Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!usc!elroy.jpl.nasa.gov!sdd.hp.com!spool.mu.edu!snorkelwacker.mit.edu!ira.uka.de!fauern!opal!wg
From: wg@opal.cs.tu-berlin.de (Wolfgang Grieskamp)
Newsgroups: comp.lang.functional
Subject: Re: Views revisited
Message-ID: <2738@opal.cs.tu-berlin.de>
Date: 22 Feb 91 21:54:02 GMT
References: <17992.9102181734@olympus.cs.hull.ac.uk>
Sender: news@opal.cs.tu-berlin.de
Reply-To: wg@opal.cs.tu-berlin.de
Followup-To: comp.lang.functional
Organization: Technical University of Berlin
Lines: 29
Nntp-Posting-Host: troll.cs.tu-berlin.de

rst@cs.hull.ac.uk (Rob Turner) writes:

>From the example given, "views" look enormously like algebraic
>specifications of abstract data types.

>Is this correct? 

I think so. But the algebraic specification framework doesnt 
say much about implementing models, rather about classes, initiality, 
and determinination of models up to isomorphism :-}

The pragmatic correlation between specification and high-level 
implementation via the functional paradigm isn't really so researched 
and especially valified in praxis as one might think after the first 
"view".

>Is this another instance of inventing a new name for
>something which already exists, in order to confuse everyone?

If there is only one view and it should satisfy reachability and is
free generating, you might be right to better call it type - if thats
what you mentioned. If views havent to satisfy all this stuff, their 
intention can be better expressed by subsorting, is my current 
conclusion.

>Rob

bye
wg