Many people have already discovered it and sent me objections, as
a result of which I've today posted an updated version. So, having
already lost the element of surprise, can I encourage porters and
other interested Inform people to take a look, and see if the code
can be compiled properly on different machines?
Bob Newell and I are already exchanging frank and full emails on
the subject of how awful the code is for small PCs, always the
most troublesome architecture for Inform!
In theory Inform 5 is very similar to Inform 4 internally, so the
process of compiling it shouldn't be too much worse.
In any case, Inform 5 is considerably better than Inform 4: I try
not to make updates unless I have something significant to add.
Try looking in:
inf5src.tar Tar directory of ANSI C source
inf5man.tar ...of the manual file on its own
inf5lib.tar ...of the Library files and sample games
There will be a proper announcement, and may well be another slight
update on the code to fix any problems people email me with.
Andrew Fabbro If laws are outlawed,
welts...@umich.edu only outlaws will
University of Michigan have laws. Fnord.
Finger afa...@churchst.ccs.itd.umich.edu for PGP public key.
PGPprint: 87 41 65 E0 C2 51 9F E5 A9 44 ED A6 6B 16 76 9E
NSA bait: assassinate uranium dreamland CIA p.o.e. zimmerman
I did both a 386+ port (DJGPP) and a generic PC port. They were pretty
difficult but thankfully they are in good shape now.