Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!wuarchive!zaphod.mps.ohio-state.edu!samsung!olivea!mintaka!bloom-beacon!eru!hagbard!sunic!ericom!tnetxa.ericsson.se!eds.ericsson.se!lmebgo From: lmebgo@eds.ericsson.se (Bengt G{llmo, L M Ericsson, Stockholm, Sweden) Newsgroups: comp.edu Subject: Re: sys & lang standardization? Message-ID: <17146@eds.ericsson.se> Date: 20 Nov 90 09:36:13 GMT References: <27318@cs.yale.edu> Lines: 63 In article <27318@cs.yale.edu>, bloom-debbie@cs.yale.edu (Debbie Bloom) writes: > In article hp0p+@andrew.cmu.edu (Hokkun Pang) writes: >>..... It seems to me that every month or so, >>there's a new (or major upgrade to an existing) operating system or >>programming language. So much energy are wasted (can I say this?) in >>learning new features and rewiting old codes. Still an undergrad, I have >>used the following languages already: Ada, Assembly, Basic, C, Cobol, >>Comal, Forth, Fortran, ML, Lisp, ML, Pascal, Prolog, Scheme. And all of >>them are still widely used. How many more do I need to learn? Do we know >>enough to know which direction(s) to go yet? ..... > Off the top of my head, I can think of two main reasons why there are so > many languages out there (still being used and being dreamed up anew): > > 1. There is a great deal of code out there written in old languages > (e.g., Cobol, Fortran) that would take an incredible amount of time to > rewrite in a new language. People don't want to spend this time (it is > not cost-efficient in many areas), so the old code gets reused and updated > in the old language. > > 2. Different languages exist for different uses. So long as computers > are used for so many widespread applications, we will need languages > geared toward those applications. Other posters have pointed to the evolutionary character of technology, with comparisons to cars etc. Why doesn't anyone admit that there is an element of religion also, in the history of programming languages? This started already in the 1950's with the Algol/Fortran dispute, which clearly had religious overtones. The credo of the Algol believers was that a programming language should be designed to be self consistent, complete, and preferably aesthetically pleasing, whether it could be implemented efficiently on then available computers or not. In Fortran, efficient implementation and practical usefulness were the top priorities, and to hell with recursive procedures (which were allowed in the Algol language but could not be implemented efficiently in those days, and were not implemented by most Algol compilers). The argument between Fortran and Algol was not resovable by rational means. It was easy enough to prove that programming in Fortran was more efficient than programming in Algol, in terms of compilation time and execution time, but that didn't impress the Algol believers one bit. As we know, Algol won in academia, Fortran in the rest of the world. Much later, when hardware and software technology had developed much further, Algol derivatives like Pascal and Ada started making inroads in industry. But, to this day, we have to live with Fortran, for Debbie's reason # 1. True, many of the languages that have appeared between then and now have been developed by someone who saw a functional deficiency in all existing languages and therefore developed a new one. This is technological development. But, sometimes one can wonder if the improvement over earlier languages is large enough to justify the introduction of a new language, and why do so many people seem to feel that their language is the ONLY sensible language, superior to all others? That, again, is religion! ----- Bengt Gallmo e-mail: lmebgo@eds.ericsson.se Telefonaktiebolaget L M Ericsson phone: +46 8 719 1940 S-126 25 STOCKHOLM fax: +46 8 719 3988 SWEDEN Sometimes a majority only means that all the fools are on the same side!