Usage idea: translations

[zkamvar]

One potential use-case that I've been thinking about involves translations of markdown prose. AFAIK, it's a hard problem because gettext is designed for standalone messages, not prose.

That being said, it HAS been done:

• https://stackoverflow.com/a/8142666/2752888
• https://github.com/carpentries-i18n/po4gitbook.

In fact, on SO user suggested an XML solution: https://stackoverflow.com/questions/30585841/is-there-a-method-for-translating-markdown-formatted-text#comment49393344_30585841

One non trivial solution if you are planning to do this in a long term would be to generate your markdown by applying an XSLT on your content XML i.e. apply your XSLT on you English XML => your English markdown, apply same XSLT on you Spanish XML => your Spanish markdown. Non trivial because the initial setup part is going to be tough. – potatopeelings Jun 5 '15 at 11:47

I think {tinkr} is a good tool for this because it can provide a way of extracting paragraph elements, adding keys, and facilitating the back and forth between po file and markdown file. This way, translators can use standard translating apps like PoEdit to facilitate translations.

[maelle]

You mean for having automated translations?

[zkamvar]

Not for automated translations, but more of translation assistance. The rationale behind this was presented at CarpentryCon: https://youtu.be/IzRCuk7XX18, https://pad.carpentries.org/cchome-translation

That being said, this is definitely not a trivial example, but it's something that could be a feasible package that uses {tinkr} on the back end.

[maelle]

😸 https://docs.ropensci.org/babeldown/