Re: Request for Comment - L10n enhancement

Kathiravelu Pradeeban <[email protected]> Tue, 6 Aug 2013 22:46:41 +0530
Newsgroups gmane.editors.abiword.devel
Message-ID <CAAarrS=xniBHEY577QsjnXb7-UFqcP9H8SvnS5Y9RYeK-FnO=g@mail.gmail.com>
On Tue, Aug 6, 2013 at 7:31 PM, Chris Leonard <[email protected]> wrote:
> On Tue, Aug 6, 2013 at 4:59 AM, Kathiravelu Pradeeban
> <[email protected]> wrote:
>> Hi Chris,
>>
>>
>> According to your proposal, both BR and PT variants of Portuguese will
>> be maintained. So to avoid confusion, I would prefer to keep the name
>> of pt-PT.po as pt_PT.po, instead of renaming it to pt.po.
>>
>>> pt-BR.po  > pt_BR.po - change to underscore
>>> pt-PT.po  > pt.po
>>
>> Did I misunderstand something?
>>
>
> Pradeeban.
>
> pt_BR is a commonly maintained and localized variant across many
> projects.  My understanding is that there is sufficient variation to
> merit it's inclusion as separate from European Portuguese and also
> sufficient interest in maintaining it (given the large population of
> Brazil).  I know from an OLPC deployment that pt_BR is the preferred
> option in Portuguese-speaking Africa (Sao Tome and Principe).

+1.

>
> As for maintaining the country code for Portugal to distinguish it, I
> sincerely do not believe it is necessary.  I base this on the
> practices of numerous L10n projects that I work with across the FOSS
> world.  It is widely understood that pt is European Portuguese and
> pt_BR is the Brazilian variant, not further distinction is generally
> needed.  If there are technical aspects I am not undersdanding in your
> request, we should discuss more.  I think there may be technical
> advantages to not including the country modifier for Portugal,
> (possible failover from pt_BR to pt if pt_BR is not complete).
> Similarly de is widely used for German and de_CH for SwitzerDeutsch,
> even though a unique code (gws) is now available for the Swiss German
> variant.

I was just curious regarding the conventions, and having pt_PT.po
named as pt.po doesn't have any issue technically.

>
> My only stronger argument (other than consistency with others
> practices) is that our Pootle instance uses pt and pt_BR and doing
> this would make direect commits from Pootle to SVN much simpler

Given that it is the common practice, and also match with Pootle, I
have no objection on this. Thanks for explaining this.

Regards,
Pradeeban.

>
> cjl



-- 
Pradeeban Kathiravelu.
Postgraduate Student,
Erasmus Mundus European Master in Distributed Computing,
Instituto Superior Técnico, Lisbon, Portugal.

Blog: [Llovizna] http://kkpradeeban.blogspot.com/