Path: news.daimi.aau.dk!news-feed.inet.tele.dk!bofh.vszbr.cz!uninett.no!sn.no!news-feed.ifi.uio.no!ifi.uio.no!mikkelor
From: mikkelor@ifi.uio.no (Mikkel Orheim)
Newsgroups: comp.lang.beta
Subject: Re: Alternative interaction between conc. objects
Date: 10 Feb 1998 19:13:57 +0100
Organization: Dept. of Informatics, University of Oslo, Norway
Lines: 32
Sender: mikkelor@nntp.ifi.uio.no
Message-ID: <6bq5d5$90$1@sessrymnir.ifi.uio.no>
References: <6b9cj3$g7n$1@sessrymnir.ifi.uio.no> <6bdj38$s35$1@nf.aau.dk>
Reply-To: mikkelor@logi.ifi.uio.no
NNTP-Posting-Host: sessrymnir.ifi.uio.no
Xref: news.daimi.aau.dk comp.lang.beta:11384


In article <6bdj38$s35$1@nf.aau.dk>, olm@daimi.aau.dk (Ole Lehrmann Madsen) writes:
> In <6b9cj3$g7n$1@sessrymnir.ifi.uio.no> mikkelor@ifi.uio.no (Mikkel Orheim) writes:
>
> I am not aware of implementation of a scheme similar to the one
> suggested by Svend Frĝlund. I will, however, be very interested
> in seeing such an implementation in BETA.
> 
> The BETA libraries do, however, have an implementation of
> Ada-like rendezvous where an object may decide wheter or not
> to accept an object. 

I am aware of this rendevouz mechanism, but I find it inadequate
when it comes to describing/implementing objects with the ability
to  "act" even when participating in a communication. What I'm 
trying to say, is that the "do-part" is "lost" because the 
communication is synchronized  - the recieving object has to 
"stand still"  until an incoming message corresponding to the 
acceptet port is recieved. 
This I find troublesome, and in my theses (which will be finished 
this monday) I propose an alternative scheme. 

--
(---------------)==========================((((((((()
Mikkel Orheim, Hovedfagsstudent v/Institutt for informatikk
URL: http://www.ifi.uio.no/~mikkelor/  
Tlf: 920 18 746 eller 67 13 07 64
-- 
(---------------)==========================((((((((()
Mikkel Orheim, Hovedfagsstudent v/Institutt for informatikk
URL: http://www.ifi.uio.no/~mikkelor/  
Tlf: 920 18 746 eller 67 13 07 64