Language support summary/discussion

[Thibaut Varene]

Hi!

On Wed, 16 Feb 2005 15:54:33 +0100, Martin Pitt <martin.pitt at ubuntu.com> wrote:
> Hi!
> 
> Thibaut Varene [2005-02-16 15:43 +0100]:
> > A quick reaction on these figures: They are completely biased.
> 
> I don't think that they are biased, there are just a gazillion
> different ways to interpret "speaker"

Sure. But I'm not dealing with how we interpret speaker, I'm standing
from the end user point of view: eg: "we want to cover as many users
as possible on the CD", if I get everything right. For that purpose,
the first set of figures was completely irrelevant, imho ;)

> > You are only accounting native speakers here, as noted.
> 
> Indeed, and I did that deliberately. Ubuntu's long-term goal is that
> everybody can use a computer in his native language. :-)

Of course, but that's irrelevant to the question "what should go on
the CD?", for languages pack availability has nothing to do with what
we put on the CD or not.
 
> > "The following is a list of these languages in terms of the number of
> > countries where each is spoken.  The number that follows is the total
> > number of countries that use that language (from Weber, 1997):
> >
[snip]
> >
> > These figures are IMHO much more relevant to the question "how broad
> > our language support will be if we include that language package".
> 
> To avoid misunderstandings, I in no way insist on taking the very
> first list, I can as well take this one. We just have to find a
> consensus which has an objective justification to avoid flamewars
> about "you don't like my country". So we must base this decision on a
> well-known, public, and objective statistics to avoid getting in
> trouble with geopolitics.

I think there are two approachs to that problem:

1) Either you don't want to go into flamewars, and put NO language
packages on the CD (not even English), leaving it to the user to get
them from the network. That's harsh, but that can be easily justified
and will certainly avoid any "you don't like my country" flamewar.

2) You want to cover as many users as possible, and then you probably
want to use more balanced figures. Hence the set of figures I've
proposed. There's also another list mentionned on the website at the
very end that's worth looking at:

"After weighing six factors (number of primary speakers, number of
secondary speakers, number and population of countries where used,
number of major fields using the language internationally, economic
power of countries using the languages, and socio-literary prestige),
Weber compiled the following list of the world's ten most influential
languages:
(number of points given in parentheses)

   1. English (37)
   2. French (23)
   3. Spanish (20)
   4. Russian (16)
   5. Arabic (14)
   6. Chinese (13)
   7. German (12)
   8. Japanese (10)
   9. Portuguese (10)
  10. Hindi/Urdu (9)"

That list can be of interest for it tries to balance several factors
WRT language "influence" over the world.

> > PS: I hope you understand why ripping off French would be a bad idea, btw? ;o)
> 
> Sure :-) FWIW, I don't really appreciate throwing out German, too, but
> I can't help it (well, I should produce some - a lot of - children,
> maybe :-) ).

hehe

> Martin,
> who goes to increase the number of German speakers now :-)

On work hours? Damn! ;o)

HTH

-- 
Thibaut VARENE
Debian, Ubuntu and Kernel developer ;)