Path: utzoo!attcan!uunet!mailrus!cs.utexas.edu!sun-barr!apple!motcsd!mcdcup!mcdchg!att!cbnewsc!lgm From: lgm@cbnewsc.att.com (lawrence.g.mayka) Newsgroups: comp.lang.misc Subject: Re: Multi-compilers Summary: Large software systems exaggerate language differences Message-ID: <1990Sep24.234531.5763@cbnewsc.att.com> Date: 24 Sep 90 23:45:31 GMT References: <2581@l.cc.purdue.edu> <1990Sep22.061027.9223@d.cs.okstate.edu> Organization: AT&T Bell Laboratories Lines: 24 In article , pcg@cs.aber.ac.uk (Piercarlo Grandi) writes: > My probably warped view of the programming scene is that the best known > general purpose language technology, applied under favourable > circumstances, seems (hand waving very prominently here) no more than > 2-3 times more productive than mainstream language technology (Fortran, > Cobol, C, PL/1); claims to the contrary make me a bit skeptical. Now I > admit that even 200-300% is a lot, in economic terms, but it pales by > comparison with the different productivity levels of *people*. I disagree with your estimate, but even if 3.0 were the correct multiplier to apply to individuals working alone, the nature of large software system development can greatly exaggerate the effect. 30 mediocre programmers who are constantly in each other's way and who have only a fair understanding of project requirements and system architecture may be an order of magnitude less productive than 3 top-notch programmers who know every inch of their system and their project, and who can work together as a harmonious team. Lawrence G. Mayka AT&T Bell Laboratories lgm@iexist.att.com Standard disclaimer.