Bearing in mind that this approach wants to serve translators, it is not hard to think of possible lines of research.
Here are a few:
• Error analysis: if we know what errors are commonly made, we can analyze them and devise methods to help translators avoid them;
• Comparing grammatical structures in different languages so as to systematize the kinds of shift that will necessarily occur in translating between two specific languages, which in turn can lead to optimum solutions for particular situations, for particular types of text or context, etc.;
• The compilation of glossaries, terminology banks etc. as aids to translators; terminology banks usually concern specialized or technical language, and they can be huge, with all sorts of useful data thrown in;
• Improving the didactics of translator training by considering what knowledges need to be imparted and in what order, what kind of drills work best, etc.