Re: Forward into the past!

Norman Diamond <diamond@jit081.enet.dec.com>
Tue, 7 Jan 92 22:17:45 PST

          From comp.compilers

Related articles
Re: What's wrong with alloca() ? wws@renaissance.cray.com (1991-12-30)
Forward into the past! drw@kutta.mit.edu (1991-12-31)
Re: Forward into the past! salomon@silver.cs.umanitoba.ca (1992-01-05)
Re: Forward into the past! rcd@raven.eklektix.com (1992-01-05)
Re: Forward into the past! diamond@jit081.enet.dec.com (Norman Diamond) (1992-01-07)
| List of all articles for this month |

Newsgroups: comp.compilers
From: Norman Diamond <diamond@jit081.enet.dec.com>
Keywords: C, design, Fortran
Organization: Compilers Central
References: 92-01-004 91-12-089
Date: Tue, 7 Jan 92 22:17:45 PST

Although another reader has already expressed something approximating
my personal opinion (below), I would like to do so more pointedly.


In article 92-01-004 drw@kutta.mit.edu (Dale R. Worley) writes:
>In article 91-12-089 wws@renaissance.cray.com (Walter Spector)writes:
>>Some compilers support arrays which are dynamically sized when a context
>>is entered. This is similar to a very useful Fortran-90 feature. It is
>>supported by the GNU C compiler.
>It's hard to resist noting that this feature was in Algol 60. That it
>has taken C and Fortran 30 years to adopt it says something about the
>computer community.


I almost agree that it says something about the Fortran community -- but
not quite, because language designers went on to try to design languages
that were better all around, rather than try to retrofit Fortran (for a
while, anyway -- of course, retrofitting did begin later).


But it says something quite opposite about the C community. I am offended
by the number of complex operations that have been added to the C language
already. Assembly languages are supposed to map fairly closely onto the
underlying hardware. Assembly languages such as C were intended to
provide simple operations that give the programmer a great deal of
control, and to be useful as tools for implementing application languages
like Fortran and Algol and Prolog.


Now it is almost necessary to go back and invent a replacement for
assembly language all over again -- one which will be used for
implementing complex features, not for using them. Sigh.


-- Norman Diamond diamond@jit081.enet.dec.com
--


Post a followup to this message

Return to the comp.compilers page.
Search the comp.compilers archives again.