1.Improving the current implementation

[Should be completed]

2.Memory management

If I have the courage, I might one day write a garbage collector for TeXmacs.

3.Graphical interface

We wish to move towards Guile-Gtk or another portable graphical user interface. Now that we have found someone for making a Windows port, this not an urgent issue anymore though.

4.Encodings and fonts

• Systematically use Unicode.
• Some changes have still to be made in the way font encodings are handled. This should make it easier to maintain fonts with characters from several physical fonts, virtual fonts, special characters, etc.
• Currently, the current logical font is determined from a fixed set of environment variables only. The current typesetting font should rather be a tree (instead of a string), which is evaluated (so that environment variables are replaced) and then passed to find_font(display,tree). The current font can then be a joined font and future fonts might depend on user environment variables (i.e. colored fonts, using more than one color).

5.Speed

In order to speed up the program, we already made the major change that not the whole document is typeset again when making local changes. However, several other main optimizations should still be made:

• Indexing file names which are in the distribution. This may accelerate booting the program.
• Encode the system environment variables; this will globally accelerate the program. Also, intermediate data during the typesetting process might be encoded in a more binary way.
• The typesetter should be made lazy in a more fundamental way.

6.Scheme

The Guile/Scheme interface should become more robust and well documented. Several things still need to be done for this:

• Implement a system for “contextual overloading”.
• Better preservation of locality.
• Systematic use of closures throughout the code.
• A clean interface for manipulating TeXmacs content (a unified interface for both internal TeXmacs trees and the associated Scheme representation).
• Documentation.

7.Plans for the future

[Should be completed]

8.Typesetting

The typesetter should be reorganized and improved as follows:

• Decouple the style-sheet language from the actual typesetter.
• Better support for agglomerated documents (from physically different documents).
• Incorporation of better linking primitives (XLink, Proclus).
• More types of dynamic objects, like animations, help balloons, etc.
• Better line-breaking of large formulas.

9.Extra facilities for editing texts

Although cursor movement, selections, etc. have already been implemented, some other standard editing facilities have not yet been completed. Let us mention a few of these:

• Searching/replacing texts, formulas, certain environments etc. and regular expressions.
• Mathematical facilities: simplification of a selected region, substitutions of formulas in other formulas, etc.
• Version control.
• Data compression and protection.
• Grammar checkers and automatic translation programs. Does someone know where to find detailed free dictionaries and stuff like that?
• Incorporation of free speech recognition program.

10.A universal spreadsheet

We would like to incorporate a “universal spreadsheet” facility into TeXmacs. The idea is that all dependencies between the cells in the sheet are analyzed by TeXmacs, but all actual computations are delegated to an extern system of your choice, like one of the currently supported computer algebra systems. Also, the data of the spreadsheet will not necessarily be formatted in a rectangular table; one can also imagine dependencies between nodes of a tree, elements of a graph, or anything else.

11.Technical pictures

I also would like to include a facility for drawing technical pictures. In this implementation you should be able to benefit from the fact that you can define macros for making geometrical constructions. It would for instance be possible to write a style file for drawing electronic circuits or chemical components with a nice icon bar for selecting circuits or components, just as you select lines and circles in usual pictures.

12.Tools for usage on the web and in networks

It would be nice to have a better integration of TeXmacs with the web. As a first step, we need a clean internal plug-in for Wget or Curl with support for cookies, security, etc. At a second stage, the Html converters should be improved to take advantage of this. After that, we may think about things like collaborative authoring via the web, integration with preprint servers, etc.

Another interesting thing would be to incorporate tools for live conferencing inside TeXmacs using the recent concept of mutator tags. We actually expect this to be quite easy and this would open a different road towards collaborative authoring, instant messenging, etc.

13.Interface with computer algebra systems

The following improvements should still be made in order to link TeXmacs to computer algebra systems:

1. Improving the layout of computer algebra sessions.
2. Add extra features to increase the interoperability between TeXmacs and computer algebra systems and to give additional control over the layout of big output.
3. More semantics for the objects being communicated. This may either be high level information (like Openmath or HTML 4.0 mathematical markup) or low level information (including information about the representation of data), depending on the required speed.
4. Further possibilities for evolution concern highlighting, debugging facilities and so on.

14.Interaction with other GNU-like projects

It might be nice to increase the interaction between TeXmacs and other GNU-like projects, such as Gnome or multiplatform GUI's. This might facilitate the incorporation of extern data into TeXmacs documents or increase the number of supported platforms. On the other hand, several TeXmacs features, such as its font handling, might be interesting for other projects too.

15.Roadmap for upcoming developments

It is difficult to give a precise roadmap for the TeXmacs development, because our plans are permanently adjusted as a function of unexpected needs, help by new contributors and humour. Nevertheless, we tend to spend a significant time on a few major objectives, while amusing ourselves with the implementation of a few new features. Roughly speaking, our roadmap can therefore be divided into three major parts:

1. A limited number of major objectives for the upcoming stable versions (1.1 and 1.2).
2. A certain number of minor objectives, subject to continuous changes, and which progress as a function of available time.
3. Other projects, which are developed by external contributors, but which occasionnaly require changes in the TeXmacs core.

15.1.Major objectives for the upcoming stable versions

Our main focus for the next stable version 1.1 (to be released during 2007) is on the improvement of the user interface and documentation. This comprises the following developments:

• Increased standardization of the user interface: use M$-style keyboard shortcuts as the default (and provide Emacs compatability as an option), further simplification of the menus and increased use of popup windows.
• Development of a markup-based graphical user interface for popup windows. This new interface might only be ready after the next stable version 1.1.
• Development of user-friendly tools for documentation and translations, which can be used in a collaborative version over the web.
• Completion of a TeXmacs user manual and make it available in book form.

The next stable version 1.1 should also contain a more robust version of the graphical drawing tool, which is developed by Henri Lesourd. Finally, we plan to create an association for the proposition of free scientific software. One of the objectives of this association is to create a simplified system for making donations to TeXmacs and sell our software and documentation.

Within a slightly longer time period of about one or two years, we have also started to reorganize TeXmacs so as to make it a stable development platform for developments. The aim is to reach as quickly as possible a point where the different parts of TeXmacs are well documented and modularized, so that they can easily be further developed in parallel by different people. Most of these deeper developments will reach their maturity only in the after-next stable version 1.2, and comprise the following items:

• Improve the quality of the TeXmacs makefiles so as to make them completely compatible with automake and autoconf.
• Improve the Cygwin port and other ports of TeXmacs and monitor the availability of TeXmacs in major distributions.
• Reshape the low-level window interface so as to make the interface to Xlib easily portable to other systems, such as Gtk, Qt, MacOsX, Windows, etc. Also provide plugins for Cairo, OpenGL, etc.
• Replace the current widget system by a markup-based system, with the possibility to use widgets (from Gtk, Qt, Aqua, etc.) from standard GUI's instead of the TeXmacs-provided style files.
• Separate the style rewriting engine from the typesetter and make both completely lazy.
• Increase the robustness and use of DRDs (Data Relation Definitions), which contain meta-information about TeXmacs or user-provided DTDs.
• Migrate as much as possible of the high-level interface from C++ to Scheme.
• Increase the robustness of TeXmacs and its Scheme APIs by providing detailed exception semantics and tools for debugging.
• Provide extended documentation for developers and intelligent interactive ways to use this documentation.

15.2.Other things we are working on

Even though our main focus is on the stabilization of TeXmacs and to make it more suitable for developers, we intend to continue some time on the development on new features. A few points currently have our priority:

• Further development of the graphical drawing mode. First, the upcoming support of Cairo, OpenGL, etc. should improve the rendering quality. Secondly, we intend to allow the user to create new macros, either by explicit constructions or constrained-based constructions. Finally, we would like to incorporate several features which are typically found in editors of vector graphics, such as Inkscape.
• Extended features for computer algebra sessions, such as automatic folding of large expressions, lazy output (with subexpressions which can be further evaluated by clicking on them), cas-aided writing, etc.
• Providing more semantics to mathematical formulas by providing a simple way to construct parsers and pretty printers and to use them for establishing a bijection between presentation and content markup for certain well-defined languages.
• Continue the development of remote TeXmacs servers for providing web-based services in combination with TeXmacs. These services comprise chatting (with mathematical formulas) and collaborative authoring.
• Development of universal spreadsheet tools, which can be connected to any external plug-in for doing the computations. In a similar vein, external plug-ins might be used to draw graphs of functions in the graphical mode.

15.3.Ongoing external developments

Some people which are not part of the core development team use TeXmacs for other projects. We do our best to support such external developments and make the necessary changes inside TeXmacs when necessary. Here follows a list of a few initiatives that we are aware of:

• Lionel Mamane is developing a plug-in for the Coq proof assistant, called Tmegg. A first version of this plug-in is about to be released; please check Lionel's homepage. Similarly, Henri Lesourd is developing an interface to the Omega system. The use of TeXmacs as a front-end for proof assistants and theorem provers raises interesting questions about asynchroneous plug-in evaluations, keeping track and appropriate rendering of the state of a prover, appropriate mathematical and proof markup, etc.
• Saugata Basu, Richard Pollack and Marie-François Roy have written an interactive book “Algorithms in Real Algebraic Geometry” using TeXmacs. Any people interested in developing addional interactive features inside TeXmacs are invited to contact us.
• TeXmacs is used in combination with Maxima in a high-school education project at the Lycée Villegénis at Massy Palaiseau (nearby Paris). In view of this project, we intend to further simplify the user-interface and to make it easier to install up-to-date TeXmacs versions under Windows.
• Felix Breuer and formerly David Mentré have considered developing a literate programming plug-in for TeXmacs. Any concrete implementation of their proposals would happily find its way into the main distribution.

If you want to start a project based on TeXmacs, then please let us know.