Path: utzoo!utgpu!news-server.csri.toronto.edu!mailrus!cs.utexas.edu!swrinde!zaphod.mps.ohio-state.edu!sol.ctr.columbia.edu!ucselx.sdsu.edu!bionet!arisia!dourish From: dourish@arisia.Xerox.COM (Paul Dourish) Newsgroups: comp.lang.lisp Subject: Re: why lisp is dead Message-ID: <7954@arisia.Xerox.COM> Date: 10 Apr 90 14:37:31 GMT References: <485@paradigm.com> <14980@cbnewsc.ATT.COM> <6008@udccvax1.acs.udel.EDU> Reply-To: dourish@arisia.UUCP (Paul Dourish) Distribution: usa Organization: Xerox Palo Alto Research Center Lines: 22 In article <6008@udccvax1.acs.udel.EDU> captain@vax1.acs.udel.EDU (Jeffrey Kirk) writes: >Support your own LISP, take one of the public-domain LISPS and add/subtract >what you want. This solves 1-3 above. Scarcely the point.... it proves some sort of "inferiority" for Lisp if I have to do this. I don't have to support my own C, for instance. It's certainly a common criticism of Lisp that the run-time support is large, making it difficult to deliver applications to end-users. What are people's experiences with distillation tools that remove unwanted code and give smaller images for delivery systems? Gjc's right that this is a vicious circle; unfortunately, what it needs is some supplier who can afford to break it, perhaps in conjunction with a Lisp house prepared to lower the financial cost of run-time licenses. (Didn't Symbolics start to address these sorts of problems with a compatability system for PC-type machines?) -- Paul Dourish, Rank Xerox EuroPARC, Cambridge UK "Ain't they got no barbers where you come from, boy?"