Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!sun-barr!apple!motcsd!mcdcup!mcdchg!ddsw1!igloo!nevin From: nevin@igloo.scum.com (Nevin Liber) Newsgroups: comp.lang.misc Subject: Re: How to make a language downward-extensible? Message-ID: <2873@igloo.scum.com> Date: 3 Oct 90 02:31:34 GMT References: <1990Sep24.160705.21113@newcastle.ac.uk> <9363:Sep2521:41:1290@kramden.acf.nyu.edu> <7935@scolex.sco.COM> <29047:Sep2816:51:129 Reply-To: nevin@igloo.UUCP (Nevin Liber) Organization: Igloo public access unix, Northbrook IL Lines: 26 In article stephen@estragon.uchicago.edu (Stephen P Spackman) writes: >We already trust the compiler to do much of our optimisation. Making >that compiler smarter, possibly interactively, is what we want to do >next. I definitely agree with you that the solution to this problem is to make our compilers smarter. However, I'm not quite sure that making them interactive is going to help. It assumes that the person who is compiling (and this person is not necessarily the author) a given program knows what is more optimal. I doubt that the majority of people who compile or build software have the time to learn the code and machine architectures well enough to make informed decisions about optimisations. Also, interactive compiling would probably take 5-10 times longer than batch compiling (just a guess; if someone knows of some real research done in this area, please post the results), and requires that someone devote a lot of time to a relatively boring job. Just some thoughts, -- NEVIN ":-)" LIBER nevin@igloo.scum.com or ..!gargoyle!igloo!nevin (708) 831-FLYS California, here I come! Public Service Announcement: Say NO to Rugs!