🚀 go-pugleaf

In article <12423@uflorida.cis.ufl.EDU> mfi@beach.cis.ufl.edu (Mark Interrante) writes:
>A question on the applicability of network window systems to do work
>from home.  More specifically, how dense is the _typical_
>communications between the host server?

GraphOn makes hi-perf graphics terminals with mice and local RasterOp
intelligence, and distributes hacks to Xlib to create a server that
will run on any BSD machine and spit GraphOn's protocol over a serial
port to their terminal.

This means that I could have, say, a GO terminal talking over a pair
of TrailBlazers doing LZ compression to a Pyramid on campus.  X
applications would run on the Pyramid or anywhere else, just as
before; it's just that the server would be running on the Pyramid,
too, rather than on the Sun on my desk.

I've wondered about how difficult it would be to do the same thing
with NeWS as with X on the GraphOn terminals.  You wouldn't have the
compactness of expression in PostScript over the slowest link (from
the server to its graphics engine), though clients would still talk PS
to the server.

One normally thinks of the server-graphics engine link as being very
fast (DMA to a frame buffer) and the client-server link as potentially
much slower (serial line or network IPC speed); perhaps turning around
those constraints would cause re-thinking of the way we view clients
and servers.

Has anybody tried any of this stuff?  The PO for our NeWS 1.1 sources
hasn't gone out yet, and I can't convince "them" to get me a GO
terminal just to play with (yet :-).  Or is there a better scheme?
-=-
 Bob Sutterfield, Department of Computer and Information Science
 The Ohio State University; 2036 Neil Ave. Columbus OH USA 43210-1277
 bob@cis.ohio-state.edu or ...!cbosgd!osu-cis!bob