It was thus said that the Great Brian Chase once stated:
Whether there are enough people with enough resources and interest to
maintain this sort of endeavour, I don't know. My impression,
especially after reading some of the _Managing uucp and Usenet_ book, is
that UUCP died out in part because of there being too many different
ways to configure a system depending on your hardware, your OS, your
version of UUCP, and your modem type. Some of the can be eliminated by
using simple, and now prevalent, TCP/IP connections. For those unable
or unwiling to taint the purity of this experiment, dial-up gateways
would be needed.
In my experience, setting up UUCP is a bitch. There are two major UUCP
systems, each with slightly different semantics and protocol support and the
whole modem thing under Unix just sucks (sucks less now, but UUCP still
probably wants things done the Old Bad Way (TM)). I remember spending hours
trying to get UUCP working for clients (at an ISP years ago) and it was
painful.
On the plus side, once it was set up and worked, it worked and you never
had to touch it again (thankfully!). But getting it working was way too
difficult. Good thing it died.
Actually, I'm sort of surprised we aren't
already running an extensive
UUCP network... The only reason I can guess as to /why that is/ would be
that everyone's too busy collecting old computer systems to actually do
something on a grander scale with them.
It's hard to configure and you have to work out UUCP routing are two
things I can think of that make it not worth the hassle. At least for me.
-spc (It always seemed more complicated than it needs to be)