Path: utzoo!utgpu!news-server.csri.toronto.edu!rpi!usc!cs.utexas.edu!uunet!mcsun!ukc!warwick!nott-cs!piaggio!anw From: anw@maths.nott.ac.uk (Dr A. N. Walker) Newsgroups: comp.lang.misc Subject: Re: Algol68 Message-ID: <1991Mar27.174552.4770@maths.nott.ac.uk> Date: 27 Mar 91 17:45:52 GMT References: <9168@castle.ed.ac.uk> <46036@ut-emx.uucp> <20109@alice.att.com> Reply-To: anw@maths.nott.ac.uk (Dr A. N. Walker) Organization: Maths Dept., Nott'm Univ., UK. Lines: 61 In article <20109@alice.att.com> bs@alice.att.com (Bjarne Stroustrup) writes: [Much that is true about the reasons for Algol's relative demise. However, ...] > - Algol68 was a decade in the making before implementations were > beginning to appear The Malvern compiler appeared in 1972, give or take a year [depends a little on what "appear" means]. There were other implementations by around 1975-6. > - and for many (most?) machines implementations > has yet to become widely available. You just have to know who to ask! There are several re-targetable implementations which could easily be put onto any [reasonable] specified machine. As there is a version for the Amsterdam Compiler Kit, for example, there is, by inference, an existing implementation for any machine which has C [and can therefore run the ACK emulator]. "Widely available" is, of course, the killer. > Algol68 did not emphasize [encapsulation > and modularity]. Somewhat to the contrary, the Algol report was the first place where I remember seeing *in a language definition* the idea that a user-program ran in an environment which was defined by the system- and user-libraries. The mechanisms of "keep lists" and of "holes" and "import/export" loom large in the user manuals for actual implementations. Admittedly, it all looks somewhat primitive today. > - Algol68 had a main-frame bias at a time where first mini-computers > and then micros emerged as the major growth areas for programmers. The 68S implementation was designed specifically for minis, at a time when, really, there were very few high-level languages available for them. > Had there been a K&R >for Algol68 and a cheap and easily portable implementation of Algol68 available >for a PDP11 in 1975 or so it would have swept the world covering the areas >where Pascal and C thrived. However, there wasn't such a book and there wasn't >such an implementation - and it is doubtful if there could have been. We were running 68S on our PDP-11 in the late 70's. It *had* been ported to Unix from something-or-other, I *think* in 1976. It failed to sweep the world. There *were* good books on Algol, there *were* cheap (at least to universities!) portable implementations. They didn't come with source code, and they were too difficult for undergraduates to play with. Compare Pascal. Shame. For that matter, compare Ada, and wonder. You omit another prime reason. There was too much bickering. People take C or Pascal, warts and all; they may wish something was different, but no-one denied the initial rights of DMR&KT and NW to define C and Pascal as they wanted. The Algol community was riven by schism after schism, and it really didn't help to have distinguished computer scientist after distinguished computer scientist quibbling, too often from ignorance, about features of Algol. -- Andy Walker, Maths Dept., Nott'm Univ., UK. anw@maths.nott.ac.uk