Re: Proposal for a Guile binary file format

[Klaus Schilling <Klaus dot Schilling at munich dot netsurf dot de>]

Lynn Winebarger writes:
 >    As for compiling procedures involving eval, I'd imagine it would have a
 > lot to do with how convolved the construction of the expression you're
 > eval'ing is, if it's available at compile time (i.e. not interactively
 > input), and how much computation you're willing to spend on abstract
 > interpretation to gather up the potential eval'ed code.  

Naos is a game that requires saving code generated at run-time and being
run possibly in one of the following sessions. Only a small part of the
code will probably be needed in a session, and it's close to unpredictable
whether and when a particular piece of code, besides the game core,
is needed.
The game administrator should also be able to edit part of the code
at run time and fix it. The size of runtime-generated objects is also
largely dependant on user interaction and pseudo-random. I don't see
any hope for such a beast being anywhere near to statically analyzable.

Klaus Schilling