I haven’t personally used APL or J for production purposes, though I studied APL in detail when I was designing SMP, the forerunner of Mathematica. I thought APL had some really interesting ideas, some of which have shown up in Mathematica.

I actually gave the keynote at the APL annual conference in 1989. I think they had some kind of conference song there, and there was a terrible moment where it said something about how APL had died and been reborn as Mathematica. From my point of view, I’d really like to see good ideas from APL propagate, and if Mathematica can help that, it’s great.

I got to know Ken Iverson at that conference, and talked to him quite a bit over the years. He explained to me his concept of APL being a notation for algorithms, analogous to mathematical notation for math. I said I’d ended up just leveraging people’s knowledge of English, and making the core of Mathematica notation be plain English words.

I talked to Ken Iverson, particularly late in his life, about why he thought people hadn’t understood functional languages better. His main conclusion seemed to be that somehow most people just weren’t smart enough (being an optimist about such things, I of course disputed this). I suppose it’s then really encouraging all around that in recent years it seems that functional programming is finally beginning to be widely understood.