The Education VECTOR

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

The previous column under this heading concluded by announcing how the idea of Public Domain APL was born at APL86. Since then, I-APL (International APL) has been born, and the embryonic concept has moved inexorably to project status. The British APL Association has promised funds subject to committee approval by 1st December of a technical specification and business plan. SIGAPL has promised similar support. An appeal has been made to individual and sustaining members of the BAA and to the other European associations to contribute to the £30,000 which is required to see the project to successful completion. Happily enough initial funds have been guaranteed to enable the project to engage Paul Chapman to embark on the I-APL interpreter, while on both sides of the Atlantic work is progressing with the supporting documentation.

The existence of an interpreter and documentation is of course by no means all. The next stage will then be to bombard our schools with APL, and although our plans include approaches to LEA Mathematics Advisers, Mathematics Project Directors and High IQ (Information Technology Quotient) Schools, this is a point at which all members of the Association have a part to play in consciously telling and persuading the rest of the world that APL is here, and available to all, and that it contains the seeds of important educational innovation in the Mathematics and Science classroom. In particular those members who are concerned with recruitment into their companies have a special role to play. An argument heard regularly from Polytechnic lecturers is that they do not teach APL because it is not used by employers. How I wish I could broadcast the telephone conversation I had recently with a representative of a large British company asking me if I knew at which Polytechnics they should look to recruit graduates competent in APL. Is there a turning point in view?

Meanwhile we continue to hear of new APL enterprise in education. At University College, London, for example, Dieter Girmes in the Statistical Science Department runs a 2nd/3rd year course on Mathematical Computation which is based entirely on APL. After a minimum of formal instruction, students are issued with sheets each containing an APL function and its associated “brother” function which gives a detailed trace of execution, thereby explaining what the various APL primitives are doing. The aim is not to impart any new mathematics, but rather to relate known mathematics and statistics to APL, and thereby focus on the power of APL to express and execute algorithms. The course is practical rather than didactic, and support is provided by terminals connected to the College’s Euclid computer.

On the school front, delegates at the MUSE (Microcomputer Users in Schools and Education) Conference showed considerable interest in APL, and we hope that by the time of the next conference I-APL will be a viable and attractive option on school micros. At Portsmouth Grammar School, the computer donated by Quaker Oats in the competition judged by the BAA has been received with enthusiasm by pupils, one of whom has been working on a mini-APL interpreter in COMAL as a project.

Meanwhile, as the dawn of I-APL is awaited, we take a look at what low-cost APL systems are currently available to the home or school user. First we mention two free APLs – and yes we do mean free! These are VIZ::APL – put into the Public Domain by Inner Product and which runs on the RML 380Z and other CP/M machines – and Acornsoft APL which runs on the BBC micros equipped with the second Z80 processor. Anthony Camacho can be contacted for details of the Acornsoft interpreter and we hope to have VIZ::APL available through the BAA by the end of the year; please do not ask Inner Product for copies.

The cheapest APL system remains the Sinclair QL, which can still be obtained; ask MicroAPL if you have problems. This typically costs around £200 with a monitor, and QL/APL from MicroAPL costs a further £100 in either keyword or symbolic notation. (These prices in common with all the others quoted here include VAT.) QL/APL is a full-blown no-corners-cut APL with file facilities and graphics links. I have been using it myself at home for a year or so and I have yet to find any flaws in the interpreter.

The next available product going up the quality ladder is the Atari 1040 for which you can expect to pay £920 with a monochrome monitor, or £1150 with colour. APL (again from MicroAPL) will cost just under £200. APL for the Commodore Amiga comes in at £260, and for the Apple Macintosh at £295. MicroAPL reports shipments of 300 per month for this product, but sadly most of these are to the Continent. PortaAPL, the other APL product which runs on the Macintosh, sells for about $200. (Ed: most of these interpreters have been reviewed in recent issues of VECTOR.)

If your budget runs to the IBM PC or its look-alikes, the best purchase is version 2 of IBM’s own APL which retails at £328 and is subject to its Educational discount. STSC’s longer-established APL*PLUS sells at £546-25; Pocket APL is STSC’s mini-version of APL, which costs about $100 but is not on sale in the U.K. It is too early to say whether the new Amstrads will run APL successfully, but the first signs look hopeful.

In summary, APL is poised at an educational cross-roads – I-APL promises the opportunity to prove once and for all that it is only cost which has prevented APL from taking its place as one of the most popular and useful computer languages. We move forward in faith and hope.