I had similar problems downloading the port of GAP for the Mac from
umanitoba : for a few times during the first 10 days after the announcement
the connection was *very* slow and broke after transmitting between
8 and 80K. I stopped trying, and tried again maybe two weeks later
without any problem. Since I am using Xferit 1.5 in hqx mode, decoding
is simultaneous with transmition - I can't help you about the unbinhex
problem.

If it can be of help, I put GAP in anonymous ftp :

machine : fpssun19.unige.ch
login : anonymous
dir/file : outgoing/GAP31.sea.hqx

I will leave it there for a while. This is the Mac port, of course.
The on-line doc and the smallest group libraries are included (e.g.
NOT the two-groups, nor the character tables). There are also
some group definitions for some sporadic groups that were
provided with the previous release of GAP but were (as far as I
know) not included in the last release. The syntax has changed,
but one may plausibly scavenge generating permutations from
the code (I am not sure this is wise, though... ).

the .hqx is 1.8M, the .sea fits on a diskette, although the
uncompressed files take close to 5M (this was my first use of
the Stuffit Lite shareware; the package then started to boast
about its impressive feat each time I started it to push me
into registering).

I have had no time to play with GAP since that time, but forwarded
it to a friend. On the phone yesterday, he "complained" on the
"added workload", but commented on the human interface of the Mac
port.

As it is, GAP outputs to a window with no sliders, and thus no
history. The THINK C wrapper allows to copy this output into
a file, but that file can only be consulted after exiting GAP.
He tells me that adding a slider to the window might simply
be a matter of putting the right resource into the GAP application.
I am doubtful. Does anyone *know* what would be involved ?
This would be a huge improvement for someone working with
an SE/30 !

BTW, is the source for the Mac port available ?

Regards,
                                    10   2+1     5  2
Boris Borcic                     1+2  -10       2 -5
School of Psy. & Ed. Sci.        ----------  = ------ = 1
U. of Geneva                         1     -1    2+5
                                 2*10 +10*2