SGML Project

[Virginia Tech]

Here is the SGML posting I mentioned in my previous message.

--David Gants


Date:    Wed, 28 Aug 1996 12:00:26 CDT
From:    Murray McGillivray <mmcgilli@acs.ucalgary.ca>
Subject: Textual Representation Group

The following message is being widely posted.  Please forgive me if you
receive multiple copies; please copy to listservs and newsgroups I may
have ignored but that you know would be interested.

Below you will find the Draft Terms of Reference of a new Working Group
formed this spring.  Most of the tasks listed were discussed as
desiderata at the physical sessions of the Electric Scriptorium
conference in Calgary (November 1995), and the initial members of the
working group were attendees at that conference.  We have since been
joined by prominent textual scholars and members of the international
text-encoding community.

The draft terms of reference are being posted to several different
scholarly lists for comment.  We would appreciate it if you would post
your comments to scriptorium@morgan.ucs.mun.ca (normal listserv
subscription procedures apply: send a single-line message, SUBSCRIBE
SCRIPTORIUM <Your Name>  to listserv@morgan.ucs.mun.ca   before
attempting to post to the list).  We are especially interested to hear
>from people who think that our plan (or some part of it) is badly
thought-out or just a bad idea.  We also would welcome new members and
offers of assistance.

Murray McGillivray

University of Calgary

______________________________________________________


DRAFT TERMS OF REFERENCE
WORKING GROUP ON ELECTRONIC REPRESENTATION OF
HANDWRITTEN AND PRINTED MATERIALS

Preamble:  The Working Group on Electronic
Representation of Handwritten and Printed Materials has been
formed to devise standards and guidelines to be used by
scholars who wish to prepare electronic editions, textbases,
image-bases, or other electronic objects which have as a
purpose the exact representation of a physical print or
manuscript of whatever form, period, language, or script.
Significant work has been done towards this goal by the Text
Encoding Initiative in its guidelines published in 1994
(hereafter "P3").  There remain, however, several areas
which the Text Encoding Initiative did not have the time or
resources to explore fully.  Chief among these is the area
of exact bibliographic description, with respect to which P3
explicitly solicits further work, but there are other areas
where P3 is either merely suggestive or where it gives too
many possible answers, so that scholars may, fairly
arbitrarily, make completely different decisions about
essentially similar problems, with resulting incompatibility
between what ought to be compatible projects.

The primary focus of the new Working Group is the exact
electronic representation of physical books (and other
printed or handwritten materials), a focus both narrower
than and different from the main goal of P3.  It is
therefore not so much a continuation of the TEI effort as it
is a new initiative, which will, however, aim for
compatibility with the TEI recommendations wherever
possible.

Draft Statement of Goals:   The Working Group sees a need
for standards both for text-file representations and for
graphics-file representations of printed and handwritten
materials, since it is clear that scholars can make use in
different ways of electronic transcriptions or encodings of
physical prints and manuscripts and of electronic images of
those prints and manuscripts.

Some goals are certain to emerge in the course of the work.
However, there are a certain number of goals that the
Working Group adopts explicitly from the beginning:

For text-files:

1.  To devise standard encodings for the most common of
those marks used by European and American scribes and
printers that are not contained in modern standard character
sets (such as brevigraphs, ligatures, and variant letter
forms) in the form of SGML-type entity sets.  These entity
sets would supplement the work of the Text Encoding
Initiative and serve as a resource for scholars using P3 for
transcription of manuscript or print materials.

2.  To recommend procedures for scholars whose encoding
needs are not met by the entity sets mentioned in 1).

3.  To devise standard ways, likely using SGML-type tags, of
encoding structural and other features of physical books and
documents that do not fall under a strict definition of
"text," such as: page layout; font, script or hand;
condition of the carrier or ink; relationship of leaves,
quires, signatures, etc. to one another; interlineal or
marginal annotation; damage, whether affecting the text or
not; and so on.  Many of these features are treated
suggestively in P3, but not considered as fully as the
working group intends to do.

4.  To develop appropriate definitions describing the
possible relationships between the tags in the proposed
tag-sets. The new encoding procedures should be as
compatible as possible with the work of TEI:  a DTD fragment
should be developed that will make documents encoded using
the new tags TEI-conformant, and further DTD fragments or
other encoding strategems should be developed that will make
it possible for encoders to combine tags from the existing
TEI sets with tags from the proposed set(s) within the same
document without loss of structural information.

5.  To develop a standard procedure for referencing a
graphics file from a text file when the graphics file
contains an image of the page that is transcribed in the
text file.

For graphics files:

1.  To develop for graphics files containing images of
printed or handwritten pages a standard text header that
would specify such things as: bibliographic information,
page of book, copy number, library where kept, size, and so
on of original the image is of; by whom and how the image
was created (camera or other device, lighting used, method
of digitizing, etc.); what transformations it has undergone
since digitized and who was responsible (resizing,
compression, dithering, adjustments of colour, etc.); the
format of the image (e.g. GIF, TIFF); copyright information
for the image where relevant.

2.  To recommend standard ways of indicating within a
graphic image itself the size and colour of the original it
is a picture of.

3.  To recommend standard ways of referring to precise
locations or areas within a graphic image of a printed or
handwritten text.

4.  To develop a standard procedure for referencing a text
file from a graphics file when the graphics file contains an
image of the page that is transcribed in the text file.

5.  To investigate and if possible recommend on ways of
annotating images by attaching text files or other graphics
files  to them.

6.  To recommend standard methods for mutually referencing
several graphics files when those graphics files are images
of successive pages of a single book or manuscript, and for
indicating the relationships between: images of pages that
succeed one another; the images of the two sides of a leaf;
images of pages related by membership in a gathering or
signature; and so on.

7.  To investigate ways of assuring the integrity of images
contained in graphics files, so that users can be reasonably
certain that they are working with images that have not been
altered without their knowledge.