Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!mnetor!uunet!seismo!sundc!pitstop!sun!decwrl!kent
From: kent@decwrl.dec.com (Christopher A. Kent)
Newsgroups: comp.sys.mac
Subject: Re: X-windows for the Mac? (question)
Message-ID: <119@bacchus.DEC.COM>
Date: Thu, 5-Nov-87 13:38:22 EST
Article-I.D.: bacchus.119
Posted: Thu Nov  5 13:38:22 1987
Date-Received: Sun, 8-Nov-87 03:32:47 EST
References: <16518@gatech.edu> <3083@husc6.UUCP> <383ec1cf.b0a1@apollo.uucp> <277@algol.sol.SPS.TRW.COM>
Reply-To: kent@decwrl.UUCP (Christopher A. Kent)
Organization: DEC Western Research Laboratory
Lines: 25
Summary: let a larger host do most of the server work

Given the power of the Mac toolbox, it seems that any X implementation would
benefit from using the toolbox. Given the "thinness" of most Mac connections
to the outside world (at worst 1200 baud, at best AppleTalk speeds, ignoring
Ethernet cards for now) it seems that sending the full X protocol to the Mac
would be a loss.

I think the way to implement X for non A/UX machines is the way it's been
done for the so-called "X terminals". Some large host (a VAX or Sun) runs
half of the X server -- the half that speaks to clients, manages fonts, etc.
It speaks to the Mac with a lightweight protocol that implements the X
functions via the toolbox. The server-side manages font translations,
intercepts bitmaps, translates input events back and forth, provides
virtual memory for keeping really large resources around. The Mac just
paints bits, keeps some small bitmaps around, and handles input events.

This would allow all properly written X applications to work just fine
on the Mac, taking advantage of all the work that's gone into the toolbox.

Comments?

Cheers,
chris
-- 
Chris Kent	Western Research Laboratory	Digital Equipment Corporation
kent@decwrl.dec.com	decwrl!kent			(415) 853-6639