APL91: Conference Announcement

Transcribed from the printed issue; not yet reviewed. Read it in the PDF of the issue, page 33

We are glad to bring readers an early and highly provisional draft of the plans for APL 91. Vector will circulate news of the plans as they develop, and will later include the call for papers and details of how you get on to the mailing list.

APL91
THE NEXT 25 YEARS

[Provisional Theme]

The International Conference on APL 4 - 8 August 1991
Stanford University, California, USA

Call for Participation¶

Sponsored by the Associating for Computing Machinery, Special Interest Group on APL ACM SIGAPL, in cooperation with Stanford University Computer Science Department. Organized by The APL Bay Area Users' Group

Conference Overview¶

The APL91 conference will take place 4 - 8 August 1991, on the campus of Stanford University, in Palo Alto, California. University accommodation will be available as primary housing with local hotels for those who prefer them. University dining facilities will be available as well, so that subsistence expenses will be quite low. We hope that the low cost, location, and time of year will make it attractive and induce strong international attendance.

The main conference will run from 5 - 8 August, with paper presentations, general sessions, poster sessions, exhibitors, vendor forums, and social events for informal exchange with other attendees. Two tutorial tracks are planned for Sunday, 4 August.

Tutorials¶

The first tutorial will be aimed at APL experts, addressing the use of APL in one or more of these areas:

Artificial Intelligence/Expert Systems:

• APL has been used in a variety of research and production applications such as medical diagnosis and factory automation.

Object Oriented Programming:

• Recent papers have shown various ways that APL can be used for object oriented programming style.

Case:

• APL can be considered a tool for Computer Aided Software Engineering with its proven capabilities and uses in generating code for ADA, assembler, FORTRAN, and C.

The second tutorial is being planned to introduce APL to a new generation of students, through informal sessions. Attendees at this tutorial would receive a package of free APLs to take away and continue learning and using. In addition to the regular conference registrants, students and teachers from local schools and colleges will be invited to participate.

Papers¶

50 papers will be presented in two or three parallel tracks depending on topic, 45 minutes each.

Posters¶

Posters will have a special session set aside for them and will address short topics. We expect to include work in progress and informal presentations which may later be prepared for publication.

General Sessions¶

Opening and closing sessions will feature prominent speakers in fields of interest to APL users, but not necessarily in APL.

Vendor Forums¶

Vendors will be invited to present products and/or participate on panels.

Exhibitors¶

Exhibitors will display products and services in an exhibit area open during the entire conference and featured during an exhibit reception one evening.

Discussions¶

Rooms will be available for ad hoc scheduling on birds-of-a-feather topics.

Events (Provisional):¶

Program Statement¶

Important early developments in implementing Kenneth Iverson's mathematical notation as an executable computer language occurred at Stanford in the 60's: the first full working version of APL resulted from collaboration of Stanford graduate students with IBM. APL has now come into wide use in academic, scientific, and commercial endeavours around the world. 1991 will be the 25th anniversary of the availability of an APL system, marked by the date IBM's 1 CLEANSPACE was saved. 1991 will also be the 10th anniversary of the annual international APL conference last held in the San Francisco Bay Area. Now that free APLs make widespread growth of its educational use it seems appropriate that we meet at Stanford to consider the next generation of APL and APL users.

APL development has diverged and converged through many paths over the last 25 years, achieving an ISO standard and enhancement to second generation APLs, and now working on an ISO standard for extended APL. APL has pioneered in many areas, including interactive timesharing; functions acting on whole arrays; dynamic data arrays; array theory; use of vector, array, and multi-processor hardware; cooperative processing; spreadsheet like applications; independence of national languages; free distribution of interpreters; use for AI and expert systems; use by non-programmers and use in education from grade school to university. Computing technology is currently facing several large issues, such as proprietary versus open systems, hardware interchangeability versus new hardware architectures, programming versus CASE, and PCs versus workstations versus mainframes. The recent availability of free APLs for PCs and school computers has made APL freshly available to a wide audience in education. Thus, it seems an appropriate time to consider the role or directions for APL in "The Next 25 Years," the theme for this conference.

APL is subject to an interesting variety of paradoxes:

• APL is often considered an isolated technology. Yet software developments continue to credit ideas from APL, rightly for example, work on the new standard for vectorized FORTRAN or Mathematica.
• It has been said that APL cannot be taught; it can only be learned. It is not often taught as such in University. Yet first courses in computer science taught to undergraduates at Yale University with APL had students writing simulators and assemblers by the end of a term. And APL is being used in grade school in parts of Europe to teach mathematics.
• APL's symbolic basis is widely criticized for requiring special handling. Yet this makes it largely independent of national languages (APL needs no translation). And graphical user interfaces based on icons for mnemonic association are becoming the norm on workstations.
• Some people think the symbols inhibit portability. Yet the strict syntax of APL easily reveals implementation-specific features, and APL is the only language that can run on PCs, mainframes, and vector-hardware supercomputers without program changes.
• APL is a powerful tool for non-programmer computer users, such as engineers, statisticians, and actuaries. Yet it still bothers professional programmers to subordinate detail in the expression of algorithms. It has been called too general a tool for production commercial applications. Yet computer science students now graduate thinking that C is appropriate for business applications of all kinds.
• APL has sometimes been thought to use too many computing resources in return for its conceptual advantages. Yet in other advanced computing technologies such as artificial intelligence, this is expected. And the result of the conceptual advantages is often better algorithms for much larger performance gains.
• APL permits unstructured programming. Yet its natural programming style is functional and it can utilize structure inherent in the data.
• APL has proven productivity advantages over other languages, yet it is not uncommon for DP managers to discourage its use. Do we know why?

We hope to receive papers addressing these paradoxes and issues, concentrating on real experiences or work in progress toward new uses. These experiences, together with conference discussions among attendees, general sessions of speakers and panels, and participation of exhibitors, will focus on where and how APL should best be used in the next twenty five years. Is APL a notation, a language, an environment, a system, or a paradigm? Where should vendors and users take it among other technologies? In which areas is APL successful, and in which areas is it not? Now that it can be accessible to schools with limited computing resources, what should the next generation of students be encouraged to think about it?

Conference Committee:¶

Co-chairman, Administration/Finance: Charles Schulz
Co-chairman, Program: Tibor Horvath
Secretary, Communications: Curtis Jones
Exhibits, Events: Nancy Wheeler
Press, Publicity: George Mendonsa

Program committee: to be formed