PM, geni wrote:
It's very pragmatic. People in general seem to want plot summaries
Pandering to popular opinion is a dodgy proposition. We regularly ignore what people in general seem to want on fiction articles. This is largely our invention - Britannica has relatively few plot summaries. Now, mind you, I agree we should include them, but we should recognize that they are not the traditional focus of encyclopedic coverage of fiction.
Depends if you are considering a general encyclopedia or a specialist encyclopedia.
Sure. The issue is, there are other things that are as obvious and easy to do as plot summaries, that are vital parts of the traditional conception of how to encyclopedically describe literature, that we do not allow, and in fact explicitly forbid for reasons that are not the pragmatic "the customer demands it" argument you present, but rather because they are "interpretation" and not "description."
Strawman.
I'd prefer a pragmatic approach comparable to what we do to solve the "we don't want citations for obvious statements" problem on WP:V - a hedge like "material challenged or likely to be challenged." That's a pragmatic approach.
What we have now is not.
Blanket denial without at least a line of reasoning behind it is not a helpful approach to debate.
On Dec 17, 2008, at 2:55 PM, geni wrote:
Strawman.
Unhelpfully reductionist response that doesn't actually explain itself and so is worthless.
Blanket denial without at least a line of reasoning behind it is not a helpful approach to debate.
I don't even understand what you're saying here.
-Phil
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@gmail.com To: "English Wikipedia" wikien-l@lists.wikimedia.org Sent: Wednesday, December 17, 2008 10:30 AM Subject: [WikiEN-l] The Community vs. Scholarly Consensus
Picking up on a thread from the anti-intellectualism thread, WP:NOR currently reads " Any interpretation of primary source material requires a reliable secondary source for that interpretation. Without a secondary source, a primary source may be used only to make descriptive claims, the accuracy of which is easily verifiable by any reasonable, educated person without specialist knowledge. For example, an article about a novel may cite passages from the novel to describe the plot, but any interpretation of those passages needs a secondary source."
At a number of points, this steps squarely into a decades-long debate in literary studies about the nature of reading and of interpretation. This is a debate that is still - perhaps permanently -unsettled. However the view Wikipedia is taking - that there is some core of knowledge that is "descriptive" as opposed to "interpretive" - is decades out of the realm of accepted. It's a discredited view.
The counter, which I've seen discussing this on the NOR page, is that the language represents community consensus, and that's good enough.
This is troubling to me - most significantly because it suggests that the community is empowered to set an official Wikipedia position on the nature of language and meaning, and to settle a decades long dispute. On the one hand, this seems to me a problem of NPOV - but it's also an intractible one, as we can't avoid having some policy on the nature of reading.
But on the other hand, surely we do not actually intend to empower the community (by which we really mean the people who wrote NOR) to rule on issues like this and ignore a huge scholarly debate.
I don't have a good answer here, so I figured I'd ask the list - to what extent is the community empowered to set a Wikipedia policy on a scholarly debate? How do we square this with NPOV? Is there a clever out to the underlying problem in NOR that sidesteps the debate?
Thoughts?
-Phil
WikiEN-l mailing list WikiEN-l@lists.wikimedia.org To unsubscribe from this mailing list, visit: https://lists.wikimedia.org/mailman/listinfo/wikien-l
2008/12/17 Phil Sandifer snowspinner@gmail.com:
Picking up on a thread from the anti-intellectualism thread, WP:NOR currently reads " Any interpretation of primary source material requires a reliable secondary source for that interpretation. Without a secondary source, a primary source may be used only to make descriptive claims, the accuracy of which is easily verifiable by any reasonable, educated person without specialist knowledge. For example, an article about a novel may cite passages from the novel to describe the plot, but any interpretation of those passages needs a secondary source."
At a number of points, this steps squarely into a decades-long debate in literary studies about the nature of reading and of interpretation. This is a debate that is still - perhaps permanently -unsettled. However the view Wikipedia is taking - that there is some core of knowledge that is "descriptive" as opposed to "interpretive" - is decades out of the realm of accepted. It's a discredited view.
The counter, which I've seen discussing this on the NOR page, is that the language represents community consensus, and that's good enough.
This is troubling to me - most significantly because it suggests that the community is empowered to set an official Wikipedia position on the nature of language and meaning, and to settle a decades long dispute. On the one hand, this seems to me a problem of NPOV - but it's also an intractible one, as we can't avoid having some policy on the nature of reading.
But on the other hand, surely we do not actually intend to empower the community (by which we really mean the people who wrote NOR) to rule on issues like this and ignore a huge scholarly debate.
I don't have a good answer here, so I figured I'd ask the list - to what extent is the community empowered to set a Wikipedia policy on a scholarly debate? How do we square this with NPOV? Is there a clever out to the underlying problem in NOR that sidesteps the debate?
Thoughts?
Wikipedia policy is written with a certain degree of pragmatism. Finding secondary sources for descriptive plot mummeries tends to be rather hard and since wikipedians can generally agree on them there seems to be little point in preventing them from doing so. Yes students of literary criticism can argue for endlessly over what descriptive means but most wikipedians are not students of literary criticism so the problem doesn't appear.
Sure you can start arguing over the validity of this approach but those kind of conflicts have been done so much better by others.
On Wed, Dec 17, 2008 at 9:30 AM, Phil Sandifer snowspinner@gmail.com wrote:
Thoughts?
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
On Fri, Dec 12, 2008 at 09:08:55PM +0000, David Gerard wrote:
Yeah. The common usage of interwiki links on en:wp is "that article over there is roughly equivalent to this one." I have seen en:wp articles pointing to two de:wp articles, quite reasonably.
I disagree with the "quite reasonably" part. Interlanguage links ought to represent a closer connection than that. If one enwiki article "corresponds" to two dewiki articles, then neither of the dewiki articles should be links from the enwiki article. Perhaps a disambiguation page on dewiki could be linked from the enwiki page.
- Carl
From: "Ian Woollard" ian.woollard@gmail.com (...)
Interlanguage links are equivalence relations (at best.)
(...) 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]