From mdp2@michaelpoole.demon.co.uk Sun Oct 31 15:49:06 2004 From: Michael Poole To: Lawrence Dickson Cc: occam-com@ukc.ac.uk Date: Tue, 19 Dec 2000 12:46:49 +0000 Subject: New language syntax MIME-Version: 1.0 X-Mailer: Turnpike Integrated Version 4.02 S <6hJWhgvJfFYyxBJ0C0BIDTGGuI> Message-ID: Dear all, I am so pleased that occam-com has for once become an interesting forum to be a listener to! I could never develop any enthusiasm for Java. A few points: 1) Could I encourage posters use the subject line properly? Why on earth conduct a discussion on syntax under a heading on automotive software? Only Jim gas got it right and I am using his subject line. 2) Ian Page mentions: >Tony Hoare's assertion that "occam really should have *been* C" I don't understand what this is supposed to mean, but do remember life at INMOS before there was an occam and one of things Tony Hoare did for us was to propose a language that looked very much like BCPL. C had not taken over the world in those days! 3) I agree with Rick that agreeing a new syntax will be much more difficult than implementing a parser. One thing that I am sure that David May got right was to keep these matters in his own tight control and stamp hardly on attempts by others to spend his valuable time in discussing their contributions to language design. There may be a few features that we second class citizens (never been to Warwick University) eventually got into the language, (eg CASE construct, counted array protocol) but not many! 4) It has not been made clear in this forum what is the impetus for this bout of syntax discussion, but I hope that when we have all had our say someone of appropriate calibre will shut himself/herself away and do a single handed job. Any language that is too big for this to be possible could not be a worthy successor to occam. 5) Finally my personal votes for points that have been discussed: a) uppercase keywords are a good thing - PHW could have a special editor that uses underlined lower-case on the screen only! b) any syntax, or stylistic convention, that has matching grouping brackets at different positions on the line is BAD. c) curly brackets are too faint/small to be used as statement brackets d) indentation is a good thing - I like Jim's suggestion on this that 2 spaces should not be the only unit allowed. e) a folding editor is the best way to look at long structured texts f) any attempt to curry favour with EMACS users will alienate vi users, etc 6) In my opinion the worst mistake we made at INMOS in promoting the Folding editor was cutting off existing programming teams who had an established tool base from these tools by not making the TDS representation of occam programs an ordinary text file. In TDS3 (IMS D700E) I tried to rectify this, but was not allowed by management to make this a documented feature. The TDS was then killed and Julian had another battle to get his folding editor into the toolsets. No wonder occam and folding got a bad name when the world saw the company that had invented them lacking the confidence to continue to promote them. -- Michael Poole 15 Heather Close, STROUD, GL5 3QY 01453 759557