Path: utzoo!utgpu!news-server.csri.toronto.edu!mailrus!wuarchive!zaphod.mps.ohio-state.edu!usc!ucsd!ogicse!milton!brucec%phoebus.phoebus.labs.tek.com@RELAY.CS.NET From: brucec%phoebus.phoebus.labs.tek.com@RELAY.CS.NET Newsgroups: sci.virtual-worlds Subject: Re: world modelling Message-ID: Date: 16 Aug 90 18:31:57 GMT References: <3084@cuisun.unige.ch> Sender: hlab@milton.u.washington.edu Organization: Tektronix Inc. Lines: 36 Approved: hitl@hardy.u.washington.edu In-Reply-To: simon@cuisun.unige.ch's message of 12 Aug 90 16:21:00 GMT > >> From: Stan Shebs > I've designed and implemented a language for building virtual worlds to be > used for multi-player games. Unfortunately, in the push to be more general, > and despite some effort, I ended up reinventing a strongly-typed Scheme with > a set of predefined object types (objects, spaces, etc). I don't believe a > procedural language is entirely appropriate for this type of model building, > so am in the process of re-evaluation of the task, will maybe do a purely > declarative language (such as a constraint logic language). > > stan shebs > shebs@apple.com This looks like an interesting topic of discussion in itself: what programming models suit world design best. My own prejudice is that Stan is right as far as saying that "a procedural language is [not] entirely appropriate", but that a purely declarative language won't cut it either. I tend towards a constraint description layer over an action layer in which actions are described in some sort of polymorphic-procedural way, i.e., you declaratively state how parts of the model are constrained, and then state what actions will be taken to enforce the constraints, the actions to be in fact dynamically bound so that knowledge of the actions' implementation is not available to and not needed by the constraint declaration. The object of the polymorphism is to allow evolution of the model both as it is used and in redesign. --------------------------------------------------------------------------- NOTE: USE THIS ADDRESS TO REPLY, REPLY-TO IN HEADER MAY BE BROKEN! Bruce Cohen, Computer Research Lab email: brucec@tekcrl.labs.tek.com Tektronix Laboratories, Tektronix, Inc. phone: (503)627-5241 M/S 50-662, P.O. Box 500, Beaverton, OR 97077 -- --------------------------------------------------------------------------- NOTE: USE THIS ADDRESS TO REPLY, REPLY-TO IN HEADER MAY BE BROKEN! Bruce Cohen, Computer Research Lab email: brucec@tekcrl.labs.tek.com Tektronix Laboratories, Tektronix, Inc. phone: (503)627-5241 M/S 50-662, P.O. Box 500, Beaverton, OR 97077