(please followup to -i18n)
We discussed the various initial issues in private with Amanpreet
(hmm, you have to tell me whether this is your first name, sorry for
that).
languagechooser and localechooser have the needed entries for handling
Punjabi when it will be ready.
There was no real problem and I expect the Punjabi team to start
commiting translations quite soon. Of course, this will put more
pressure on us for getting some progress on the graphical installer
as, like Hindi, Punjabi will require it.
We have concluded that the Punjabi translation will officially be
called "Punjabi (Gumurkhi)". The code used will be "pa_IN" and not "pa".
The reason is that the Punjabi language is spoken both in India and
Pakistan : the spoken language is the same, but the scripts are
different. Punjabi in India (sometimes called "Eastern Punjabi") is
written with the Gumurkhi script, an Indic type script, while Punjabi
in Pakistant (called "Western Punjabi") is written in a Persian-style
script named "Shahmukhi", like Urdu (which is the official langage in
Pakistan, though only spoken by 8% of the population).
Given that difference, which is close to te difference in the ways of
writing the Chinese language ("Simplified" used in Continental China
and Singapore and "Traditional" used in Taiwan, Hong-Kong, Malaysia
and overseas), I think the best way to sort it out is using "pa_IN"
for Punjabi using the Gumurkhi script while we will maybe someday use
"pa_PK" for Punjabi using the Sahmukhi script.
It has to be noted that the current Punjabi translations of software
indeed use "pa" rather than pa_IN. However, Amanpreet Alam agreed that
using pa_IN is far more logical in case some translators start working
on a pa_PK translation (Punjabi is the language of half the population
in Pakistan!).
