[comp.dcom.modems] obvious opinion on What Needs to be Done

ulmo@ssyx.ucsc.edu (scritzifchisted ulmo qzutvchsxik) (10/03/88)

Most of the modems out there today fall into what could be said to
be three categories; great, so-so, and horrible.  That is not perhaps
what most system designers are used to ... they probably used to
usually frown down on what today are better than so-so modems.
However, because of cost, need and popularity, I would think that
these qualms could be overcome by much more intelligent software
which would have not only a robust set of tools for
configuring dynamic crappy (and high quality) modem use, but a sort of
awareness of the ups & downs of any particular modem.  Such software
could conceivably be simple enough to implement that it could come
as part of standard OS distributions.  Thus, we wouldn't have to constantly
be fixing a myriad of poorly and averagely manufactured modems
and OS's.

This obviously gets into the realm of changing source code for uucp,
many kernels, etc.  But, that's what I think needs to be done,
despite the fact that many modem manufacturers are kicking us in the face
(for a cheap price).