About Thot

Thot is a generic system for the development of document centered applications based on the concept of structured active documents. It can be used for building interactive system as well as automatic processors. It's the result of many years of research at INRIA, CNRS and the University of Grenoble (UJF), with a number of academic and industrial collaborations. The Grif products, for instance, are a result of these collaborations.

The most significant contributions are acknowledged here.

Initial research

The earliest work at the origin of Thot is a joint research project launched in 1983 by INRIA, CNRS and the University of Grenoble at the IMAG Institute. At that time, producing documents with a computer was much more complex than today. The most common tools were batch formatters which imposed a complex process to their users. However, some of them, such as Scribe and later LATEX, had a clean document model based on the logical structure of documents that presented a number of interesting features. The project aimed at exploring the problems to be solved to combine the advantages of these formatters and those of interactive systems, hence the name of the project: Grif (GRenoble Interactive Formatter).

During the first three years, the work was mainly carried on by a small team leaded by Vincent Quint (INRIA) and Irène Vatton (CNRS, now at INRIA). Three languages were defined for specifying respectively the logical structure of documents, their presentation and their external syntax. An editor called Grif was also built to implement these languages in a single interactive editor-formatter.

Industrial collaborations

After publication of the first results [1], several collaborations started with the industry and the academia.

In 1986, GSI Tecsi was the first software company that worked with the research team. They used the Grif prototype for developing an ODA editor for the PC. They also stimulated the research group to add missing features. This cooperation was the opportunity to integrate a page model in the editor and a PostScript output, to support WYSIWYG printing. Hassan Bedor, a PhD student, was instrumental in this resarch.

Another project, called HYP, started in 1988, again with GSI-Tecsi, funded by the French Ministry of Industry. The main objective was to combine a rich hypertext functionality with structured editing. A large part of the hypertext features [2] of the Grif editor come from this work.

Most of these extensions aimed at improving the original Grif editor-formatter. Cooperation with SFGL opened a new perspective to the structured approach to document editing. This company was interested in using Grif as a user interface component for a software development environment. This request helped the research team to consider a more modular approach, where structured editing is a functionality added to a broader software environment. This was also the opportunity for the research team to extend Grif in the field of structured graphics.

In 1988, Gipsi, a start-up of INRIA, adapted the Grif editor to their X terminals. They made improvements, specially to the user interface, to turn it into an industrial product. They then decided to use it as the kernel of a new SGML editor, as most of the concepts of SGML were already implemented. The research team extended then the document model and the editor to make them (almost) fully SGML compatible. The resulting software was called Grif SGML Editor. This work was done (and funded) as part of ESF (Eureka Software Factory) an Eureka project. In 1991, the group that developed the Grif product at Gipsi founded a new company, called Grif SA, to develop and market that technology.

In 1990, the joint research team from INRIA and IMAG counted 7 persons and received the status of INRIA project with the name Opera. Structured documents and related issues were the main topics of project Opera, but other issues such as digital typography, hypertext and cooperative editing were also part of the project charter. Among the new activities launched by project Opera, one should mention the research of H. Richy on spelling checking, as the checker she developed with E. Picheral [3] is now part of Thot.

In 1991, the Euromath trust, which developed a working environment for mathematicians using the Grif SGML Editor from Grif SA, asked project Opera to study the issue of structure transformations in structured documents. This work was funded by Euromath and accomplished by E. Akpotsui [4] as part of his PhD. This is the origin of the work continued later by Ph. Claves and more recently by S. Bonhomme and C. Roisin, who developed the structure transformation mechanism for Amaya [5].

In 1993, project Opera contributed with Grif SA and O2 Technology in a project funded by France Télécom for studying the problems posed by cooperative editing of structured documents stored in an object oriented data base.

At the beginning of 1995, INRIA started a major joint project with Dassault-Aviation in the area of concurrent engineering. Project Opera was involved in addressing the issue of interactive structured documentation, and took that opportunity to complete a task that started one year before: the design of a new generation of structured editing tool, Thot. Thot is not an editor, but a set of libraries that can be used for building applications based on the concept of structured active documents [6].

The latest industrial collaboration of project Opera started in September 1996 with SGS-Thomson and is funded by the French Ministry of Industry. In this project, Storia, Thot is used as a component of the Alliance cooperative editor [7] that is experimented for improving cooperation between remote sites.

Academic cooperations

All activities presented in the above section have been done in the context of formal cooperations with industrial partners. Obviously, project Opera had also a number of less formal contacts that had a strong influence on its research activity. It is difficult to mention all people and organizations that deserve some acknowledgement but we hope that the most influential are listed below.

First of all, the following members of project Opera should be cited, as they have strongly contributed to the development of Grif and Thot: J. André, S. Bonhomme, D. Decouchant, N. Layaïda, V. Quint, H. Richy, C. Roisin, I. Vatton.

As most of the research that led to Thot has been done in a joint team from IMAG and INRIA, many people of these institutes should also be mentioned, but the joint research team Bull-IMAG deserves a special mention, as many collaborations have been done with them, through a number of student projects and PhD theses where Grif and Thot were involved. These projects helped a lot the Opera team to develop Thot and to specify and implement its API and its application definition language (language A). Other IMAG or INRIA projects were also very helpful in this respect, especially Spectre (Ph. Schaar), Chloe (A.-M. Vercoustre) and Acacia (Ph. Martin).

Project Opera had a number of contacts abroad during its research on structured documents and hypertext. In the early work on structured editing, preparing a winter school and several publications [8] [9] with R. Furuta (Texas A&M University) and J. André (INRIA-IRISA) was very helpful. Several discussions and seminars prepared with G. Coray (EPFL), Ch. Vanoirbeek (EPFL) and R. Ingold (Freiburg university) were also a way of refining our approach, as well as contacts with M. Harrison and E. Munson (UC Berkeley).

Xerox (PARC and RXRC) should also be mentioned, as the Opera project had many contacts with researchers in that company. In particular, experiments for interchanging documents with Tioga helped a lot to improve the T language.

From Thot to Amaya

Project Opera started its work with the Web by two simple experiments. Language T was used to make a filter that generates an HTML form for Thot documents. The HTML document model was also entered in Thot. After these first steps, a prototype HTML editor/browser, called Tamaya was developed on the basis of the Thot library.

This work was then continued in the context of project Wedi (Web EDItor), funded by the European Commission (DG XIII), whose goal was to develop a browser/editor for the Web. Tamaya was used as a workbench for experimenting new ideas. Grif S.A. participated in the project with the mission of building a commercial product, which is now known as Symposia.

While Symposia is based on the Grif SGML Editor, Tamaya was specifically developed for HTML. It contains a parser that is flexible enough to cope with most existing Web documents [10] and it implements a subset of the Cascading Style Sheets (CSS). It was demonstrated at the 3rd and 4th WWW conferences, and, at the beginning of 1996, W3C proposed to project Opera to carry on the development of Tamaya within the Consortium.

Some people from INRIA then joined the W3C and, with other W3C staff members, defined the specifications for the new Web client derived from Tamaya. A new name was chosen, Amaya, and a new user interface was designed that should make the editor as user friendly as a word processor, while keeping its structured representation of documents.

The first beta version of Amaya was released to W3C members in July 1996 and to the public one month later. The source code is available since January 1997.

The first version of Amaya does not take advantage of all features available in the Thot library, on which it is based, but as Amaya is essentially considered as an open-ended platform for experimenting new web features, this rich set of editing and formatting functions is an advantage for further developments.

References

[1] V. Quint, I. Vatton, ``Grif: an Interactive System for Structured Document Manipulation'', Text Processing and Document Manipulation, Proceedings of the International Conference, J. C. van Vliet, ed., pp. 200-213, Cambridge University Press, 1986.

[2] V. Quint, I. Vatton, ``Combining Hypertext and Structured Documents in Grif'', Proceedings of ECHT'92, D. Lucarella, ed., pp. 23-32, ACM Press, Milan, December 1992.

[3] H. Richy, P. Frison, E. Picheral, ``Multilingual String-to-String Correction in Grif, a Structured Editor'', Proceedings of Electronic Publishing 1992, EP92, C. Vanoirbeek et G. Coray, ed., pp. 183-198, Cambridge University Press, April 1992.

[4] E. Akpotsui, V. Quint, ``Type Transformation in Structured Editing Systems'', Proceedings of Electronic Publishing 1992, EP92, C. Vanoirbeek et G. Coray, ed., pp. 27-41, Cambridge University Press, April 1992.

[5] S. Bonhomme, C. Roisin, ``Interactively Restructuring HTML Documents'', Proceedings of the Fifth International World Wide Web Conference, Computer Network and ISDN Systems, vol. 28, num. 7-11, pp. 1075-1084, May 1996.

[6] V. Quint, I. Vatton, ``Making Structured Documents Active'', Electronic Publishing - Origination, Dissemination and Design, vol. 7, num. 2, pp. 55-74, June 1994.

[7] D. Decouchant, V. Quint, M. Romero Salcedo, ``Structured and Distributed Cooperative Editing in a Large Scale Network'', Groupware and Authoring, R. Rada, ed., pp. 265-295 (chap. 13), Academic Press, May 1996.

[8] J. André, R. Furuta, and V. Quint, Structured Documents, Cambridge University Press, 1989.

[9] R. Furuta, V. Quint, J. André, ``Interactively Editing Structured Documents'', Electronic Publishing -- Origination, Dissemination and Design, vol. 1, num. 1, pp. 19-44, April 1988.

[10] V. Quint, C. Roisin, I. Vatton, ``A Structured Authoring Environment for the World-Wide Web'', Proceedings of the Third International World-Wide Web Conference, Computer Networks and ISDN Systems, vol. 27, num. 6, pp. 831-840, April 1995.