I have an ideas list, a sort of todo list for when I have time to burn. One of the ideas on my list is a website that can help authors of open source software to translate their software into languages other than english. I have recently agreed to teach part time at Seneca and also I decided to take a break from looking for a full-time job for a couple of months. Which means I have time to burn :)
There are similar strings in many pieces of software, so theoretically if it’s translated in SoftwareA it should be easily translatable in SoftwareB. In practice however this is not the case. It is impractical for a project maintainer to find strings in other software that will look like his own. In fact in most cases it is difficult to even find the pot/po files in existing software because there are so many version control and release systems.
So the core of my idea is a database of existing translations that can be used to automatically translate some strings in new software. The open source maintainer would submit a .pot (template) file and get back partially translated .po files in a bunch of languages.
Over time this website can be expanded to provide general internationalization (i18n) advice, advice for which english strings to pick which the service will translate more easily, a network of translators who may be interested in volunteering to help a particular project, etc.
I’m going to try and record in this blog the issues that come up with designing, building, and marketing this website. That shouldn’t be much of a problem since I don’t have a hope in hell of ever making money from this project, it’s just community service :)