3.1250 programming languages, texts, and a course (75)

Willard McCarty (MCCARTY@vm.epas.utoronto.ca)
Mon, 2 Apr 90 19:56:40 EDT

Humanist Discussion Group, Vol. 3, No. 1250. Monday, 2 Apr 1990.

(1) Date: Sat, 31 Mar 90 11:43:55 EST (25 lines)
From: Stephen Clausng <SCLAUS@YALEVM>
Subject: programming languages

(2) Date: Sun, 1 Apr 90 11:14:10 EDT (28 lines)
From: "Eric Johnson DSU, Madison, SD 57042" <ERIC@SDNET>

(1) --------------------------------------------------------------------
Date: Sat, 31 Mar 90 11:43:55 EST
From: Stephen Clausng <SCLAUS@YALEVM>
Subject: programming languages

I think too much emphasis in this discussion is being made on whether a
given programming language has, or does not have, string capabilities,
and whether the resulting programs are portable. In particular, the
question is whether Pascal has such features. Most implementations of
Pascal have quite adequate string procedures. Standard Pascal does not.
But such procedures can easily be written, if necessary using ANSI data
structures, and it is silly to reject a programming language because you
might have to write a 10 line procedure to duplicate the function of
another language. The real value of any programmng environment is the
general ease of use. Does the language encourage good programming, does
it prevent errors, does it make it easy to avoid errors, is there a good
debugger for your particular enviroment, etc. In all of these regards,
Pascal is hard to beat. Incidentally, there are two excellent
introductions to Pascal programming which are designed specifically for
Humanists, namely: Nancy Ide "Pascal for the Humanities" and Mary Dee
Harris "Introduction to Natural Language Processing". Pascal textbooks
are also repleat with examples of text processing, and algorithms are
readily available for all the major string functions using standard ANSI
Pascal, which is one of the most portable, dialect-free languages ever
written. Actually I find that computers have become so specialized, in
terms of toolbox functions, that I believe it is almost pointless to try
to port programs directly from one computer to the next. The hardware
is the problem, not the software.
(2) --------------------------------------------------------------35----
Date: Sun, 1 Apr 90 11:14:10 EDT
From: "Eric Johnson DSU, Madison, SD 57042" <ERIC@SDNET>

After reading the HUMANIST discussion of my idea to offer a
course designed to teach academic humanists to program, I am unsure
that anyone could design a course to attract sufficient numbers to
make it practical to offer it. Following is the course as I would
teach it.

PROGRAMMING FOR THE HUMANITIES An introduction to programming for
applications in the humanities such as analysis of texts, arranging
data from research, and formatting for printing and desktop publishing.
BASIC will be taught as a means of understanding how computers
process instructions. The majority of the course will be an
introduction to the use of SNOBOL4: a powerful and economical
language for non-numeric computing. At the conclusion of the
course, students will be encouraged to work with other versions
of BASIC (such as QuickBASIC), with high-performance SNOBOL4
(SPITBOL), and with Icon.
Students will be sent course materials over BITNET and will
return assignments via BITNET.
At least as it is first offered, the course will be non-credit.

Eric Johnson