[Foundation-l] How to dismantle a language committee

Sat Jan 10 23:37:40 UTC 2009

Hoi,
This is a personal attack, an attack that is the result of discontent of the
way in which the policy of the language committee has been implemented.

So let me show where Mohammed is wrong. First of all, the language committee
is multiple people. Recently two high powered people were added to the
language committee. They are Michael Everson, who is so much of an expert in
this field that he rates his own Wikipedia article and Milos Rancic, a
linguist and a steward of the WMF. It is also incorrect that I would be the
only one "doing" things. The last request to the board for a Pontic
Wikipedia was written by Pathoschild. With such people actively involved,
the argument that I can force my will on them is ... a bit off.

Second, Mohammed is upset about one issue only. The existence of the
Egyptian Arabic Wikipedia. Mohammed is on one side of a dispute. He is under
the impression that I am on the other side of this. The people who are on
the other side of his dispute are the people who requested the Egyptian
Arabic Wikipedia and who are making the most of their project.

At the time when the request for the arz.wp was made, I asked the members of
the language committee if Egyptian Arabic was eligible. I got the reply that
it should be because it is recognised as a separate language. Nobody
objected to this and consequently the language was given the status of
eligible. The reason why I consulted the other members of the committee was
exactly because I foresaw people objecting forcefully. I have not been
disappointed. I am disappointed by Pathoschild's assertion that there was
not enough discussion, he should have spoken up at the time.

As far as the localisation for Egyptian Arabic is concerned, there is a
request for an Egyptian Arabic Wiktionary and it is to fulfill the
requirement for a secondary project that the localisation is doing so well.

Key in the current policy is that we have looked for objective criteria to
base the policy on. Using the ISO-639-3 standard is at the basis of how
languages are identified on the Internet. We publish our content on the
Internet so it was an obvious choice. The procedure and the requirements are
well published. It does not make everybody happy, including myself but it
does its job.

I have discussed illiteracy at some point. It is a well established fact
that learning to read and write is best done by learning to read and write
in the mother tongue. I have been involved in the translation of a study
about the use of SignWriting in Saudi Arabia that proved this point. There
are many such studies, and they make it quite clear that a close connection
between the written language and the spoken language improves literacy. When
the basic reading and writing skills are learned, a second language is
learned much easier and this has a lasting impact on the abilities of a
student.

It may surprise you, but I have a long history of trying to achieve things
for languages like the Arabic language. At some stage there was an
opportunity to have the complete English language Wikipedia translated by a
state of the art machine translaton engine. For all kinds of reasons this
did not happen. One of the reasons was the availability of servers.

The Arabic Wikipedia is doing really well. It has a consistently excellent
localisation, and it is growing nicely. If anything I am surprised that it
is not bigger then it is. In the discussions about Egyptian Arabic, I have
made the point repeatedly that the best way of proving the point that
standard Arabic is superior is by making sure that the Arabic Wikipedia is
indeed superior.

As to Latin script in the arz.wikipedia, that is news to me. If this existed
while in the Incubator, the project would not have had approval. I am
disappointed about this. If anything, this damages the project considerably.

Given that Egyptian Arabic is indeed a separate language, it is not feasible
to do full justice to the language by using machine translation. Machine
translation only work up to a point. One such tool that is likely to do a
great job is called Apertium, this tool is particularly good at translating
closely related languages, and I am sure that you agree that this applies...
I would be happily surprised if a real effort is given in creating such a
machine translation.

Thanks,
     GerardM

2009/1/10 Mohamed Magdy <mohamed.m.k at gmail.com>

> Hi all,
>
> I would like to propose the dismantling of the language committee and
> creating a new one (not including Gerard, of course).
>
> Why?
> Because it is chronically malfunctioning.
> Manifested in:
> # Gerard is forcing all his opinion, anything else is going nowhere.
> # Other members don't really care and leave it, unfortunately for us, to
> Gerard.
>
> Background:
> I read about how unfair the LangCom before but I didn't really care because
> it wasn't affecting a language I care about. Then came the dreadful
> proposal
> for a dialect Wikipedia in my dialect, Egyptian dialect. At first, I wasn't
> sure in the beginning if I should support it or not, then I became sure if
> this should happen, it shouldn't happen on a platform like Wikipedia (for
> many reasons laid out in detail in the proposal page). I don't care if
> Ghaly
> and company (the people who made the proposal) started that on an
> independent website (Wikia or on an own domain for their campaign) but on
> Wikimedia, we should do the right thing (I hope). The proposal was approved
> (Gerard requires that you have the relevant ISO code and everything from
> there could be done, he is a bit annoyed now becuase of all the current
> proposals for dialect Wikipedias which were brought up by the Egyptian
> dialect Wikipedia proposal) and the technical team had no option but to
> create the wiki because Gerard gave it his blessings and the foundation
> didn't say a word (I heard that people were happy at Wikimania (Florence?)
> because of that proposal but I fail to understand why the Egyptian people
> there didn't express their opinion about it (it was in Egypt :!).
>
> Trivia (I like structure but..):
> * Gerard is talking about how good the localization of the Egyptian
> dialect,
> well, that is a natural thing when the localization is a matter of
> copy-pasting Arabic translation and converting it to a slang form or
> English
> words in Arabic (nothing wrong at all in that of course, we do it all the
> time, but we don't do it for the sake of looking hip (there is a certain
> language charisma we have in Egypt, that is, if you can speak English and
> mix English with Arabic to look cool. don't know if other countries have
> it), we do it only to introduce new words that we are unable to find their
> equivalent in Arabic (e.g. Acetylcysteine which is أسيتيل سيستئين in
> Arabic,
> basically English (latin) in Arabic).
>
> * May be ISO is wrong: why people are taking ISO codes as absolute,
> don't-discuss matter? in our case, we have 22 dialects of Arabic and the
> pathetic decision to call them languages of the supposedly "Macro" language
> Arabic, that is nonsense and it should be amended, not the blind (if not
> stupid) opinion of making all these sorts of dialectical projects (
> http://www.sil.org/iso639-3/documentation.asp?id=ara). I tried to contact
> the ISO, they say to contact the local office in my country (
> http://www.eos.org.eg), and as always, they have dead emails, don't know
> about the phone numbers, I'm not even sure that anyone there would listen
> to
> a word of mine, besides, I wish to see changes before my expiry date is
> due.
>
>
> * Gerard have the false delusion of protecting the freedom of Egyptians and
> taking us out of illiteracy into the light of knowledge by making a new
> Wikipedia in slang and dialect. well, you are *wrong*, you are doing quite
> the opposite and other people are helping you alas. hope you understand
> that
> someday.
>
> * Wave of ignorance: a new wave of ignorance are upon us and I don't like
> Wikipedia being part of it.
>
> * Did you know that when I tell people about this new Wikipedia, the
> consiperacy theory of the west dividing us is brought up? like it wasn't
> enough that the ar.wiki isn't appreciated because of the several issues we
> have. no, now we have another big issue created because of the carelessness
> of some people. arz.wiki is a regression, making people think of Wikipedia
> as an enemy is a regression.
>
> * Did you know that what was rejected before, is being done on that
> arz.wiki? I'm talking about Arabic in latin characters Wikipedia. they have
> no objections there if you write Arabic in latin (a big no no in ar.wiki or
> any another respectable venue). dialect writing/Arabic in latin writing is
> for fun only, nothing serious.
>
> * They have a template on arz.wiki which is placed on articles copied from
> ar.wiki that says ~"this article needs more egyptianizing" like the one on
> uncyclopedia "this article needs to be more uncyclopediac" or something
> like
> that (sorry for the lack of links).
>
> * I think it would be doable to make a tab that Egyptianizes (or any other
> dialect) the Arabic article, that is, if we have some sort of conversion
> memory, that is if the dialect is stable (or standard), the dialect differs
> from a place to another, from a muhafazah to another (
> http://en.wikipedia.org/wiki/Muhafazah). if anyone knows the technical
> method we could make a trial instead of the great mess of dialect
> Wikipedias. I'm not too sure about this compromise yet.
>
> So, to sum it up:
> # Dissolve the current committee and make a new one of people who care.
> # Make all the discussions of the committee public and allow community
> members to comment and the committee really reads what they have to say.
> # Make sure that Gerard isn't on the new committee.
> # Treat ISO codes flexibly, they could be amended, they could be ignored if
> appropriate.
> # Undo the arz.wiki.
>
> Pardon the long email, but I had to say what I have on that important
> issue,
> may be the new year would bring something else besides massacres.
>
> --alnokta
> _______________________________________________
> foundation-l mailing list
> foundation-l at lists.wikimedia.org
> Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l
>