Report from EuroTeX '94 in Gda'nsk (fwd)

[Petr Sojka]

Michel Goossens <Goossens at cernvm.cern.ch> was recently in attendance at the
1994 EuroTeX Conference in Gda'nsk.  He has kindly provided an overview
report of the activities at that meeting (similar to his earlier report
about TUG'94 in Santa Barbara) and asked that they be passed along for
everyones reference.  The attached ASCII text file has been placed in the
newly-created usergrps/info/ directory on the CTAN as eurotex94.txt; the
LaTeX2e source file is present there as eurotex94.tex.  If you missed it,
the report for TUG'94 is in tug94.* in the usergrps/info/ directory.  Other
reports for other meetings of users groups will also appear in this new
directory.  Also already present (and to be posted in my next message) is
Michel's report from the 1994 CyrTUG meeting in Dubna.

Regards,   George
==========================================================================
            The 1994 EuroTeX Conference in Gda'nsk
            ======================================
                   September 26-30 1994
                   ====================

          Michel Goossens, CERN, Geneva, Switzerland


After a Fokker 50 took me in about one hour from Copenhagen to
Gda'nsk, it was about midday on Sunday September 26th, when I
stepped out of the plane and was greeted by a beautiful blue sky
and summer-like temperatures. A trip by taxi of 41 km around the
southern part of Gda'nsk brought me in about 30 minutes to the
Orle holiday center in Sobieszewo, a resort on the Baltic some 20
km east of the city. The hotel was located 100 meters from the
beach, and already straight after lunch I had the pleasure of
walking along the wide sandy beach in search of amber and shells,
while discussing TeX and other text processing related problems
with several colleagues and friends of the TeX world, who had
already arrived the day before. The hotel had nice rooms,
comfortable lecture theaters; and the organizing Committee of the
Polish user group GUST made available two PC's, a printer and a
photo-copying machine, so that those who wanted to exchange
software had the necessary infrastructure, while many of the
participants came with their own notebooks (PC's, Mac's, a
Sparcbook). The drop in price of these machines (I now have one
myself!) allows users to readily take their development or
production systems with them and demonstrate their work. They
allow allow one to test new ideas immediately, and that is very
important in developers' environments, like the ones that most of
those attending TeX conferences are dealing with. It also makes it
possible to make notes, without the need to retype them
afterwards, and to distribute documentation in a flexible way. In
fact the hard disk of my Notebook PC contained a lot more Mbytes
of data and source files when I left Sobieszewo than when I
arrived, and it will take some time before I can put to good use
all the goodies I got from various friends at the conference.


The next day, the Monday, was kept free to allow people to
register and to meet one other. On arrival everybody was given a
copy of the Proceedings, and extremely useful typographic ruler,
and the tradional mug, with the specially designed EuroTeX94 logo,
showing a paper origami boat in the foreground, symbol of Gda'nsk
and its famous shipyard, on a dark background (the Baltic sea),
and the words EuroTeX'94 in a light sky at the top. Very nice
stylistic work, indeed!


I also be glad if you mention that mug and rules were given to the
all participants. My friends put a lot of work preparing these.
What about the EuroTeX '94 logo. That also cost a lot of
voluntarily work? Could you mention that too?


The conference had a total of 57 participants coming from 15
different countries, with the Polish (19) and German (13)
representations being the largest, while there were also TeX users
from Belgium, Canada, the Czech Republic, France, Hungary,
Lithuania, the Netherlands, Norway, Russia, Slovenia, Switzerland,
the United Kingdom, and the United States. All those who had
arrived by 1 o'clock in the afternoon were taken by bus on a
guided tour of the most interesting parts of Gda'nsk, a northern
Polish city of 400000 inhabitants with a proud cultural history
and a promising future. Thanks to the extremely competent guide
who showed us the various places where events had become history,
we got an extremely good idea of the life and death, the joys and
tragedies, the wars and peace in this famous city on the Baltic
during the last millenium.


The formal opening of the conference was on Tuesday morning, and
after the usual opening speeches by the organisers, it was Petr
Sojka who walked us through the various aspects of hyphenation
with TeX and described the significant success that has been
obtained recently, especially in the case of multi-lingual
documents. Bernd Raichle demonstrated how useful it is to use
TeX's mouth to process data and showed how he had applied these
ideas to implement a quicksort algorithm. Klaus Lagally, the
author of the multi-lingual ArabTeX system, explained in his talk
how he solved, staying within standard TeX, the problem of
line-breaking inside paragraphs with text that runs both from
right to left (like English) and from left to right (like Arabic
or Hebrew), thus providing a really portable solution. The
techniques described were successfully used to process a
forty-page paper containing mixed English and Hebrew with some
Arabic and even Latin or other short language fragments, and it
showed how flexible and powerful the basic TeX typesetter really
is. Just before lunch Marion Neubauer told us about her
experiences with converting Word and WordPerfect documents from
and to TeX and LaTeX and I am sure many of those present would
agree with her finding that unless the elements of the document
are already clearly marked up in the source, the LaTeX document
obtained was hardly usable, and that converting rather complex
documents is in any case a time-consuming process. The answer
might be using an editor in conjunction with an internal
conversion program.


The afternoon session started with a description by Olga Lapko of
the METAFONT package developed at Mir Publishers in Moscow and
distributed as part of the CyrTUG-emtex package. It contains the
METAFONT sources of a completely redesigned Russian Computer
Modern-like font family, which is more adapted to Russian
typographic tradition than previous Computer Modern Cyrillic
fonts. Yannis Haralambous then gave a detailed overview of the
Omega   system, a 16-bit extension to TeX that uses the Unicode
standard as internal encoding and allows multiple input and output
character encodings. He described various applications, including
calligraphic poetry, mixtures of languages with many special
characters, vowelized Arabic, fully diacriticized Greek, and
correctly kerned Khmer. Finally, Kees van der Laan gave us an
overview of his BLUe's (Ben Lee User) Format. At the user level
this new format is supposed to facilitate the creation,
formatting, exchange and maintenance of compuscripts during the
complete lifetime of a publication. The format is easily
customizable and provides for the possibility of having
cross-references using a one-pass process. I found it an
interesting approach since Kees introduced many ideas from modern
software engineering practice.


That evening, we had the traditional EuroTeX banquet, with a lot
of atmosphere, champagne, wine, plenty of beer, good food, guitar
playing and singing by several of the participants --- in a word
(or two) --- a hell of an evening, with TeX and LaTeX (almost)
forgotten and other themes such as family, children, politics,
philosophy, ``real life'', in short, becoming the main subjects of
the evening, and it was not before we were all convinced that we
had solved all of the world's problems that we went to bed in the
early morning hours.


So, the next morning at breakfast, it came as a shock to many of
us that there were still a few unsolved problems left (mainly in
the area of TeX, of course) and so we decided to continue the
conference and turn our attention to the niceties of colour
techniques and their realization in LaTeX2e. I emphasized in my
part of the first talk of the day that colour is rarely needed in
normal text, and that when it is used, the function of each colour
should be unambiguously clear. There exist many (empirical) rules
about colour harmonies and only a lot of experience and practice
allows one to become an expert in this field and apply colours
efficiently. Sebastian then showed how LaTeX2e implements a few
simple tools for obtaining colour and he presented some nice
examples. Janusz Bie'n gave an overview of different standards
connected with the Polish language (keyboards, character
encodings, localisation, fonts layout) and put them into
perspective relative to developments of the international
standards bodies. Yannis Haralambous described his Tiqwah
(``Hope'' in Hebrew) system for typesetting Biblical Hebrew, going
into some detail on issues of font design, classical Hebrew
typography and the user interface. He hoped that his system would
help to revive the interest in Biblical Hebrew typography.
Vladimir Batagelj gave an introduction to the PostScript language,
and presented some of his experience in combining TeX and
PostScript. Karel Hor'ak gave an overview on how one can decompose
large METAFONT pictures into smaller fragments and described
techniques to place them on a page in a seamless way. He stressed
the importance of resolution dependence and hoped that the new
versions of some of the drivers would eliminate most of the
constraints of this powerful approach where one only uses TeX and
METAFONT to generate pictures, thus making the whole document
fully portable.


Boguslaw Jackowski and Marek Ry'cko showed some extremely nice and
pleasing pictures made with METAFONT (and their paper was rightly
given the prize of Best Paper in a ballot amongst all conference
participants at the end of the conference). They demonstrated some
examples of non-standard METAFONT programming and advocated the
creation of libraries of METAFONT routines, that would make the
use of METAFONT as a universal drawing tool much more appealing.
'Eric Picheral, who looks after the Unix part of the GUTenberg TeX
archive, gave an overview of the history of that archive, the
various steps required to adapt TeX and its companion programs to
the needs of the French-speaking TeX user community, and the way
the various versions (Unix, PC, Mac) are made available to users
worldwide via the Internet (ftp, http/www, gopher). Lutz Birkhahn
discussed his work on developing debugging tools for TeX and
presented Tdb, an extension to TeX that provides an interface to
the Tk/Tcl X11 toolkit. This allowed him to set up a graphical
user interface to allow one to set breakpoints, have stepwise
execution, and to look at macro definitions and the value of
variables. The last talk of the day was by Philip Taylor, who
advocated the virtues of defensive programming for TeX since in
the real world one cannot assume that user code or input is
correct. Hence it is the task of the programmer to make sure that
the results of developed macros are as close as possible to those
the users expect. Defensive programming techniques let the
programmer anticipate both errors in data and flaws in algorithm
design.


During the first talk on the last day of the Conference I had the
pleasure of giving a 20-minute talk about the lessons learned when
writing The LaTeX Companion. I once more tried to emphasize the
importance of generic markup for all logical elements of the
document. Also, the global design of the book should be discussed
at an early stage, while formatting decisions should be left to
the final stage of running the chapters into pages. Wietse Dol and
Erik Frambach then gave a very impressive talk-demonstration of
their 4TeX workbench, that also forms the basis of the extremely
succesfull NTG CD-ROM. It is without doubt the best integrated TeX
system for the MS-DOS world, and many participants who wanted to
know more about the system also took part in the full day tutorial
that they ran on the Friday. J"org Knappen discussed work going on
to standardize the IPA characters, and advocated the creation of a
256-character IPA font for use with TeX. Jir'i Zlatuska talked
about work he was doing within the framework of LaTeX2e to allow
different languages and encoding schemes to be used together in a
same format, at the same time providing mechanisms to switch
freely between languages and encodings. Friedhelm Sowa showed his
approach to generate colour pictures, especially on cheap
printers. He discussed how the dvi driver must be colour conscious
and gave as an example the dvidjc drivers and the latest version
of his BM2FONT program, that provides the four primary colours of
the pictures by generating four different bitmap images. He showed
some quite impressive pictures as examples of his approach but he
pointed out that colour is not simple to realize and great care
must be taken to obtain the effects one really intends.


The afternoon session began with two presentations about LaTeX2e,
the first by Johannes Braams, who gave a clear introduction to the
principles of class files and packages, as he showed using simple
examples how it is possible to transform LaTeX 2.09 styles into
LaTeX2e classes and packages. Dag Langmyhr, in the second talk,
gave an explicit example of how to roll one's own complete LaTeX2e
document class, and detailed the various stages of building up the
necessary ingredients by borrowing from existing examples,
introducing (small) changes into existing constructs, and
incorporating the functionality of supplementary packages. He
showed convincingly that customizing LaTeX classes to obtain a
certain house style or specific look and feel is not very
difficult and he hoped that his listeners would now be able to
create their own document classes.


Before the official part of the 1994 EuroTeX Conference came to a
close, Philip Taylor and the varepsilon   TeX and NTS team presented
an overview of the present status of these two projects. The first
one is based on the existing TeX code, and plans to extend it in
various areas, while keeping 100% backward compatibility with TeX
for those who want it. The NTS project, on the other hand, seeks
to first reimplement TeX in a list language, so that several
alternative approaches to the various components that build the
system can be more easily tested. In the longer run it might thus
be possible to develop a New Typesetting System (hence the name)
that will be at least as good as TeX, but that extends or improves
TeX in areas where the latter is considered too limited.


These last eighty minutes or so about futures were followed by
closing remarks from Wlodek Bzyl and Philip Taylor of the
Organizing Committee, who announced the winners of the Best Paper
contest (see above), who were given a bottle of vodka with tiny
pieces of gold floating inside (a local specialty), and the venue
of the next Conference, that is to take place next year somewhere
in the Netherlands (possibly in the (now) famous town of
Maastricht).


On Friday, most participants stayed on to attend one or more of
four tutorials, namely on 4TeX (Wietse Dol and Erik Frambach),
Manmac BLUe's (Kees van der Laan), Book design and Typography
(Marek Ry'cko and Philip Taylor), and LaTeX2e (with Johannes
Braams and myself). In the afternoon, a 20-hour course by Marek
and Phil on advanced TeX macro writing started.


The LaTeX2e tutorial and advanced TeX course went on the Saturday,
while on the Sunday only Marek and Phil had enough energy to go on
speading the TeX word, this time no longer at the Sobieszewo
Center, but on the computers in Gda'nsk University. Thus, when I
left around 11 o'clock on Sunday October 2nd, I left behind me
three floors of (almost) empty rooms. Also a Babel-like mixture of
East and Western-European languages no longer floated through the
corridors. Yes, it had been a good conference, and quite different
from the 1994 TUG Annual meeting, whose theme was ``innovation'',
so that many papers described more or less exotic, front-line
developments (colour, sophisticated page layout, object-oriented
techniques(  See my report on the Santa Barbara TUG Annual Meeting where a
similar talk was given by this team, or the Proceedings of that
Conference that will be available in November from the TUG Office,
Balboa Building, Room 307, 735 State Street, Santa Barbara, CA 93101,
USA, for a price of 30 USD), while in Gda'nsk more attention was given
to practical issues of typesetting in multiple languages and working
with cheap printers and machines (hyphenation, the use of
METAFONT, MS-DOS related developments). I can only congratulate
the local organizers, especially Wlodek Bzyl and Tomek
Przechlewski, for their nice work, and I hope that this conference
has contributed to making TeX better known in Poland, and that
those present will take with them the ``spirit'' of Sobieszewo,
where it was shown how to put principles into practice to make
progress in the field of applying TeX in real-world applications
(a copy of the 200-page Proceedings of the EuroTeX94 Conference,
which, as mentioned above, were available on the Monday thanks to
the hard work of Tomek and Wlodek, can be obtained by sending 15
DEM (postage included) to Wlodek Bzyl, Instytut Matematyki, Uniw-
ersytet Gdanski, Wita Stwosza 57, PL 80-952, Poland).


When I stepped into the Fokker 50 to fly back to Copenhagen (and
from there on to Geneva), I realized that I had witnessed how
Summer had become Autumn. The trees were putting on their nice
yellow-brown dresses, and were waving their heads in the breeze,
that had become definitely cooler in only a week's time. They were
preparing for their last party, before going into a long winter
sleep.


See you next year, some time, somewhere, they were saying. And
indeed they will.