Humanist Discussion Group, Vol. 14, No. 715.
Centre for Computing in the Humanities, King's College London
Date: Mon, 05 Mar 2001 20:35:46 +0000
From: Charles Ess <email@example.com>
Subject: Indexing sorrows and joys
Humanists may remember that early this year I issued a panicked plea for
help with indexing software. Several helpful replies resulted, including
recommendations for software as follows -
Marc Wilhelm Kster for TUSTEP
You can consult the homepage:
http://www.uni-tuebingen.de/zdv/tustep/tustep_eng.html (in English)
http://www.uni-tuebingen.de/zdv/tustep/ (in German)
There you will also find a list of indices produced with TUSTEP.
As you will see from this list, TUSTEP is a highly versatile tool for
constructing indices from structured data. It does need a moment to get
used to, however.
As to obtaining TUSTEP, you can contact Mr. Fuchs
(firstname.lastname@example.org) for details on how it could be shipped
to you (a shipping charge applies).
>From Eric Johnson:
..... I wrote a program called BITZER. You can download the program and a
copy of an article that I wrote about my program from the web:
BITZER may or may not be what you are looking for -- but the (free)
price is right. -- Eric
>From Kevin Roddy:
In my book on UNIX nroff/troff from 1986 there is a chapter on doing just
the sort of indexing you describe, in fairly straightforward steps. UNIX,
unlike DOS, lives, and I think even Linux should be able to handle it.
The book is
<I>UNIX nroff/troff: A User's Guide</I>. New York: CBS College Publishing
[Holt, Rinehart and Winston], 1986.
One lead, from Brian Nielsen, didn't work when I tried it:
Adrien Miles recommended:
I use Endnote, it is mac and windows, not perfect but good. its
available via http://www.niles.com/
Finally, one good correspondent suggested several good programming languages
I might use to write my own program...
In what was perhaps hasty misjudgment on my part, I decided, however, that I
did not have the time to write my own program, wait on software from Germany
and/or to learn a new program (especially as a Linux novice). Nor did I have
the spare change for Endnote.
At least one Humanist (Francois Lachance) has kindly asked
> Just how did you produce the index?
Briefly - the hard way. Using a standard wordprocessor, I simply started
entering entries, alphabetizing "by hand" (i.e., without help from the
computer) as I went along.
One trick perhaps worth passing along. My introduction to the volume, as
one would hope, already identified several key themes and figures. I indexed
this section first, followed by entering additional terms and concepts I had
jotted down by hand from a previous read for copyediting purposes. Between
the two a good portion of index terms and categories were established and
entered, providing a substantial skeleton for the rest of the work.
Francois Lachance further asks:
> Were you successful in automating parts
> of the markup?
Again, briefly: no. I had toyed with the idea of embedding bookmarks, for
example, to speed finding various places in the index. But this seemed to
only add overhead, both in terms of time for creating the bookmarks, and in
terms of cognitive demands - the bookmarks would have been a second layer on
top of the original alphabetical armature. In the end, between my (not
especially good) memory and simple global searches through the
wordprocessor, I got to be fairly adept at remembering and finding what I
was looking for.
>Was there a lot of cross-referencing between the entries?
Yes. Here again, the global search capability helped considerably - as it
also did in subsequent searches for the odd phrase, name, term, etc., that I
knew was in the document but couldn't find by eye.
After that, the wordprocessor was useful for rearranging entries that were
not properly alphabetized the first time, and for checking for stylistic
errors (e.g., no periods at the end of an entry).
Perhaps if I'd only bought Endnote?
And/or: does it seem odd, now nearly two decades after the PC revolution
began; after mind-boggling advances in computing and storage technologies
(e.g., this now somewhat dusty G3 is still nearly a thousand times faster
than the first Macintoshes, and my pocket Visor is the computational
equivalent of a 1985 desktop PC); after thousands of ambitious and sometimes
remarkably successful humanities computing projects -
the admittedly mundane but important process of indexing (where an index is,
after all, an important form of hypertext) - a process, moreover, that lends
itself rather well to computation -
we find ourselves faced with:
a) strong advice _not_ to use indexing software bundled with even the "best"
b) something of a lag, shall we say , between an important humanities
process and standalone software (at least for those of us not ready to write
our own programs and/or missing the spare change for a commercial product)?
I don't know if this disparity between need and computational possibility,
on the one hand, and software reality on the other, means anything other
than indexing is something only a very few souls concern themselves with,
and hence fail to constitute a robust market that would make software
development commercially worthwhile. If this disparity goes beyond that - I
wonder if it would inspire a listing by HUMANISTS of similar tasks
important to our teaching, research, and publication that, however important
they may be and however well-suited they may be to computational algorithms,
likewise remain under- or non-supported in the midst of the computer
Chair, Philosophy and Religion Department
900 N. Benton Ave. Voice: 417-873-7230
Springfield, MO 65802 USA FAX: 417-873-7435
Home page: http://www.drury.edu/Departments/phil-relg/ess.html
Co-chair, CATaC 2000: http://www.it.murdoch.edu.au/~sudweeks/catac00/
"Egos appear by setting themselves apart from other egos. Persons appear by
entering into relation to other persons." -- Martin Buber, _I and Thou_
This archive was generated by hypermail 2b30 : Mon Mar 05 2001 - 16:11:50 EST