On Wed, 2004-24-11 at 08:51 -0700, Mark Williamson wrote: I think the big thing to understand is that each wiki takes time, effort, and money to maintain. Each wiki is a security risk, both technically and legally; an active community can offset this risk as well or better than fancy technological security measures. An analogy: leaving the door to your house unlocked when you're having a backyard barbecue is the right thing to do. There's lots of people around, and they'll need to get in and out, and there's little risk of problems. It'd be really inconvenient to have to keep locking and unlocking it. Leaving the door unlocked when you are the only one home is probably OK. Leaving your door unlocked when no one is home is asking for trouble.* A wiki without an active user community is an unlocked house with no one home. You've repeatedly pointed out that you don't want to spend the time, money, or effort to create and maintain wikis without any active participants, yet you insist that the Wikimedia Foundation does so. You haven't really made a case for why, except that these languages exist and have speakers. I don't think anyone debates those points; I also don't think that "a language exists" has "there should be a Wikimedia wiki for it" as a necessary consequent. Even if the disadvantages are low, the advantages of having empty, unused wikis don't seem to outweigh them. I find it hard to believe that, say, gv: is a huge source of pride for Manx speakers. If that was my native language, I'd think to myself, "Man, nobody really gives two shakes about us, do they?" And the idea that any unused Wikimedia wiki is a scholarly resource for the language is absurd. I guess there's a intellectual exercise in reading the list of ISO 639 codes, seeing which ones don't already have a Wikipedia, looking up the language, and then making a request on this list, but... I don't think that gratification is enough to offset the disadvantages. I'd say that having a clear set of rules about how and when to start a Wikimedia wiki would obviate the need for a Board vote on each one. Like, I dunno, say: having a single person step forward willing to work on the wiki in the language, and maybe having one edit per 30 or 60 days. That seems like a pretty low threshold to me. ~ESP * In some cities and countries. In other places, it's perfectly OK to leave your door unlocked when you're not home. -- Evan Prodromou <evan(a)wikitravel.org>

All the different Wikipedias have things like this that others would not find NPOV. This comes from the fact that any given language will have a different proportion of people with any given point of view than any other language, and as you may have observed, people often see reality through POV-coloured glasses, even if they try their best to be NPOV. For example, given recent disputes involving the Balkan area on en.wikipedia, I think there is a very good chance that, for example, the sr.wikipedia equivalents of some of these hotly-disputed articles (as well as those at any other areal language Wikipedia) may be tainted with a local POV. I'm sure people such as Unitifler from the Tatar Wikipedia have known for quite awhile about NPOV policy, but they did not see what they did as a problem, just as fr.wikipedia now has holiday themes for diverse holidays but I still object to it on the grounds that I believe that any such display is POV (but really, that issue is not important to me, especially if nobody at fr.wikipedia cares, and also since while I can read French to a certain degree, my written French is at best the laughingstock of all French speakers who read it). The problem here stems from people thinking something is or is not POV when really there is no absolute standard by which we can definitively decide for all cases without some degree of judgement on the personal level. If we really think the previously-noted demographics issue is such a problem, perhaps we should attempt to foster article synchronisation between Wikipedias and less fluidity in the changing of articles, so that if somebody sees something wrong they can make an objection and it can be discussed first. However, I do not think this is such a problem as eventually all the sharp corners will be shaved off in some way or another, at least I believe as much. So I think there is a good chance that there are a few articles on de.wikipedia which have community consensus to being NPOV that I personally would find tainted with POV, or that might raise objections in some other-language Wikipedia. Ultimately this stems from having different Wikipedias for different languages, even from different content for different languages. However, at present I see no reasonable workable solution, and since we do not have communities agreeing that an article with the text "'''Wisians''' are a very mean peoples, from Wisia. They like to lie and steal and cheat, and they all deserve to die." is NPOV, I don't think it's big enough that me /must/ take action immediately. Mark

Henry H. Tan-Tenn wrote: Not a single Wikipedia is NPOV. At first, all support the creation of a free, open encyclopædia. I'm sure there are many other systemic biases as well. I don't think it is such a terrible disaster if some small language Wikipedia has NPOV on the frontpage. If the alternative is locking, I would support the former. If a user comes across the wiki, knowing the language and seeing that there is a bit of content, he/she may become enthousiastic and edit the content. However, if it's locked, he/she may become disappointed at the whole concept of a wiki, discriminating small languages, and "quod licet Iovi, non licet bovi". The major languages have wiki's, the small ones have not. Let small languages be. :-) regards, Gerrit Holl. -- Weather in Lulea / Kallax, Sweden 26/11 10:50: -7.0°C wind 6.3 m/s NW (34 m above NAP) -- In the councils of government, we must guard against the acquisition of unwarranted influence, whether sought or unsought, by the military-industrial complex. The potential for the disastrous rise of misplaced power exists and will persist. -Dwight David Eisenhower, January 17, 1961

But, as noted by Henry, this is difficult even for established Wikipedias and community standards vary depending on the values of the community in question. Thus, what is NPOV to the Basque Wikipedia might seem very POV to the Chinese Wikipedia. Mark On Thu, 25 Nov 2004 08:48:26 -0500, Evan Prodromou <evan(a)bad.dynu.ca> wrote:

Mark Williamson wrote: I think this is the single biggest problem with minority-language Wikipedias ever being useful NPOV encyclopedias. If all speakers of a language are relatively homogenous, in terms of location, culture, and values, then their encyclopedia will with high probability by quite biased. By contrast, wikipedias like en:, de:, and fr: have people from multiple continents and backgrounds contributing. An "American English" or "Australian English" Wikipedia would, I would hazard a prediction, be much inferior in terms of neutrality to the current English Wikipedia, simply as a result of the fact that all Americans (or Australians) editing a Wikipedia would represent a much narrower set of viewpoints than the current large and diverse group of editors. I'm not sure this is surmountable for languages in which a large and diverse group of language speakers don't even exist. Note that not only very tiny languages are affected by this: I imagine that the Turkish and Greek Wikipedias are less likely to present a reasonable account of the Turkish-Greek conflict than en:, fr:, or de: are. -Mark

It may seem to you that en.wikipedia is totally devoid of this sort of systemic bias, but in fact no Wikipedia is. While one the one hand English speakers come from diverse environments and have diverse cultural experiences, the majority of en.wikipedians are American (from the USA), and after that are British, Australians, and Canadians. While this may not be a single nation or a single culture, the collective cultural experience of the populations of each of those countries are very similar, especially when compared to that of, say, Laos or the Maldives. But even the cultures of Laos, the Maldives, and the aforementioned Anglophone nations seem very similar when compared with the cultures of the Navajo, Apache, and Hopi (which are also very distinct from one another, to the same degree that American culture is different from Laotian culture), part of the reason being that the former 6 cultures are patriarchal and patrilocal (the patrilocal part I am not sure of for Laos and the Maldives, in the Anglophone countries it is, nowadays, the norm for most people to establish their own home but traditionally they are patrilocal), while the latter three are matrilocal and matriarchal. If you ask a Laotian, Maldivian, American, Brit, Australian, or Canadian who he or she is, after you get a first name, if you say "Yes, but /who/ are you?" chances are the reply after that will have something to do with a name inherited from the father and his father in turn and so on (thus they are patronymic, which seems to be the convention for all UN memberstates, but I am not sure), while if you ask a Navajo, Apache, or Hopi you will get a response which includes a name inherited maternally for generations untold (thus they are "matronymic"; I believe Tibetan culture is matriarchal but I am not sure, I do know though that it does allow polyandry in some situations which is fairly rare among world cultures and usually only occurs in matriarchal societies, just as polygyny rarely occurs in matriarchal societies). In addition to this most fundamental cultural experiential difference, there are plenty more differences which mean that ANY Wikipedia will have some form of systemic bias. In en.wikipedia, many articles about strictly American concepts do not note that fact, and many articles - especially those from 1911 EB - speak negatively of taboo subjects such as incest, cannibalism, pederasty, and other things which should be dealt with neutrally. For example, before I made edits to it, the Ainu article was very racist and offensive in that it said the Ainu were "primitive" - human cultures are not primitive if they achieve the requirements a culture must achieve: telling people where they came from, explaining natural processes (some cultures, more recently, have replaced this with an *independent* concept of religion, for example French culture has no single explanation for the origin of the world and individual natural processes: some people believe the world was created by a single god, others may believe it was created by gnomes, others believe in more complex modern western scientific theory, some may believe it was sculpted out of clay) - it talked about how hairy they are; note that not only is this a stereotype, it is from a British point of view: the Ainu of the time might've described the authors of that article as bald or almost completely bare, or lacking of any significant amount of body hair, etc. while they may have actually been what I personally would consider to be relatively hairy. Little words like "unfortunately", "fortunately", "tragically", "interestingly", "peculiarly", "strangely" and short phrases like "one strange thing about ___ is that it ___" are one of the largest sources of systemic bias in en.wikipedia. If we are describing a language and we say one feature is peculiar and interesting, people for whom that is the native language will beg to differ because to them it is completely ordinary. We might say it is fortunate that nobody was on a bridge when it collapsed, but another culture may believe that people who are standing on a bridge when it collapsed fall not to their death but to an island paradise. We might say somebody's death is tragic, but some cultures may see it as a blessing (some cultures view death positively). And if you find yourself wanting to respond to any of this with "but ______ IS _____" (for example, "death IS tragic!"), that just shows the extent of the problem. So even experienced, well-respected Wikipedians may let through their POV filters things that do not strike them as POV, because of their collective cultural experiences and because of their own POV. Thus this is not just a problem for minority language Wikipedias, it is a problem for ALL Wikipedias which cannot be completely solved in the near future (though we can begin to combat it by for example searching for all occurances of the words noted above and remove them, as I began to do [removed about 20 occurances of "fortunately" and "unfortunately"] but it is very repetitive and I had to constantly go forward and backward between Google and Wikipedia, though this does not mean I won't do it in the future) -Mark (just to make things more confusing for you all) On Fri, 26 Nov 2004 21:50:24 -0500, Delirium <delirium(a)hackish.org> wrote:

On Fri, 26 Nov 2004 23:03:56 -0500, Stirling Newberry &lt;stirling.newberry(a)xigenics.net&gt; wrote: And thus, the fact that as a general rule more pages at en.wikipedia are viewed and edited by Americans than anybody else, and by a much larger margin by more people from Anglophone countries than anywhere else, is systemic bias. As I noted, a consensus version will often retain systemic bias. And as I also noted in this e-mail, there is currently no viable solution to this problem, and thus we must live with this systemic bias to a certain point until international cooperation becomes more possible, either through loss of linguistic diversity (which I hope isn't how it ends up) or through improved machine translation techniques, or through both (If there are only two languages spoken on Earth, all people who work on machine translation engines will obviously be working towards the same goal, Lang1<->Lang2 translation, rather than many different goals and thus the rate at which the quality of translation improves would be significantly higher). Consensus versions only help to remove the personal biases of the author, not the collective bias shared by all those who participate in this process. And what might that work be? I don't work on machine translation engines, and as a general rule I do not try to proactively speed loss of linguistic diversity (if I wanted to, I don't think I could bring myself to do it). One of the main purposes of this e-mail was to note that when people say that "...it may be difficult to build a quality NPOV encyclopedic resource in a language with such a small population..." or similar things, or when they reveal oh-so-scandalous occurances of POV on Wikipedias other than their own and imply it couldn't happen on their own Wikipedia, they are in fact not entirely correct because all Wikipedias contain systemic bias. Mark

I explained why it is correct. In addition, I said "systemic" and not "systematic". When you tell somebody they're still using it incorrectly even though they attempted to validate it on the principles you yourself stated, it'd be nice if you could tell them why it's still incorrect. Mark On Fri, 26 Nov 2004 23:55:20 -0500, Stirling Newberry &lt;stirling.newberry(a)xigenics.net&gt; wrote:

I just joined this list less than an hour ago (and I'm very new to WP too), but I thought I'd drop in my 2cents on the msg by Mark Williamson. (Yes, I was being UScentric; no, I'm not in the US; yes, I was being intentionally ironic.) Really, every piece of information, every datum, every idea, everything is biased to what we (or the authors) know. When we find that there are new planets with intelligent life out there, we will discover that everything we wrote is Earth-centric; when we discover new parallel universes or time-travel, that it was universe- or time-ist. What about WP being NPOV- or truth-centric? Doesn't the imaginary world in my head have a right to be documented to? For these reasons, I think using this as an argument for saying that WP is biased is an invalid argument due to its absurdity; and misunderstands that bias is a sliding scale, so the semantics of the word, unbiased, always imply a certain amount of what you would call bias (in the same way that the term, biased, does not imply that something has total bias and is meaningless as this is also impossible). WP is IMO the least biased document that it is possible to create because it is based on a free self-selecting meritocratic democracy. A truly unbiased encyclopedia would be infinite (and contain totally random content--although true randomness implies infinity and vice versa) and contain everything thereby deeming it as useful as an empty encyclopedia; though content is king--conciseness can be more important than sheer quantitiy. A finite encyclopedia (or other piece of information) can never be truly unbiased. So what? It would be absurd (and impossible) to make it so. Mark's last post was suggesting any changes to Wikipedia to solve the problem, because it can never be solved really, so let's all not worry about it. The only possible way to reduce bias is to try and increase the number and range of editors on WP (and to some extent by increasing the content). Also, your points about systematic bias are kind of moot, because in order to say there is systematic bias in something due to bias in a set of individuals, A, there must be a larger set, B, which contains A, that you are comparing A with. That set, B, is part of a larger set, C, which means that set B is biased wrt C, therefore when you talk about systematic bias, your idea of systematic bias is itself biased. Just thought I'd join in and make that point. Hopefully it makes some sense as I've been on this machine for 13 hours and havent slept for several days (and I don't know anything about bias or stats). Happy wikipediaing, Joe Ll. G. Blakesley. -- Get FIREFOX: the brower you can trust <http://www.spreadfirefox.com/?q=affiliates&amp;id=20750&amp;t=1"> -- Sign the petition against the FBI's illegal seizure of Indymedia's London-based WWW servers. Protect freedom of expression in the UK. <http://solidarity.indymedia.org.uk/> -- This mail sent through http://webmail.bangor.ac.uk

I personally believe it is a problem, but that it cannot be solved at the moment and that Wikipedia may seem a bit "sterile" without it. However, I think we *do* need to be a bit more careful about things which can be seen if you check carefully, such as "unfortunately"s and "interestingly"s. Mark

Mark Williamson wrote: This was not my argument---my argument was that systematic bias can be minimized on Wikipedias which are in languages for which a diverse body of speakers exists, but is unlikely to be minimized on Wikipedias in a language spoken by only a small and homogenous group of people. It may not be eliminated on any, but it will be much less severe in languages with more diverse speaker populations. The current en Wikipedia does not have solely contributors from English-speaking countries either---there are a large number of English-as-a-second-language contributors who contribute valuable information and viewpoints. Our Beijing article has contributors from Beijing; our Gdansk article has contributors from Gdansk; and so on. This is very unlikely to happen in languages which are rarely learned as second languages. -Mark

"You've repeatedly pointed out that you don't want to spend the time, money, or effort to create and maintain wikis without any active participants, yet you insist that the Wikimedia Foundation does so. You haven't really made a case for why, except that these languages exist and have speakers. I don't think anyone debates those points; I also don't think that "a language exists" has "there should be a Wikimedia wiki for it" as a necessary consequent." How much money does it cost to create a new Wiki? Technical efforts aside (and Tim Starling seems entirely willing to give technical assistance to Wikipedias for living natural languages), there is very little wrong. You are exaggerating this very much, and perhaps you should also check http://bloglines.com/public/inactivewikipedias . Your allegations that I don't want to spend the time on these things are pure crap because I already spend time on them - for the last few days I have closely been monitoring the activity on every single one of these Wikipedias, and I don't mind it one bit. If somebody does something bad, I can revert it; if it becomes a problem I can't handle, I can bring it up with a developer or steward or whomever it would be appropriate to bring it up with. As to being upset that nobody cares about your language, from what I have learned the usual reaction is quite the opposite - in the time I have been monitoring inactive Wikipedias, I have seen ne.wikipedia become active, in a good way, and although I am not sure it is permanent, I think it's good. Also, take for example ka.wikipedia. The sysops there now, Malafaya and Sopho, found it through a search engine and when they saw it had 0 content, they decided to remedy that. And what on earth are you talking about, a "scholarly resource"? When did I argue that inactive Wikipedias were scholarly resources? The only time I ever brought up anything of the sort was in regards to the Gothic Wikipedia (it looks dismal now, but I have after-holidays commitments from a few people), in which case I did not say "Oh golly gee, I know this Wikipedia will have two articles and 0 users, but somehow I think it'd be a scholarly resource anyways", what I said was basically that, if it accumulated a large number of articles, it would be a scholarly resource of sorts because it would be such a large corpus. Note the "large corpus" and "large number of articles". I think that your views are extremely Draconian and such extreme measures are not needed at this point, the benifits are not outweighed by the risks with the current solution. I think that - and I am serious here - you have a real problem with minority languages. Mark On Wed, 24 Nov 2004 12:26:20 -0500, Evan Prodromou &lt;evan(a)wikitravel.org&gt; wrote:

Except that, you know, everyone who has ever supported cutting down on small language Wikipedias has noted that they would never object to a Wikipedia like German, Japanese, or another major world language that's used for business, that the great works of civilization have been written in, and that has several million people who use it as their primary language. The idea that these policies would restrict any of those Wikipedias is a straw man, plain and simple. To my knowledge, there is no major language Wikipedia we currently lack. There are no major language Wikipedias in danger of being locked. The question only applies to Wikipedias that are unlikely to have more than a handful of users in the next five years. The Japanese Wikipedia always had an obvious and large population to draw from. But Limburgish? Not so much. Personally, I increasingly advocate a moratorium on new language Wikipedias. Instead, we should start one "small language" Wikipedia to be run like Wikibooks, with various subsections. So articles would be in the form of [[Limburgish//Article title]]. And, should one of the small languages suddenly take off to have a real and sustainable community on it, it could be moved to being its own independent Wikipedia. But I can see no plausible reason to waste limited developer resources on creation and support of projects that will have no lasting userbase except for occasional surges when someone threatens to take them down. And since no one is actually suggesting any of the strawmen you're bringing up, I have to ask, with all due respect, that you stop trolling this listserv. -Snowspinner On Nov 25, 2004, at 12:29 PM, Mark Williamson wrote:

On Nov 26, 2004, at 11:36 AM, Mark Williamson wrote: For the same reason you hate arguing without resorting to strawmen.

Mark Williamson wrote: One point that you seemingly fail to comprehend is that by advocating for all and sundry minor languages you become a part of the problem. Analogously we can't have a meaningful SETI project until we can filter out the radio noise from earth. The problem with squatting has its roots in the fact that most of us don't understand these languages, and don't monitor events there. If there is any silliness going on we have no way of knowing. It makes sense to put these languages into cryogenic support if it has been inactive for an extended period of time. I think that there is very little support for that position. Ec

But why do we need to change anything when the current system works fine? Mark On Wed, 24 Nov 2004 09:58:24 -0800, Ray Saintonge &lt;saintonge(a)telus.net&gt; wrote:

I agree about conlangs, but other than that I think no policy change is needed. This is not to say the Board agrees with me. But, really, "if it ain't broke, don't fix it." Mark On Thu, 25 Nov 2004 00:54:01 +0000, Angela &lt;beesley(a)gmail.com&gt; wrote: