Current Situation A Future for APL Currently, a handful of A P L vendors are sharing a market that is probably not growing, at least not in relative terms. They do their best, but they have failed to develop a state of the art programruing language and environment that can compete with the likes of Borland, Microsoi~, etc. The state of the A P L community has s o m e striking resemblances to Axelle Zabou's view of the state o f the African c o n t i n e n t (Et si l'Afrique refusalt le d~eloppernent?, L ' H a r m a t t a n , Paris, 1991): Complacency by an establishment; an idea o f cu.It u r a l specialness while ignoring progress m a d e by others; particularism, with declared, b u t ineffective, cooperation. Martin Gfeller Financial Applications Development Group Reuters Kleinstrasse 6 CH-8008 Zurich, Switzerland Tel: +41 1 2 5 1 2 3 0 0 Fax: + 4 1 1 262 17 92 E-mail: mgfOrmx, risk. r e u t e r . d e There Is No Future for APL I believe that some drastic steps are needed to secure a future for A P L and to save it as a tool of choice for the coming years. I believe that the A P L community has been too complacent for too long, and that the A P L establishment has ignored the progress in mainstream programming languages and systems for too long. W e run the risk that A P L will no longer be accepted by employers, schools and colleagues, not because they are stubborn, but because other languages and systems have overtaken A P L in fitness for the task (productivity, interoperability, and performance, amongst others). A Few Action Points 1. Vendors, unite! Competition is e x t e r n a l . T h e A P L suppliers should merge, or e n t e r s e r i o u s cooperation agreements. Maintaining half a dozen major i n t e r p r e t e r s for a l a n g u a g e o f APL's m a r k e t share is a t r e m e n d o u s w a s t e o f resources. Develop a new APL t o g e t h e r w i t h a c k n o w l e d g e d experts in modern p r o g r a m m i n g l a n g u a g e s , sof~ware engineering, and object-oriented t e c h n o l o gies. Bring outside and c u r r e n t expertise into the APL community. Release full product suites, w i t h tools, a n d possibly a compiler, at competitive prices (made possible by vendor cooperation). A t t a c k the professional m a r k e t . Programmers are the main users of p r o g r a m m i n g l a n g u a g e s , and professionals have decision power. Provide education for professional p r o g r a m m e r s . 2. It Is Not Enough To secure a future for APL, it's not enough: ¢ to provide a good language interpreter with some interoperability facilities. ¢ to provide a language that w a s essentially defined in the eighties (nested arrays and operators), whereas the mainstream languages since have adopted and absorbed completely n e w paradigms. ¢ to compete with other A P L implementations. ¢ to lag behind other tools by at least one year with regard to seamless interoperability (Windows, D D E , O L E 2, RPC, C O R B A , Distributed Objects, etc.). ¢ to teach the language to a few students at a few selected schools, at great individual efforts. 3_ 4. 5. I wonder if I will still be able to use A P L in five y e a r ' s time. I f y e s - - a n d I s i n c e r e l y hope s o - I'm certain it will be v e r y different f r o m t o d a y ' s APL: In a world in w h i c h c o m p u t i n g is done by combining objects, APL will be one o f the l a n g u a g e s t h a t can be used to implement services of objects, and it will intermix and i n t e r o p e r a t e freely w i t h object services w r i t t e n in o t h e r l a n g u a g e s . I f A P L doesn't live up to that, it will be a dead l a n g u a g e . ¢ ¢ to address the elementary and intermediate educational and "amateur" markets. These markets are not the prime consumers of programming languages, and they are not where the decision makers are. ¢ to claim A P L suffers only from bad marketing, and simply needs proper selling. At every A P L conference I have attended, I have heard this. If most marketing attempts failed to take the product really off, maybe the product itself is wrong. A n s w e r to Editor's Quiz on P a g e 14: lJ~l m,~ =ql ;v e!ssn'~l ~.uo - p r o pu~ ~o I s~! ol ~ s ~ I V q l ~ 'o.m~5~t oq~ j o op~ 1~.u uzolloq oql le epeue3 m o q ~ o u o z r ~ o ~ ,~ffsvo~zouz ~ u z no,~ pure 'u~op op!sdn ~ v d ~q~ uznl 'dol Otll le q ~ o N ql1Ax polu~uo sdem ~tnoos ql!,~ . r e . . ~ j lsom oze oq,~ sn j o osoq~ i o ~ June 1994 - V o l u m e 24, N u m b e r 4

/lp/association-for-computing-machinery/a-future-for-apl-cuidMpT0hf