What was rejected was Wikipedia in 'Traditional Greek', which encompasses
(at least) three different periods of the Greek language, including the
Ancient one, and had no ISO code.  It also had only one supporter.
Ancient Greek, however, has a lot of supporters, and the language is unified
I really thought I was starting to have acid flashbacks, stuff
like remembering I spoke with Zeus on the Mount Olympus yesterday :-) Now
it’s okay. It’s an important datum for me because the decision about
traditional Greek is one of those I use as “precedents” in our decisional
history (the archetype of “too little users”), so if that moves I happen to
base my precedents on my senile delirium only :-)
I propose denying and closing the following requests:
- Belorussian Wikisource: let alone the small wikisize, there is too much conflict from both belorussian sides. Until they do not start to have normal relations I'd freeze them both for good. It is NOT a linguistic argument but rather a disciplinary intervention. It can be done only if we all share the idea that we can put constant pressure on conflict situations in order to force them to serious and public pacificatory moves. A less open way consists in never answering their request, but I’d rate this really unfair. People have the right to know what we think of them, if we do think something.
- Insubric Wikipedia: they already have all the space they want in insubric at LMO wiki. LMO has 5-6 constantly active members, plus a number of occasional editors. No ISO code for insubric. I see no point in letting them there as an open request.
Add-on : one of the arguments for insubric was (quite paradoxically) that lmo.wiki has... too much insubric (the interface is) so it would be better to divide orobic and insubric. It is weird that the argument comes from insubric speakers and it says a lot on the nature of the request. Anyway, this much is true. Could we possibly have an alternate user interface in orobic issued for lmo.wiki? It would be possible to choose it at user level, while the wiki would traditionally open in insubric.
Well, I’d say that wikisource is the one case in which we do not use codes as fences, but simply as descriptive labels. I quite agree with Yann in what he writes about the project. Only, when seen in this perspective one thing remains unclear: why to have national wikisources at all, instead of one big “commons” wikisource with selectable linguistic tags in the search field. None of our business, obviously, I’m just speaking for the sake of logics.
I propose the immediate rejection of Värmlandic Wikipedia; there are no
sources given indicating that this is a separate language from Swedish, and
I have never heard talk of it either. It has no ISO 639-3 code, and the
example articles in the Incubator use some sort of phonetic spelling.
It’s about how contact (transitional, according to other classificatory
systems) speeches get mapped into ISO 639-6. A “transitional speech” is what
arises along the contact line of two (or more) major speeches. In my native
language map we have one such case with Orbasch, that is spoken all along
the borderline between Ligurian (LLJ) and Piemontese (PMS). It’s got
elements of both speeches but it cannot be reduced to none of the two. Don’t
look for it in ISO, you won’t find it.
It’s often easy to classify such entities as individual languages, but if
you have to build a hierarchic classificatory system it’s not simple to deal
with them. That’s why I was asking whether ISO 639-6 admitted multiple
inheritance (i.e. more than one parent in the tree). This is also
interesting when you have to classify creole languages.
As per English I’d say it’s more like a stratified job depending on
invasions/migrations, as per Walter Scott’s works. We have something similar
in Piemontese, with the periodic shift of the “high language” towards either
Tuscan (Italian) or French, depending on which nation we were part of (we
twice were annexed by France). I suppose Savoy and FRP have a similar
history. Although there is no such thing as a 100% “contact” or “stratified”
language (all border regions show both behaviors, for obvious historical
reasons), contact behavior is brought about mostly by spontaneous activity,
like trade, passing voyagers, etc.
Following this convention all current “anglified” versions of the modern
languages are showing a “contact” behavior. Just a DB-designer curiosity,
I proceeded to reject the request for insubric as there was no opposition,
but I'm probably not aware of a number of steps in the procedure. What do I
have to do apart from placing the rejection template? It seems that requests
are also to be moved, are they?
Hello all. I propose the following approvals and rejections.
===Upper Sorbian Wiktionary===
I propose the conditional approval of this project. There is an interested
community, the test project is active, and there are even a few translated
localization files already available (thanks to the Upper Sorbian
Wikipedia). All that remains is to finish translating the rest of the
I propose this request be rejected. It has been open for some time now
(October from the looks of it), and it has received only 4 simple support
votes. The supporters have not offered any arguments as to why this request
should be approved; they are always welcome to open a new request, with
LOL I had a look at it, some words are really close to ours, although they have a different graphic convention… I suppose I can fake some hillbilly Cajun by moving some piemontese vowels once I get the idea of how it should sound like :-) They should add audio recordings for us to learn the sounds of it :-) I’ll write an article about it on pms, faraway cousins are always interesting to general public :-)
I can’t see how we can pass this thing without violating the policies. Would a request to open a Hanja namespace within Korean wiki do? BTW, it could serve as a template for other such cases. Thinking of Belarusians but not only, even pms DOES have cases of local orthographic variations that sometimes originate small edit wars. I proposed using separated namespaces for alternate versions of a single article and the pms community loved the idea. I guess it’s because I entitled it as “let everyone pray as they wish”, which is the famous (for us) phrase that marked the end of our 500 year long religious war.
Yet I do believe that it could be used in a number of cases to have formal fences without making separate communities. Only… how do you add a namespace?
Fine for me, I volunteer PMS wiki for a test case, if Erik thinks it can be dangerous and he needs a guinea pig to start with. We can have a public vote about it in days, so it can be regular from all formal POVs, but since we have been talking exactly about this kind of problems and we are all looking for a solution I can tell you it’s a big YES immediately.
I would grant MLMW to Hanja, too. It would mean no exceptions to the policy and yes for them. BTW, it can also solve the Latgalian problem, if we cannot find any better way out of it. And it can be the final cut for Belarusians, too. Tell Erik I love him :-) As we say in Piedmont “He’s taking the chestnuts out of the fireplace” :-)
So, if nobody opposes the idea I’ll tell Hanja that we can conditionally approve their project with the condition that they can be released only as an additional linguistic entity to the Korean database. Hmmm… On a second thought, I smell problems here… they will need a separate admin structure for Hanja, as they say most Koreans cannot read Hanja anymore they cannot even be expected to be able to administrate such pages.
May MLMW allow separate admin structures? It would also do VERY good for Belarusians, while it’s irrelevant for PMS … In the meantime I’ll tell Hanja people that we are working on a technical solution that involves using new software, without entering in details.
This request seems like a tough one. The multilanguage feature sounds
like it would be perfect, but I can understand that they are getting
annoyed (and bored) waiting for it. It seems unfair to me too, to make
them wait god knows how long for that feature.
Sabine mentioned it being "only" a script question, but looking at the
request it is not so simple. It seems that most Korean Wikipedians are
opposed to automatic conversion. Secondly, and more importantly, the
request also states that the vocabulary and orthography are quite
I really don't know enough about it at the moment to comment. I'm not
sure what we've decided, or if we have even decided anything, but the
people who made the request in the first place want an answer. I think
they'd appreciate it if we had *something* to tell them.
in some way you are right - but there is that but ... that is: if we
create exceptions we will get into trouble since anybody will ask us to
So we need a solution and the only one I can think of now is that they
or wait for Multilingual Mediawiki or they agree with ko.wikipedia in
some way so that both ways of writing can co-exist on one wikipedia ...
Automatic conversion seems not to be an option ... so manual conversion
should do the trick then, right? But how to get things there without
creating too many hassles and long discussions?
Nope, no exceptions. They
wait for MLMW, as stated many times. BUT… it would be nice to find a way to
encourage them. Like a barnstar or something, and tell them that the project is
already good for approval, and will be as soon as MLMW is useable. BTW, I’m
installing a copy of MLMW on our servers, tonight, so I will soon be able to
give more details.
The project is okay and
deserves space, it’s the idea of an exception that is not good. This is a
problem we have in many contexts and it must have a regular solution, not just
a quick patch.
Jesse Plamondon-Willard (Pathoschild) 14 January 2008 13:23
Just to clarify for the archives, GerardM and Berto are referring to a
dialect (with no ISO code of its own) separating from a wiki (with an
ISO code) that uses another or multiple orthographies. Such a project
fails the policy requirements <
http://meta.wikimedia.org/wiki/WM:LPP#Requisites >, but the
Multilingual MediaWiki (MLMW) would allow them to coexist on the
Indeed, it cannot have a wiki on its own. This is not even another language or dialect, but mostly another script (although formal expressions may also differ). Hanja is based on Chinese characters and it used to be the Literary Language in Korea, being taught as such in schools.
In time, the role of the script in school teaching has diminished, to the point that nowadays a vast majority of Koreans aren't able to read it. This leads to the current situation. A large number of admins will simply zap content they cannot read, although formally "legal" in Korean.
I don't think imposing a rule that would make such content "good" on top of the community would make good internal relations. It's much better if they both remain inside a single ISO code and wiki (as per policy), but get given an IT that allows them putting things into separate linguistic namespaces.
This way people do not get to clash for practical problems (that are quite understandably annoying in nature) while allowing everyone to use the communication style they like better. This will never solve situations in which the clash is politically based, yet it can IMHO do wonders for those cases in which frictions come out of practical life only.
The preconditions to use such a system are:
1) alternate flavors of a code (be it scripts, dialects or whatever) apply for an obtain a valid IANA code
2) MultiLanguageMediawiki is made available to them
In perspective, this can be used to finally reunite spawned communities like BE, and it can solve an impressive quantity of trouble for macrolanguages (or for those languages that while not being tagged as such by ISO still use a number of alternate orthographies).
If from one hand we try and impose a correct cataloguing system, we shall not forget that people have to live with it, so it's nice if we can offer a system that allows everyone to "be themselves".
The Turkish wiki is doing not bad: http://stats.wikimedia.org/EN/TablesWikipediaTR.htm and the language has a great historical relevance. While I quite agree with some of the opponents that maybe a Wikisource project would best fit this kind of languages having a standard wiki going would help forming a community that could later successfully run a Wikisource. Besides, deciding what exactly they should want from life is none of our business. Let’s see what they can do in the incubator.
You know what? I’m getting to think exactly like you. I wrote the proposal before seeing how many revival requests we have. It really makes more sense to include them as spaces inside the cultures they belong into, unless they cannot be included, like latin.
I was acting based on Ancient Greek, and now I’m sorry we conditionally approved it, because it will put ourselves in contradiction with a precedent decision. Yet we did not know about MLMW when we did, so I guess we can simply reroute it into gr.wiki once it’s ready to leave the incubator.
It would be nice to publish a page about historical linguistic entities and guidelines for them, but we need to have tech details about MLMW before doing it.
Ancient Greek is a completely different scenario than Ottoman Turkish.
Ancient Greek is a classical language, while Ottoman Turkish is "simply" a
historical language. The difference is subtle, but quite big.
Yes, I thought about after posting. It’s actually thought all over the
planet. The reason why I always hate this kind of exceptions is that they
imply a judgment in value. Most of the fights we manage our due to
nouns/adjectives/rules that imply a value statement. Besides
consider it “classical” if it was not a classical language for us
Well… can’t speak even basic German, so I’m at a loss in judging. I suppose seeing what they can do in the incubator can be an answer. When ISO codes (like LMO) are fragmented into a huge number of dialects because of historical reasons the situation may become explosive because of personal conflicts, so quite a lot depends on how they can manage their internal diversity.
I’d say that it would be much better if they did NOT speak the same variety, so we can have the bomb explode immediately, if a bomb is there. Apart from that… I find it difficult to evaluate the chances of success based on the number of actual speakers. Germany has a lot of internet connections, so the density of possible users for this wiki is probably much higher than the density for a multimillion speaker project in Ghana.
Besides… to remain in Italy: fur.wiki is the best of us all with just 750k native speakers, while a double number of speakers make only a semi-dead lij.wiki. The script they choose and their capability to market the project will mean a lot.
Hmmm considering that it is part of this region:
http://de.wikipedia.org/wiki/Rheinland-Pfalz and some parts are in the
federal states near it (original terriroty of the Pfalz was:
http://de.wikipedia.org/wiki/Kurpfalz) I would estimate approx.
1.000.000 speakers, but cannot be sure about it. The situation should be
simiar to Kölsch, that is many "flowing" variations from one place to
the other ... well also spoken Neapolitan is a bit like this. I am
wondering about the written language ... ah just found this page:
http://de.wikipedia.org/wiki/Pf%C3%A4lzische_Dialekte so approx. one
million speakers is correct. In Germany you have that very particular
situation that there were hundreds of small cities being considered own
countries with borders in the middle ages. These cities were delimited
by walls and had their own currency and of course also "language"
therefore it can happen that from one small city to the other you find
people who have a very different dialect.
This means whatever German language you take you always have to deal
with a continuum in itself. Please consider that nds for example has
around 400 varieties, but the nds wikipedia applied only one of those as
"being allowed" and that is writing method according to Sass. We had
loads of fights about that time ago (with two users in particular since
these more or less hijacked the project then). I don't know how things
are going now. Anyway: we should consider to tell these people that they
need to accept any variety, otherwise we will connect the iso code to
one of them only and people often consider wikipedia to be the non plus
ultra of language knowledge :-( and this would mean that the other
variations over time get discriminated.
Uhmmm ... not a nice situation, but it very likely will happen if the
ones that start the wikipedia are all from the same place.
Yes it is, we already conditionally approved it. But in the motivation it’s written that they should go to a “localization” section in the talk page to get further instruction, and there’s no such section there.
Can't help with this. You have a 200% probability that I would say something
stupid, since all my knowledge of the japanese culture comes translated
literary works. But we do have a japanese native speaker in the Commission.
There are actually two "Old Japanese" requests on [[[[Requests for
new languages|m:Requests for
new languages]]]] at the moment: Bungo Wikipedia, and the Archaic
As I understand it, the Bungo Wikipedia uses Bungo, which was the
written form of Japanese until the early Showa period (around the
1920s and 1930s). Bungo is largely a written form of Japanese, and
while it was popular in the pre-WW2 era in legal documents and such,
it has largely died out since then. I do know some pieces of
classical literature that are written in Bungo, but it is not a
popular way to write Japanese nowadays.
The Archaic Japanese Wikipedia, on the other hand, uses a much
earlier form of the Japanese language, called Kobun (literally "Old
Language"). While types of Kobun span many centuries and there is no
one defined language, it was basically the language used in Japan
during the Heian period (8th century - 12th century).
Kobun (Archaic Japanese) is currently taught in junior and senior
high schools in Japan, so I would guess a large proportion of the
Japanese population is familiar with this type of Old Japanese,
although nobody grows up speaking it or uses it in daily speech.
(It's like learning Shakespeare in an English-speaking country. No
one speaks or spells like Shakespeare in daily life.)
Both types of Old/Classical Japanese are either difficult or
impossible to read for the average Japanese speaker without a
lexicon. None of them can be automatically translated from one
another, as the grammar and diction are significantly different.
I do believe that preserving these two forms of Japanese is an
important undertaking; world-famous classics like "The Tale of Genji"
were written in Kobun, for example. However, I am not really sure
writing an encyclopedia, covering modern topics like computers and
the latest anime cruft, is a plausible task for these two
encyclopedias; I feel that there is a real danger that these two
projects will also become dumping grounds for mutilated jawiki articles.
I agree they are important. My doubt is whether we would have a better
result in routing them towards a wikisource project, in which they could
manage all in Bungo and Kobun (talk pages and interface included) instead of
growing up a wiki about nuclear rockets in Kobun and Bungo.
The current state of Latin wikipedia is a bit funny, from this point of
view. Translating "Steven Spielberg" into "Stefanus Ludimontius" doesn't
seem to make for anything better than a students' joke, while if they
started to discuss Lucretium in latim it would probably end up in producing
That's my doubt, too. I'd rather give them a wikisource than a wiki.
Producing new material in a dead language does look like Original Research.
My idea is that possibly these requests would better fit as subsets of the
current state of the contemporary culture(s) they gave birth to (the
MLMW format would really help in this).
Let alone Belarus' we have an increasing number of similar problems waking
up in eastern wikies, and I suspect that this is but the start of a tide.
Hanja wants out of Korean, Trad Chinese wants out of Simplified Chinese,
etc. LMO just had it, Palatinate maybe a possible next spot. All this stuff
clearly involves a number of personal conflicts that get incubated in a
local political background. There also is a strictly technical part, though:
most korean admins, for example, simply cannot read the Hanja script. Same
may at least partially apply to entities that come from a fragmented
historical background, like LMO/Palatinate.
IMHO we could look for a solution that while allowing "more space" to lower
interpersonal frictions would also keep communities toghether. I guess it
would also help into marketing projects that otherwise risk getting very
I suppose that many more Japanese and Turks would make use of such projects
if they could see it as evidently present in their usual wiki environment.
In a fully indipendent state they would attract only a limited number of
literary fans/ancient language students but they would make a poor service
for the general public.
I believe that if "using another script" did not imply a "recognition of
independence" the number of such requests would lower to reasonable limits.
They would not mean drawing a border between you and the "enemy admin". They
simply would add "a specialized museum space". Existing communities would
not react to this as if it was a secession/betrayal (it would not be, in
fact), but rather as an expansion of their project and possibly we could
save lots of nerves to quite a big number of people.
Only we need a clear criterion to do this, it cannot be based on motivations
like "X is a dangerous place/culture". I'd rather use something like
"Historical phase" or the like and put it as a blanket policy for all such
I'm aware that this is more of a community management issue. Yet, who else
can act on how domains are formed? It's only us doing it, and since the
problem is centered around the way domains are formed it could probably be
addressed at the same level.
I'm not clear on this stuff, though. It's just an early impression I get
from reading meta here and there. Let's keep an eye on the conflicts and
mostly on WHAT they use to bang each other on their heads. So far mostly
it's requests for secessions, but it still may change.
If I haven't been very active on the LangCom since I was invited, it's
because I was traveling and now I'm working quite a bit on frwiki's
Wikicontest and handling real life crises. However, now that things have
calmed down a bit in my life (but only a bit :-p ), I have a request to make
of the veteran members of the LangCom. You often talk of ISO codes and other
mailing lists and rules and regulations. Perhaps a list of useful links
could be provided for us new members to read and consult when giving our
opinions on the requests at hand. I cannot speak for the other new members,
but I know I often feel out of place when reading your e-mails, wondering if
I could comment without saying anything stupid or redundant.
It's just an idea. I hope it is not too difficult a task, given how
complicated the ISO issues seem to be.
You might want to consider that some academic linguists I spoke with
consider ethnologue sometimes ‘aproximate’ in its data. Yet we have no
better universally recognized source, AFAIK. As per an history of the
linguistic ISO standards I still have a request from that student pending.
So if anyone can point me to a source...
Rotuman has already been conditionally approved as well. I would prefer to
leave the Ingush request open for now, as it looks like the test project has
no articles. For the Niuean Wikipedia, it would be better if they had at
least one other interested user, but if the language is under threat of
extinction I can understand the lack of interested users. Palembang has no
test project, as far as I can see, and has no support or oppose arguments. I
think this request should stay open. Everything else I agree with.