BABYL OPTIONS:
Version: 5
Labels:
Note:   This is the header of an rmail file.
Note:   If you are seeing it in rmail,
Note:    it means the file has no messages in it.

1,,
Received: from ATHENA-AS-WELL.MIT.EDU by po7.MIT.EDU (5.61/4.7) id AA22731; Tue, 13 Apr 93 14:30:39 EDT
Received: from W20-575-48.MIT.EDU by Athena.MIT.EDU with SMTP
	id AA23241; Tue, 13 Apr 93 14:30:37 EDT
From: kkkken@Athena.MIT.EDU
Received: by w20-575-48 (5.57/4.7) id AA11896; Tue, 13 Apr 93 14:30:34 -0400
Date: Tue, 13 Apr 93 14:30:34 -0400
Message-Id: <9304131830.AA11896@w20-575-48>
To: gill@ai.MIT.EDU
Cc: kkkken@Athena.MIT.EDU
Subject: Comments

*** EOOH ***
From: kkkken@Athena.MIT.EDU
Date: Tue, 13 Apr 93 14:30:34 -0400
To: gill@ai.MIT.EDU
Cc: kkkken@Athena.MIT.EDU
Subject: Comments


I'm finally going over my thesis...!

The simple corrections you made I agree with 100%, and have
incorporated into the document.

Other corrections I don't fully understand, and I'd like to go over
with you sometime.  Specifically:

On page 10, you ask why don't I describe a debugging environment where
the target hardware is connected to the workstation with a "generic
connection mechanism", or where the workstation is connected to a
"generic hardware simulator", and the answer is because I don't know
what either of the things in quotes are!  :) so I'd like to talk about
that sometime.

On page 16, you ask why I don't explain events in a time-ordered way,
so that the next event done is the earliest scheduled.  Answers are
that there is no global ordering of events possible, since events can
be generated simultaneously, and that even if one is generated before
another SOTEST does not guarantee that they will be processed in that
order, and there is in fact no need for such a guaratee.  SOTEST's
scheduling is the most subtle aspect of the project, and I thought
about it carefully before settling on what I have.  I'd like to talk
about this more.

You said that I should flush out SOTEST implementation just short of
actual code.  I am unclear on the level of detail that you mean here.
Certainly I wouldn't want to describe every data structure the code
uses, or describe the algorithm used parse .ABS files or to build an
internal representation of Microsoft debugging information.  I'd like
to talk about this too.

Thanks!!

	-Ken