Daniel, all just to underline my previous answer, a response from one of the two authors of the original M4:
Begin doorgestuurd bericht: > Van: Brian Kernighan <XXX> > Datum: 17 juli 2014 17:10:46 GMT+02:00 > Aan: Raphael 'kena' Poss <[email protected]> > Onderwerp: Antw.: A little question about the history of M4 > > Hi -- > > I don't know anything specific (and was not part of any > discussions), but I think your guess is right: programs > were small, memory was very limited (the first C compiler > ran in something like 12K bytes), and processors were > slow. Namespaces would have been serious overkill for > that kind of situation. > > Sorry I can't give you anything more definitive. > > Brian > > On Thu, 17 Jul 2014, Raphael 'kena' Poss wrote: > >> Dear Brian, >> >> I am teaching various courses related to programming languages, their >> features and history and a question has been popping up recently: why was >> there little consideration for namespace isolation in the design of C, M4 >> and other programming technology from that time? >> >> My first guesses are that software was not yet large enough that namespacing >> was a concrete issue, and that memory was still scarce enough that the >> overhead of namespace support was too costly. But perhaps you have another >> opinion? >> >> Would love to hear your thoughts on this. >> >> Best regards, >> >> -- >> Raphael 'kena' Poss ? [email protected] >> http://staff.science.uva.nl/~poss/ >> >> >> >> >> >> >> >> -- Raphael 'kena' Poss ยท [email protected] http://staff.science.uva.nl/~poss/
