Fwd: [Ltru] Punjabi

["Mark Davis" <[e-mail]>]

This got bounced from LTRU for having too many recipients, so am forwarding.

---------- Forwarded message ----------
From: Mark Davis <[e-mail]>
Date: Mar 14, 2007 5:23 PM
Subject: Re: [Ltru] Punjabi
To: "[e-mail]" <[e-mail]>
Cc: John Cowan <[e-mail]>, "Sarmad Hussain, Dr." <
[e-mail]>, LTRU Working Group <[e-mail]>, Nayyara
Karamat <[e-mail]>, [e-mail], [e-mail],
[e-mail], Abbas Malik <[e-mail]>,
[e-mail]

A few points of clarification, since this arose with regard to CLDR. The
CLDR data is divided into linguistic data and non-linguistic data, and what
are called "locales" are only used for linguistic data; so in that sense
CLDR locales are really collections of language data, not locale data in a
POSIX sense. We use inheritance in the data model, so we only need to add
script subtags when necessary (when the language is customarily written in 2
scripts), and only need to add region subtags when the language+script is in
common use in more than one country AND some of the linguistic data differs
between them. (Although sometimes for compatibility we add empty country
locales.) So what we would need at a minimum right now is:

1. pa (containing data appropriate for pa-Guru-IN)

If in the future we get data submitted for pa written in Arabic, we would
add

2. pa-Arab (containing data appropriate for pa-Arab-PK)
3. pa-Guru (containing data appropriate for pa-Guru-IN)

Our policy is to have each parent locale's data be what is appropriate for
the most populous (measured as literate, first or second language speakers)
child locale. So what we'd need to have at that point is the population of
pa users (as opposed to lah) in Pakistan vs India, which would determine
whether we change the contents of pa at that point. It sounds like, from
what you say, that the bulk of the population of pa users (measured as
literate, first or second language speakers) would be in India, while
Pakistan would be more lah users than pa users. Is that the case?

Mark

On 3/14/07, [e-mail] < [e-mail]>
wrote:
>
> > The current plan is to map "Eastern Panjabi" onto "Panjabi", and all
> > the others onto "Lahnda".  It sounds like you are proposing to map
> > both "Eastern Panjabi" and "Western Panjabi" onto "Panjabi"
> > and the others onto "Lahnda".  Is that correct?
>
> No, I'm saying that a significant number (i.e. millions) of people speak
> "Eastern" Punjabi in Pakistan.  The only major difference between this
> spoken language and that spoken in Indian Punjab is the written script (
> i.e.
> Shahmukhi/Gurmukhi).  The sources that say that Western Punjabi and
> Eastern
> Punjabi suddenly stop at the border are simply wrong.  The Maajhi dialect
> is
> centred on the border - leading to significant portions of the population
> speaking it on either side.  Hence the need for both 'pa-IN' and 'pa-PK'.
>
> "Grierson [in the Linguistic Survey of India] defined Western Punjabi as
> being west of a line running north-south from Montgomery and Gujranwala
> districts." (This is directly from Wikipedia, but I was the original
> contributor for this in the article. If you wish, I can probably track
> down
> the exact page number for this, but I don't have it at hand at the
> moment.)
> This is well within present day Pakistan, and Lahore alone has nearly
> seven
> million people.
>
> Anyway, that is my take on things.  I would be interested to see what NU
> professors/researchers think on this because I've never been to Pakistan
> so
> I could be way off base here.
>
> Regards,
> Sukhjinder Sidhu
>
>


-- 
Mark

-- 
Mark

_______________________________________________
Ltru mailing list
[e-mail]
https://www1.ietf.org/mailman/listinfo/ltru