Date: Thu, 21 Nov 1996 21:53:39 GMT Server: NCSA/1.5 Content-type: text/html
Computer Science Department Boston University 111 Cummington Street, MCS-211 Tel: (401) 781-5137 dgd@cs.bu.edu
I'm a graduate student in Computer Science at Boston University, working on collaborative editing. I'm more interested in text and hypertext than multimedia and hypermedia, and I'm most interested in how people can collaborate by using and editing shared hypertexts.
On my practical side I'm interested in multi-user collaboration on shared documents, on real networks (that is to say unreliable, regularly-disconnected networks). The experience of writing a book with a co-author has added a first-hand depth to what were originally theoretically-motivated beliefs.
On my radical side, inspired to a great degree by Ted Nelson, I'm interested in shared writing spaces where free writing and re-writing of evrything can take place, with a careful trail of what has happened as insurance against misrepresentation.
Fortunately for both sides of my professional personality, the same kind of technology is required to solve both problems. Real-time collaboration is something that I think is very important, but it's not one of my current areas of activity.
I'm also an SGML (or at least content-markup) bigot.
This page contains descriptions of my work on versioning and markup. You'll have to find some other way to scope out my hobbies and favorite links.
I'm working on a model I call Palimpsest after the overwritten scroll from which scholars extract previous versions of texts. Palimpsest is a general data model that provides very fine-level control of version management states and policies. It makes an explicit architectural assumption that a version management system is controlling data, and not processes for accessing data. This approach allows finer control at the cost of greater application overhead.
The Palimpsest paper was presented at the Workshop on Collaborative Editing Systems at CSCW '94. It is the most up-to-date documentation of my most general work on version control. I've done a partial HTML conversion of this paper. This converted version is complete except that it is still missing the first 2 illustrations. If you'd prefer a fully formatted version, download the file at the end of this link to get compressed postscript of the Palimpsest paper.
Fabio Vitali and I have written a paper proposing a system of version management for the World Wide Web. This is designed as a relatively easy to implement method to bring "artifact-based" collaboration to the WWW by supporting independent editing (via version control) over the Web. The recent announcement of widespread availability of Java-based Web browsers makes implementation of cross-platform client-side editors seem much more feasible. This paper will was presented at the Fourth World Wide Web Conference in December 1995.
I've recently written a little screed on versioning and the WWW for the Versioning Working group run by Jim Whitehead at the University of California.
With Fabio Vitali and the others in the group, I've been working on a draft of functional requirements for versioning on the web.
Content markup is technique for describing texts for future processing. The current baseline for all content markup systems is SGML (ISO standard 8879). However there are many questions about markup that are unanswered (and perhaps unanswerable), for instance, for all the talk of SGML being semantically based, there is no good description of what it means to mark up a document. There are also interesting problems with handling non-hierarchical markup in SGML, as well as good arguments that non-hierarchical structures are important in texts
I've been following up the work on non-hierarchical markup with Elli Mylonas and Steven DeRose in a paper abstract accepted at the 1996 ACH/ALLC annual conference in Bergen. This abstract describes some advanced directions for describing the structure of the kinds of complex texts and textual theories that come up in humanistic research.
The Text Encoding Initiative has a variety of lessons to offer to the HTML effort. In a paper to appear at the Fourth WWW conference, four fellow TEI alumni and I offer some Lessons for the World Wide Web from the Text Encoding Initiative.
The Versioning Working Group is a group of (currently four) people who are all concerned with the problems of version management in hypertext systems. Currently, we are working on a paper (a sort of union framework of the strategies for version and change management of which we are aware). We held a workshop at ECHT '94, which we hope will help to expand and solidify the "hypertext version management community."
Recently we held another workshop at ECSCW '95, whose proceeedings are not yet available.
Another interesting workshop was the Workshop on Collaborative Editing Systems held as CSCW '94.
Steven DeRose and I have written a book on the HyTime standard, which describes an SGML-based hypermedia interchange language.
More on Making Hypermedia Work: A User's Guide to HyTime.
Steven DeRose and I have also prepared some input to the ongoing HyTime review process -- A technical Corrigendum has been accepted at ballot, but The final wording of the upcoming changes has not yet been determined.
Steve and I have had a number of discussions of changes that might be made, and I've prepared a document discussing our proposed changes. Most changes proposed in this document have been accepted in principle (i.e. some things have been changed, but in a different way from what we suggested). A few were rejected (Our annex C comments, mostly), and some others deferred for more careful consideration.
This is my daughter Despina.
Last updated 6/5/96