European Developer's Conference

By Jörg Langowski, MacTutor Editorial Staff

Letter from Europe

Happy Holidays. As MacTutor is entering its fifth year, we see more and more activity on the “other” (my) side of the atlantic, and many great articles have been sent to my address which we just couldn’t publish yet. Coincidentally, the first European developers’ conference has just been held in Paris; so we decided to dedicate part of this issue to the European Mac developers’ scene.

First I would like to express my thanks to all of you who have contributed to this issue and waited patiently for your article to appear. We are looking forward to more great contributions from overseas - keep them coming in!

European Developers’ conference, Paris

If Apple’s figures are correct (9000 certified developers in the US and 1000 in all of Europe), quite a significant proportion of those 1000 showed up at the conference. The main lecture hall was packed with 600 people, and there was just not enough time to listen to all the talks or see all the exhibits. This report is therefore necessarily selective, but I hope I can give you an overview anyway.

This being the first developers’ conference I attended and the first such event in Europe, I cannot draw any comparisons. To me it seemed rather focused on marketing rather than technical aspects of development. [Apple’s developer support right now seems to emphasize commercialization of “almost-ready” projects; more and more talented Mac programmers that I know start having difficulties getting certified at an early stage].

The list of speakers included mainly people from Apple Europe, but also a few famous names from the outside scene. All in all, it was a well-organized event, at a conference center in a park at the city limits of Paris, with good food included; one could meet quite a few people [for instance, long-standing MacTutor contributor Steve “Prof. Mac” Brecher and I met for the first time in person] and learn a couple of new things.

The main drawback of the conference was its lack of time. 2 days packed with lectures and exhibits on the developers’ fair changing twice a day (there were only 20 stands for 80 exhibits) was just too much, and Apple took a little too much time for its slick visionary presentations. Which were, anyway, interesting to listen to. Some highlights: “UNIX is an operating system neither my nor your wife would use - or want to” (Mike Spindler). “There is no artificial intelligence”; “A real personal computer is a computer you’d want to take to bed”; “When we’ll finally succeed in replacing the mouse with a pen, those used to three-button mice will have to take saxophone lessons” (Jean-Louis Gassée). However, to make room for those presentations which I’d file under “entertainment”, the technical talks had to be distributed into two parallel sessions, and I couldn’t split myself, so I had to miss quite a bit.

Neil Minkley, of the European Developer Services Group, emphasized that international compatibility still left a lot to be desired; for instance, most character recognition programs won’t deal with accented characters - a simple neglect on the part of developers who are not used to thinking “multinational”. Language, he said, might be considered an attribute of a piece of text, just as script, font or style. That way one could greatly simplify hyphenation, spelling checkers, etc. The perspective of a European market without trade barriers after 1992, he summarized his talk in one more visionary quote, should not have to wait till then - why not 1989.

Interesting to note was that French developers will finally get access to AppleLink. Not so hot: the access level will probably only allow for sending and receiving memos, all the other services won’t be accessible for a while. Whatever sense that makes.

We also had a chance to get some “first hand” information on the Mac IIx by Peter Orr, who made (involuntarily) a convincing point against upgrading for Mac II owners. It’s not quite clear to me how one can manage to squeeze a 30-minute talk out of a one-page press release, but he did. At least I learned nothing new compared to what I had read already, and simply a 68030 and 68882 at 16 MHz didn’t seem that revolutionary to me. An interesting question - which was not addressed - was if and when the Multifinder will finally require the PMMU. From all the promises about compatibility that were made at the conference, I can only hope-guess that we won’t have to trash our Pluses and SEs for a while.

Very thorough presentations were given on the two mainframe connection products, MacAPPC and MacWorkstation. Fergus Murphy, who showed MacAPPC, outlined the IBM LU 6.2 networking protocols and then described the intelligent APPC communication card developed by Apple in great depth. This card sits in a Mac II, contains its own 68000 and operating system and will provide the APPC network service to any Macintosh on the same Appletalk network.

Alain Andrieux presented MacWorkStation, the program that will provide a front end to any application running on a mainframe connected to the Mac by a network or serial link. With my programs still running on a VAX and having a full Mac “look and feel”, this seems to be the ideal toy for me. Later, on the Apple Expo, I could watch two demos: one was an application especially written for MacWorkStation, with menus, dialog boxes and window appearance controlled by the mainframe application, and one “plain vanilla” interface to a database where only few changes could be made to the program and therefore only text was displayed in various windows. Setup of the various Mac interface elements was amazingly fast, even though the mainframe connection was only at 1200 baud.

Alain Andrieux also gave a talk about MPW 3.0, which should be - hopefully - available in France by the time you read this. The points that impressed me most at first glance were the new symbolic debugger, SADE, which allows source-level debugging with Smalltalk browser-like windows across different programming languages under MPW, and the perspective that tools (like compilers) will finally run in the background under Multifinder. As a whole, it was confirmed that the Mac system is moving towards 32-bit addressing, and distinction between user and supervisor modes. We’re looking toward some more compatibility headaches here

Speaking of compatibility, Jim Friedlander from Apple Cupertino gave a long list of do-s and don’t-s. I guess since there still exist some programmers who write directly into the menu bar, such talks are justified. A couple of things in the discussion were interesting to note: The old Graf3D library won’t be supported anymore (who would have guessed?), the SANE engine won’t be speeded up much, they still think (or at least they say) precision is more important than speed, and Macintalk won’t exist for languages other than English unless one of the foreign developers writes another version.

With all the do-s and don’t-s, I missed the why-s. Developer guidelines seem to have a way to be stated rather than justified. Now, since modern software technology is largely object-oriented, and object orientation has to do a lot with information hiding, I kind of understand Apple’s way of thinking - to hide some of the information behind the developer guidelines that is considered “too dangerous” for the lowly “restuvus”. Rest assured that MacTutor will continue to try its best to break this information barrier.

Developers’ fair

Parallel to the conference, an exhibition was held to give participants a chance to show their products. Well, at least for three hours. Since space was so limited, each exhibit was shown only half a day, then someone else took over the booth. I was able to see at least some things.

The Macintosh with its easy localization is especially adapted to a multi-lingual environment like Europe, and there were numerous examples of products making use of these properties.

Winsoft - our Grenoble local heroes - showed their multi-lingual multi-script word processing system, Wintext. With a word-processing functionality somewhere between MacWrite and MS Word, this program will allow you to mix left-to-right and right-to-left writing, e.g. in combined English/Arab texts. On top of that they added an equation editor for scientific texts, and spreadsheets that can be placed into the text similar to Ragtime. The whole program is a wonderful example of what MacApp can do in the hands of a competent person. Kamel Gaddas, one of the creators of Winsoft, shared his MacApp experience in his conference presentation.

Unfortunately, Wintext is copy-protected, and in a stupid way: even when installed on a hard disk it will ask for the master disk on each launch. Sorry to say, much as I’d like to buy the program, they won’t get my money for a while.

[Note on the side - on copy protection, my favorite subject: while last year the standard answer was still ‘well, in the US we might be able to do without copy protection, but for France there is no other way’, this year we could hear ‘well, in Europe we might be able to do without copy protection, but for the middle East markets there is no other way’ OK, so the targets are shifting. Let’s hope they disappear].

AllPage was another multi-lingual word processing/ desktop publishing system by a company from Israel. Interesting feature: any selection, comprising text and graphics, can be copied as a picture and then pasted and re-sized.

Artificial intelligence (or whatever runs under this name) was represented in several exhibits: Procyon Research from Cambridge (UK) presented its Common Lisp implementation, which Expertelligence will distribute as a replacement of Exper Common Lisp. They distributed a very thorough introduction brochure and a demo diskette with a HyperCard stack. POP-11, the new AI programming language developed in Edinburgh and Sussex, was of course presented in its Mac implementation, AlphaPOP. Texas Instruments showed their Mac II-based AI workstation, microExplorer.

Mainstay, the makers of VIP, Think ‘n Time, and other things, presented MarkUp, their new approach to text revision; sort of an electronic “red pencil”. This is a system where a group of people can share a text document and make their comments on transparent “layers” which can be later selectively displayed on top of the document. So Joe can type a text, then have Debbie and Al revise it, later look at their comments on top of his text independently, change the text, make his own comments, etc. Comments can be entered into the text by typing or by using a set of graphics tools. Of course, color is supported. A pretty unique idea, and hopefully a success.

One particularly interesting contribution was an image processing application for microscopy Optimage, developed in Paris by university researchers and now being commercialized. It is the first program that I’m aware of on the Mac that allows digital filtering, Fourier transforms, particle detection and counting, or size determination of objects in microscopic images.

Apple Expo 1988

The annual Paris Apple Expo was held during and after the conference. This time, the exposition hall, an old glass-steel construction, had been entirely transformed into a giant Mac II. 10-feet high floppies greeted the visitors in front of the building (how many gigabytes on that one?), and the - nonfunctional - 30- by 40- foot monitor on the outside served as a cloakroom. In the belly of the Mac II, the usual exposition come-and-go. Interesting the stand of Informix, in the form of a giant spaceship, where they introduced the long-awaited “hyper-spreadsheet” Wingz; this time they were announcing shipping in November. Vaporware special was the Gigatape, a 1.2 Gbyte SCSI tape backup which was supposed to work with DAT cassettes. When I asked for a demo, they said “we’re still working on the driver, but it’ll be ready Very Soon”. Oh well.

Apple occupied - naturally - a lot of floor space, and many of the more innovative applications were displayed on their stands; for instance a Labview-like data acquisition system, Personal Writer, a handwriting recognition system using a digitizer tablet, tons of educational HyperCard stacks, among them a system that will automatically find the best subway connection between points A and B in Paris - and also tell you how long it’ll take (usually too long).

This year a contest was held; the Apple Trophy was awarded to seven programs selected by a jury from Apple France. 23 programs were on the short list, and I was promised to get a copy of that list so we could print it here; alas, it never arrived before our deadline (unless David receives a link from Apple France directly, in which case it might be printed after all). I’ll just proudly mention that one of the members of our local Mac club, Yves Baulac, was among the prize winners with his geometry program, Cabri Geometre. The other six were Fileguard by Olivier Merenne, a hard disk protection/encryption facility that will protect your documents transparently; Hyperpage by Bernard Meunier, a page layout XCMD for HyperCard; Emploi du Temps, a planning utility by Jean-Luc Deleage; two medical utilities, Hypermed by Almanza and Thesaurus Urologie by Lardennois; and a graphing program, page de Graphe by Robert Elbeze.

Otherwise, the usual exhibitors showed their usual products (hello, Microsoft; bonjour, ACI). Most of the real new stuff had already been shown at the developers’ fair. It was fun, anyway, to spend some time with friends at the MIDI stand.

Mach 2.14 news

Just as I write this, one day before the deadline, I’ve received the final release of Mach 2.14, where a lot of things have been added even with respect to the beta version. Finally, the editor allows you to PRINT your text as well! I realize that this might seem funny to anybody not familiar with Mach2; most Mach2 users had just resigned and used Afterthought, MacSink, Joliwrite, Mockwrite or some other DA text editor. Don’t laugh. Mach2 would be a great product without its integrated editor, and making that part fully functional has only improved it.

I have talked about the changes of the Forth system itself in my last column; however, they now added an assembly source file with some definitions from the Forth-83 Standard’s Controlled and Uncontrolled Reference Word Set, along with several common Forth extensions. This should make the system even more Forth-83 compatible. New words include: W/MOD , W/ , W* , WMOD , UNDER+ , U.R , TUCK , TEXT , ROTATE , ON , OFF , OCTAL , O. , NOR , NIP , NAND , MASK , K , INCR , H. , ERASE , DECR , CELL/ , CELL- , CELL+ , CELL* , BLANK , B. , @BITS , >= , >C< , >< , <= , -ROT , -MATCH , (-MATCH) , -TEXT , (-TEXT) , ** , !BITS , CELL , FALSE , TRUE , BL .

Next month it’s back to Real Forth again. Till then.