3.507 Offline 25 (382)

Willard McCarty (MCCARTY@vm.epas.utoronto.ca)
Wed, 27 Sep 89 17:35:20 EDT

Humanist Discussion Group, Vol. 3, No. 507. Wednesday, 27 Sep 1989.

Date: Tuesday, 26 September 1989 2104-EST
Subject: OFFLINE 25

<<O F F L I N E 2 5 by Robert Kraft>>

guest columnist Robin Cover

[[There are many things that could be reported in OFFLINE at the end
of this very short and full summer (including the misprint in the header
to the previously published column, which should read "24," not "23").
I recently returned from a papyrological conference in Cairo at which
there was opportunity to demonstrate the magic of the IBYCUS Scholarly
Computer searching the Duke Papyri Data Bank CD-ROM as well as the
Thesaurus Linguae Grecae CD-ROM to many people quite unfamiliar with
such technology. Some discussions on the use of computerized (digitized)
images for paleographical research, on the one hand, and of data base
compilations of prosopographical information from the papyri, on
the other, also took place there. The various pieces of regular
and electronic mail that awaited my return included numerous
items of potential interest for OFFLINE readers, from the announcement
of new or improved hardware releases to relevant software developments
and new electronic texts and data sets.

But with the annual November meetings of SBL/AAR/ASOR approaching
fast, it seemed most appropriate to devote this column to a preview
of some of the computer-related aspects of the Anaheim scene.
For those of you who are able to attend, and who can take advantage
of the information and expertise available at those sessions,
some of the new developments can be seen and discussed in person,
along with the old. Thus I asked Robin Cover, co-chair (with Alan
Groves) of the Computer Assisted Research Group of the SBL,
to provide us all with an overview of the activities and interests
of CARG, with a specific eye to the Anaheim meetings. Robin agreed to
this request, and his contribution follows. Please check it all
out for yourselves on 18-21 November in Anaheim!]]

<CARG Past and Present>

In the first part of this column I will offer a summary description
of the current goals and activities of the Computer Assisted Research
Group (CARG). In the second part, I will suggest areas in which CARG might
provide additional computer assistance and service to the Society of
Biblical Literature. Readers interested in helping enrich our vision for
the use of computers in individual research and within the Society's
corporate activities are invited to respond in writing.

In the most general terms, CARG's primary task has been to promote
the use of computing technologies in the professional and scholarly work
of SBL members. The specific activities of CARG have never been guided by
a canonical "mission statement," at least to my knowledge. Rather,
several factors have contributed to CARG's historic maintenance of a
flexible identity. (1) CARG has no permanent base of funding, but has
employed adaptive strategies for its financial existence. Contributions
from the SBL and from private donors are deeply appreciated, but funding
based upon good will renders CARG's program contingent upon uncertain
economies and fortunes. (2) CARG pursues its goals in relation to the
rapidly-evolving role of "academic computing" centers in colleges,
universities and seminaries, where institutional support for humanities
computing is highly variable. While CARG cannot duplicate every function
of an institutional "Academic Computing User Services" department, we do
attempt to assist in some domain-specific problems encountered by biblical
and classical scholars. (3) CARG has attempted to meet the needs of a
highly diverse group of interested scholars -- scholars having widely
divergent computer literacy skills and widely divergent computing
applications. Adding to this complexity the impact of periodic computer
hardware revolutions, we find no shame admitting that CARG's goal is a
moving target.

The Annual Conference of the SBL/AAR is the locus of CARG's visible
activity, though a steering committee maintains electronic mail discussion
throughout the year. On an annual basis, we attempt to identify
technological developments (hardware or software) which have lead to
applications that are "ripe" for promotion among the SBL constituency. In
the main CARG session (usually on Saturday of the Annual Meeting), we
invite two or more individuals to discuss these new applications in terms
of their own research, and if possible to demonstrate visually the
results. These invited lectures are meant to capture the imagination of
scholars in biblical and classical research, and to help them visualize
the new computer applications in related areas of study. At the upcoming
Anaheim meetings, for example, we have invited three scholars to speak on
the general theme "Scanning Technologies and Archives in Humanities
Computing." Terrance Erdt (Villanova University) will speak on "Scanning
and Character Recognition, New Tools in Humanities Computing;" Theodore
Brunner (University of California, Irvine) will speak on "Machine-Readable
Text Archives for Classicists: The Thesaurus Linguae Graecae Project;"
Robert Kraft (University of Pennsylvania) will speak on "Text Archives:
Why you Can't Find/Use the Texts you Need."

A second part of the main CARG session at the Annual Meeting is
dedicated to reports on recent or ongoing computing activities at academic
institutions. Institutional research and development often requires
several years for the introduction of a mature computer product or for a
major work of data preparation. The report session provides an
opportunity for institutional representatives to describe databases and
programs that are available for public use, to announce new research
endeavors, to solicit cooperative working arrangements with other
institutions, etc. This November in Anaheim we hope to hear reports from
or pertaining to the following institutions and projects:

Biola University (Virginia Doland and Don
Wilkins: CAI Software for Biblical/Classical Greek); Harvard University
(Greg Crane and Elli Mylonas: PERSEUS Project; Richard Saley:
Photogrammetry Project); Johns Hopkins University [with Hebrew Union
College-Jewish Institute of Religion] (Stephen Kaufman: Comprehensive
Aramaic Lexicon Project); Hebrew University, Jerusalem (Emanuel Tov:
Computer Assisted Tools for Septuagint Studies [CATSS]; Michael Stone:
Armenian Inscriptions Data Base); Manchester University (Tony Smith and
Gordon Neal: Greek Syntactic Parsing Project); Maredsous Centre:
Informatique et Bible (R. F. Poswick: Maredsous Biblical Databases);
Oxford University (Susan Hockey, Lou Burnard: Oxford University Computing
Center and Oxford Text Archive); Packard Humanities Institute (David
Packard, Wilkins Poe: Greek & Latin Texts [with Micro-IBYCUS]); Princeton
Theological Seminary (Richard Whitaker [also with Claremont Institute
for Antiquity and Christianity] with Jim Roberts: [Electronic] Hebrew
Lexicon Project; and with James Charlesworth: Qumran Machine-Readable Texts);
Summer Institute of Linguistics (Steve DeRose: CELLAR [Computing
Environment for Linguistic, Literary and Anthropological Research]);
University of California at Irvine (Theodore Brunner: Thesaurus Linguae
Graecae Project); University of California at Los Angeles (Giorgio
Buccellati: Computer Aided Analysis of Mesopotamian Materials; Andrew
Dyck, Bernard Frischer: Classicist's Workbench); University of California
at Santa Barbara (Randall Smith: CD-ROM Retrieval Software for Textual
Research); University of Pennsylvania (John Abercrombie, Alan Humm, Robert
Kraft, David Louder, Jacqueline Pastis, Jay Treat, David Rech: Center for
Computer Analysis of Texts [CCAT]); University of Sheffield (David
Clines: [Electronic] Dictionary of Biblical Hebrew); University of
Stellenbosch (Walter Claassen: Research for Computer Applications to the
Language and Text of the Old Testament; Johann Cook: Syriac Peshitta
Project); University of Toronto (John Hurd, Trinity College: Center
Coordination, Software Library); Vrije Universiteit, Amsterdam (Eep
Talstra: Werkgroep Informatica [Hebrew Bible Syntactical Analysis]);
Westminster Theological Seminary (Alan Groves: Westminster Computer
Project [Hebrew Bible Morphological Parsing]); Wooster College (J. Arthur
Baird: Computer Bible Project).

CARG supports other computer-related activities in specially
designated CARG Demonstration Rooms during the Annual Meeting. These
scheduled demonstrations and tutorials serve the personal computing
interests of SBL/AAR members. Commercial software developers, academic
institutions and hardware companies are invited to schedule 30-minute
demonstrations of their academic product at no charge. These small-group
demonstration sessions are used to introduce new products and sometimes to
provide personalized support. Late-afternoon discussion sessions focus on
common problems of text- or word-processing -- frequently the problems of
data conversion, document markup, file formats, multi-lingual
wordprocessing (foreign-character fonts), concording, text retrieval and
desktop publishing. These discussion forums permit computer users
collectively to register their complaints and wish-lists with the software
developers present. The CARG Demonstration Rooms also contain literature
tables for promotion of academic software and provide a meeting place for
computer user-groups and special-interest groups.

<CARG Future and Some Present Concerns>

The CARG Steering Committee hopes that the current program supplies
vital computer-related information and assistance to members of SBL/AAR
who may otherwise be un-supported or under-supported by their own
institutions. We recognize, however, that CARG could provide assistance
and leadership in other areas of computer technology relevant to the
scholarly and professional work of SBL members. In the following
paragraphs I will identify two broad computer-related concerns which I
feel could be formally addressed by the Society through the help of CARG
and/or other groups.

<Electronic Networking>

Perhaps no aspect of computer technology has affected scholarly
research more dramatically than the international academic networks
(BITNET, CSNET, Internet, NSFnet) which permit rapid communication and
data sharing. Electronic networking permits scholars on different
continents to work on collaborative research projects almost as easily as
they might if located at a single institution. The administrative offices
of the SBL and AAR are now connected electronically via BITNET, so that
business communication can be conducted over the networks as well.

If these electronic networks are in place, why is a majority of
scholarly communication still confined to paper? (I would include the FAX
technology as a category of paper communication, since data sent over FAX
is characteristically just printed on paper, not delivered to the
recipient in editable machine-readable format.) Two major barriers stand
in the way of the full democratization of scholarly networking. The first
is education and training: electronic mail and networking services are
sometimes not adequately promoted or supported by institutions which have
these resources, particularly within humanities departments. CARG could
help in education and training, and indeed, I began a joint effort with
AAR members this summer (plans initiated by Lewis Lancaster and Andrew
Scrimgeour) which may result in useful documents on academic networks. We
must demonstrate that e-mail communication and networking can be integrated
into the electronic scholarly workspace as easily as wordprocessing.

A far more serious barrier to networking, I suspect, is that too few
members of SBL/AAR have institutional access to network resources. Large
research universities, doctorate-granting universities and comprehensive
colleges of course support BITNET, Internet and other research networks.
But a significant number of SBL/AAR members belong to smaller liberal arts
colleges, professional schools and seminaries which do not support the
academic networks. In other cases, institutional networking resources may
be under the control of engineering schools or computer science
departments, and thus not readily accessible to departments of religion
where SBL/AAR members work. CARG may be able to coordinate assistance at
various levels for SBL/AAR members who face these "access" difficulties.

Of several research networks that might be designated as the
"recommended" (or official) network for SBL/AAR, BITNET and Internet are
the most prominent candidates. BITNET (now merging with CSNET under the
auspices of a new Corporation for Research and Educational Networking
[CREN]) is currently the network of choice for most humanities scholars,
and would probably be the easiest for SBL to adopt. The popularity of
BITNET among humanities scholars is due, in large measure, to the fact
that institutional membership fees (fixed annual fees, determined by E&E
budgets) are very reasonable and that fees are based on access-only. On
BITNET, no per-usage fees may be passed on to end users. The Internet is
a more modern, high-speed network which supports gateways to BITNET; its
installation and support is more expensive, and per-usage fees sometimes
make it financially inaccessible to humanities scholars. In a subsequent
article I may survey academic networks more broadly, indicating the
hardware/software requirements for each and the respective fee structures.
I visualize that SBL/AAR could play an intermediary role (perhaps jointly
with the APA and related societies) with the BITNET administration in
helping medium-size and smaller institutions overcome obstacles to
acquisition of BITNET membership.

<Electronic Publishing and Text Archives>

With dramatic decrease in the costs of data storage and electronic
publication (especially CD-ROM) and increasingly powerful microcomputers,
information management specialists are faced with the problem of
comparatively crude, antiquated and otherwise inadequate software.
Humanities scholars likewise confront multiple difficulties in the use of
electronic tools to create, publish and maintain their written research:
(1) inadequate support for multi-lingual authoring and text processing;
(2) lack of clear standards for use of foreign-language character sets and
fonts; (3) incompatibilities between file formats used in commercial
software packages, and inadequate file-conversion utilities; (4)
inadequate software support for management of document formats required in
various publishing houses; (5) personal desktop-publishing software, or
publishers' electronic typesetting software which actually corrupts data
from the standpoint of information retrieval. When the work of the
international Text Encoding Initiative is completed (1990-91), we may hope
for clearer standards and for the compliance of software developers who
are committed to serving the academic community. In the interim, we
cannot expect quick solutions to these problems, either from the business
world or from the special efforts of humanities computing initiatives. On
the other hand, the existence of an affiliated "Scholars Press" gives the
Society of Biblical Literature a unique opportunity to support emerging
standards and to develop goals for an electronic publication division
which makes scholarly research available in machine readable format. The
SBL would be in a position to set high ethical standards for the
protection of intellectual property contained in its scholarly document
archives. A few examples are offered below.

First, the Society could encourage or require the submission of all
major work in electronic as well as paper format, whether academic or
administrative data. Microcomputer diskette may be the preferred medium,
although simple and standard means are available for mailing binary data
in encoded format (uuencode, binhex) over the electronic networks.
Encouraging the submission of electronic data will serve to heighten our
collective awareness of several vital facts. Of foremost importance: the
goal of simply printing information on paper must now be understood as a
shortsighted, inferior goal. With some rare exceptions (ephemeral data),
any information worth typing and printing on paper is probably worth
preserving in machine readable format: for subsequent editing, for
information retrieval, for archiving. Similarly, preserving electronic
data files and submitting information in electronic format to others will
constantly remind us that highly proprietary ways of managing information
are usually counter-productive to our research goals. Data submitted to
the Society or Scholars Press in electronic format may or may not be of
immediate benefit, but it should all be archived for use in future years.

A second goal of the Society could be to promote standards (including
recommended hardware and software choices) which optimally support our
long-range electronic information retrieval objectives. Brief reference
was made above to the international Text Encoding Initiative. To judge
from the preliminary efforts of this initiative and from inertia in the
broader electronic publishing industry, it now appears highly probable
that a form of descriptive markup (e.g., SGML = Standard Generalized
Markup Language) will be recommended, at least as a standard for document
interchange. As an ISO standard, SGML is being required by several
government agencies, and is receiving broad acceptance in the publishing
and information science industries. Descriptive document markup is a
strategic choice, for it permits document content and document structure
to be represented independently of document "appearance." The traditional
fixation on "document appearance" (viz, the printed page) has usually
worked to the detriment of other scholarly concerns, especially electronic
publication and information retrieval. More than once in recent years,
prestigious publishing firms have thrown away electronic typeset tapes and
kept the lead printing plates -- with obvious consequences for electronic
publication of that data. Robert Amsler has spoken of getting paper
printout as a "transient joy," and Ted Nelson (father of "hypertext") has
described "getting on paper" as a shortsighted obsession. Of course,
electronic publication will not replace paper publication for many genres
of scholarly productivity. But we must begin to believe that authoring,
typesetting or electronic publishing schemes which corrupt or obscure
information content are inadequate tools, antithetical to our other goals
of communication and research. The Society can assist the progress of
scholarship by supporting the ideals and standards of the Text Encoding
Initiative or other agencies which have clearly articulated the inadequacy
of current document processing and publication methods.

A third kind of support by the Society would be to specifically
target selected publication projects for simultaneous print-copy and
electronic-copy formats. I have met on two occasions with the SBL group
which is producing the multi-volume anthology series "Writings from the
Ancient World." Scholars on this translation team have agreed in
principle to the publication of machine-readable editions of these texts
along with the bound volumes. Other SBL publication series might be
selected for similar treatment, particularly where scholars find it highly
desirable to search, concord or index the machine-readable text data.

A fourth type of support would be for SBL to sponsor or subsidize
software development for particular humanities computing applications that
present special problems for scholars in biblical and classical studies.
A review board could be set up to referee competitive proposals for
funding; recommendations from the review board could be made to other
granting agencies which support the software development. Political or
legal obstacles may prevent achievement of this goal, but perhaps not.
For example, computer applications used by textual scholars in the SBL/AAR
arena are usually demanding in that they require a level of multi-lingual
support not anticipated by designers of operating-system software
(Macintosh Script-Manager notwithstanding). It appears that Donald
Knuth's new implementation of TEX and Version 7.0 of Apple's Macintosh
operating software may provide more robust font support for selected
applications, but what about font support (screen and printer fonts) for
some of our popular, standard DOS applications? I know of no full-
featured DOS wordprocessor (e.g., Nota Bene, WordPerfect, Microsoft Word)
which provides native support for pointed Hebrew and accented Greek in
Adobe PostScript. It's not obvious that PostScript support in these cases
would be commercially feasible. On the other hand, many needs of
scholarship are met through funding of "commercially infeasible" projects,
and foundations exist specifically for these purposes. Even if the
example offered is problematic, I can visualize other applications for
which SBL/AAR subsidy might make a significant difference in solving an
annoying humanities computing problem.

<Concluding Invitation for Response>

Anyone wishing to contribute to the formation of the goals of the
Computer Assisted Research Group is invited to respond in writing with
pertinent suggestions. I will present all recommendations and requests to
the CARG Steering Committee in our e-mail forum or at the Annual Meeting in
November. If you have a special need in your own computer applications,
and feel it would constitute a common problem among SBL/AAR members,
please let me know.

Professor Robin C. Cover
Program (Co-) Chair, Computer Assisted Research Group

Assistant Professor of Semitics & Old Testament
Dallas Theological Seminary

BITNET: zrcc1001@smuvm1
UUCP: attctc!utafll!robin
FAX: (214) 841-3540
MCI: 332-1975
SNAIL: 3909 Swiss Avenue
Dallas, TX 75204
VOICE: (214) 296-1783 [h]; 824-3657 [w]

<Please send information, suggestions or queries concerning
OFFLINE to Robert A. Kraft, Box 36 College Hall, University
of Pennsylvania, Philadelphia PA 19104-6303. Telephone
(215) 898-5827. BITNET address: KRAFT at PENNDRLS. To request
printed information or materials from OFFLINE, please supply
an appropriately sized, self-addressed envelope or an address
label. A complete electronic file of OFFLINE columns is also
available upon request (for IBM/DOS, Mac, or IBYCUS).>