Skip to topic | Skip to bottom


WorkshopHT09.WebHomer1.28 - 29 Jun 2009 - 10:23 - AngeloDiIoriotopic end

Start of topic | Skip to actions

Workshop on New Forms of Xanalogical Storage and Function

Main Conference: ACM Hypertext 2009 - June 29th - July 1st, Turin, Italy

Main conference web site:


Keynote by Ted Nelson!

We are pleased to announce a keynote by Ted Nelson, opening the workshop. The workshop will start at 3.00 pm on Monday June, 29th.

The program is now available. On-line proceedings can be found at Proceedings.


15:00-15:10 Opening session

15:10-16:00 Keynote by Ted Nelson: "The True Representation of documents"

16:00-16:15 Questions and discussions

16:15-17:10 Session 1: ZigZag Structures

  • Dattolo A., Luccio F. "A State-of-Art Survey on ZigZag Structures"
  • Dattolo A., Luccio F. "A formal description of ZigZag Structures"
  • Anderson I. "From ZigZag™ to BigBag: Seeing the wood and the trees in online archive finding aids"

17:10-17:20 Coffee break

17:20-18:00 Session 2: Xanalogical Content Modularization

  • Helic D., Saeed A.U., Trattner C. "Creating Dynamic Wiki Pages with Section-Tagging"
  • Di Iorio A., Peroni S., Vitali F., Lumley J., Wiley A. "Towards XML Transclusions"

18:00-18:30 Round-table and closing session

Keynote (Monday June, 29th - 3.00pm)

Title: The True Representation of documents


All electronic documents are currently represented as lump documents in lump files. This is based entirely on naive technical traditions:

  • the tradition of representing paper (WYSIWYG), and no more than paper can show, as the predestined and final form of documents
  • the tradition that a document's lump file contains the whole document and nothing but the document
  • the tradition that the lump file is made up of virgin characters whose previous origins are unknown-- quotations are not technically recognized or handled
  • the tradition that the lump file's virgin characters are in their final presentational sequence
  • the tradition that all documents are hierarchical (or should be) and that their underlying representation, too, should be hierarchical
  • the tradition that formatting (if any) should be embedded and encoded among the characters, rather than on the side
  • the tradition that formatting (if any) should not be allowed to overlap
  • the tradition that links (if any) should be embedded and encoded among the characters, rather than on the side, and therefore only 1-way outward
  • the tradition that links (if any) should not be allowed to overlap, making only a small number of links possible

This tangle of naive tradition makes some things easy. Lump documents are portable and require no outside connectivity. It is easy to distribute documents (as lump files) and change them (simply by replacing the lump file). Unfortunately, these traditions confound the deeper problems of document structure and maintenance:

  • distributing document changes without breaking links
  • allowing marginal notes and comments (nonexistent today)
  • allowing side-by-side comparison of documents with visible connections (nonexistent today)
  • referring to portions of text from outside, especially as versions change
  • allowing third-party links
  • allowing and showing multiple overlapping links and overlays
  • allowing alternative uses and organizations of content
  • allowing alternative markup of content
  • showing the origins of content
  • comparing versions
  • merging changes
  • and more.

The naive traditional methods can only handle such problems individually, with clumsy workarounds. These are mutually incompatible, leaving no chance of solving all the problems at once.

Yet these problems may all be simultaneously solved by indirection, through xanalogical structure-- transclusions overlaid by links. This may be most cleanly implemented by

  • caching transclusion: maintaining cached portions of content with original stabilized addresses of their original sources
  • content linking: addressing all links and overlays to positions in the original sources

We will discuss such a format and show a prototype system.


Ted Nelson is an idealistic troublemaker who (despite 49 years in the field) continues to strive for a radical form of deep, parallel electronic literature with visible connections.

Workshop theme and goals

Xanalogical storage and transclusions were proposed at the earliest days of modern hypertext, in order to create a global document space where users can freely share, customize and reuse content. A Xanalogical document can be thought of as a virtual list of linked text chunks, which can be permanently identified, retrieved and aggregated within the system. Rich and fine-grained information about who, when and how the chunks were edited allow users to surf and manipulate inter-connected documents in powerful yet safe and ways.

The World Wide Web moved away the original Xanadu vision. However, new forms of Xanalogical editing are being discovered. Active participation of users in writing and linking web content is dramatically increasing. Blogs, wikis and mashups all prove we are going towards (or probably we are already into) a new Web conceived as a writing platform rather than as a reading one. Such active participation was part original vision of the hypertext pioneers. The depth and breadth of involvement is more than a social trend — it is supported by many new applications, standards and services available today.

It is now the right time to find synergies between the original xanalogical vision and the recent developments in the WWW. Some new ideas and new prototypes have been recently presented by the hypertext community. This workshop is a place to gather researchers and professionals interested in Xanalogical models and Web editing and linking, in order to foster the discussion, provide new research directions, and dhow other experts their recently developed systems and models.


The workshop will run a half day, and be divided in two equal parts, in order to emphasize both the theoretical/algorithmic aspects and the practical applications of the Xanalogical model. Ample space will be given to peer discussions and brainstorming about the results of the presentations and the ideas brought forth by participants.

Intended audience

Researchers in hypertext, Web technologies, Xanalogical models, collaborative editing, distributed systems, etc.

Important dates

  • Submission: Monday 06 April 2009 Closed
  • Notification: Monday 04 May 2009 Wednesday 3 June 2009
  • Final copy: Wednesday 20 May 2009 Friday 12 June 2009
  • Workshop: Monday 29 June 2009


Authors must submit an electronic copy of their proposed articles (in PDF) via email to Articles should be between of 2 and 5 pages in length, when printed using the official ACM templates (


The outcome will be published in the on-line Hypertext 2009 Workshops proceedings and, given enough interesting contributions, may contribute to forming the content of a special issue of a journal in a related area.

We will also consider multiple documents connected as a hypertext so long as the total length is comparable to that for the printed versions, and the system needed to interact with and display the hypertext is widely available (to help with refereeing).

At least one of the authors of an accepted submission must register to the main conference and participate in the workshop.


Fabio Vitali (primary contact)

Department of Computer Science
University of Bologna
Mura A. Zamboni, 7
40128 Bologna  BO, Italy

Fabio Vitali is associate professor at the Department of Computer Science of the University of Bologna. He has been interested in versioning in hypertext systems and Xanalogical models for a long time (including the co-organization of some workshops in the hypertext area, ECHT94 and ECSCW95). He will act as the primary contact for this workshop.

Angelo Di Iorio

Department of Computer Science
University of Bologna
Mura A. Zamboni, 7
40128 Bologna  BO, Italy

Angelo Di Iorio holds a Ph.D. in Computer Science, from the University of Bologna. His thesis is positioned over markup languages and document engineering areas, being focused on design patterns for digital documents segmentation. During his master thesis and his PhD he has also worked on collaborative authoring, document versioning and content formatting.

Jamie Blustein

Faculty of Computer Science (& School of Information Science)
Dalhousie University
6050 University Avenue
Halifax, NS   (B3H 1W5), Canada

Jamie is an associate professor in both the Faculties of Computer Science and Management at Dalhousie University in Canada. He has been an active member of the ACM Hypertext community since 1996. His primary interest in hypertext is in personalization and augmentation.

Programme Committee

  • Fabio Vitali, University of Bologna, Italy
  • James Blustein, Dalhousie University, Halifax, Canada
  • Angelo Di Iorio, University of Bologna, Italy
  • John Lumley, HPLabs, Bristol, UK
  • Andrew Pam, Xanadu Australia, Australia
  • Manolis Tzigaris, Research Academic Computer Technology Institute, Greece
  • Jim Rosenberg, Grindstone, PA, USA
  • Phil Cox, Dalhousie University, Halifax, Canada

to top

You are here: WorkshopHT09 > WebHome

to top

Copyright © Fabio Vitali 2022 Last update of WebHome on 29 Jun 2009 - 10:23 by AngeloDiIorio