Robin
We are using the “output” files of a process (that we don’t know)
Are you sure about that? How to explain then the structural errors present in them?
why did we start translating the “output” files?
We started translating with the HTML files, since I have no clue about the meaning of this .conf file you mentioned, nor know anything about .adoc and ASCIIDoc. For example, I have written the antiX Help Video resource directly in plain HTML using a plain text editor, so I assumed the antiX FAQ files were written the same way. And from what I learned while writing a script which cumulates the strings from the Help video HTML to gettext .pot format and allows after translation the replacement of the strings in a copy of the original file set to create a localised version, I understood that this approach would exceed my time budget when applying it on the huge antiX FAQ file set with way more translatable strings. This is why I started testing the translatability of direct HTML files at the community test bed, and it worked fine, once you have understood how to manage them; so I started a full test translation of one of the files in the test bed, and it worked fine, so I had no doubt it would work for all the other files in the set the same way. I was not aware of the fact that the original HTML is buggy, nor that the texts are partly outdated and need maintenance.
OK, since we all do know that anticapitalista is short of time, I guess it’s up to us community people to improve them. You stated above, that the en_GB translation would be insufficient. My plan was, to use this “translation” as a place to fix the issues we have identified, and re-inject the results into the original file set later, once completed, e.g. by using the en_GB “translation” as the resource file at transifex then. Actually I noticed you can easily add complete paragraphs this way if missing, by including the respective HTML structure into a translation. This works, at least to a certain degree, if you know HTML.
The correct process to update EN text would be to edit the “.adoc” files, and run the page generating again,
If the existing HTML FAQ files actually are a result of running an automated process, I’d not trust in a recreation from scratch; the arbitrary bugs we have spotted in HTML structure will then be repeated another time simply.
Windows is like a submarine. Open a window and serious problems will start.