Need a 2-way alternative to popen()

Don Libes libes at cme.nist.gov
Fri Mar 2 03:46:08 AEST 1990


The following article originally appeared in comp.unix.questions.
In article <9539 at portia.Stanford.EDU> bugboy at portia.Stanford.EDU (Michael Frank) writes:
> [discussion about how to do 2-way communication - deleted]
>
>(Alternatively, I wish I could just have my program pretend to be me
>logged in on some tty; that would accomplish all of this in one step.)

We have submitted a paper to the upcoming USENIX describing a very
elegant system that solves bugboy's "pretend" problem as a simple case
of what it is capable of.  We will release the code, but not before
the paper is accepted.  (Otherwise the committee can say the work
isn't "new".)

Even if the paper is accepted, we're not sure it would be good style
to release the code before the conference itself.  Thoughts on this?
(We have sufficient beta testers, but I suppose we would be willing to
take on a few more desperate individuals, like bugboy.)

If the paper is not accepted, we're going to wait 6 months and
resubmit it to the next USENIX (knowing that the program committee
made a mistake - hey, I've been on the committee, I know how it works).
Admittedly, we realized the deadline at the last moment and scratched
together an abstract that might not be very good.  Plus, we've got new
results that don't appear in the submitted abstract.  So our next
submission can be better.

(We would not normally spill our private thoughts out, but it really is
nice work, we are quite pleased with it, and it is hard to contain
ourselves.  And we just want to be recognized as the egotists we are!)

Wish us luck.

Don Libes          libes at cme.nist.gov      ...!uunet!cme-durer!libes



More information about the Comp.unix.wizards mailing list