Cobrin.SBDERX@Xerox.COM (01/13/88)
Answer-To: Cobrin.sbderx@Xerox.com Key-Words: C, Unix Source, Unix History, Camouflaged Languages In article <149@ateng.UUCP> chip@ateng.uucp (Chip Salzenberg) Writes: > ... Peter Bourne wrote the Bourne shell using macros that make the >program itself look like a shell script: and in <7055@brl-smoke.ARPA> gwyn@brl-smoke.arpa (Doug Gwyn) wrote: >Any relation to Steve? yep he means Stephen >The Bourne shell sources were turned back into non-ALGOLized C several >years ago. (Thanks, Dave!) Back when I was at college in 1980 and looking at Unix sources, I noted the name Bourne. Being an ex Algol68 junkie (a real computer language), the name Bourne was known to me for having worked on the Algol68C compiler at Cambridge, England (68C being a subset of true Revised Report Algol 68). I found out that these people were one and the same, so one of the reasones the Bourne Shell looks like it does is because he was most probably still enthusiastic about Algol 68, and hearing that the original sources were structured as Chip and Doug say is very interesting. By the time I saw them Unix 2.7 (Version 7 Unix) they must have already been reformatted. The formatting of one language to look like another is nearly always something people new to a language try to do, and this is *always* detrimental to the maintainers of that code later on. Usually most of these discussions occur because the language definition didn't rule on layout, the language itself was flexible, and early example programs were 'badly' laid out. I refer to the 70's Pascal Layout Style Wars, and all the PrettyPrinter discussions.