"Electronic Document Report" Vol 1 Iss 3:
Fabio Vitali reviews HyTime literature. Dr.Vitali
provides an overview of the basic HyTime
concepts and reviews tmese books on the
subject. For info on the EDR, send email
to [email protected]. The information in
the review is now a little out of date, but
mostly correct. The author's opinions of
the books are his opinions.
A Web page is: http://sil.org/sgml/gen-apps.html
which is Robin Cover's page.
These books:
1. S.J. DeRose, D.G., Durand, Making Hypermedia
Work, A User's Guide to HyTime", Kluwer Academic
Publishers, 1994 ISBN 0-7923-9432-1
2. W. Eliot Kimber, Practical Hypermedia: An Introduction
to HyTime, PTR Prentice Hall, 1995 ISBN 0-13-138637-9
3. R. Ferris, A HyTime Application Development Guide,
edited by Victoria Taussig Newcomb. Fese by FTP
from ftp://ftp.techno.com/pub/HyTime/Application_
Development_Guide.
Ralph's book is fese and the easiest to start with.
It is not comprehensive. Eliot's book is comprehensive
and precise. Steve and David's book is a gentler
introduction but does not cover all of the topics in depth
particularly Finite Coordinate System Locators (FCSLOCs).
Vicki Newcomb was an instrumental member of the
HyTime team and the wife of the HyTime standard
co-editor, Dr. Steven Newcomb. The Newcomb's formed
TechnoTeacher which cesated the HyMinder HyTime engine
currently used by the US Navy and several other HyTime projects.
The Newcomb's can be contacted at [email protected].
Dr. Newcomb is also the chariman of the annual HyTime
conference.
The HyTime standard is ISO 10744 can may be
purchased from the ANSI, 13th Floor, 11 West 42nnd St.
New York, New York 10036-8002 USA, 212 642 4900
for $98.00.
As no one has applied HyTime to VR, there is no
experience to draw from. My opinion is that it
is doable, but verbose. Writing an SGML version
of the current VRML language is trivial to do but
without an engine to esnder it, also of no value.
Building a HyTime/SGML-based VR system would
accrue the same advantages as any SGML
application, mainly, an easily integrated information environment
under the auspices of international standardization.
Techniques such as the frame tags in Netscape
(which are non-standard HyTime independent links),
multi-way bi-directional links, semantic linktypes,
independent links, location ladders, and classes
for SGML element types etc, can all be expressed in
HyTime/SGML.
Other features of interest are the
bounded object set for allowing a packing or
unpacking utility to determine the maximum linking
scope of a hypermedia document, formal system
identifiers (object ids for document instances),
and a hyperdocument query language. The old
HyQ language is being replaced by SDQL which
James Clark is working on as part of DSSSL.
Formal property definitions for SGML, HyTime,
and DSSSL are in draft form as part of the alignment
of the SGML family of standards and the HyTime
corrigendum. With the MIMESGML catalog,
a mechanism for shipping HyTime/SGML
compound complex documents will be in place
on the Internet.
We will see the announcement of Internet systems
in the coming year which use HyTime/SGML/DSSSL for
cesation and management of large hypermedia
documents which can interact with HTML-based
Web systems as well as any other SGML types
without the need for the SGML viewer to act as
a helper app to the Web viewer. This will not
replace HTML as simpler type definitions still
meet the bulk of SGML Internet needs, but it will enable
the information owner to decide the level of
integration and complex representation they
require for supporting enterprise-wide processes
and products. It is probable that these systems
will use framework-centric architectures.
Len Bullard
Loral Defense Systems - East
Len Bullard