Xref: utzoo comp.lang.misc:2528 comp.lang.fortran:1697
Path: utzoo!attcan!uunet!lll-winken!ames!joyce!sri-unix!garth!phipps
From: phipps@garth.UUCP (Clay Phipps)
Newsgroups: comp.lang.misc,comp.lang.fortran
Subject: Stone Knives And Bear Skins (was Re: FORTRAN Lament)
Summary: Tain't a computer science research question.
Keywords: FORTRAN,dynamic-arrays,PL/I
Message-ID: <2401@garth.UUCP>
Date: 13 Jan 89 07:42:06 GMT
References: <117400003@uxa.cso.uiuc.edu>
Reply-To: phipps@garth.UUCP (Clay Phipps)
Followup-To: comp.lang.misc
Organization: INTERGRAPH (APD) -- Palo Alto, CA
Lines: 49

In article <117400003@uxa.cso.uiuc.edu> gsg0384@uxa.cso.uiuc.edu writes:
>
> ... I think we, stone-age FORTRAN programmers, 
>usually physicists, mechanical engineers, and civil engineers, 
>need [dynamic *local* array sizing] desperately.
>Otherwise, we should always provide the source code to the customer 
>so that he can redimension for the memory size of his hardware.
>Too often I cannot help thinking that the CStists do research 
>on what they like to do, but not on what most users want them to do.  
>The result is that we still don't have a language any better than 
>FORTRAN 77 for our purpose, ...

The net's Algol partisans have already posted their comments, I think.

Workable solutions were supplied by a language designed 
initially under the name FORTRAN VI in the early 1960s.
So many changes were made away from the Spirit of FORTRAN
that the language was eventually renamed PL/I.
It's still possible to find PL/I compilers laying around
for many minis and mainframes.  The language user community you 
identified chose to stick with FORTRAN, rather than switch to PL/I.

Time on the order of a decade elapsed between the 1st release of PL/I
and completion of the FORTRAN 77 standard.  Ask the FORTRAN committee
why the features you wanted aren't *already* a part of the language.

If you want the Cray features on another machine, complain to
your computer's manufacturer's marketing and sales organizations.
Computer manufacturers depend heavily on their marketeers to decide,
*in effect*, what extensions and features, *if any*, 
their compiler folks will provide in later releases of their compilers.
Compiler people can talk until they are blue in the face to management
about the features they think are "needed" in an existing compiler, but
until some potential customer waves big bucks in front of sales people
("no feature, no sale"), those "needed" features often just won't happen.
[It may not be this way where I now work, but it was at other companies.]
A favorite "feature" of customers, marketing, and sales alike, is *speed*.
Compiler writers working to get the fastest compiler for the "next 
release" typically aren't spending time adding other features.

Adding extensions to FORTRAN that duplicate features available
in other well-known languages is not generally considered
*computer science research*.  Politics, maybe, but not research.

-- 
[The foregoing may or may not represent the position, if any, of my employer]
 
Clay Phipps                       {ingr,pyramid,sri-unix!hplabs}!garth!phipps
Intergraph APD, 2400#4 Geng Road, Palo Alto, CA 93403            415/494-8800