Actually, I meant the existing MS-DOS/MAC/whatever client programs.
I suppose that the fact that I have looked at these documents makes me
ineligible to give anyone hints as to how things work, or even to start
on a project to independently reverse-engineer the data, since I have
some idea (it's been a while since I've looked at them) how the data is
structured.
I've been trying to think if there's any way at all that I can help, hint,
whatever, but everything I can think of ends up violating the non-disclosure.
So, I guess I'll probably just end up lurking on the list... maybe I'll
end up using the software that someone else finally comes up with :-)
I do have one suggestion, albeit a nasty one: create two teams of programmers
working on the project -- one team reverse-engineers the data stream or one
of the client programs, and writes their own spec sheets on them. The other
team writes a new client based on the specs.
Unfortunately, as Lewis mentioned, reverse-engineering the data is actually
more difficult than I optimistically thought (I looked at one of my old
captured data streams and realized that it was more complex than I
remembered), and reverse-engineering a client program itself is a full-time
painful job, since you'd only have a disassembled listing to work from.
This is a difficult situation to be in, but I think that all I can do now
is wish all the developers luck with the project.
Bill
-- Bill Fenner fenner@jazz.psu.edu ..psuvax1!hogbbs!wcfpc!wcf wcf@hogbbs.scol.pa.us (+1 814 238-9633 v.32bis)