(leaving the un-adable subject line intact to keep this stream separate from the one about symbols)
BobGraf [mailto:[EMAIL PROTECTED] wrote: > I sincerely believe that APL, and its successor, J, have both failed in > popularity for reasons having nothing to do with technical issues, but > rather with issues related to sales and marketing of the APL and J products. I think it is true that marketing of APL and J has been less successful than it could have been. But the reasons are quite complex, and in my opinion they have little to do with the popular myths regarding character sets, lack of open source versions, and other ideas that have been suggested as "THE" reason why the languages have not become more popular. One big mistake that I think we have been making for the last 20 years is to continue to think of APL and J as languages which should be competing head-to-head with languages like C. This sort of thinking did make SOME sense back in the 70's and 80's, when all software development was a pioneering activity and the APL environments were very competitive in a general sense. But nobody wants to be prototyping their own General Ledger system any more, although lots of large companies did this with APL in the 70's. The APL market lost a lot of value when many of these products matured and became commodities (and left the mainframe). But this is how it should be: The infantry has arrived to secure the beachhead, and the special forces move on. If I had to maintain that Ledger system today, I would have quit anyway. This shrinking of this "unnatural" market says nothing about the true value of these notations. To me, APL/J/K programmers allowing themselves to be depressed by the fact that their language has not replaced C would be like rocket engineers retiring from NASA after deciding that rockets have no future: For inter-continental bulk shipments, the market share is pitiful when compared to container vessels. The "(mostly) infix functional array languages" will NEVER become the software engineers tool of choice. I personally believe that they will become VERY much more popular than they are today, but mostly because "software-assisted problem solving" will grow in importance relative to "programming", which is becoming an activity for "unskilled" labour. Our languages appeal to people who like to solve interesting problems. The languages appeal less to people who are out to make quick and fast money selling things at a higher price than they can buy them ("business people"), or to people who like to think about the beauty, generality and completeness of information-processing algorithms ("computer scientists"), or to people who like to think of programming as an activity which can be compared to pouring concrete ("software engineers"). The fact that our communities have an over-population of "problem solvers" who are willing to depart from the beaten path DOES cause a fundamental problem with respect to marketing, because the problem solvers are only infrequently motivated to explain their solutions to business folks, computer scientists and software engineers - as they do not easily relate to these folks and feel uncomfortable in their presence. Also because "finding the beaten path" is one of the most important things that businesses generally want to buy, and we have little wisdom to offer on that front. "Most people would sooner die than think. In fact, they do." - Bertrand Russell This doesn't mean that I don't think there are people using Java, VB or C# (or Python, Ruby, etc) who could not benefit from using APL, J or K. There are tens, maybe hundreds of thousands of people out there that we should be trying to attract. Increasing our market share by one or two orders of magnitude is realistic. But we should never expect "popularity". Only interesting times and for those who focus on money, handsome profitability when ideas are brought to the market just ahead of their time :-). I believe that current market conditions favour significant growth for technologies which are enable to involve "domain experts" more directly in the software development process. We've been doing this for 40+ years now and have some real value to add. I'm placing my personal bets on APL because it has the broadest commercial and technical base from which to grow, but I feel that the entire family of languages should have a good decade or two ahead. Morten Kromberg ---------------------------------------------------------------------- For information about J forums see http://www.jsoftware.com/forums.htm
