Re: a history question

From: Kevin G. Rhoads (kgrhoads_at_alum.mit.edu)
Date: 09/28/04 

Date: Tue, 28 Sep 2004 12:35:31 +0000

> The design of C can only be
>understood if you assume they had no knowledge of the
>state of the art (of the early 1970's).

I think that we will just have to agree to disagree on
this. I will comment that IF you assume the intent was
to design a general purpose language for most uses on
all processors, then I think you'd have somewhat more of a
case. However, it is clear that C started out life as
a systems programming language for the PDP-11 (once the
attempt to reproduce BCPL, which was also aimed more
at systems programming than general purpose, had been
abandoned).

That it then sort of evolved to general use was a totally
different matter. Remember that early Un(ixes, ixen, ices?)
also had F77 as well as C standard, and that AWK &c were
also developed for other uses on them.

THat is not to say that at that time those people were
experts in the history of computing, language design &c.
I had said:
>And not knowning
>one is re-inventing the wheel tends also to be the rule not the exception.
>
>The original developers of C were smarter, and better educated in their
>history, than that.
I do believe that Plauger, Kernighan, Ritchie &c definitely knew
that some of what they were doing was reinventing the wheel.
That "Unix" was named as hack on "Multics" points to one definitive
data point of knowing wheel-reinvention. Just as C starting life
as an attempt to recreate BCPL is another definitive data point
of KNOWING wheel-reinvention.

They d*mn well knew they were reinventing the wheel. They knew
enough about the history and/or the state of the art in 1970's
to know that much/most of what they were doing was re-inventing the
wheel. Knowingly re-inventing the wheel. That's what I claimed
for them, and unless/until you can provide countering data,
I will continue to stand by it.

Perhaps I wasn't clear in my early post. I hope this is clearer.

Relevant Pages

  • Re: a history question
    ... > to design a general purpose language for most uses on ... > that some of what they were doing was reinventing the wheel. ... > data point of knowing wheel-reinvention. ... BTL yanked its folks out of a project they had lost faith in. ...
    (comp.lang.fortran)
  • Re: Reinventing the wheel!
    ... One time I asked about making something on the knifemaking newsgroup and got a response about reinventing the wheel. ... There is a gray middle ground between sticking with the old proven design and going with the 'bleeding edge' technology. ...
    (rec.crafts.metalworking)
  • Re: Reinventing the wheel!
    ... There is a gray middle ground between sticking with the old proven design and going with the 'bleeding edge' technology. ... One time I asked about making something on the knifemaking newsgroup and got a response about reinventing the wheel. ...
    (rec.crafts.metalworking)
  • Re: Another argument for disk brakes?
    ... And, BTW, John Forester is an engineer. ... I see no reason a design should punish that particular mistake by ejecting a front wheel. ... right hand threads are used because in practicebecause it's more idiot proof than having some space cadet tighten the lugs to failure in their efforts to "unscrew". ...
    (rec.bicycles.tech)
  • Was Semi-OT, now Cowichan-style project
    ... Next is to design a Cowichan-style cardigan as a ... couple of kids telling me that their grandmothers have a spinning wheel ... idea of the design. ...
    (rec.crafts.textiles.yarn)