Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!sdd.hp.com!decwrl!shelby!lindy!jon@hanauma.Stanford.EDU
From: jon@hanauma.Stanford.EDU (Jon Claerbout)
Newsgroups: comp.lang.fortran
Subject: what good is the fortran?
Message-ID: <10724@lindy.Stanford.EDU>
Date: 31 Jul 90 04:27:38 GMT
Sender: root@lindy.Stanford.EDU (Rooter)
Organization: Stanford University, Department of Geophysics
Lines: 16

I use ratfor (rational fortran) and C equally often for 10 years.
I'm writing a textbook that exhibits about 40 subroutines
that illustrate engineering concepts for imaging.
The subroutines make about 120 figures.
I chose to show the subroutines in _RATFOR_ because:
	1. Ratfor looks more like mathematics than C
	2. The do-loop in C is obtuse and unnecessarily verbose.
	3. Matrices in C don't look like matrices in mathematics
	4. Ratfor can be easily converted to C, but not vice versa.

To make fortran bearable,
I wrote a preprocessor to allow memory allocation statements:
	temporary real x(n1,n2)
which I did with awk and UNIX system calls.
Maybe someday C++ will be a suitable alternative,
but meanwhile it still has the clumsy do-loop's.