Received: by ATHENA-PO-2.MIT.EDU (5.45/4.7) id AA05910; Wed, 22 Feb 89 17:12:43 EST
Received: by ATHENA.MIT.EDU (5.45/4.7) id AA17118; Wed, 22 Feb 89 17:12:17 EST
Received: by decwrl.dec.com (5.54.5/4.7.34)
	id AA19133; Wed, 22 Feb 89 14:11:58 PST
Date: Wed, 22 Feb 89 14:11:58 PST
Message-Id: <8902222211.AA19133@decwrl.dec.com>
Received: by decwrl.dec.com (5.54.5/4.7.34)
	for bgardner@athena.mit.edu; id AA19133; Wed, 22 Feb 89 14:11:58 PST
From: treese%crl.DEC@decwrl.dec.com (Win Treese)
To: ackerman@ATHENA.MIT.EDU (15-Feb-89 1843 EST)
Subject: Re: to some of Ack's comments, questions, etc. 

 
> I think the most reasonable way to do the scheduler to avoid your
> intra-message problems is to just give the person some time slice
> (fixed beforehand) of the disks.  They can do anything they want
> within that time-slice.  After that, they have to request again, 
> I suppose.  (I haven't thought about wanting more time.)
 
They get a single time slice, for the purposes of SIGGRAPH.
 
> This may lead to some UI ugliness.
>  
> Also, I hadn't thought about multiple windows (as in many multiple
> video windows) in a single message.  Only one can be moving video
> but any might go active in the time-slice.  This is a problem,
> so I'm glad you pointed this out.
 
Again, for the purposes of SIGGRAPH, there is a single video window, and only
one video segment per message.  We can think about the more general cases, but
I don't think we should spend too much time on them.
 
Anyway, I'm back for a few days -- I'm leaving Tuesday the 28th for the rest
of that week, and I'll be back at work on Monday.
 
	- Win
 
