🚀 go-pugleaf

In article <BvAz89.Cop@unix.amherst.edu>, twpierce@unix.amherst.edu (Tim Pierce)
writes:

|>In article <7441@lib.tmc.edu> jmaynard@oac.hsc.uth.tmc.edu (Jay Maynard)
|>writes:
|>>In article <1992Sep26.151349.1962@psg.com> randy@psg.com (Randy Bush)
|>writes:
|>>>mathew@mantis.co.uk (mathew) writes:
|>>>> It causes UUCP systems using slow modems on noisy lines to waste ...
|>>>           ^ broken
|>>            ^ standard
|>             ^ standard broken?

Perhaps we should all go back to using JCL and programming in FORTRAN IV?

This argument is a bit pointless since there is noboby who can make a decision
on the matter - execpt people who start posting large articles.

From my point of view it is rather daft to expect the whole USEnet to be driven
by the needs of modem users with crappy software. Equally it is unreasonable to
expect everyone on the net to have a 100 spec mark workstation.

However it is amazing what happens when people are forced into changes. Take the
IBM group here. After years of demanding that the whole computing environment of
the site be organised round their obsolete FORTRAN (its "standard" - right?),
their crappy communications etc etc, people finally said "sod it" and bought
workstations. Then the strange thing was that it suddenly became possible to
have long variable names, TCP/IP connections etc.

The reason why long articles *ARE GOING TO START APPEARING REGARDLESS* is that
we haven't even touched on some of the things we could (and should) start doing
with the net.

For example we could easily integrate hypertext into the communications
protocols, there are already World Wide Web browsers which are capable of
reading SGML and which have NNTP connections for reading USEnet. Within a year
or so they should be quite common. (telnet info.cern.ch, password web for demo)

When articles get split up they often can only be pieced back together manually.
I have programs which assist in this process, but even those only get to about
90% accuracy. There are many occasions in which one part in a ten part post gets
lost. In this case the whole article is useless. It would waste fewer resources
here if either the whole article arrived, or none of it did.

--

Phill Hallam-Baker