The interest for a reverse migration should be placed not on a migration between formalisms, but on a migration taking into consideration linguistic engineering aspects. The fact is that the LFG community disposes of enough German LFG grammars, and so the situation we had for ALEP is not present: for the German ALEP grammar we wanted to reuse as much as possible already existent grammar resources. But on the other side, it would be interesting to see how some work on linguistic engineering done within the context of the LS-GRAM project could be reused within a LFG grammar development environment. This is actual since the step done from the GWB to the XLE platform definitely permits to place the task of further developing a LFG grammar into the context of the broader aspect of linguistic engineering. But for sure, I didn't have time to experiment anything for this. So I will just shortly present some results of the LS-GRAM project, wondering what can be done in order to achieve a general strategy for the integration of linguistic engineering tools into unification-based grammar development environments.