From adrian.lawrence@computing-services.oxford.ac.uk Sun Oct 31 15:49:06 2004 From: A E Lawrence To: jm40@ukc.ac.uk Cc: occam-com@ukc.ac.uk Date: Tue, 05 Dec 2000 21:44:38 +0000 Subject: Re: (now completed) Slides from my talk at CPA2000 Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset=us-ascii MIME-Version: 1.0 X-Mailer: Mozilla 4.76 [en] (X11; U; Linux 2.2.18pre23 i686) Message-ID: <3A2D61C6.5993D78@oucs.ox.ac.uk> jm40@ukc.ac.uk wrote: > > > -----Original Message----- > > From: Campbell, John [mailto:John.Campbell@siinet.trw.com] > > Sent: 05 December 2000 17:44 > > To: 'Tom Locke'; occam-com@ukc.ac.uk; java-threads@ukc.ac.uk > > Subject: RE: (now completed) Slides from my talk at CPA2000 > > > > > > Tom and Others > > > > > >The problem is that it's not much of an object if you have to sever > > > >its existing channel connections before sending it. > > > > > > Could you elaborate on this? > > > > What I'm struggling to understand is how you avoid the problems > > attendant to aliasing if you transport an object with open > > connections. It seems to me that even if you manage to avoid > > aliasing of the *object* you're going to get it on the open > > *channels*. > > David May did some work on mobile processes which he presented at WoTUG-21. > I asked someone at Bristol for a copy of the relevant papers (an occam-like > langage called Icarus), but never heard anything more. It was pretty > interesting stuff though, worth looking at for this case. It is on Henk Muller's pages, I think. Try also http://www.cs.bris.ac.uk/Tools/Reports/Ps/2000-may.ps.gz , http://www.cs.bris.ac.uk/Tools/Reports/Ps/1998-may.ps.gz, http://www.cs.bris.ac.uk/Tools/Reports/Ps/1998-muller-0.ps.gz and there is "Using Channels for Multimedia Communication." with a note saying that if you want a copy, you should email David (David.May@bristol.ac.uk). Ditto "A Simple Protocol to Communicate Channels over Channels." and "Icarus language definition." > > I'm supposing that when you move the channel object (as opposed > > to a conventional OO object) out of its original context, the > > original context cannot communicate with it. That's what I meant > > by "sever its connections". Is there a way to limit access to > > a sent object such that you don't get aliasing, but it can > > still interact with its original environment? I think that you can get around any aliasing problems by using the same methods that I used in dealing with mobile variables. I can't remember how David and Henk managed that. Maybe Henk is listening and will remind us? Adrian -- Dr A E Lawrence