website translation experiments

[Dirk Hohndel]

This may not go anywhere - but it might, so please bear with me...

I have spent a few hours (mostly because I suck at Perl) and wrote a couple of scripts to try to connect our current git based translation system for the website with Transifex.

There's a new resource on Transifex that is called "about" and that contains the strings of our landing page. For the website we currently have translations to de_DE fr_FR es_ES it_IT nl_NL pl_PL ru_RU pt_PT fi_FI so these are the ones that have been populated. The closer the translations were to the English version, the better these imports will be. While eventually we can consider other translations, right now I'm not asking for people to add more translations!

If you would like to help me figure out if this is feasible and you speak (doesn't matter at this point how fluently) one of these languages, you could help me by logging into Transifex and cleaning up the translations of that "about" resource. The biggest issue likely will be "translations" that are on the wrong source string. Because of the way I extract these preliminary translations from the existing translations, if the "grouping" of the HTML in the translation doesn't match the grouping in English, things will go poorly. I had to break up a couple of longer paragraphs in German to match the paragraph breaks in English, for example.
Then tomorrow I'll try to pull these translations again and see if I can feed them into git and push them from there into WordPress in order for them to show up on our website.

This initial process is labor intensive on my part, that's why I did it for only one of the pages. But once I have this part taken care of (= one time investment of my time) the continued maintenance should be rather scriptable / low touch. Which is REALLY the part that I want to test. Because if this does work then we can do the website translations on Transifex as well, we can translate announcements before they go out and overall create a much better experience for our users.

Admittedly, the initial process is also a bit annoying for the translators, but having done it for German it's actually not too bad because it is pretty obvious what is happening (since you have the source strings right there on the web site).
The main reason for this convoluted process is that I want to preserve as much of the existing translation work as possible. It would have been MUCH easier for me to throw away all the existing translations and start from scratch, but that seemed entirely unreasonable towards the translators and all the efforts they have already put into translating the website.

Anyway, if all this works out, maybe the AMAZING group of translators that we already have for the application can help us keep the website better translated as well.

If you have the time and want to help, respond here, tell us when you are happy with the translation (and what language) and I'll take the experiment from there. If you don't have a Transifex account, signing up is easy and then ask to join the subsurface team - I or one of the language coordinators will add you (there is a manual approval step involved - sorry).

For now, none of the automation is hooked up, so none of these translations will show up on the website unless I process them here (just to avoid unmet expectations) :-)

Thanks as always!

/D
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.subsurface-divelog.org/pipermail/subsurface/attachments/20180715/0cffa7f9/attachment.html>