Markup: encoding variants (44)

Thu, 16 Mar 89 00:02:25 EST

Humanist Mailing List, Vol. 2, No. 727. Thursday, 16 Mar 1989.

Date: Wednesday, 15 March 1989 2229-EST
Subject: Encoding Variants

Chaim Milikowsky asks how CCAT has been encoding variants.
For the initial experiment, we (myself and one of my graduate
students) did it by hand, producing the file for the book of
Ruth in Greek, based on the Cambridge Larger Septuagint apparatus.

Thereafter, we have been scanning the apparatuses (both Cambridge
and Goettingen) with a KDEM III and running a series of reformat
programs to reshape the material as much as possible automatically.
Because of inconsistencies, idiosyncrasies, etc., in the way the
published apparatuses are set up, some things remain to be done
by hand in the final reformatting and verifying stages.

For the Sahidic Coptic project, however, which in most instances
must work from individual copies of manuscripts (or fragments),
we scan in whatever we can (e.g. Budge's Psalter edition of a
British Museum MS), then we use those verified results as the
base to be modified in order to produce running texts of other
manuscripts of the same material. This reduces considerably the
time that would be needed for typing anew and verifying, although
it may foster certain types of errors if the encoder is not
alert (e.g. minor spelling variations). Thus the Budge Psalter
is kept as a separate file, but it is also used as the base
for encoding other Psalter material. In some instances we also
use the alternative approach and simply verticalize the base
text (e.g. Budge) and fit in the variations from any other
MS. This is even quicker, in the long run, and provides the
sort of multifaceted flat file that I described earlier in
the HUMANIST discussion, from which individual MS texts can
be reconstructed easily. (A hypertext base in flat form!)