Path: utzoo!utgpu!news-server.csri.toronto.edu!rpi!usc!snorkelwacker.mit.edu!bloom-beacon!eru!hagbard!sunic!mcsun!ukc!icdoc!qmw-cs!keithc From: keithc@cs.qmw.ac.uk (Keith Clarke;W208a) Newsgroups: comp.lang.misc Subject: Re: Algol68 Message-ID: <3009@redstar.cs.qmw.ac.uk> Date: 21 Mar 91 13:45:41 GMT References: <3787@bruce.cs.monash.OZ.AU> Sender: usenet@cs.qmw.ac.uk Lines: 22 Nntp-Posting-Host: aux32 In <1991Mar20.154216.26337@cl.cam.ac.uk> nmm@cl.cam.ac.uk (Nick Maclaren) writes: >Overall, it was the fact that people REGARDED it as unbelievably complex >that killed it, because the few compilers that were produced were rarely >brought up to production quality. They could have been, and the world >of programming languages would now be totally different .... Totally? I'm surprised at people at Edinburgh & Cambridge being so, well, downbeat about this. My feeling is that ML is everything one could want in a child of Algol68 - the things that aren't there are the horrible stropping rules, the coercions, the i/o.. (Nick's list, I think). I can add one more Algol68 misfeature that ML fixes, & that's the stack-implementation of closures. I once got a 5 cm hex dump from an ICL 2900 for having the temerity to try out whatever the Algol68 is/was for fun twice f = fn x=>f(f(x)); (twice sin 0.5) I can't think of anything Algol68 has that I miss in ML. Oh, perhaps multi-dimensional arrays & slicing, if I still did that kind of programming. -- Keith Clarke QMW, University of London.