make reproducible-builds.org translatable?

[Hans-Christoph Steiner]

Daniel Shahaf:
> Hans-Christoph Steiner wrote on Fri, 01 May 2020 10:08 +0200:
>> Translations are no different.  What we're proposing for this website
>> is something that is achievable with the small level of contributor
>> time that is available.  We can definitely deliver something that is
>> useful without being a maintenance burden using established software
>> and patterns.  It will not be perfect, and it would not be able to
>> provide the specific feature you describe.
> 
> Okay.  Thanks for clarifying this.
> 
>> I know of no translation workflow software that will provide this
>> specific feature that you are asking for.  I know that people can do it
>> manually, but that's a lot more work.  po4a is a great tool, its written
>> in perl, and the maintainers welcome contributions, if anyone wants to
>> implement such a feature.
> 
> IIRC, there was at least one translation toolkit that presents
> translators with one paragraph at a time, and then, if a paragraph's
> translation was missing, the English version would be used, exactly as
> in the git-reset(1) usage output I posted upthread.  (I don't remember
> its name, but in any case that was years ago.)  Such a design would
> involve zero ongoing overhead and address the out-of-datedness concern,
> though I grant it has not only upsides but also downsides.

You can get this using po4a by setting the --keep to 0, meaning po4a
will always generate a page for every language, even if only a single
string is translated, e.g. only the title.  po4a also breaks up
documents into paragraphs for translation, that's the best practice for
long form text.  Some editors will have a sentence sub-display though,
which has pluses and minuses.

> I don't have anything to add.  The concern is clear, I hope, as are the
> existing ways to address it.  Beyond this point, I defer to the list.
> 
> Thanks for the information!

Sure, happy to share.  Translating websites has recently gotten a lot
easier, but there is still much work to be done, so this kind of
discussion and feedback helps towards that goal.

.hc

-- 
PGP fingerprint: EE66 20C7 136B 0D2C 456C  0A4D E9E2 8DEA 00AA 5556
https://pgp.mit.edu/pks/lookup?op=vindex&search=0xE9E28DEA00AA5556