At 23:58 +0200 2001/05/13, Lars Hellström wrote: >Read Sections 8--12 (Section 12 in particular) of the Omega draft >documentation---that will answer you question more thoroughly that I bother >to do right now. Marcel's summary contains a reference to it. I saw it. Using the lingo of section 12, I think that Omega needs a few more modes, 32-bit characters, perhaps a few variable characters modes; possibly also the option of being able to read compressed files. As for sections 8-9, I think that reading byte codes might still be slow, because one is down on such a low level. On a fast machine, perhaps this does not matter. But this suggests that one may need to use C/C++ code translators, at least for some common character encodings. One possibility might be, instead of using byte codes, generate C/C++ source files as output, which can be compiled and linked (perhaps as DLL's) with the original program. The byte codes would still be great in situations where C/C++ compilers are not available. But only profiler tests can tell whether it is worth the effort. Hans Aberg