13.0270 markup & scholarly practices

Humanist (humanist@kcl.ac.uk)
Tue, 16 Nov 1999 22:02:58 +0000

Humanist Discussion Group, Vol. 13, No. 270.
Centre for Computing in the Humanities, King's College London

Date: Tue, 16 Nov 1999 21:00:42 +0000
To: humanist Discussion Group <Humanist@kcl.ac.uk>
From: Patrick Durusau <pdurusau@emory.edu> (by way of Humanist
Subject: Markup & Scholarly Practices
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"; format=flowed
Status: O


There was a lively discussion at the recent Electronic Publication of Ancient
Near Eastern Texts conference (Oriental Institute, University of Chicago,
Oct. 8-9, 1999) on the topic of how much markup should be imposed in the
creation of electronic versions of historical texts. After thinking about the
presentations and informal discussions it occurred to me that the
organization of encoding projects is having a profound influence on the level
of markup that is seen as appropriate for such texts.

Mark Olsen (ARTFL) and Jeffrey Rydberg-Cox (Perseus) made strong arguments
for what I term middle-level markup in terms of delivering "large" bodies of
texts to users by relatively small groups of scholars. Anyone familiar with
either of these projects recognizes the benefits that markup even at this
level has brought to the scholarly community. The objections (as I understood
them) to imposing more markup on these texts were ones of time and personnel.
(Some scholars dispute the usefulness of extremely detailed markup but
usually in defending traditional research contexts that make such markup

In their respective projects such objections are certainly valid but what if
scholars change the context in which such projects were operating? What if
scholars began following the open source model that has lead to much of the
software infrastructure that supports the Internet for the production of
scholarly texts? Expanding the number of scholars involved in such projects
would certainly answer the objections concerning time and personnel.

If electronic texts were shared like the source code for Internet software
then if an individual wanted to impose character level markup or
morphological/syntactic information on a text they could do so without having
to duplicate the prior efforts of others. If we develop a culture of sharing
such additions to a base text, then other scholars can build upon that work
as well. That would avoid determining the level of markup to be imposed on a
text by what resources the project has on hand or decisions made within the
project. Scholars would be free to impose such additional markup as they deem

I am not advocating that we simply adopt the open source model uncritically
from the software development community but I would like to see a discussion
of how such a model could be used to produce scholarly electronic texts for
research and teaching.


Patrick Durusau
Information Technology Services
Scholars Press
Manager, ITS