A report from euroFORTH '95

Len Zettel
zettel@acm.org

What is a conference? In this case, 32 registered delegates, (three from the US), 21 scheduled papers in five sessions, four workshops, a jazz concert, and a programming contest. Three days and two nights (fourth day optional) of good food and drink, good conversation, all held at a very well appointed conference center (International Meeting and Research Center for Informatics, Schloss Dagstuhl) in the boondocks behind Saarbruecken. And you didn't have to explain to anybody what Forth was. If I had known I was going to prepare a report on the doins maybe I would have paid more attention to what was going on instead of just having fun. More information on the 95 conference and euroFORTH in general is available on the euroFORTH web site. Mention should also be made that Dr. Peter Knaggs was one of the conference chairs. Marina Kern was conference organizer and very much all over the place keeping things running smoothly. The organizational contact was Delta t Gmbh of Hamburg. They claim they had help from the University of Saarland in Sarrbruecken and two working groups of the Gesellschaft fuer Informatik: Architekturen fuer hochintegrierte Schaltungen (FG 3.5.5) and Alternative Konzepte fuer Sprachen und rechnen (FG 2.1.4).

It was an interesting experience, not least because it was my first trip across the Atlantic. My wife and I were staying in Frankfurt. I shipped her off to Paris and the Louvre while I tooled down die autobahnen to Schloss Dagstuhl. When I say boonies I mean boonies. Got a bit lost and had some interesting views of the scenery (spit and image of western New York or Pennsylvania) around Wadern, which at least was on most of my maps. Stumbled across a sign post aimed at a paved cow path and drove down it. Came to a beautiful bunch of buildings that this American would characterize as a chateau with a number of cars parked outside it. Did likewise, walked across a front lawn with drive and fountain to a locked door. Peering through door window could see, well down a hallway, what looked like a registration desk. Knocked and got admitted. This was indeed the place. These guys not only talk computers, they use them. All the doors are locked, but have push button combination gadgets. Conference guests get the combination. If you had a good memory, they worked better than Rochester's cards (sorry, Larry).

In some ways the formal conference sessions and presentations are the (necessary) excuse to hold a conference. The important part is being there, seeing people in the flesh, and getting to gauge the man (or woman) behind the name and the paper. It may be unscientific, but it helps greatly in judging the validity, the strengths and weaknesses, of what is presented. There is also the chance to make new friendships, kick things around informally. Thus the axiom that the important stuff happens in the hallways. I did everything I could to seize my opportunities.

So this is going to be a very personal, informal report. You can get the formalities (or most of them) by ordering the conference proceedings from delta t. MY overall impression was that the attendees were somewhat more serious as a group than at the American conferences I have attended. Conspicuous by their absence were the religious hacker anarchists you get here as a prominent minority.

Some highlights:

• Meeting fellow JFAR (plug) editor Peter Knaggs for the first time and getting to thrash out some of the issues in getting it started again. I also found some delicious irony in this guy, whose appearance is essence of hippie (if you don't believe me, check out his home page) talking very much like the professor he is and complaining about how students nowadays don't want to put any effort into their studies.

• It was also good to see Ulrich Hoffman, whom I had known only from comp.lang.forth postings and email.

• Stephen Pelc, whom I had met before, was also very much in evidence, as was Microprocessor Engineering ltd.

• The formal presentation that really made an impression on me was Paul Frenger's "Forth in an Object Oriented Android Control Network". He had constructed an artificial hand largely out of tongue depressors, beads, dowels, plywood and nitinol "muscle wire" for a total cost of about $20. This thing had all the degrees of freedom of a human hand and was run off a New Micros board. Forth based, of course. Wow!

During the conference (Friday afternoon or evening, I think) we were asked for suggestions for workshops and then voted for the ones of most interest. Selected were Object-oriented programming, Literate Programming, Using Forth as a Target Language, and Industrial Networks.

Since I had suggested the Object-oriented programming topic, I got to lead it. My heart sank a bit when I heard somebody remark that there has been and object-oriented programming workshop at every euroFORTH since time immemorial. Oh well - I hadn't been at them. The workshop was wonderful, an intense exercise in group creativity. We set ourselves three tasks - to specify the characteristics of object-oriented programming, to invent a syntax for doing that in Forth, and to develop sample ANS Forth code implementing the syntax. As it turned out, we did pretty well with one and two but didn't get around to three. There was general agreement, however, that what we had come up with was implementable and had not yet been done in Forth to that level of generality. One of the items on my personal to do list is to get it written up, with the Forth column in SIGPLAN notices for the first try at publication. Even if nothing comes of that, it was worth it as the best head trip I have had in a coon's age.

The programming contest was for "the most elegant way to crash a Forth system". Runner up, that would have won if it had been less wordy, was

VARIABLE MOST_ELEGANT_CRASH MOST_ELEGANT_CRASH EXECUTE

1 DUP BASE ! .