> How can companies protect their bytecode, at least their modules, from reverse
> engineering?

I fear that there is probably no way to prevent this: as soon as you have a
virtual machine that can (somehow) decode the byte code - it has to if it
wants to execute anything useful - you know the internal structure of the
code.

Since all the sources of INRIA are open (i.e. also the protocols),
everybody could "fake" a virtual machine that, instead of just executing
the code, emits the decoded byte code.

That's also the reason why you can (in theory) always break any kind of
copy protection: as soon as a program runs on your machine, you can track
any information about it you want. That may be lots of work (admittedly
less work for byte code) - but it can be done.

Regards,
Markus Mottl

-- 
Markus Mottl, mottl@miss.wu-wien.ac.at, http://miss.wu-wien.ac.at/~mottl