Path: utzoo!utgpu!news-server.csri.toronto.edu!rpi!zaphod.mps.ohio-state.edu!wuarchive!udel!rochester!pt.cs.cmu.edu!zardoz.club.cc.cmu.edu!ddj
From: ddj@zardoz.club.cc.cmu.edu (Doug DeJulio)
Newsgroups: comp.sys.next
Subject: Re: (Ne)X(T) Terminals---a hot product idea?
Message-ID: <12805@pt.cs.cmu.edu>
Date: 26 Apr 91 12:53:23 GMT
References: <1991Apr25.084827.1475@math.ucla.edu>
Distribution: na
Organization: Castle Anthrax, Pittsburgh
Lines: 25

In article <1991Apr25.084827.1475@math.ucla.edu> barry@pico.math.ucla.edu (Barry Merriman) writes:
>(2) a home machine for people with a NeXT at work; if your server
>has SLip or PPP, a NeXT Terminal at home plus a 9600 baud
>modem would be a great way to stretch one NeXT computer to other
>locations.

Alas, no.  I've got a NeXTstation connected to my campus by a
19.2kbaud VJ CSLIP connection.  I've occasionally run apps over the
network.  It's *S*L*O*W*.  Unfortunately, NeXT apps seem in general to
be *much* worse about network bandwith than even X11 apps.  At
19.2kbaud, it's not usable.  That's the reason I keep MouseX around --
it's usable over a 19.2kbaud network, although barely.  I really need
to get MGR working on these things.

For folks writing apps, if you want them to be efficient over
low-performance networks, remember that minimizing the bandwidth is
far more important than sending efficient PostScript.  Use
dictionaries and small PostScript commands that do a lot.  Compress
bitmaps and have PostScript display routines that do the
uncompression.  Perhaps we can change the appkit shared library to do
all this stuff, and then a NextStep terminal could be viable over a
slow network.
-- 
Doug DeJulio
ddj@zardoz.club.cc.cmu.edu