Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!usc!apple!agate!ucbvax!CUNYVM.CUNY.EDU!PIRARD%vm1.ulg.ac.be
From: PIRARD%vm1.ulg.ac.be@CUNYVM.CUNY.EDU (Andr'e PIRARD)
Newsgroups: comp.protocols.tcp-ip.ibmpc
Subject: Re:  Desqview/X
Message-ID: <9011210539.aa12430@louie.udel.edu>
Date: 21 Nov 90 10:18:16 GMT
References: <fks@FTP.COM>
Sender: daemon@ucbvax.BERKELEY.EDU
Organization: The Internet
Lines: 26

On Wed, 14 Nov 90 11:11:02 -0500 Frances Selkirk said:
>Desqview/X runs over our PC/TCP kernel. I can't tell you any more than
>that, because we haven't got a copy, yet. It isn't on our third party
>software list, because we haven't tested it. Hopefully, we will get to
>test it ourselves, soon.

(The kernel is the TCP driver.)
It seems rational indeed that they use an (existing) API to open X connections.
In fact, they say DESQview/X not only provides an X server, but also
an API for multitasked DOS X applications to display locally or remote.
And that a "plain" DOS application output can be converted to X and display
on a remote terminal (but too slow for bitmaps). Multi user DOS in fact...
What I am very anxious to know is the way they do provide multi access to TCP.
Does the TCP multiplexing inside DESQview occur at the X API, and DESQview
itself would be the only user of the TCP API (excluding its use by other
TCP/IP applications, very sad).
Or does the multiplexing occur at the TCP API level (and the X API uses that
multiplexing on behalf of its callers)?
The latest would mean that we would finally have got true TCP/IP under DOS,
I mean usable by multiple applications concurrently.

I find this even more important than the announced X capabilities.

Andr'e PIRARD             SEGI, Univ. de Li`ege
B26 - Sart Tilman         B-4000 Li`ege 1 (Belgium)
pirard@vm1.ulg.ac.be  or  PIRARD%BLIULG11.BITNET@CUNYVM.CUNY.EDU