[comp.lang.fortran] Our favorite language

jf@threel.co.uk (John Fisher) (04/05/91)

In article <16149@chaph.usc.edu>, 
echeverr@sal-sun8.usc.edu (The Black Sheep) writes:

> Excuse me, but is it just me, or is fortran a language so old it is becoming
> obsolescent?	Really, i just can't handle doing programs while tripping over
> small details and phantom errors from nowhere and tricky compilations and
> core dump after core dump after execution...

Lots of people use the language without having the problems you mention. 
You might find it profitable to compare this fact with your own experience. 
If you call yourself a programmer, you should consider it a matter of
professional pride to be able to master *any* language without too much
difficulty.  If you can't, perhaps you should consider whether you should
find another field of endeavour. 

In more general terms, I can't see the point of postings like this.  I
don't read this newsgroup because I *like* Fortran, but because Fortran is
an important aspect of my work, and I need the news and the help.  People
who work in university environments may be suprised to hear that it is not
the case that everyone can chose the language they work with; to many
people it is a given. 

To me it seems out of order to bust into a newsgroup and simply say, "You
guys are wasting your time", without giving any substantive reasons and
disregarding everybody else's agendas. 

--John

vsnyder@jato.jpl.nasa.gov (Van Snyder) (04/06/91)

In article <16149@chaph.usc.edu>, echeverr@sal-sun8.usc.edu (The Black Sheep) writes:
>
> Excuse me, but is it just me, or is fortran a language so old it is becoming
> obsolescent?	Really, i just can't handle doing programs while tripping over
> small details and phantom errors from nowhere and tricky compilations and
> core dump after core dump after execution...

and lots of folks have since responded with various apologies for Fortran.

Isn't it interesting that the "experts" have been criticizing Fortran since
1956, and it STILL hasn't gone away?  The problem is, the "experts" who
design alternatives don't know what's needed for _Computing_.  Sure,
Fortran doesn't have records or recursion.  But Pascal and Modula-2 don't
have open arrays or generic intrinsic functions.  And C, until recently
standardized by ANSI, coerced all intermediate floating point expressions
to double precision; C still lacks generic intrinsic functions.  With Fortran
90 having records and recursion, AND array processing, the "modern" languages
touted as alternatives by the "experts" seem to be falling further behind.

I've criticized Fortran 90 because it's not all it should be, but I don't
see any REAL alternatives for _Computing_.

It's interesting that Guy L. Steele, Jr., the main designer of Scheme (an
especially clean Lisp-like language), and the editor of "Common Lisp, the
Language", editions 1 and 2, is presenting a paper "Compiling Fortran at
10 Gigaflops" at the upcoming SIGPLAN conference on Programming Language
Design and Implementation.
-- 
vsnyder@jato.Jpl.Nasa.Gov
ames!elroy!jato!vsnyder
vsnyder@jato.uucp

mac@cis.ksu.edu (Myron A. Calhoun) (04/06/91)

In article <16149@chaph.usc.edu>, echeverr@sal-sun8.usc.edu writes:

> Excuse me, but is it just me, or is fortran a language so old it is becoming
> obsolescent?	Really, i just can't handle doing programs while tripping over
> small details and phantom errors from nowhere and tricky compilations and
> core dump after core dump after execution...

One of my colleagues who doesn't read this group had this to say about
this thread:

>I wonder whether the language addicts can step away far enough from
>their addictions to see what they are hollering about.

>The fortran addicts are hooked on fortran because it gives them
>the freedom (and the responsibility) to do 'risky' things, i.e.
>the freedom do do (almost) anything the machine can be made to do.
>They may have to pay the price by reading crash dumps in hex ...

>The 'structured-language' addicts are addicted (not to misnamed
>'structured languages' but) to structured code, because they can build
>or buy compilers that can proofread much of that code for adherence
>to a safety dogma (thou shalt not do arithmetic on address variables;
>thou shalt not compare characters to cardinals ...).  Those addicts
>have forfeited their claims to freedom in exchange for the promised
>land of make-believe correctness ....

>Just as adherence to style guides guarantees uniformly mediocre
>(mass-produced?) 'correct prose' so use of 'structured languages'
>guarantees 'stodgy but reliable code'.  Neither style guides nor
>structured languages leave room for genius.  Because genius is scarce
>and style guides and structured languages are popular, both claim to let
>people join the select class of genius without paying the entry fee.
--Myron.
--
# Myron A. Calhoun, Ph.D. E.E.; Associate Professor   (913) 539-4448 home
#  INTERNET:  mac@cis.ksu.edu (129.130.10.2)                532-6350 work
#      UUCP:  ...rutgers!ksuvax1!harry!mac                  532-7353 fax
# AT&T Mail:  attmail!ksuvax1!mac                   W0PBV @ K0VAY.KS.USA.NA

macq@miguel.llnl.gov (Don MacQueen) (04/06/91)

The original posting was:
|> 
|> > Excuse me, but is it just me, or is fortran a language so old it is becoming
|> > obsolescent?	Really, i just can't handle doing programs while tripping over
|> > small details and phantom errors from nowhere and tricky compilations and
|> > core dump after core dump after execution...

The best response would have been silence.

-- 
--------------------
Don MacQueen
macq@miguel.llnl.gov
--------------------

tom@itc.univie.ac.at (Tom Kovar) (04/08/91)

Since some days I have been following this discussion and I must say I am more
and more delighted...
I am an experienced both fortran and C programmer and have heard many dis-
cussions and arguments against fortran - the most interesting thing has always
been, that especially those people are sharp against fortran, who had never
written a programme in fortran, many of them being wonderfull high computer
science theoretitians, who had never written a reasonable running programme
at all - in any language.
I claim that for many tasks in the field of numerical mathematics and number
crunching fortran is a good language, not to speak about practical reasons;
in many cases only fortran coding results in the best optimized and fastest
programme. Ever seen a Cray, big CDC or IBM? Probably not...

In article <27fc5084@ThreeL.co.uk> jf@threel.co.uk (John Fisher) writes:
>Lots of people use the language without having the problems you mention.
>You might find it profitable to compare this fact with your own experience.
>If you call yourself a programmer, you should consider it a matter of
>professional pride to be able to master *any* language without too much
>difficulty.  If you can't, perhaps you should consider whether you should
>find another field of endeavour.
>
>In more general terms, I can't see the point of postings like this.  I
>don't read this newsgroup because I *like* Fortran, but because Fortran is
>an important aspect of my work, and I need the news and the help.  People
>who work in university environments may be suprised to hear that it is not
>the case that everyone can chose the language they work with; to many
>people it is a given.
>
>To me it seems out of order to bust into a newsgroup and simply say, "You
>guys are wasting your time", without giving any substantive reasons and
>disregarding everybody else's agendas.
>

 ... I fully agree with John Fisher in all these points - except for one.
I *am* stil working in a university environment, and anyhow, there are also
many situations here where it is given to you which programming language
to use.
There are millions of code lines written in highly optimized Fortran 
programmes; and if you need some hours of CPU on a Cray for one single small
run, you'll probably not risk rewriting them into a language popular with
CS-theoretitians and thus cause them need twice as much CPU time... You'll
rather continue developing them in fortran, utilizing the best optimized
fortran compilers...
 And if you have a task for which C or pascal or modula or devil knows what
is better, you'll use it; as a programmer, you are supposed to be flexible
and capable of adapt yourself to the needs of your a) programme b)operating
system c)HW. Or not?

  I think, that "Real programmes are no quiche eaters..."
								Tom
PS: last but not least, now a little more philosophically: if I do not like
a thing, it is by far no proof for the thing to be bad. In most cases, it is
*my* fault that I do not like it...