Language codes (and specifically for Armenian)
armenzg
Soon we will be having Firefox in Armenian and I know that at some
point I will have the discussion with people with regard the three
known different types of Armenian:
* Armenian from Armenian (hy-AM)
* Western Armenian (?)
* Eastern Armenian (?)
What is the standard that we follow for Mozilla?
I will try as hard as I can to come to a consensus to don't split
among variants but I already know that somebody came up with Western
Armenian langpack.
I tried these pages to find the right info without luck:
http://en.wikipedia.org/wiki/Armenian_language
http://en.wikipedia.org/wiki/List_of_ISO_639-1_codes
http://en.wikipedia.org/wiki/Language_code
Seth Bindernagel
armenzg wrote:
> Hi all,
> Soon we will be having Firefox in Armenian and I know that at some
> point I will have the discussion with people with regard the three
> known different types of Armenian:
> * Armenian from Armenian (hy-AM)
> * Western Armenian (?)
> * Eastern Armenian (?)
>
> What is the standard that we follow for Mozilla?
> I will try as hard as I can to come to a consensus to don't split
> among variants but I already know that somebody came up with Western
> Armenian langpack.
>
Hi Armen,
It's probably a good idea to think about this now. As you probably
know, we prefer to use a two letter locale code like "hy". But, as you
have said, and as we have seen across the world, many languages need
more specificity because there are variants that different subcultures
speak.
In some cases, we map to a variant to a country level (like es-ES,
es-AR, es-CL, and es-MX), but that is not necessarily always correct.
In other regions, we have different languages within the country.
Looking at Norway, for instance, we have two languages named nynorsk
and bokmål. These locale codes are nn-NO and nb-NO respectively.
There, you can see that the country specification /Norway/ remains the
same, but the local name is different.
So, perhaps you need to name the eventual Eastern and Western variants
based on references to the dialects. Something like a two letter
abbreviation for Arevelahayeren and Arevm'tahayeren followed by AM.
hv-AM and hm-AM? I don't really know. You might be writing the book on
this case. :)
But, why not just move forward with hy-AM and we'll discuss to do with
Eastern and Western later.
-Seth
Robert
There are 3 types of Armenian, but one is old and as I know it's very
common with Western Armenian, but Armenian in Armenia it's Eastern
Armenian
1) Eastern Armenian (hy-AM), it's State language of Republic of
Armenia
2) Western Armenian, Armenian Diaspora mainly use this type (Except
Russian Diaspora)
3) Grabar Armenian, which is old Armenian and not usable anymore.
Armenian diaspora will not use Eastern Armenian.
It's important to separate Eastern and Western Armenian.
But because Eastern Armenian is a State language of Republic of
Armenian we can call it Armenian without mentioning Eastern.
P.S. though there are peoples who want to rename RP Armenia and call
it Republic of Eastern Armenian, I'm among those people , but it's
just a dream.
Axel Hecht
There's an authorative answer, linked to from the deep ends on
wikipedia, and you gotta know which links to follow.
It boils down to, if we *really* need them to be separate,
%%
Type: variant
Subtag: arevela
Description: Eastern Armenian
Added: 2006-09-18
Prefix: hy
%%
Type: variant
Subtag: arevmda
Description: Western Armenian
Added: 2006-09-18
Prefix: hy
Per http://www.iana.org/assignments/language-subtag-registry,
http://www.ietf.org/mail-archive/web/ltru/current/msg06986.html,
http://tools.ietf.org/html/bcp47.
Technically, I'd be interested to, as always, learn what actual
differences we encounter within the browser. And if there are ways to
share work in one source to generate both.
Axel
armenzg
> It boils down to, if we *really* need them to be separate,
>
> %%
> Type: variant
> Subtag: arevela
> Description: Eastern Armenian
> Added: 2006-09-18
> Prefix: hy
> %%
> Type: variant
> Subtag: arevmda
> Description: Western Armenian
> Added: 2006-09-18
> Prefix: hy
>
> Perhttp://www.iana.org/assignments/language-subtag-registry,http://www.ietf.org/mail-archive/web/ltru/current/msg06986.html,http://tools.ietf.org/html/bcp47.
>
> Technically, I'd be interested to, as always, learn what actual
> differences we encounter within the browser. And if there are ways to
> share work in one source to generate both.
>
> Axel
As much as I can I will try to have a shared source. We will see as we
go.
> There are 3 types of Armenian, but one is old and as I know it's very
> common with Western Armenian, but Armenian in Armenia it's Eastern
> Armenian
> 1) Eastern Armenian (hy-AM), it's State language of Republic of
> Armenia
> 2) Western Armenian, Armenian Diaspora mainly use this type (Except
> Russian Diaspora)
> 3) Grabar Armenian, which is old Armenian and not usable anymore.
>
Thanks Robert. I will not consider the 3rd one.
I will bring more discussion later in time but from your comments I
have two preferences:
* hy-AM and hm (without -AM since it is not spoken in Armenia)
Something like a two letter abbreviation for Arevelahayeren and
Arevm'tahayeren followed by AM
* hv (for hy-AM) and hm
I will try to dig more on the internet to see what other projects are
using (if they are).
Axel Hecht
You can't make up locale codes, there's a standard, and we follow that.
hy-arev??a might be subject to
https://bugzilla.mozilla.org/show_bug.cgi?id=525494, though.
Axel
armenzg
> You can't make up locale codes, there's a standard, and we follow that.
>
> hy-arev??a might be subject tohttps://bugzilla.mozilla.org/show_bug.cgi?id=525494, though.
>
> Axel
I will research more wrt BCP 47 (RFC 4646 and 4647).
Are these the standard you are referring to?
Robert Kaiser
> Are these the standard you are referring to?
See https://wiki.mozilla.org/L10n:Simple_locale_names for how we build
our locale names and what standards we're using.
Robert Kaiser
armenzg
> armenzg schrieb:
>
> > Are these the standard you are referring to?
>
> our locale names and what standards we're using.
>
> Robert Kaiser
Thanks Kaiser.
From looking at it seems that these should work:
* hy-AM-arevela
* hy-AM-arevmda
The problem from the second one is that the "arevmda" is not really in
spoken in Armenia but I guess we could deal with it.
Is there an XY for unspecified region?
* IETF language tag: hy-arevmda: http://www.ietf.org/mail-archive/web/ltru/current/msg06986.html
* Generic variants and Armenian dialects (long) -
http://www.alvestrand.no/pipermail/ietf-languages/2006-September/004978.html
* Forum discussion about it - http://forum.hyeclub.com/showthread.php?t=1632&page=2
armenzg
> On Apr 8, 9:39 am, Robert Kaiser <ka...@kairo.at> wrote:
>
> > armenzg schrieb:
>
> > > Are these the standard you are referring to?
>
> > our locale names and what standards we're using.
>
> > Robert Kaiser
>
> Thanks Kaiser.
>
> From looking at it seems that these should work:
> * hy-AM-arevela
> * hy-AM-arevmda
> The problem from the second one is that the "arevmda" is not really in
> spoken in Armenia but I guess we could deal with it.
> Is there an XY for unspecified region?
>
> * IETF language tag: hy-arevmda:http://www.ietf.org/mail-archive/web/ltru/current/msg06986.html
> * Forum discussion about it -http://forum.hyeclub.com/showthread.php?t=1632&page=2
Awesome. I just found this:
http://www.ethnologue.com/codes/LanguageIndex.tab
hye AZ D Western Armenian
hye AM D Eastern Armenian
Would this work for all of us? (In the case that we have to split)
* hye-AM
* hye-AZ
Cheers,
Armen
Axel Hecht
- Don't touch that dial until we have to.
- Use hy-AM for now.
- Iff we have to touch that dial:
-- Migrate hy-AM to either hy-arevela or hy-arevmda (no AM, region codes
suck), pending support for that in gecko
-- Create an intelligent fork of that for the other version
That's the least hassle in terms of string reuse, historic compat, IT
tickets, and community work force.
Axel