9.307 programming & knees

Humanist (mccarty@phoenix.Princeton.EDU)
Wed, 22 Nov 1995 08:45:14 -0500 (EST)

Humanist Discussion Group, Vol. 9, No. 307.
Center for Electronic Texts in the Humanities (Princeton/Rutgers)

[1] From: Tom Horton <tom@sunrise.cse.fau.edu> (33)
Subject: approaches to software development

[2] From: Willard McCarty <mccarty@epas.utoronto.ca> (31)
Subject: muscular computing

[3] From: Ian Lancashire <ian@epas.utoronto.ca> (10)
Subject: Re: 9.305 programming

Date: Tue, 21 Nov 1995 11:53:30 -0500
From: Tom Horton <tom@sunrise.cse.fau.edu>
Subject: approaches to software development

Harry Gaylord in Humanist 9.305 alludes to a difference in approach towards
programming for the humanities scholar. He mentions the "Toronto example"
and contrasts that with efforts of himself and other individuals.

I'm responding to this not in an effort to start an argument or anything --
I'm just trying to understand what the needs of researchers are in the area
of text-based computer software. First, it's probably not realistic to
think that there is just one community of users with one common set of
needs! So it's no surprise there might be more than one approach.

I presume the Toronto example refers to a "large" software package (e.g.,
TACT, WordCruncher, Collate) that requires many person-years to develop
(perhaps), is produced by a third-party, is bought or shareware, has
signficant documentation, and is designed to serve the general or common
needs of a wide audience of users.

I'm not sure I can guess at a good description of the other as easily, but
Harry mentions UNIX, SNOBOL and PERL, so these are clues. (Or maybe they're
"signifiers"? I never did understand that stuff. :-) ) Perhaps this second
approach is characterized as follows: individuals (not teams or
organizations) write software, perhaps to meet general needs, but perhaps
more often to meet a specialized need. This leads to a set of smaller
programs that form a toolset for an individual or smaller group of
specialists, but may not be adopted by a wider audience.

Is this accurate? Again, I'm not trying to express any judgements here.
I'm trying to think about user and developer communities and their needs.
So mail me (or post) your comments or corrections to these descriptions.

Some further questions:
a) If my descriptions of these two approaches are correct, does anyone see
text-processing programs being developed under yet another approach?
b) Does anyone have a sense of the size of these two communities? In other
words, how many people are using software developed in either one of
these ways?
c) Does either community of developers need anything they're not getting
(in terms of programming or software methods, tools, environments or
components) that would help them meet their goals more easily?


Date: Tue, 21 Nov 1995 00:34:11 -0500
From: Willard McCarty <mccarty@epas.utoronto.ca>
Subject: muscular computing

I am sorry that my friend Harry Gaylord thinks my note gives evidence of
weak knees, i.e. a weak mind, with respect to computing. (Those who are
close to a copy of Richard Broxton Onian's invaluable book, <t>The
Origins of European Thought...</t> may, however, wish to check out the
wider meanings given to that bodypart, as in O.E. cneow and related
forms, in various ancient cultures. If I thought Harry had THAT in mind,
I'd be on the next plane to Groningen....) So may a pioneer
regret the passing of those strenuous days when the frontier had to be
tamed. If now I am soft, I certainly once was a fearless programmer of
assembly languages, which ran on computers so daunting that only the truly
tough could survive their use. (6000 vacuum tubes -- who here remembers
what THEY were? memory on a drum, etc.) And then there was FORTRAN....
Lest I give evidence here only of anecdotage, however, I must stop these
reminiscences of youth and get to a more valuable point.

As an activity, programming is undoubtedly a good mental discipline. I for
one would be most pleased to see a course in programming become one of the
requirements for an undergraduate degree, along with a few natural
languages, including Latin of course. I mean, let 'em sweat, good for the
character. I agree that there will always be a need for programming, and
that humanists should be able to do it. I thought I was arguing for a
much "higher-level" language than any we have seen so far, that's all, so
that for many of the kinds of things a humanist needs to do with a
computer the conceptual middleman could be eliminated even more
than has already happened. Surely we can all see that perl -- a newly
purchased textbook on which sits in my bookbag now -- is itself an instance
of a high-level language that makes active use of a computer easier for
those who have not the time or inclination to be professional programmers.

Ah, UNIX. The operating system we humanists prefer and use here in Toronto.
Down on our intellectual waterfront, among the academic toughs who hang out
in the dockside seminars (Genet's <t>Querelle de Brest</t> comes to mind),
unless you can grep your way in the dark you won't survive. But I speak
only of those I know, and my views are only my own. Perhaps it would be
better if Toronto were a unanimous entity with a unified outlook on the
world, but it isn't. If only everyone here shared my mind on all
subjects.... Then we'd really be in trouble.


Willard McCarty, Centre for Computing in the Humanities
Departments of Classical Studies and Italian Studies (Toronto)
(416) 978-3974 voice (416) 978-6519 fax mccarty@epas.utoronto.ca

Date: Tue, 21 Nov 1995 11:36:15 -0500 (EST)
From: Ian Lancashire <ian@epas.utoronto.ca>
Subject: Re: 9.305 programming

Apologies to Harry Gaylord if somehow "Toronto" has seemed less than fully
enthusiastic about snobol, perl, UNIX, and SGML. We ("they"?) use all
these splendid things and wouldn't be without them.

Let me put in a special plea for a humanities programming book on perl.
(Have I missed it?)

Perl and UNIX scripts have made my work a lot easier. If even an English
professor like me can run with perl, I can only imagine what
strong-kneed Harry and his colleagues are able to do with it. More
power to them all.

Ian Lancashire