Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP Posting-Version: version B 2.10.1 6/24/83; site pucc-i Path: utzoo!watmath!clyde!burl!mgnetp!ihnp4!inuxc!pur-ee!CS-Mordred!Pucc-H:Pucc-I:ags From: ags@pucc-i (Seaman) Newsgroups: net.lang.c Subject: Re: Selecting a Prog-Lang: Support for C Message-ID: <279@pucc-i> Date: Fri, 18-May-84 01:00:35 EDT Article-I.D.: pucc-i.279 Posted: Fri May 18 01:00:35 1984 Date-Received: Sat, 19-May-84 00:43:28 EDT References: <132@ihnet.UUCP> <2590@ncsu.UUCP> <278@pucc-i> <55@stat-l> Organization: Purdue University Computing Center Lines: 31 > Dave, no one mentioned Modula-2 because it's still an experiment, > and, besides, the article was titled (see above) "...Support for C"; Then why did the article mention Pascal, Ada, FORTRAN and (*gasp*) COBOL? Seriously, I recognize that this group is net.lang.c and not net.lang.mod2, but as long as other people are comparing languages, it seems worthwhile to have some good languages available for comparison. Modula-2 combines the simplicity and readability of Pascal, the low-level facilities and expressive power of C, and the information-hiding, high-level structuring and separate compilation capabilities of Ada in a single language. I listened to Brian Kernighan's talk here in which he mentioned C++, the new experimental version of C. C++ looks like an attempt to add some of Modula-2's features to C (primarily information hiding), but it lacks Modula-2's built-in facilities for separating the "definition" and "implementation" portions of a module and providing automatic version control, both at compile time and at load time. There is a fundamental difference between "adding on" features to a language and designing them in from the beginning, a point which Wirth obviously appreciated when he abandoned Pascal and started over with Modula-2. Remember the programmer's axiom: "Build one to throw away." -- Dave Seaman ..!pur-ee!pucc-i:ags "Against people who give vent to their loquacity by extraneous bombastic circumlocution."