==================================================================== @LaTeX3-article{ LaTeX3-L3-004-txt, filename = "l3d004.txt", archived = "ctan:/tex-archive/info/ltx3pub/", author = "Frank Mittelbach and Chris Rowley", doc-group = "Project core team", title = "The {\LaTeX3} Project; Report 1991--92", version = "1.00", date = "01 April 1993", time = "13:37:20 GMT", status = "public, official", abstract = "This report describes the current activities of the \LaTeX3 project. After a short project overview, the major events of 1991 and 1992 are described. Finally the project plan for 1993 is presented. ", keywords = "", project-address = "LaTeX3 Project \\ c/o Dr. Chris Rowley \\ The Open University \\ Parsifal College \\ Finchley Road \\ London NW3 7BG, England, UK", project-tel = "+44 171 794 0575", project-FAX = "+44 171 433 6196", project-email = "LTX3-Mgr@SHSU.edu", copyright = "Copyright (C) 1993 LaTeX3 Project and Frank Mittelbach and Chris Rowley. All rights reserved. Permission is granted to make and distribute verbatim copies of this publication or of coherent parts from this publication provided this copyright notice and this permission notice are preserved on all copies. Permission is granted to copy and distribute translations of this publication or of individual items from this publication into another language provided that the translation is approved by the original copyright holders. No other permissions to copy or distribute this publication in any form are granted and in particular no permission to copy parts of it in such a way as to materially change its meaning.", generalinfo = "To subscribe to the LaTeX3 discussion list: Send mail to listserv@vm.urz.uni-heidelberg.de with the following line as the body of the message (substituting your own name): subscribe LaTeX-L First-name Surname To find out about volunteer work: look at the document vol-task.tex which can be obtained electronically, see below. To retrieve project publications electronically: Project publications are available for retrieval by anonymous ftp from ctan hosts: ftp.tex.ac.uk ftp.dante.de ftp.shsu.edu in the directory /tex-archive/info/ltx3pub. The file ltx3pub.bib in that directory gives full bibliographical information including abstracts in BibTeX format. A brief history of the project and a description of its aims is contained in l3d001.tex. If you only have access to email, and not ftp You may use the ftpmail service. Send a message just containg the word help to ftpmail@ftp.shsu.edu for more information about this service. For offers of financial contributions or contributions of computing equipment or software, contact the project at the above address, or the TeX Users Group. For offers of technical assistance, contact the project at the above address. For technical enquiries and suggestions, send e-mail to the latex-l list or contact the project at the above address.", checksum = "30097 902 5006 49192", docstring = "The checksum field above contains a CRC-16 checksum as the first value, followed by the equivalent of the standard UNIX wc (word count) utility output of lines, words, and characters. This is produced by Robert Solovay's checksum utility.", } ==================================================================== The LaTeX3 Project (C)1993 Frank Mittelbach Chris Rowley Report 1991--92 Abstract This report describes the current activities of the LaTeX3 project. After a short project overview, the major events of 1991 and 1992 are described. Finally the project plan for 1993 is presented. Project overview The aim of this project is the development of a new version of LaTeX with the following enhancements. * It provides an enhanced user-interface, facilitating the expression of a large range of typesetting requirements from fields both within and outside the natural sciences. This includes the requirements of technical documentation (offset layout, change bars, etc.) as well as requirements from humanities (critical text editions, etc.), many of which are not supported by the current LaTeX. * It provides a more robust user-interface. For example, restrictions on the nesting of commands will be removed; error handling will be improved. * It provides a syntax that allows (semi-)automatic translation from popular SGML DTD's into provided LaTeX document style options. The new syntax will include, for example, attribute and short reference concepts that can be mapped directly into corresponding SGML features. * It provides a style-designer interface that allows easy specification of a wide variety of layouts. This includes, for example, access to arbitrary font families. * The style-design language should be as natural as possible. It will be based on a hierarchical collection of generic formatting functions, so that layouts can be specified by parameters without involving difficult TeX coding. * The new interfaces will be described in detail, with carefully chosen examples, making the learning time LaTeX3 Project --------------- for new users (both designers and authors) as short as possible. To achieve this aim it was necessary to review large parts of the user-interface and to design a complete new language for style-design. One result of this work was to show the necessity of completely rewriting the kernel of the current LaTeX. Therefore the project is divided into six partly overlapping phases. Phase 1: Review and redesign of the user-interface. The requirements imposed on a typesetting language by various applications are analysed. This includes the proposed translation from popular DTD's into the language of the LaTeX user-interface. Phase 2: Prototype implementation of new concepts. The new concepts developed in phase one are implemented and tested. The tests should not only prove that the concepts are implementable within TeX but also test the functionality of the concepts. Larger real-life applications from different fields will show to what extent the new concepts are suitable and might lead to some refinements in the design of the user-interface. Phase 3: Design of a designer-interface. This phase is more or less independent of the first two phases. The current LaTeX is considered unsatisfactory since it forces the designer to become deeply involved with TeX's algorithmic philosophies and these are likely to be unfamiliar to him. An analysis will be made of the extent to which it is possible to support a layout-description language that is natural for a designer. Phase 4: Prototype implementation of the designer interface. This is similar to phase two. The concepts resulting from phase three are implemented in prototype versions which are used to test the acceptance of the 2 LaTeX3 Project --------------- new designer-interface. In this area implementation details are relatively uncritical, whereas the development of the interface might pose difficult questions since there is no previous research in this area. Phase 5: Re-implementation of the LaTeX kernel. The results from phases two and four are combined and a system kernel that supports both interfaces is implemented. At this point beta-testing at selected sites is started. Phase 6: Documentation. While internal documentation is an integral part of every phase, the main work of phase six (manuals and conversion) is devoted to the writing of informative manuals covering both the user- and designer-interface. Milestones '91 Shortly after the start of the project three important events took place: a workshop with professionals from the publishing industry in Dedham, followed by working week at the American mathematical Society (AMS) in Providence and later on a meeting with Professor Donald Knuth in Stockholm. In the last quarter of 1991 substantial effort was put into the development of an internal LaTeX3 programming language. A prototype kernel, written in a first approximation to the final language, formatted its own documentation (a 120 page document) for the first time in December. Workshop After the annual meeting of the TeX Users Group in Dedham Dedham (July 91) a two-day workshop was held to discuss design questions for LaTeX3 with TeX experts, compositors and other professionals from the publishing industry. In this workshop (lead by Chris Rowley and Frank Mittelbach) questions of page layout specifications and user interface design were discussed. It was based on material gathered and collated by David Rhead from previous electronic mail discussions about 3 LaTeX3 Project --------------- these topics on the LaTeX3 discussion list. This workshop was sponsored by the TeX Users Group. Working week Following this workshop Chris Rowley, Michael Downes at AMS (AMS), and Frank Mittelbach worked for one week at the AMS headquarter at Providence on the specifications for the LaTeX3 error recovery and help system and on some problems related to the efficiency of TeX's approach to mathematical typesetting. Paris At the European TeX conference in Paris (September), conference Chris Rowley and Rainer Schoepf presented an overview of the enhancements to be made to the user interface. Stockholm In November Frank Mittelbach was invited to Stockholm to attend the celebrations for the Honorary Doctorate of Donald Knuth and to give a paper on LaTeX3 at the Nordic TeX conference. The handouts for this paper were later published in TeXline [7] and in [9]; later an enlarged version, co-authored by Chris Rowley, appeared in [14]. Whilst in Stockholm Frank took the opportunity to discuss for two days with Donald Knuth design decisions he had taken in the development of the TeX system. Milestones '92 The last year has seen substantial progress in three major areas: establishing structures to consolidate support for LaTeX 2.09 (these structures will therefore be well tested before they are needed for LaTeX3); deciding which of the many requirements of a high-quality document formatting system can be implemented using TeX as the formatting engine; research on models of document formatting which support high-quality typography. These last two are, of course, closely related and mutually supportive. SGML Users In February Chris Rowley gave a presentation to the Group inaugural meeting of the UK chapter of the SGML Users Group. This gave an outline of the project and its importance to the formatting of documents encoded in accordance with SGML. 4 LaTeX3 Project --------------- Hamburg At the Dante Meeting in Hamburg (March 92) Frank Mittelbach presented an overview of the LaTeX3 project; this was later published as [8]. After the conference a two day workshop with various TeX and LaTeX experts from several European countries was held to discuss questions of the user and style-designer interfaces for LaTeX3. Questionnaire Rolf Lindgren, a student in the field of psychology, volunteered to develop and analyse a questionnaire on the subject of site-specific usage, including additions and changes, of popular extensions to LaTeX 2.09. The project team hopes that this will provide important data about such areas as the current usage of LaTeX, which of these additions are the most important for inclusion in the main system of LaTeX3, etc. A first draft of the questionnaire was handed out to the participants at the Hamburg meeting and Rolf Lindgren used the results of this to write a preliminary report [5] and also to redesign the questionnaire itself. The final questionnaire will be sent out to a representative sample of sites in 1993. Lausanne At the Electronic Publishing Conference EP92 in conference Lausanne, Chris Rowley and Frank Mittelbach gave a paper about potential problems when trying to produce high quality output using automated document typesetting systems [18]. The results presented in this paper were influenced by their work on the LaTeX3 system. On the other hand, this basic research into models for document formatting has direct benefits for the LaTeX3 project since it leads to the development of useful new concepts and methods. CERN Frank Mittelbach and Chris Rowley were also invited to CERN to give a talk about the LaTeX3 system and to discuss problems encountered in the production of high-quality output from SGML sources. Addressing such problems is a major goal of the LaTeX3 project. 5 LaTeX3 Project --------------- Mainz At the end of April Frank Mittelbach invited the core workshop development team for LaTeX3 to an 11 day workshop in Mainz to discuss data structures for the kernel and for the style designer language. This workshop was made possible through support from the Zentrum fuer Datenverarbeitung der Universitaet Mainz (providing working facilities), the AMS (facilitating Michael Downes attendance), and DANTE, the German TeX Users Group (help with other expenses). The main results of this workshop were summarized in an internal project paper by Denys Duchier [2]; they are mainly of a somewhat theoretical nature, e.g., discussions about what algorithms to use, the nature of internal interfaces, etc. Some of them are now implemented in prototypes, others are scheduled for further work during 1993. Paris meeting In May GUTenberg (the French TeX Users Group) invited Rainer Schoepf and Frank Mittelbach to a one day meeting in Paris to discuss various aspects of the LaTeX3 project. They presented an overview of the user interface, a summary of the structure of the LaTeX3 system, the layout specification interface and the coding of tabular material. They also discussed details of the LaTeX3 programmer's environment, the short reference mechanism, the float mechanism, the mark mechanism, the I/O interface and the error recovery mechanism. GUTenberg also made a much appreciated contribution to the project's expenses. Later in the year the possibility of a collaboration between the LaTeX3 project, the University of Mainz and one or more French institutes was discussed. This collaboration is now becoming concrete and it is planned that, in the second half of 1993, French students of informatique will spend periods of three or more months in Mainz and assist Frank Mittelbach in the development of the LaTeX3 system. Macro Together with David Carlisle, Frank Mittelbach developed validation a system for automatic validation of changes in large system macro packages [13, 1]. The purpose of this system is to 6 LaTeX3 Project --------------- detect errors introduced by incorrect bug fixes to the macro packages before such changes are distributed. The quality of the automated checking system depends largely on the number of test files in it. It was first tested with the New Font Selection Scheme (NFSS) package [22] where it was successful in preventing some mistakes from getting into the distributed version. After this successful test a group of volunteers under the direction of Daniel Flipo started to write test files for the LaTeX 2.09 base system; these will improve the quality of its maintenance. It is intended that they will also be used as a test base for the user interface of the LaTeX3 system. New A considerable amount of time during 1992 was devoted to LaTeX 2.09 reorganizing the distribution and support of the styles style which have been developed by Rainer Schoepf and Frank distributions Mittelbach for the LaTeX 2.09 release. A lot of work was done on the code-documentation package, consisting of the program docstrip and the style doc. The program docstrip was reimplemented and enhanced by Johannes Braams, Denys Duchier and Frank Mittelbach. The documentation of the doc style was brought up-to-date by David Love [11]. The array package, which contains an enhanced reimplementation of the array and tabular environments of LaTeX 2.09, was improved by the addition of code written by David Carlisle [12]. Frank Mittelbach rewrote parts of the multicol package to support \marks for use in running headings and to allow for up to ten columns [10]. These packages, and all others, have been reorganized to make use of the extra features of the new docstrip program; this will ensure their smooth and easy installation at every site having a TeX installation. LaTeX 2.09 At the same time, a new system for maintaining the maintenance LaTeX 2.09 base system was installed at Mainz by Rainer Schoe pf and the whole base system was transferred to this site so that it can be maintained by him and Frank Mittelbach. 7 LaTeX3 Project --------------- BibTeX reim- One of the important programs in the LaTeX system is plementation BibTeX, a system for managing bibliographical material and producing bibliographies in documents. At the TeX Users Group Conference in Cork (1990) Reinhard Wonneberger and Frank Mittelbach presented a paper discussing several proposals for a reimplementation of this program [24]. During the summer Oren Patashnik, the original author of BibTeX, was contacted to discuss with him a reimplementation of BibTeX. He agreed to reimplement this program so that it will interface smoothly with the new LaTeX system. These reimplementation proposals are now being discussed by a small group of experts and work is under way to finalize the user interface for providing bibliographical information in the LaTeX3 system; this will support the many distinct citation conventions (author-date system, etc.) that are in common use. In addition to providing the LaTeX3 system with a bibliographical system which is also freely available, it is planned to provide, wherever possible, interfaces linking LaTeX3 with commercially available bibliographic software. This will, for example, enable users to integrate into LaTeX documents information extracted from large bibliographic databases. Much of the work in this area is being carried out by David Rhead. TUG annual At the annual meeting of the TeX Users Group in meeting Portland, Oregon, Chris Rowley gave a talk which surveyed the current status of the LaTeX3 project [23] and ran a mini-workshop at which some of the more technical aspects were discussed. This meeting was important in establishing the project as being of central importance to the community of TeX users and developers and to the future of TeX as a formatter for high quality typographic output. Volunteer One of the results of the meeting of the core tasks development team in Mainz was to single out tasks that could be taken up by volunteers whose work could be to a large extent independent of the core team. During the summer, Michael Downes and Frank Mittelbach prepared a task list and working procedures for such volunteers. 8 LaTeX3 Project --------------- George Greenwade from the Sam Houston State University kindly agreed to serve as a task list manager; the Volunteer Task List was therefore presented to the public at the Prague conference in September [15, 17, 16]. Prague In September the European TeX conference took place in conference Prague. Chris Rowley and Frank Mittelbach reported on the status of the LaTeX3 project and described some aspects of the formatting model which it will implement. In a separate presentation they described some results of their research on more general models for document formatting. Series During the conference in Prague Frank Mittelbach agreement concluded with Peter Gordon (Addison-Wesley) publication agreements concerning the documentation of the LaTeX3 system and related books. They will appear in a book series with the tentative title `Tools and Techniques for Computer Typesetting' under the editorial supervision of Frank Mittelbach and with the participation of Leslie Lamport and other key players in the LaTeX3 project. BCS meeting In September Chris Rowley was invited to give a talk about LaTeX3 at a meeting on Structured Documentation organised by the British Computer Society EP Specialist Group in Nottingham. This contained a brief history and outline of the project, followed by some discussion of the requirements of the designer's interface. Working week Leslie Lamport visited Frank Mittelbach for five days in in Mainz October to discuss the work in progress. They covered many aspects of the user interface including the important mechanism for attaching attributes to commands and environments. 9 LaTeX3 Project --------------- Washington During the Lausanne Conference Chris Rowley and Frank conference Mittelbach had been invited to present further results of their research on document formatting models at an international `Workshop on Principles of Document Processing' in Washington DC, organized by the Xerox Webster Research Center, Cornell University and Syracuse University. This talk, which has now been written up for publication, led to many fruitful discussions. These covered many aspects of the relationship between their approach and that of other work on models of document processing. The workshop also showed the need to investigate the relationship between two types of model: those similar to that of Mittelbach/Rowley, which are oriented towards data structures and operations that can be effectively implemented, and those of a more general nature, which describe a large range of document processing applications in terms of formal structures such as attribute grammars and trees. Bell-Labs During their stay in the US they also visited AT&T Bell Laboratories, invited by John Hobby, who worked under Knuth on the TeX project in Stanford. In addition to giving a talk about their research in document formatting models and discussing this with researchers there, they had the opportunity to discuss with Brian Kernighan his recent work on page-makeup by post-processing [4]. Multiple As part of the LaTeX3 kernel Frank Mittelbach developed marks prototype code for multiple independent marks. This allows information, such as the first or the last index entry for each column in a multi-column layout, to be recorded for later use in the final page make-up. Due to the limited support for such layout structures in TeX, the current LaTeX only provides a very simple mechanism. Bremen Following the Lausanne conference Frank Mittelbach was colloquium invited to give a colloquium talk on the subject `Computereinsatz in der Typographie: Kann die hohe Qualitaet des Handsatzes durch Computerprogramme 10 LaTeX3 Project --------------- erreicht oder uebertroffen werden?' at the Computer Science Institute of the Bremen University. During his two-day visit to Bremen Frank Mittelbach had several fruitful discussions with the research team working with Professor Frieder Nake on research problems in computer controlled typesetting. Meppel In the second half of November Frank Mittelbach was conference invited to give a talk at the meeting of the Netherlands TeX Users Group in Meppel. He spoke about possible directions in research on typesetting problems, discussing tasks which are impossible to achieve using state of the art typesetting systems such as TeX. This talk was based on a paper published in the conference proceedings of TeX90, [6]. Nordic TeX The Nordic TeX users group invited a representative of conference the LaTeX3 project to speak at their meeting in Kobenhavn so Chris Rowley presented a brief history of the project and discussed several aspects of the designer interface. He also discussed future collaboration with other research projects in the area of document processing. Euromath Chris Rowley has been exploring the relationships between the LaTeX3 project and the Euromath project. This latter project was started in 1986 and is aimed at the production of an integrated system which provides a homogeneous working environment for mathematicians. It is based on the Grif structure editor, which provides an object oriented system and applications programming interface. Thus LaTeX is their natural, and preferred, choice of application for the production of publishing-quality typeset output. alpha-test In the last quarter of 1992 Frank Mittelbach rewrote the NFSS2 NFSS system (New Font Selection Scheme for LaTeX [20, 21]), based on a prototype extension to NFSS written by Mark Purtill in 1991. This reimplementation extends the released version in several ways including encoding support (multiple languages), scalable font 11 LaTeX3 Project --------------- support (PostScript fonts) and math font support. In addition Frank Mittelbach wrote new documentation for the package, which will appear as a chapter of [3]. The alpha-testing of the new release was started at the end of December and the official release is planned for summer 1993. Error For error recovery a prototype had already been recovery and implemented early in 1992, as part of the prototype help system kernel for LaTeX3, but the interface to an interactive help system existed only as a specification written during the working week at the AMS in Providence. A first prototype was implemented by Michael Downes in December, so that it is now possible to do small field tests; this implementation is, however, currently only available as a LaTeX 2.09 style file. It is planned to reimplement this in the LaTeX3 programming language during 1993. Current status Phase 1: Review and redesign of the user-interface. There is now enough information about the requirements to be able to produce a testable version of the user-interface in the near future. Phase 2: Prototype implementation of new concepts. As outlined above, several prototype implementations have been tested during the last year. However, further tests will be necessary once the user-interface, the design language and the internal LaTeX3 programming language are in their final form. Phase 3+4: Design and prototypes of a designer-interface. The main concepts for the design interface are now developed and some proto-typing has already taken place. Nevertheless, this phase of the project will need considerable additional time as these concepts are new and thus cannot rely on experience already gathered. Phase 5: Re-implementation of the LaTeX kernel. As mentioned above, a prototype version of the LaTeX3 12 LaTeX3 Project --------------- kernel, written in the current version of the LaTeX3 programming language, has already been completed. However, as long as the work on phases 3 and 4 is still under way, this forms only an approximation to the final product. Plans for 1993 In the first quarter of 1993 the project team plans another workshop in Germany to work on the style design language. The core project team will investigate the possibilities for collaboration with related research and development projects such as Euromath. They will also establish contact with the ISO working group on the Document Style Semantics and Specification Language (DSSSL). The beta-testing of NFSS2 is planned for March up to May and the official release of the system is scheduled for summer 1993. The project team will work on a prototype of the designer-interface. This will lead to testing of this interface by working with designers to discover how easily it enables them to realise practical designs. It is planned that in the summer a French student will come to Mainz for three months, to help with the project's work. Finance There has been a modest but steady flow of contributions into the project fund set up by the TeX Users Group. Many of these come from individual members and they are particularly valued by the project team since, in addition to their financial significance, they represent a very high level of personal commitment to the success of the project amongst the TeX community---much thanks to everyone involved in these fund-raising efforts, especially to Alan Hoenig, Doug Henderson and the TUG office staff. The project's finances will also benefit greatly from the generosity of the authors of the forthcoming book A LaTeX Companion [3] to be published by Addison-Wesley in 13 LaTeX3 Project --------------- 1993: they are donating 50% of their royalties to the project. This book will describe the contents and use of many of the extensions to LaTeX 2.09 which are now widely available. The following institutions have made substantial contributions to the fund administered by TUG or to the project directly and we are very glad to acknowledge them here with thanks: Addison-Wesley, Blue Sky Research, Dante, GUTenberg, NTG Nederlandstalige TeX Gegruikersgroep, Nordic TeX Users Group, UK TeX Users Group. The following organisations have also assisted the project in various ways, including the provision of books, equipment, software, staff time, etc: AMS, ArborText, ArchaeoInformatica, Aston University, Blue Sky Research, CERN, Digital Equipment Corporation, EDS Electronic Data Systems, Elsevier Science Publishers, Life Science Communications, Manchester University, Open University UK, PCTeX, Royal Institute of Technology Stockholm, Royal Military College of Science, The Royal Society, TCI Research, Southampton University, TeX88, Zentrum fuer Datenverarbeitung der Universitaet Mainz. The project is still in need of financial assistance since demands on the fund are certain to grow substantially when wide-spread testing begins. References [1] David Carlisle and Frank Mittelbach. A test system for LaTeX-2.09. LaTeX3 project paper, June 1992. [2] Denys Duchier. Mainz, May 26--Jun 3 1992. Internal LaTeX3 paper, June 1992. [3] Michel Goossens, Frank Mittelbach and Alexander Samarin. The LaTeX Companion (tentative title). Addison-Wesley, Reading, Massachusetts, 1993. to appear. [4] Brian W. Kernighan and Christopher J. Van Wyk. Page makeup by postprocessing text formatter output. Computing Systems, 2(2):103--132, 1989. [5] Rolf Lindgren. Preliminary report on LaTeX style options. Internal LaTeX3 paper, August 1992. 14 LaTeX3 Project --------------- [6] Frank Mittelbach. E-TeX: Guidelines to future TeX extensions. In Lincoln K. Durst, editor, 1990 Conference Proceedings, pages 337--345, September 1990. Published as TUGboat 11#3. [7] Frank Mittelbach. LaTeX2.09 -> LaTeX3. TeXline, (14):15--18, February 1992. [8] Frank Mittelbach. LaTeX3. Die TeXnische Komoedie, 4(2):15--22, August 1992. [9] Frank Mittelbach. LaTeX3 project. TeX Gebruikers Group, 92(1):87--90, May 1992. [10] Frank Mittelbach. The multicol package. Distribution of style options that allow multi-column layout with LaTeX, November 1992. [11] Frank Mittelbach, Johannes Braams, David Love and Denys Duchier. The doc package. Distribution of the utilities for the documentation and distribution of TeX macro packages., May 1992. [12] Frank Mittelbach and David Carlisle. The array package. Distribution of style options that extend LaTeX's array and tabular facilities, May 1992. [13] Frank Mittelbach and David Carlisle. The validation package. Distribution of tools for automatic validation of changes in large macro packages, June 1992. [14] Frank Mittelbach and Chris Rowley. LaTeX2.09 -> LaTeX3. TUGboat, 13(1):96--101, April 1992. [15] Frank Mittelbach, Chris Rowley and Michael Downes. Volunteer work for the LaTeX3 project. LaTeX3 project paper, September 1992. [16] Frank Mittelbach, Chris Rowley, and Michael Downes. Volunteer work for the LaTeX3 project. TUGboat, 13(4):510--515, December 1992. [17] Frank Mittelbach, Chris Rowley and Michael J. Downes. Volunteer work for the LaTeX3 project. TeX and TUG NEWS, 1(3):18, November 1992. [18] Frank Mittelbach and Chris A. Rowley. The pursuit of quality---how can automated typesetting achieve the highest standards of craft typography? In C. Vanoirbeek and G. Coray, editors, Electronic Publishing '92, pages 260--273, Cambridge, April 1992. Cambridge University Press. [19] Frank Mittelbach and Chris A. Rowley. The pursuit of quality---How can automated typesetting achieve the highest standards of craft typography? MAPS, 15 LaTeX3 Project --------------- 92(2):50--56, November 1992. Paper already published in Electronic Publishing '92, April 1992, Cambridge University Press, (ISBN 0-521-43277-4); Reprinted with permissions. [20] Frank Mittelbach and Rainer Schoepf. A new font selection scheme for TeX macro packages---the basic macros. TUGboat, 10(2):222--238, July 1989. [21] Frank Mittelbach and Rainer Schoepf. Reprint (with corrections): The new font family selection --- user interface to standard LaTeX. TUGboat, 11(2):297--305, June 1990. [22] Frank Mittelbach and Rainer Schoepf. The nfss package release 1. Distribution of a new font selection mechanism for LaTeX, May 1992. [23] Chris Rowley. LaTeX2.09 -> LaTeX3: an update. In Mimi Burbank, editor, 1992 Annual Meeting Proceedings, pages 390--391, October 1992. Published as TUGboat 13#3. [24] Reinhard Wonneberger and Frank Mittelbach. BibTeX reconsidered. In Mary Guenther, editor, TeX 90 Conference Proceedings, pages 111--124, March 1991. Published as TUGboat 12#1. 16