From - Thu Nov 12 18:12:48 1998
Received: from db.stanford.edu (Seaotter.Stanford.EDU [171.64.75.89])
	by DB.Stanford.EDU (8.8.8/8.8.8) with ESMTP id SAA01844
	for <chaims@db.stanford.edu>; Thu, 12 Nov 1998 18:04:57 -0800
Message-ID: <364B947C.F8EC214B@db.stanford.edu>
Date: Thu, 12 Nov 1998 18:07:56 -0800
From: Dorothea Beringer <beringer@DB.Stanford.EDU>
X-Mailer: Mozilla 4.05 [en] (Win95; I)
MIME-Version: 1.0
To: chaims@DB.Stanford.EDU
Subject: minutes of last chaims meeting
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
X-Mozilla-Status: 8001

Agenda[ Discussion of control data

Minutes[ 

Contradictions in the basic concepts:
- optimization  -- just composition
- opaque data   -- user data can become control data, CHAIMS type

Suggestion, clear seperation between user-data and control-data:
- user can be computed into user
- control can be computed into control
- user can become control
- control can never become user
Questions:
a) control-data can it be only boolean (==> this would be me solution 1
with the only control-data type decisionflag and operations only on
desicionflag?), or also more types?
b) ESTIMATE: does it always return control-data or always return
user-data
c) megamodules and control-data: seperate megamodules for control-data?
control-data allowed as parameters for INVOKE, and EXTRACT, but all of
them have then to be control-data 





-- 
Dorothea Beringer   Stanford University   beringer@db.stanford.edu 
          http://www-db.stanford.edu/people/beringer.html
--