openoffice-l10n mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Rob Weir <robw...@apache.org>
Subject Re: Proposal: communications/workflow for new languages
Date Mon, 25 Feb 2013 13:00:45 GMT
On Mon, Feb 25, 2013 at 3:13 AM, J├╝rgen Schmidt <jogischmidt@gmail.com> wrote:
> On 2/23/13 11:56 PM, Rob Weir wrote:
>> On Sat, Feb 23, 2013 at 12:58 PM, Michael Bauer <fios@akerbeltz.org> wrote:
>>> A common translator's tuppence on the issue. Most translators are not
>>> programmers and the more technical and convoluted an approach you take, the
>>> more you narrow your potential translators down to that much smaller pool of
>>> people who can do both code and translation.
>>>
>>> They also, for the most part, don't give a fig about what license something
>>> is under. I realise that for the people running the project in question,
>>> issues of license exist but the translators on the whole just want to see X
>>> in their language released.
>>>
>>> Thirdly, duplication is hugely frustrating for translators, especially for
>>> people working in smaller languages. Having just "rescued" some Firefox
>>> translations from Launchpad, I can vouch for that.
>>>
>>> So what would be actually nice is if AOO and LO could agree on a neutral
>>> Pootle server which handles both AOO and LO strings and from which each
>>> project can extract whatever translations they require. Minimum duplication
>>> and standard signup and translation process (none of this Bugzilla stuff or
>>> the need for someone to extract po files, email them, translation offline
>>> and the the whole thing backwards to commit, if ever there was a #facepalm
>>> process, that's it...), if needed, slap a notice email in the process about
>>> how these will be used.
>>>
>>
>> These are all good points (and Claudio's points are good as well) but
>> they are not related to my proposal.  I was just suggesting a way in
>> which we can use Bugzilla improve communications and track progress on
>> the offline translation of the PO files.
>>
>> For example, suppose right now a new volunteer posts a note saying
>> they want to help with the Greek translation.  Where are the PO files?
>>  What other volunteers have started on this effort?  Who else has
>> expressed interest?  Today we need to search through list archives to
>> find these facts.  But if we had one Bugzilla issue per locale, then
>> we could track status there, including attaching files.  Volunteers
>> can also add themselves to the cc-list for the Bug to stay "in the
>> loop".  And they can even create sub-issues if they want to divide up
>> the work formally.
>
> Michael said very clear that he believe Bugzilla is to technical and too
> heavy for many translators. Besides all the other good points he made we
> should think about a way to simplify the usage of Bugzilla in this use
> case. We don't want to lose any translators because of too technical
> process.
>

Actually, Michael didn't say anything specifically about Bugzilla.  Go
back and read his note.  He was complaining about the entire off-line
translation process and saying that it was too complicated.

And maybe that starts getting too geeky when we post PO files in
tar.bz2 format, a format unheard of to Windows users who comprise 90%+
of our users?

In any case, I even question whether that is our main problem.  If
volunteers were confused or could not figure out how to do something
then they would ask.  We see that on every other list where project
tools are involved.  Maybe not everyone would ask.  Some would just
leave.  But we would get questions.  Look at the dev list, the qa
list, even the marketing list.   But where have we received questions
from translators saying that they are confused about off-line
translation?  Zero, right?  Where we have seen questions is about the
anonymous Pootle usage.

IMHO, our problem today is about lack of coordination.  It is not
about the technology being too hard.   Look around.  How many
volunteers have come through asking to help about Hebrew, Greek and
Portuguese?   Isn't a non-optimal that they haven't managed to
coordinate on the translation?  Isn't it non-optimal that we have no
idea whether this is even being worked on?  That is the problem we're
facing.

We're wasting our time arguing whether typing into a box in Bugzilla
is harder or easier than typing into a box in MWiki.  Etiher would be
fine.  But our problem is that we're failing to provide the tools and
processes for small groups of new volunteers to coordinate on a
translation.  This is not about technology.  It is about our disregard
for the fact that technology is not enough, that we need to help with
the coordination side as well.  And this might mean smart use of
technology rather than just complaining that technology is hard.

-Rob


> Juergen
>
>
>
>>
>> Regards,
>>
>> -Rob
>>
>>
>>> Michael
>

Mime
View raw message