Big Meeting Proceedings Editor Macros (LaTeX)

The story of a unique scientific meeting series allowing all participants to submit post-conference proceedings totalling thousands of pages.

Bob Jantzen, Dept of Mathematics and Statistics, Villanova University, Villanova, PA 19085-1699 USA

The original macro package begun in 1995 for MG7 (1994), the present LaTeX form developed for MG9 (2000), this documentation started August 2006, current date: November 2020.

The Marcel Grossmann Meetings on General Relativity and Gravitation (abbreviated MGXX, where XX = 1,2,...,15) are a series of international physics meetings that take place every 3 years with roughly between 500 and 1000 participants, begun in 1977 by Remo Ruffini and Nobel Prize winner Abdus Salam at the International Center for Theoretical Physics at Trieste, Italy. Their goal is to bring together scientists for an exchange of current ideas in a somewhat egalitarian atmosphere in which almost everyone gets an opportunity to speak in one of the many parallel sessions, if only for a few minutes to give others interested in the topic a chance to approach the person for individual discussion. The proceedings of the meeting contain about 35 long plenary talk articles (up to 40 pages) and a few long talk contributions from many of the approximately 80 parallel sessions (up to 20 pages), while ordinary parallel session speakers get 3 (later 6) page contributions which are basically extended abstracts with references. The distinguishing feature of these proceedings is that the number of contributions is in the many hundreds, so even if all the contributions were submitted using LaTeX and the publisher's macros, a single master document might not be practical. For smaller conferences, the publisher we deal with (World Scientific) after many years finally developed Editor macros for this purpose. However, in our case because of the graphics in computer presentations by experimental and observational scientists, some speakers find it most convenient to use the publisher's MS Word template to produce their contribution, so this requires disjoint compuscripts for all participants which must then be organized into 2 or 3 volumes of proceedings for a total of several thousand pages.

Participants, with lots of begging and extended deadlines, submit their contributions on-line over an extended period following the meeting, filling out on a web form for the upload process a complete author list and correct title which is then entered directly into the meeting database. After the final deadline, a report is exported to form the raw data for the LaTeX table of contents file described next. It should be noted that although this discipline (physics/astrophysics/etc) is rather abstract and requires nontrivial intelligence to master, getting physicists to follow simple directions or respect deadlines is not an easy thing, and it is important for a large enterprise like this run by volunteers that the great majority of participants cooperate. As an example, the World Science proceedings style for many iterations of the meeting had the title and author of each contribution in uppercase, but their macros didn't automatically make that happen, and many contributors do not notice this. Changing the .cls file by inserting the macro \uppercase around the title and author will almost work, unless authors use the word "and" in their list, which should be in lower case apparently. This would require recognition of the string "and " to make it active and remain lower case. More than bob can handle. Also the World Scientific proceedings latex class file replaces \caption for tables by a new macro, catching unsuspecting users unawares, since the idea of latex is to change the class file macros and not change standard macro names, other than titlepage matter which is always specific to the type of document. Some authors put the upload page title and author list used in the latex process into all uppercase, even though an example on the upload page is only capitalized according to the usual rules in English for capitalizing a title. Eventually the editor (usually bob) has to edit these by hand. Other authors don't put their first names as explicitly requested and someone has to search them out on the internet when they do not reply to email requests for information. [In the internet age where we can reach out and touch someone virtually, a full name is very useful---it is no longer a time when a handful of people did science so last names and first initials were sufficient to identify a person!]

Another serious problem involving author ignorance of instructions is the simple fact that local LaTeX environments can be customized but the local customization does not follow the LaTeX document. PDF graphics files were increasingly popular, but our 2000 designed robot software could not implement allowing this option (no desire for financing a consultant!), so often the robot would fail to compile compuscripts because of this fact which authors conveniently were unaware of from not reading the few instructions. These would either be converted by the author or manually posted in our database for the publisher to handle with no problem. And additional macro packages were also rarely included in the submitting zipped archives in spite of being requested by the instructions. Often unnecessary packages!

The editor organization tool is a table of contents file generated from input files listing the contributions in sections (the front matter, plenary talks, review talks, remaining talks grouped by parallel session and the back matter: participant list, author index) with author list, title and number of pages given for each contribution. My amateur LaTeX master file then generates actual page numbers and the actual table of contents used in the published proceedings, which the publisher can then use to organize and paginate all of the individual contributions in the production process. [See this draft mode example for MG11.] One also needs the front matter macros which produce the material up to the table of contents pages (list of past proceedings, sponsors, acknowledgements,  preface, opening ceremony remarks, awards announcements, awards talks) and the back matter macros which consist of a simple participant listing formatting file and input data for the participant list and an amateur author index generator which processes an auxiliary file produced from the table of contents into an unordered list of author names and page numbers of their contributions. The latter then requires some hand editing to eliminate duplicate names and a clever Unix editing pass to alphabetize the results (or using a spreadsheet reordering by last name if the authors are in the LastNames, FirstNames format), which is then formatted by an author index master file. All of these files were generated by an amateur TeX user (first version by bob in 1995 for MG9 with some advanced TeX knowledge and some crucial TeX wizard help from guru Barbara Beeton when that knowledge proved insufficient) rarely finding the time to clean up the quick and dirty solutions demanded by the task and lack of time.

The masterfile mgXXtoc.tex is the main organizational tool for the publishing of the proceedings from the collection of computer files representing the source files for all contributions, themselves organized via a database. [With the help of a wizard, IT friend Vittorio Vannini, a suitable exported report from this database creates the data input files for the table of contents.] This masterfile generates the actual typeset table of contents of the proceedings from input files for the contribution data produced from the proceedings database, with the start page number for each contribution, and in draft mode reveals the computer numbered ID (a 3-digit number unique to each contribution, assigned as the filename and correlated in a data base) for each contribution to identify the source computer files for each contribution for the publisher as well as the contribution page count from which the global pagination is generated. When LaTeXed, it creates an author listing file which acts as a raw file that requires alphabetization and then editing, from which the typeset author index of the proceedings is created, listing authors alphabetically, each followed by the page number of the first page of each contribution of that author to the proceedings. Multiple contributions from a given author (as lead author or coauthor) in the raw file after alphabetization (via a spreadsheet say) require hand editing to remove the repeated author names and separate these page numbers by commas (they come out ordered by first digit of the start page number so this too must be reordered by hand) to collapse them all to a single entry on a single line for each author. LaTeXing the mgXXauthorindex.tex then produces the actual author index. The final pagination should only be done when once the contents are fixed, since some effort must be invested in the manual steps to manipulate the raw author index data file. Were bob a computer scientist, he perhaps could have figured out ways to automate some of these manual steps, but alas he am a mere physicist/ mathematician/ ordinary LaTeX user.

In progress... perhaps never to be finished? [yes]

complete example files: [MG11 files, MG12 files]

Documented history of these macros: ../mgprocs.htm
[begun in 1995 for the MG7 proceedings, evolving into their present LaTeX form for MG9 in 2000]

2017 update.
For MG14 held in July 2015 in Rome with at one point 1054 confirmed participants in our database (the date of the Crab Nebula supernova) was  the largest MG ever, and open access e-book proceedings finally made these proceedings widely available instead of being limited mostly to the participants of the meeting as in the past. Most scientific articles are accessed electronically these days in any case, so it is a natural evolution of these proceedings. One can see the e-book versions of MG14 and the past few meetings before it at the publisher website:
Of course there is still a lot of work the publisher had to do to get this e-book published, only the paper printing and binding step is saved. They have always been very meticulous at examining carefully each compuscript and comparing it to the table of contents listing from author supplied data. The publication 2.5 years after the meeting had about 600 articles (19 plenary, 33 reviews, 560 parallel) in about 4400 PDF pages.

2020 update.
MG14 came out at the end of 2017 as e-book proceedings, but MG15 is currently stalled due to local politics in Rome involving the cofounder and owner of the MG Meetings: Remo Ruffini.  The system manager Vittorio "Vic" Vannini who managed the computer file collection and organization with bob retired from the University of Rome La Sapienza in November, 2020. All things eventually come to an end. As will this website collection within the next decade when bob retires.