\def\Title#1{{\sl #1}} 
\def\Speaker#1{{\bf #1\/}} 
\def\awk{{\tt awk}}


% front matter
\title{Cork90}
The fifth European \TeX\ conference (10--13 September) was also the first \TeX\
Users Group Meeting in Europe, that is, the first international meeting of the
users group outside North America. The conference was held at University College
Cork, a university campus with a charming mix of old and new buildings among
lawns and trees, in Eire's second largest city.

The organizer of the conference, Peter Flynn, of the computer
bureau at the college, had done a great job on organizing this
big conference with about 175~attendees from 23~(!) different countries. As for
the social side of the conference: there was a get-together party after
registration on Sunday evening, a dinner at the university restaurant on Monday,
and on Tuesday an excursion to Blarney Castle, followed by dinner with live
music and folk dancing in a restaurant nearby. Apart from this there was also
plenty of opportunity to meet people and discuss various things during coffee,
lunch or tea breaks. The programme was not overfull and there was more than
enough time to discuss \TeX\ and related matters with \TeX\ users from all over
the world.


\subsection{Monday, September 10}
After not-so-formal opening speeches by the Minister for Science \&
Technology of Ireland and the dean of University College Cork, the
first international \TeX\ conference in Europe was a fact.

\Speaker{Erich Neuwirth} (University of Vienna) had the honour (unpleasant
task?) of giving the first talk. He showed that you can create
quick-and-dirty databases with basic tools such as \TeX\ and the \Unix\
tool \awk.
Erich seemed to advertise the use of \awk\ to larger parts of the \TeX\
community: \awk\ has been ported to the MS-DOS world and, apart from several
commercial versions, there is a version available from the Free Software
Foundation.

In her talk on \TeX\ and hypertext, \Speaker{Christine Detig} (University of
Darmstadt) gave several definitions of hypertext, and discussed advantages
and disadvantages of the hypertext approach. The conclusion of her talk was
that the future of electronic publishing {\em could\/} be an interwoven
electronic process with the following elements: (1)~hypertext and logical
markup, (2)~\TeX\ for typesetting, (3)~drivers for various screen and printer
types.  One criticism: discussions about `hypertext' can only be useful if
all people attending the talk have {\em seen\/} hypertext -- describing it on
paper or transparencies is not enough to get an idea of what it is,
or what it can be.

\Speaker{Les Carr} (University of Southampton): \Title{Experiments with
\TeX\ and hyperactivity}.
The authors reported from their work based on the question:  How can existing
(\TeX) marked up copy be used in a hypertext system?

Problems tackled include the use of \TeX\ to format text being displayed on
screen in variable sized windows, embedding generic hypertext navigation
tools in the source and using a single source to generate both printed and
hypertext versions of a document.

The authors concluded with: `It is clear that \TeX\ in a hyperactive world
can only survive in a generic markup form such as \LaTeX. Much of its power
(such as pagination) is irrelevant in this context, and its implementation
language will continue to put off many potential designers. But \TeX\ retains
a raw beauty of its own, and if we decide that the hyper systems we build
must have a formatting engine behind them, we are confident that \TeX\ will
continue to be the first choice for many years to come.'

The talk by \Speaker{Johannes Braams} (PTT Neher Laboratory), \Title{The Dutch National
\LaTeXsl\ effort} was a report on the work done by working group~13 of the
Dutch \TeX\ Users Group.  He discussed the development of a set of document
styles that are compatible with the standard document styles of \LaTeX, but
have a layout that appeals more to Western European tastes, and the work on
`internationalizing' \LaTeX, that is,   adding tools to \LaTeX\ that allow an
author to switch between different languages, a project that bears the
appropriate name `Babel'.

\Speaker{Adrian Clark} (Essex University): \Title{Documenting a \TeX\
archive}.  The key issues addressed were: An archive is only as good as its
documentation: Where is the archive? How to locate the software? How to get
it across the networks?  The emphasis was put on the second question. Experience
has it that {\em functionality\/} requests precede name requests. Keywords and
help systems are the tools, along with the good old systematic index, provided
there is a generally accepted one. The author reported  a proto system
for automatically maintaining the `index' of the archive, the help system
etc. In order to have this kind of systems working, contributors must be
persuaded to obey the submission rules:  software documentation must be supplied
in a certain standard format.  Perhaps overlooked is the issue of
a general accepted classification index.  Even within the field of numerical
software several indices are around.

During the lively discussion after the presentation, useful ideas about what
users would like to see were proposed, ranging from:
\bi the ideal situation, just your own local archive `at the corner' with
{\em virtually\/} everything, to
\bi documentation requirements (time stamps, version indication, minimal
tests, quality, refereeing, etc.).

\noindent
Nelson Beebe mentioned the synchronisation effort of the various archives and
the TUGlib project. These projects have the same kind of problems.


\subsection{Tuesday, September 11}
\Speaker{Thomas Kneser} (GWD G\"ottingen), the first speaker on Tuesday
morning, presented an adaptation of the work of Thomas Reid to a \LaTeX\
context.  Thomas Reid has developed macros that allow to have paragraphs of
text to `wrap around' figures. Thomas Kneser showed that these macros can be
successfully adapted to the \LaTeX\ case. However, in some cases a certain
amount of cheating is necessary to get the desired result.

In his talk, \Title{The document style designer as separate entity},
\Speaker{Victor Eijkhout} ({\TeX}Techniek) discussed the need for a
separation of all tasks in a \TeX-based system in three `layers': (i)~the
author, who uses a document style prepared for a certain class of documents,
(ii)~the document-style designer, who creates a document style from a
\TeX-based toolbox, and (iii)~the \TeX\ expert who creates and maintains the
toolbox.  He argued the need for a programmable toolbox, so that the style
designer could write \TeX\ macros without programming in \TeX; this point was
illustrated with examples from a format the speaker developed.

\Title{QuickDraw, PostScript, \TeX} by \Speaker{Tim Murphy} (Trinity College
Dublin) was a rather confusing talk. Afterwards I was unable to summarize
what it was he had tried to tell us. A lot, I~know, but it lacked coherence. 
What I did pick up was this: he loves GRIF, but hates \SGML, for reasons that
are still unknown to me. He appeared to be in favour of the work, done by a
Euromath committee, on defining a complete syntax of mathematical formulae
for GRIF.  He advocated the basic idea of GRIF of letting a document be
generated by a context-free grammar.

\Speaker{Malcolm Maclenan} (South Bank Polytechnic) considered himself to be
very new to \TeX\ and definitely not an expert, but he showed some very
interesting {\TeX}nical stuff. At the South Bank Polytechnic there arose the
need for a tool for drawing circuit diagrams for text books.  \PiCTeX,
although it requires a lot of memory and is rather slow, can produce
impressive full-page circuit diagrams.  The speaker showed that a form of
object-oriented programming can help a lot.  In the discussion after the
presentation, speaker and audience came to the conclusion that an interesting
extension -- and a possible time-saver -- would be a collection of symbols
and parts of diagrams created with \MF. Another future development could be
integration of \TeX\ with a CAD package: extract a list of $(x,y)$
coordinates from the CAD programme and use this as input to a \TeX\ macro
package.

The talk by \Speaker{Rainer Sch\"opf} (University of Heidelberg),
\Title{Towards \LaTeXsl\ 3.0}, was a list of ideas for the new version of
\LaTeX. He outlined some ideas for a future version of \LaTeX, but cautioned
that none of these ideas were definite yet. Much importance will be given to
the style designer interface.

For about a year now, Frank Mittelbach and Rainer Sch\"opf, have been
discussing ideas -- mostly via electronic mail -- for a new version
of \LaTeX\ with a group of \TeX, \LaTeX\ and document-style experts
in Europe and the United States. A proposal for the new design and a
few prototype parts of the new package are expected shortly.

The presentation of \Speaker{Brian Hamilton Kelly} (Cranfield Institute of
Technology),  was the last one before the visit to Blarney castle in the
afternoon. His presentation focussed on a new public domain implementation of
\TeX~3.0 and \MF~2.0 for VAX/VMS.  The distribution contains normal and `big'
versions of \TeX, \LaTeX\ and \SliTeX.  Useful facilities: \TeX's {\tt e}
option really works, \TeX, \LaTeX\ and \SliTeX\ now available as VMS command
verbs.  It has recently been extended by Don Hosek and is now also the DECUS
implementation.


\subsection{Wednesday, September 12}
\Speaker{Malcolm Clark} (Imperial Cancer Research Fund),  \Title{Post Congress
Tristesse}. His
presentation was an enlightening one for those {\TeX}ies interested in
doing book projects with \TeX.  The topic  was the process of
making the proceedings of the \TeX88 conference at Exeter ready for
publication. To discourage anyone with ideas of being an editor in the near
future, here's a list of the problems Malcolm described:
\bi variations in medium (disk, electronic mail and even paper)
\bi character coding (\ASCII, \EBCDIC)
\bi late submission (please wait -- wait some more -- nag -- threaten to print
just the abstract) 
\bi input form (plain \TeX, \LaTeX). In the end, everything
was converted to plain \TeX\ with additional macros to reproduce a layout
similar to the layout of Addison-Wesley's computer science series.
\bi a professional indexer was hired to compile the index afterwards:
the result was not perfect.

An important decision: do you referee papers or include all of them? The
speaker's view was that you should not edit or referee a paper: if
someone wants to present work with \TeX\ or \MF, he or she should be
free to do so. Removing syntactical errors is acceptable, any further
editing of the English -- or almost English -- isn't.

A matter of style: who determines the layout, the editor or the author?
Answer: the editor, so the authors should not markup (too much). In some
cases plain \ASCII\ is better than $\ast\ast$\TeX.

Finding a publisher: Addison-Wesley (not interested), Springer (no
contacts with them), John Wiley \& Sons (never publish proceedings).
Finally Ellis-Horwood published the proceedings. Advantages of E-H: no
control, you can do whatever you like as editor. Disadvantages: you have
to do everything, upto and including producing the bromide!

Conclusions:
\bi Publishers like it when authors do all the work {\em for them},
which is a really funny view!
\bi Publishers should not accept 300\,dpi camera-ready copy, since
this is without exception of appalling quality.
\bi Computer Modern is an excellent font at 1270\,dpi, but looks
horrifying at 300\,dpi. Books produced from low-resolution output are no
advertisement for the high quality of \TeX's typesetting.
\bi Adobe and other font producers still have to come with a nice font
that includes math!
\bi Copy editors are essential. However, some publishers accept
manuscripts in electronic form, or even offer the possibility of
electronic mail submission, and then eliminate technical editing from
the publication process.
\bi Working with other amateurs on any project means sooner or later
that they withdraw as soon as other things become more important.
\bi Expect no thanks.
\bi It's fun!
\medskip

\Title{\TeX\ \& SGML} by \Speaker{Kees van der Laan} (University of Groningen)
was a talk that could have been more interesting if more people in the
audience would have had some experience with \SGML. It is not enough to give
the following description: `\SGML\ is a meta-language that is a means to
express your thoughts on the entire lifecycle of documents', that serves as a
standard for tagging and interchanging documents, and that provides a
mechanism for representing special characters, tables, and mathematical and
chemical formulae, using \ASCII\ coding.  The speaker has recently investigated
the problems with conversion of material coded in \TeX\ or \LaTeX\ to \SGML,
especially mathematical formulae and tables.

If \SGML\ becomes a household word in electronic publishing and publishers
provide their authors with {\em document type definitions\/} (dtd's),
authors have to be aware at all times of this dtd and the tags, entities
and attributes it defines. The publishers need standardized
general-purpose dtd's or \SGML\ experts who develop new dtd's.

Another problem of using \SGML\ for the coding of mathematical formul\ae\
and tables is the apparent impossibility of separating form from
content.

Publishers and other users of \SGML\ often state that one of its advantages  is
that it allows information to be used for other
purposes later on. One of the conclusions of the speaker was that re-usage is
maybe not be that important since scientists tend to continually rewrite and
update their articles and books. Another conclusion was that the meta-ness of
\SGML\ is a strong point of \SGML, but at the same time also one of its weakest
points.

\Speaker{Olivier Nicole} (INRA), presented another use of the
\PiCTeX\ package. He used it to printing graphs that were generated by the
statistical package~S.  From both talks on \PiCTeX\ at this conference it became
clear that it is a very useful and powerful package: the \PiCTeX\ macro package
can be found in several electronic archives, and there is a very good manual
available.  However, it is very slow, you need a big version of \TeX\ and a
10\,kbyte {\tt tex} file can result in a {\tt dvi} file that is ten times as
large!

\smallskip

The rest of the day was devoted to font design with \MF\ and
applications of such fonts.

\Speaker{Alan Hoenig} (CUNY),  presented a solution to the problem of
labelling figures in \TeX\ documents. His message was: graphics are no longer a
problem for \TeX\ (?), but the problem is to get labels in the figures that use
the same font as the text.  The basic strategy is the following: create a picture
with \MF, write the coordinates of points to be labelled as {\tt fontdimen}'s
to the {\tt tfm} file. In the \TeX\ document you pick up the $x$ and $y$
coordinates from the |\fontdimen|s.

\MF\ lacks read and write operations, so (ab)using the {\tt tfm} file for
this purpose is the easiest solution. Furthermore, writing a parser in \TeX\
to decode information in the log file of a \MF\ run would be, to say the
least, cumbersome.

Final remark: to make it easier to use \MF\ for creating all sorts of
pictures, one needs a macro package `on top' of it, similar to \LaTeX\ `on
top of' \TeX.

\Title{Typesetting Old German} by \Speaker{Yannis Haralambous} (Universit\'e
de Lille) was one of the nicest talks of the entire conference. He was
awarded the prize for most memorable presentation at this conference. The speaker
started out by observing that many people believe that old (early) music can
be best enjoyed when played on old instruments. If you draw a parallel with
reading old texts: these can maybe be enjoyed more when they have been
printed with old typefaces. The speaker showed pages from the Gutenberg
42-line bible and a book by C.P.E.Bach.  To re-create these texts, the
speaker has designed \MF\ fonts for Gotisch, Schwabacher, Fraktur and
ornamented capital letters -- the ornamented capitals were beautiful and
extremely well done, enough for a big round of applause.  While working on
the Fraktur it turned out that the old designers applied Bezier curves
without knowing it:  a lot of the curves can be specified by only a few pen
positions.\looseness-1

As an aside the speaker has also created an Arabic font, of which he also
showed examples. His next project will probably Renaissance Greek, where you
need a font with $\approx 400$ ligatures.

\Speaker{M\'{\i}che\'al \'O Searc\'o\i d} began his presentation \Title{Irish
letter forms with \MF} with a short history of writing in Ireland. 
Elizabeth~I ordered the design of an Irish typeface, modelled after written
letter forms.  Use of this typeface started around 1571.  Through history
there have been several re-designs. After the historical introduction, the
speaker showed an example of modern Irish letter forms, created with \MF.

The next presentation, on \Title{An international phonetic alphabet}, was by
\Speaker{Dean Guenther} (Washington State University) and Janene Winter. Dean
is manager of the project and Janene was the \MF\ specialist. The project was
started because of frequent requests from the humanities departments. At
first, some papers were published with Computer Modern fonts and handwritten
phonetic symbols. Later on, phonetic symbols were created as bitmaps. Howver,
these only looked good with Almost Modern Roman, so a new font was needed
when people started using CM. There were lots of obstacles: MF84 came when
Janene had just learned MF78, there was no help on campus, there was no
previewer. But the result is an IPA font, to be precise: a 128-symbol subset
of one of the varieties of IPA.

Since Janene now works for the AMS, the work on the IPA project has
unfortunately stopped. At the State University of Washington a different IPA
project has started, partially based on the work done at WSU.

\Speaker{Adrian Clark} (Essex University): \Title{Halftone output from
\TeX}. Adrian presented a paper which summarised the various techniques
for including images (halftones) in \TeX\ documents.  The main
techniques are by means of a `picture font', via a halftone font and
external support software, or by inclusion at the \DVI\ driver stage.
Some of the advantages and disadvantages of all three approaches were
described.  He then went on the describe how the one can generate the
four colour separations required for colour printing, and showed the
output of one of his early experiments (the colours were not too
impressive, but it's getting there).  Finally, he mentioned that
preparing camera-ready copy with colour separations requires precise
information concerning the positioning of figures on the page, which
requires modification to the {\tt output} routine.



\subsection{Thursday, September 13}
\Speaker{Nico Poppelier} (Elsevier Science Publishers): \Title{SGML \& \TeX\
in scientific publishing}. This presentation was non-technical. The speaker's
idea was not to discuss the `What and how?', but the `What and why?'. What
can \SGML\ mean to a publisher, i.e. what role can it play? \SGML\ and \TeX\ can
be be a nice combination, but so can combinations of \SGML\ with other
sophisticated text-processing systems and/or typesetting systems. 

The speaker also discussed
\bi some of the advantages of \LaTeX\ over plain \TeX, both for journal and
for book publishing, and
\bi other uses of \TeX\ within Elsevier Science Publishers, namely as the
back-end of a database-publishing system

The presentation given by \Speaker{Amy Henrickson} ({\TeX}nology), \Title{Getting
{\TeX}nical} was a 30-minute crash course in \TeX\ macro writing. A series of
useful techniques, lots of examples, a few basic and a few not-so-basic \TeX\
tricks. Interesting talk, but with a very high pace.

\Speaker{Frank Mittelbach} (EDS, R\"usselsheim) intended to give a talk
\Title{New Bib\TeX\ requirements} instead of having an open discussion with
this title, as was announced in the program. The main points of Frank were
that 
\bi\BibTeX\ is not suitable for applications in the
humanities or, for example, for books with bibliographies in every chapter 
\bi
\BibTeX\ does not fit well with \LaTeX\ 3.0 
\bi adapting \BibTeX\ to other
languages than English is not always possible 

\noindent He proposed
several major changes to \BibTeX; the complete list is too big to be repeated
here, but the paper will appear in the proceedings.

The title of the presentation by \Speaker{Angela Barden} (City of Cork
Vocational Education Committee) was intriguing: \Title{Purchasing pain with
all that joy}. It turned out to be an interesting, and sometimes amusing,
examination of several books on \TeX. The speaker, who is a teacher of
reading skills, examined the following books:
\bi `The \TeX book' by Don Knuth. It tells lies and jokes (even in
the index, where it is out of place), hides information in exercises,
contains exercises that are too difficult, the often confusing
dangerous bends, and conflicting instructions. On the whole, a
user-unfriendly book.
\bi `The Joy of \TeX' by Michael Spivak. The first exercise in the
book was pointless, so why do the others? And the answer given in the
back was wrong and unintelligible. The term `macro' is never defined.
`Maybe it helps mathematicians, but not me', was the speaker's
conclusion.
\bi The PC-\TeX\ manual, by the same author. A page-numbering scheme and a
confusing structure. For example: there is an appendix A and an appendix F,
but no other appendices. The speaker abandoned the book at an early
stage.
\bi `\LaTeX\ user's guide and reference manual', by Leslie Lamport. The
speaker was grateful to the author for his straightforward way of presenting
information. No tedious exercises, a no-nonsense way of writing, clear
concise prose.
\bi `\TeX\ for the impatient', by Abrahams, Hargreaves and Berry.
Not radically different from the other books. A dull layout, an
enormous amount of print on a page, turgid style, but not as bad as
the other books.
\bi `A gentle introduction to \TeX', by Michael Doob. Not a bad
book, but unfortunately there are exercises.

Advice for future books: tell no jokes, be careful about the imagery
you use (if \TeX\ has eyes, mouth and stomach, what is the analogy of
|\shipout|?), use shading or another technique to distinguish input from output,
use illustrations to clarify concepts, separate text book from manual, give more
examples, don't hide information in exercises.

In his presentation \Title{\TeX\ in schools, just say no}, \Speaker{Konrad
Neuwirth} argued against teaching \TeX\ in schools. There doesn't seem be a
class where \TeX\ would be appropriate: it doesn't teach you anything about
math, writing math or physics reports with \TeX\ in class takes more time
than the usual 50--60 minutes, and as an instrument in teaching \ae sthetics
\TeX\ is also questionable, since most \TeX\ documents look horrible to a
trained eye.  Furthermore: \TeX\ is not the most user-friendly of programs.

\Speaker{Philip Taylor}'s talk \Title{Improving the \ae sthetics of mixed
font documents} was an account of the work on the book `Principles of
nutritional assessment', which was recently published by Oxford University
Press.  According to the speaker, Don Knuth's final exhortation in the \TeX\
book, `Go forth and create masterpieces of the publishing art', has developed
into `Go forth and create horribly looking books that shriek \TeX (\LaTeX)!'

Letting \TeX\ typeset a book with a pleasing design, using Postscript
fonts (Times Roman) for text and Computer Modern for math, and with
lots of tables and illustrations turned out to be a far from trivial
task. The resulting book was passed around during the presentation
and proved to be an attractive book, impeccably reproducing the sort
of looks one expects from the Oxford University Press, where it was
published.  One small criticism: Philip had forgotten to turn on
|\frenchspacing|.

After a rather long introduction, the next speaker, \Speaker{Alan
Witt\-becker} (DEC), came down to business: \Title{Archi\TeX\ as an
international page pattern maker}. Archi\TeX\ is a macro package that he
developed to re-create the variety of page patterns he described, using
\TeX.  For that, a rectangular grid is put `on top of the page'. To every
intersection point of the grid a box of text can be attached.

The last presentation of the conference was \Title{Integration of graphics
into \TeX}, by \Speaker{Friedhelm Sowa}. The idea behind the speaker's
approach is to convert the output of graphics packages, in this case files in
{\tt tif} format, to \TeX's {\tt pk} files. A picture is divided into tiles
of certain dimensions, and the tiles are converted to characters in a
\TeX-type font.  \TeX\ is then used to glue the tiles together to produce the
original picture. 

The speaker discussed various methods of obtaining
half-tone pictures, with varying number of grey shades and with or without
error distribution, and showed several examples.

\subsection{Birds-of-a-feather sessions}
The organisers of a \TeX\ conference had a golden moment when they
came up with the idea of bof.  The idea is simple: on every day of
the conference programme the organization allocate one or more time
slots of, say, half an hour. Anyone with ideas for a discussion can
allocate a time slot and  just go ahead! In other words: the
programme provides time especially for informal or improvised
discussions.\looseness-1

At the Cork conference, there were birds-of-a-feather sessions, or
bof's, on graphics, the future of \TeX, and \LaTeX~3.0. The bof on the
future of \TeX\ was especially interesting, since it became a three-hour
discussion that resulted in a petition to the board of TUG, signed by
some 25--30 concerned TUG members. It is this letter which stimulated \DEK's
announcement on the future of \TeX\ and \MF, printed on the first page of this
\TeXline.

One of the major problems at present is the fact that the creators of
\TeX, \LaTeX, \BibTeX and MakeIndex have, for various reasons, decided
to stop working on their programs. In some cases they have announced
that they consider their programs as mature, that is, they will not develop them
any further. This is a matter of great concern to many TUG members and
to the board of TUG as well.
\author{Nico Poppelier}
\author{Kees van der Laan}
\author{Johannes Braams}

\endinput