Comment 2 for bug 4551

We'd be happy to oblige when majority of free software stack switches to anything like that. Until that point, there's no reason to bother and lose advantages of the current system we have (i.e. if everything uses the *same* base language [doesn't matter which one it is], we can easily find matching translations and offer them as suggestions).

Unfortunately, a majority of free software stack (and especially, GNU gettext) is designed for English (and especially en_US). If we are to allow any other language as the base language (like en_GB), we'd have to allow others as well (eg. why not have Serbian or Chinese?), which is sometimes impossible without changing the core infrastructure.

Also, this is so low on our priority list that we won't have time to even consider this issue in the next few years.