3.1253 e-texts, stemmatology, scanning (107)

Willard McCarty (MCCARTY@vm.epas.utoronto.ca)
Tue, 3 Apr 90 21:38:31 EDT

Humanist Discussion Group, Vol. 3, No. 1253. Tuesday, 3 Apr 1990.


(1) Date: Tue, 3 Apr 90 01:10:10 EDT (24 lines)
From: Robert Hollander <bobh@phoenix.Princeton.EDU>
Subject: Re: 3.1247 electronic texts, paradigms, scanning (175)

(2) Date: Tuesday, 3 April 1990 1921-EST (40 lines)
From: KRAFT@PENNDRLS
Subject: Computer Assisted Textual Stemmatology

(3) Date: Tuesday, 3 April 1990 1953-EST (18 lines)
From: KRAFT@PENNDRLS
Subject: Optiram Scanning

(1) --------------------------------------------------------------------
Date: Tue, 3 Apr 90 01:10:10 EDT
From: Robert Hollander <bobh@phoenix.Princeton.EDU>
Subject: Re: 3.1247 electronic texts, paradigms, scanning (175)

O'Donnell is barking up a better tree, but leaves unresolved who (1) will
decide what's to be done; (2) who will be responsible for the accuracy of
the scanning. From our experience at the Dartmouth Dante Project, THAT's
the most expensive part of the process. And, Michael Hart, we're not doing
much by way of formatting, tagging, etc. What has not been heard often
enough in this discussion is pleading for the accuracy of these texts. If
one has in mind "armchair" browsing of e-texts, accuracy doesn't matter
that much. However, if we are issuing texts which will be _searched_
by users who expect (and need) them to be ACCURATE, who will be basing
conclusions on their searches, the whole exercise becomes a question-
able one if the issuers are issuing error-filled texts. It currently
costs us about $20,000 to do a single Dante commentary (ca. 4 megabytes
of material). Were we starting now, we could probably do it a bit
cheaper. But there is still a major amount of contributed time (I
myself must have 3,000 editing hours in the project by now), which cuts
costs considerably. If there is anyone out there who'd like to do
a commentary (200 to 600 hours, depending on the text and the editor's
capacity) for us, I hope that person will let me hear. In the mean-
time we slog along. And in a few weeks we'll have about thirty
commentaries on-line--and miles to go before we sleep.
(2) --------------------------------------------------------------43----
Date: Tuesday, 3 April 1990 1921-EST
From: KRAFT@PENNDRLS
Subject: Computer Assisted Textual Stemmatology

Regarding the use of computers to collate and explore the
stemmatic relationships between manuscripts (I decided that
"stemmatology" was more appropriate than "stemmatization" to
describe this endeavor, since one possible result is no clear
stemmata), an issue raised by Peter Robinson (Oxford) and
responded to briefly by Charles Fulhaber (referring to the
software developed by Francisco Marcos Marin), some work has
been done in relation to ancient texts. John Hughes provides
bibliography (but no discussion) in his BITS, BYTES & BIBLICAL
STUDIES treasure trove, with special reference to the following
authors (see p.492 n.11):
J. Burch (NT, 1965)
Vinton A. Dearing (NT, several articles, 1974 onward)
M. P. Weitzman (general?; ALLC Bulletin 10, 1982)
Bonifatius Fischer (NT, 1970, 1973)
Karen A. Mullen (general?; CHum 5, 1970-71)
Wilhelm Ott (general, 1973)
Gian Piero Zarri (general, 1976, 1977)
Jacques Froger (general, 1965, 1970)
John G. Griffith (Gospels, 1969, 1973)
If someone really cares and does not have access to BBBS
for the details, I can provide them. Although I do not have
a copy of Susan Hockey's Guide to Computer Applications in
the Humanities at hand, I suspect that she will have addressed
aspects of this subject as well. Here at Penn, we are trying
to get the million or so variants to the ancient Greek translation
of the Jewish scriptures encoded so that appropriate detailed
analysis and classification can follow. A recent PhD from
the Hebrew Union College in Cincinnati, Bernard Taylor, did
his dissertation on the analysis of one subgroup of manuscripts
in the Greek tradition to the biblical book of 1 Samuel. Other
students are working on similar projects with Emanuel Tov
at the Hebrew University in Jerusalem.

Bob Kraft (Univ. of Penn., Computer Assisted Tools for
Septuagint Studies Project)
(3) --------------------------------------------------------------21----
Date: Tuesday, 3 April 1990 1953-EST
From: KRAFT@PENNDRLS
Subject: Optiram Scanning

In response to Roy Flannigan's recent query about contacts
with Optiram, a British company that has very protected
processes for scanning handwritten materials into text
files (not graphics), I did get some estimates from them
several months ago for scanning my great grandfather's
journals. I sent samples, and they responded with an
estimate of the degree of accuracy they anticipated and
the cost of such a job. I do not have the details at hand,
but my conclusion was that it would be significantly less
expensive to have a good typist encode the text at a cost
of $10 per hour. I'm sure Optiram would have done an
acceptable job, but it would have cost a great deal more.

Bob Kraft