\centerline{\bf\TeX as -- west of the Brazos}
\smallskip\noindent
There are at least three things to be evaluated at any
conference: in order of importance -- the social
activities, the organisation, and lastly, the
presentations. Basically, the people you meet are what
determines whether a conference is likely to be of
lasting value. The presentations are the reason that we
manage to obtain funding to go, but, judging from the
relatively small proportion of attendees who go to all
the presentations (at {\it any\/} conference), they are
clearly not thought to be the key `unmissable' items.

This year's TUG conference was held at College Station,
Texas, the home of Texas A\&M University. The host was
{\bf Bart Childs}, TUG's previous president. There had been
many dark mutterings about the choice of site, since June
in Texas is warm. Americans are so much fussier about
these things.

The conference started on the Sunday evening with the
`\TeX as ethnic feast' (beer and wine courtesy
ArborText, Vrazel's Polka Bank courtesy Kinch). We were
bussed to the local community hall and had some `ethnic'
food openers as we started: eventually, and I do mean
eventually, the main course arrived and we tackled it. It
was Tom's barbecue (`southern' barbecue is not your
hibatchi in the garden stuff, but a style of cooking --
very slowly), but it's actually better at Tom's place. And
once the band were fed the dance began. Now, not too many
\TeX ies can polka.  Mostly people talked to people.
Basically it was a  slow start. 

At last we get to the conference.  The lecture theatre
was  strange. It was flat, so those at the back couldn't
see the  screens.  The conference `started' quite early
with a pair of presentations aimed at `new users'. This is
quite a sound idea since most (but not all) of the  people
attending courses stay on for the conference, and there
are also a fair smattering of novices. {\bf Alan Hoenig}
gave a talk similar to the one he did at Exeter. It's a
fair, sound, worthy, etc.\ presentation.  {\bf Nelson
Beebe} talked about some aspects of the broader picture of
\TeX\ and allies. So fair enough. But there was no fire, no
excitement, no thrill. We also had an introduction to the
TUG staff and to the Officers. 

Then the real start was made. Or almost.
{\bf Nelson Beebe} and {\bf Frank Mittelbach} made some
remarks on the new versions of \TeX\ and the forthcoming
\LaTeX\ 2.10. Basically update information. Good to know,
or to be told, but essentially reportage. Then for about
half an hour there were reports from the european \TeX\
user associations. I do feel that they could have
made an effort to get the UK \TeX\ Users Group's name
right. But since they can't get my address right, I guess I
can't expect too much. The reports were again, very
worthy. Bernard Gaulle of GUTenberg wasn't there, so
Christina Thiele read a version of his report; {\bf
Joachim Lammarsch} (\dante) gave his; someone read 
Kees van der Lann's report (Dutch group) since he wasn't
there either; {\bf Jan Michael Rynning} gave the Nordic
one and had the sense to bring a map showing where
Scandinavia is. I'm saving the best to the last. I gave
two reports, one as UK  representative, and one as
European coordinator.  I deviated from the subject a bit.

Then back to site coordinator reports. Now, I find these
awfully dull, and this year was hardly any exception. 
Really there are only three site coordinators reports
which are of widespread interest: \Unix, VMS and \MSDOS.
{\bf Pierre MacKay} takes most of the fun out of \Unix;
{\bf David Kellerman} (VMS) was useful; and the `small
systems' coordinator, Lance Carnes, has not done any
visible small systems coordination for years -- so no
report. All the others are minority interest with a
vengance. Coffee at last. 

Now, a curiousity. On the floor above was the vendor's
display. It wasn't open yet, and would not be until
about 1.30. This was a mistake since the morning was
so {\sl slow\/} and the organisers kept mentioning things
which would be available upstairs from the TUG display,
but which no-one could get to till later.

{\bf Don Hosek} gave a report on behalf of the drivers
standards  committee.  They
are currently at level 0. Even then, some drivers will
likely not achieve this dizzy height. Then archives and
electronic bulletin boards. I must have snoozed then,
since I remember little about it. Nelson mentioned {\tt
tuglib}, a kind of electronic archive he has started at
Utah. There is a note about it elsewhere in \Texline.

Lunch was a `networking lunch'. The notion was that people
marked a list of topics and on the basis of these figures
a number of tables at lunch were identified for people
with these similar interests. After lunch I wandered up to
the exhibits and only by chance remembered the business
meeting. It seems to me that having the exhibits open at
the same time that the business meeting is on is a
fundamental mistake, almost as if you didn't want people
to come to the business meeting.  Rick Furuta was
unopposed as Vice President, just like David Ness as
Treasurer. On useful innovation was that the two
re-appointed officers gave a brief account of themselves. 

At last we got to the papers. So far, things had been
somewhere between sluggish and slow. There was nothing to
get excited about. The morning was definitely moribund.
The afternoon's papers were an interesting bunch. Had we
run to schedule it would have been Frank Mittelbach,
{\bf Hisato Hamano} and {\bf Yoshiyuki Miyabe}. It is
true that their subject matter was related. It is also
true that none of them were native american or english
speakers. Thus it was hard work to follow as the
afternoon crept on. Yoshiyuki did not actually present
his paper that afternoon, since the other talks overran.

Frank Mittelbach's talk on `E-\TeX' was a look into the
future. Frank looked at the current deficiencies of
\TeX3, especially in typographic terms and made a number
of specific proposals on what \TeX\ needed in order to
face the next century. 

Hisato Hamano described the vertical typesetting function
swhich the {\sc ascii} Corporation of Japan (the
publishers of the \TB\ in Japanese) had developed. This
required the introduction of V-text and H-text, and a
`new' line break rule -- {\it kinsoku}.

But many people missed them anyway because they were
upstairs with the vendors.  A couple of bofs (birds of
a feather) were organised after lunch. Or rather, convenors
were found. I was  convenor of one bof. The bowling bof.
There was a 10 pin bowling alley within walking distance
(walking is a novel concept in Texas) of the main
accommodation.

There were two workshops organised (again, quite a good
idea) back at the hotel for after we got back and before
the wine and cheese, but I seemed to end up in the pool
for a while instead. I had actually signed up for a
workshop (on tables!), but I was sidetracked. Jackie
Damrau later gave me the notes which went with the
workshop and made me regret lying languidly in the pool.

The wine and cheese (traditionally a Personal \TeX\ Inc.\
event) was quite acceptable, although the food did run
out a bit too quickly. As the PTI representative, poor {\bf
Claire Kahan}  was on tenterhooks. Come about 10 or
10.30 my bof got together and bowled the night away.
Twenty two I counted. Easily the biggest bof. I rest my
case. 

Tuesday started with the vendors' presentations. Again, a
bit scrappy, but when you have umpteen short
presentations, that's not surprising. I think they went
in reverse alphabetical order. Amy Hendrickson was
disappointing in her short presentation; Kinch had his
kids handing out informational material (very cute);
David Kellerman took up a bit too much time; Lin Tay of
Micro Press was elegant, exquisite and brief; and Betsy
Dale and John Gourlay did a neat double act. Barry
Smith did a  live demo of the new \Textures.
Yes, there were more. In particular I've missed out  Mike
Spivak's presentation on \LamSTeX.

The first paper was {\bf Amy Hendrickson's} intimidating
`Getting \TeX nical'. Given her poor vendor presentation
I was a bit worried; should I stay, or go to the vendor
displays? I stayed. That was smart. Her presentation was
excellent and truly insightful. Her examples will soon be
incorporated into my course notes. What higher praise can
I give? She really left people feeling that they
understood what was going on in her example macros. They
had the benefit of being comprehensible and non-trivial.
A rare combination. The next paper dropped down from
overdrive to a more mundane level. `Where's the Greek
Shift Key?'\ asked {\bf Stephen Fulling}. Good cannon
fodder and well within the grasp of everyone. I had hoped
that {\bf Micah Beck's} `TransFig: portable graphics for
\TeX' might be just that. He discussed the history of
Transfig, but it is so \Unix dependent that I would query
the claim to portability. All my \Textures\ documents (and
their graphics)  are portable across Macintoshes, but I
don't really regard that as portable. I still don't
believe that something which is operating system or page
description language specific can be described as
portable. Sorry. Still, worth knowing about.

After lunch {\bf Andrew Marc Greene} amazed everyone with a
Basic interpreter written in \TeX. I'm reminded of Dr
Johnson's remarks about women preaching (or was it about
a dog walking on two legs?). The next query was whether
we could have a C compiler written in \TeX. Now that
might be useful. Perhaps Andrew will have one by this
time next year. 

The after lunch session began with {\bf Helen Gibson}.
Helen's was a multi media approach (slide projector as
well as ohp). Basically she described what was going on at
the Wellcome Institute. There were endearing photographs of
proud authors (including Dominik Wujastyk). The main
value of this talk was really the demonstration of \TeX\
use in a production environment, although the Wellcome's
production is not one determined by a commercial
requirement. Helen demonstrated that 300\,dpi CM was
inferior to 1270\,dpi Times-Roman. Surprise surprise.
This did draw some comment from the audience.

This CM bashing  has got to stop. Naturally at 300\,dpi it
is not acceptable -- Knuth never intended that it should
be. He planned to use it in publishing quality
manuscripts. At 1270\,dpi it is a perfectly acceptable
typeface. I personally have grave reservations about
Adobe's Times-Roman at 300\,dpi. At 10 point it looks
blotchy and uneven. The fact that it is a 12 or 14 point
design size cannot help either.

{\bf Robert Adams} discussed solutions on the
\TeX\slash\PS\slash graphics interface, rather than
problems. The books he described were remarkable for
their sheer quantity of graphics. Someone noted one of
the perennial problems of incorporating graphics -- the
annotation is almost always in an inconsistent typeface.

{\bf Stephan von Bechtolsheim} discussed some aspects of
the production of his {\it magnum opus\/}. While it will
take Knuth about 7 volumes to discuss everything there is
to know about computer programming, Stephan is taking four
volumes to describe \TeX. It is currently running at
about 1380 pages. For all those owners of the semi-licit
photocopies of this book, (who should have
agreed to buy the final work), Stephan will be happy if
they buy just one volume. It will be published (soon) by
Springer-Verlag, and Stephan expects the cost to be
about \$30-\$40 per volume (with some reduced price for
the set). I think he was relieved to have it finished. He
described a number of software tools he had developed to
help ensure that the book kept itself in step.

There was some muted criticism later that this
represented an advertisement, but I thought that was
mean. Stephan isn't going to get rich
on this. Well-known, or even notorious, perhaps. The more
people who know about the series, the better. It's only a
pity that there were not another couple of volumes on
\LaTeX. Similarly, when a few moments were found in the
program to allow {\bf Mike Spivak} to say a bit more,
there were also mutterings about unfair advertising. How
petty. 

In `Textbook Publishing', {\bf Mimi Lafrenz} described the
way that her company had risen above the \Unix\ tools to
production quality. Hers was a very positive talk,
perhaps the most invigorating of the whole conference. At
last, someone who was excited by \TeX, who thought it a
good thing, and who wanted to convince others. We could
do with many more like Mimi. 

Yoshiyuke Miyabe was re-squeezed into the program at this
point to describe {\it AutoLayouter\/}, a structured
document preparation system -- basically a structure
editor, to my mind not entirely unlike ArborText's {\it
Publisher\/}, but without any reference to \sgml. The
avoidance of international standards was a hallmark of
the conference.

Food intervened, in the shape of a TUG Country Buffet, a
sit down affair, where the Executive Board and Conference
Dignatories mostly sat at the top table. An unfortunate
division. This was followed by  poolside reception hosted
by Northlake Software (David Kellerman's bunch). This was a
good idea, but marred by the extremely high humidity.
However, some  decided that the `poolside reception'
should become a `pool reception'.   The skinny dipping
rumour was unfounded. 

Wednesday started with  {\bf David Ness'} diagnosis of
\TeX\ errors with a preprocessor (actually, not \TeX\
errors, more user errors). Interesting, but I'm uneasy with
preprocessors. It seems too easy to string them all
together and to loose something in the middle. Some of
his preprocessors were only at the planning stage.

{\bf Linda Williams} is this year's Knuth scholar. She
presented a paper in which she had customised the 
editor, EDT (often found on VAX/VMS, but also in  \MSDOS\
and \Unix\ incarnations), so that it made \TeX\ input much
easier. I had seen this done with {\tt emacs}, but even
the humble EDT could be made to perform quite creditably.
As an {\tt tpu/eve} person myself, I would have used the
more modern VMS editor, but that in no way underestimates
the usefulness. At least EDT is available on several
platforms, making the  transition from VMS to
\MSDOS\ or \Unix\ a little less painful.

{\bf Charles Martin} talked about training \TeX nical
typists -- in his case, including squaddies (grunts in
american). I've always maintained that secretaries are
\TeX-teachable, but the notion of training your average
squaddy had never ocurred to me. It provides a base level
which must surely demonstrate either the ease of \TeX\ or
the teaching skills of Charles. Or both.

Just before lunch {\bf Barbara Beeton} gave a Help Session,
based on a number of pre-arranged questions culled from
various electronic forums. This was a good idea, but
unfortunately the inestimable Barbara tends to get
overwhelmed with ideas and can become difficult to follow.

After this lunch time meeting, {\bf Michael Vulis} 
described his V\TeX\ system. Basically he has a sytem for
dynamically scaling fonts. Therefore you no longer require
these thousands of bytes of {\tt pk} files. He has also
added some other bits and pieces. Most seemed to be aimed
at the advertising industry. I can't imagine me ever
having to make a headline exactly span a given distance,
and varying the size of the type to do it. His scaling
technique seemed to ignore the notion of design size, but
then, so to does most of the desktop publishing world.
Much of the debate centred around whether this was \TeX\
or not. Michael says it passes |trip|. Frank Mittelbach
still considers it not to be \TeX. This introduces the
question of whether |trip| is the final arbiter of \TeX
hood, or whether it is merely a necessary, not a
sufficient condition. 

And lastly, Alan Hoenig, who started the conference, 
finished it. He, like many of us, was fascinated by
D\"urer's alphabet. This was a sort of geometric
construction, but Alan concluded that although
Albrecht described the components by drawing and word, he
also cheated quite a bit, and the correspondence between
`algorithm' and realisation was not exact. Nevertheless,
it was fascinating to see \MF\ breathe new life into this
typeface, especially when Alan exploited the {\it meta-}%
quality to create, for example a sans serif alphabet based
on D\"urer's original. He agreed that the results were
mixed in quality. 

\section{Summing up}
 The social activities were quite successful. They
threw people together and encouraged them to talk.
Perhaps the polka was a little beyond many of us (imagine
{\it touching\/} your partner in {\it public\/}!), but the
ample quantities of local beer help lubricate social
intercourse. Don't mock the local beer. Shiner Bock
turned out to be one of my fondest memories of Texas, and
the source of some amusement when one of the local
restaurants offered it as a `foreign' beer. Brewed fairly
close to College Station, and practically unavailable
outside Texas, this gave an interesting view of geography
(not an American strong suit at the best of times). I
suspect that the level of interaction was high because it
was not a huge conference, and because it was in a small
town. Last year's conference in Stanford was too big and
there were too many other distractions. 

The organisation of the social events and the conference
itself sometimes left a little to be desired. I concluded
that the pace of life in Texas was not fast.  The heat
obviously has an effect -- temperatures were low when they
were in the low nineties -- say about 33\degrees C, and
humidity was high too. And it did not cool down when the
sun went down.  I think the conference needs
a great deal of re-organisation. There were too many
lacun\ae. 

In particular the first day was a sadly missed opportunity.
Sieze the first day and stamp it firmly with the theme of
the conference `Forward into the 90's'. We were just
ambling along. The introductory stuff  could be a whole lot
snappier. After all, Alan and Nelson are
presenting material which has been presented more than
once. It is worth tightening it up so that it
really sings. And where were \LaTeX\ and all the other good
document tools? Then, when the conference proper starts,
let's have a keynote address which says where we are,
where we're going, not just in \TeX\ terms, but also in
TUG terms. A sort of `State of the User Group'. This
doesn't have to be a president's report, but it could be.
It has to be the hallmark of the conference. The short
reports ought to be dispersed throughout the program, if
they were needed at all. After all, the site coordinators'
stuff appears in TUGboat quite regularly. Getting the
foreigners to talk is good public relations, but again, it
has to be handled carefully. Too much broken english (or
read broken english) is painful.

The business meeting and elections are a farce. I am
beginning to sense that other attendees felt the same
way too. They are only there because the constitution
requires it. I have the strong feeling that people are
discouraged from attending. This is wrong. 

The vendors' display should be set up before the
conference starts. It should be closed when the general
meeting takes place. Under no circumstances should it
ever open at the same time that the AGM starts.

The networking
lunch was a good idea. It helped to stimulate the birds of
a feather groups.

The papers were a mixed bunch, ranging from the trite to
the excellent. I like it like that. Everyone has to have
the opportunity to present something, and in an
organisation whose membership varies from graduate
students, through computer support staff, faculty, to
various types of publishers and out to secretarial support
staff, one person's fascinating talk is bound to be
another's yawn. It would have been better to have a few
more papers, but obviously the organisers are constrained
by what is offered. Given the American reluctance to come
to College Station, the papers will be `thinner'.
However, I'll select the best papers: Mimi Lafrenz, Amy
Hendrickson and Charles Martin. 

One of the great successes of the conference must have
been {\bf Tom Reid's} remarkable note pads. Every
participant received two customised notepads. Each page of
the pad had a character from the cmr17 font, scaled 20000
for a 15\,dpi `writes-light-grey' device. And every one was
personalised with the attendee's name and address.
Remarkable.

Next year's conference will be held in Providence, the
home of TUG, from July 28th to August 1st. If I go, I
hope to organise a sailing bof. Put your name down now!

The verdict? Won on penalties in extra time.
\smallskip
\rightline{\sl Malcolm Clark}

\endinput

Doug and I
had left the drafters of the `rest of the board's'
statement with a Mac portable. It was a strange sight to
see Reggie Girouard, Barbara Beeton, Christina Thiele and
Allen Dyer sweat over using a mouse. They had never seen
a Mac before. What's the editor Barbara asked? But they
coped admirably, and when, at 2.30 Doug and I went back
to the campus to typeset and print the document (not
without some security problems), it turned out quite
nicely. I set it in Lucida. The notion was that it would
be distributed to all the board by about 8 o'clock in the
morning. I'm not sure that our distribution was as
succesful as our production. A moral there somewhere.

Lunch was another board meeting. It was anxious. The
Executive felt under pressure. So they should. With some
suitable rewording the first two motions were passed and
the third was agreed to be postponed to Cork. That was a
minor victory for me actually. I had read the by laws and
noted that a board meeting could be called by two board
members informing the president. A quorum is a quarter of
the board -- about 7 and a bit people. I knew that
Joachim would back me and that at Cork we would have at
least 7 board members, including a few hostile to TUG.
Ray said it wasn't possible. I showed it was. He hasn't
said a lot to me since. The most contentious motion was
one regarding the site of next year's meeting. It turned
up on the back of the program as Providence. No bad thing
actually, close to Cape Cod. I could go sailing like I
did last time I was there. It was where I found Rolling
Rock beer. What had irritated many was that the
announcement on the back of the program was the first
hint that that was the site. There had been no
discussion, although many of us knew that Ann Arbor
(ArborText) had offered. The gossip is that Ray was
`fired' from Math Reviews in Ann Arbor and won't go near
the place. Anyway, the decision had been Ray's. The board
felt that he was overstepping his power. That although in
the past he had selected sites, he should nevertheless
refer such a decision to the board. Lance put a motion up
requiring that the decision be deferred to the Cork
meeting. It was defeated ny 10 votes to 9, with at least
three abstentions. As far as I can see, the abstentions
were the three European board members (myself included).
I abstained on three counts: let the north americans
decide where their conference is; deciding in september
left too  short a time to organise; and I didn't want
Nelson to have to use a casting vote, either alienating
him from his Executive or his board.  Afterwards I
discovered that some of my colleagues wanted to have a
meeting targetted for publishers in New York. I don't
know that that would have swayed me.

After this we returned to birds of a feather groups. A
new one appeared -- the non-Executive board members. I'm
not quite sure what I can say about this. It stemmed from
some earlier dissatisfaction, but turned out to be a very
positive meeting. Basically the `rest of the board' only
see one another  once a year as a group. The real impetus
of the Board meeting is carried by the Executive,
including the Executive Director (himself an ex officio
member of the Executive). The Executive meet about four
times a year, and therefore are much more aware of their
`group dynamics'. They have some feeling for the way one
another will react in meetings. At the full board meeting
they {\it seem} fairly monolithic. Now, this is likely a
bit unrealistic, but since they have already discussed
the matters which come up on the agenda, they've been
exposed to their own points of view and can give the
appearance of cohesion. The rest of the board tend to
exhibit erratic, sporadic, ill-coordinated behavious.
Only a couple of them cabal into little meetings
beforehand, and they don't have the same sort of
juggernaut head of steam as the Executive. In a sense
they operate a bit like an upper chamber -- in theory. In
practice, they don't really come together, and rarely
challenge the Executive. Not that I am suggesting that
the Executive {\it has\/} to be challenged. For me it was
a remarkable meeting. Some of my colleagues on the board,
who I have never heard put two words together, were not
only vocal, but eleoquent. We realised that we had a
great many common frustrations and aspirations in common.
One of the frustrations was on the `long term planning
committee'. At last year's board meeting a decision was
made to appoint a committee to look at the long term
future of TUG (and \TeX). Sadly, this had not got under
way by this year's meeting. The many TUG committees, who
seem to exist only in name, were another complaint. Since
many of the board are on these committees (myself
included) this was a criticism of all board members. And
there was the feeling that we just were not doing enough.
TUG is a voluntary organisation. The board are even more
voluntary, and to a person, they want to do more for the
organisation and for \TeX.
\author{\mwc}