Hi, I am considering a new project involving a Discourse instance with Catalan as primary language.
The first thing I have noticed is that there are two “languages” in Transifex related to Catalan: ca and ca_ES. They have different strings translated and reviewed. Is this normal? Do they act as completely separate localizations or is i.e. ca a fallback for ca_ES (or vice versa)? Is it possible to merge them in order to have one version only (since ca_ES is in practice the default for ca anyway and there are no other regional variants so far?)
Once these questions are solved, I would be curious to know what would it take to enable Catalan here (or at least in the demo version) so myself (and other Catalan speakers around?) can see which are the strings needing translation more urgently.
I saw that pinned topic and I started reading it, but silly me thought that these were instructions to add a language to your own instance, not to Discourse upstream. Now it’s clear.
I checked with Transifex support. ca and ca_ES act effectively as separate languages, unless someone find a way to make one become the fallback to the other in a specific software product. For now, I am assuming that they are simply separate languages with little use and a potential for confusion and duplication of work. Improving the % of completion of ca should be the goal, recycling work done at ca_ES if/where feasible.
I just contacted @foobar1234 via Transifex, since still today he is the maintainer of both versions and he might know what happened years ago to end up in that situation. Hopefully he will be still around and able to reply.
Once the plan for next steps is clear(er), I will contact the rest of translators listed in both teams, pointing the to this thread for coordination.
For everyone’s info, I am in contact with @quimgil on transifex already. I’m glad there will be another contributor to the Catalan translation! Hopefully we will sort out this ca/ca_ES duality and bring one translation up to speed.
After contacting ~20 translators only 2 got back to me (one of them @prullan, gràcies!) and both had no strong opinions about the plan.
I might be wrong, but it seems that as of today nobody is desperately waiting for a new version of the Catalan locale. Discourse installations that are using Catalan patched their Catalan locale manually and their users probably are happy enough with it.
@foobar1234 , given that we are all very busy, I propose a simpler change: just rename ca_ES to ca in Transifex (I can request this, no idea about the process) and then propose a pull request to Discourse devs to integrate the ca locale.
After looking a bit further into ca and ca_ES, the latter has much better coverage, and I am not seeing translations in ca missing in ca_ES.
In other news, I am teaching discobot to speak Catalan. This is an amazing feature, key player in the out of the box experience, and it was 0% translated.
Hi @quimgil, I do have Catalan users. None of them are complaining but I am not sure of the status of my up to date discourse instance. Is Catalan supported out of the box or do I need to add it somehow?
At this point it is better that you wait until Catalan is integrated. I’m working on it. I’ll let you know when it is available out of the box. Thank you for your willingness to make Catalan speakers feeling at home!
Is there a volunteer willing to go through How to add a new language in order to send a pull request adding Catalan support to Discourse?
A translator is working on narrative-bot/server.en.yml and the goal is to have that file translated this week. After that, discobot would speak Catalan for the first time, and what a better way to celebrate this than committing Catalan support to Discourse?
PS: If no volunteer appears but someone is interested in doing this work for a fee, I’m open to suggestions. I still haven’t managed to install Discourse locally and I’m so busy these days anyway.
Thank you @foobar1234, most of the strings translated are still yours. I hope that now that Catalan ships with Discourse more instances and more users will use it, find problems, contribute translations…