[tom_b]

Despite only occasionally hacking around with array programming languages, I have always enjoyed the terse syntax. But I admittedly find weird syntax somewhat engaging.

I have wondered what was happening with statistics in the larger group of array languages. Do other HN readers know of J, Q/KDB+, Klong, or others in use by teams outside of the normal financial folks using Q/KDB+? I'd be quite interested in hearing any stories from the business world where it isn't a "lone hacker" situation, but rather even a small group of programmers were using array languages in production situations.

[nerdponx]

The terse syntax is totally off-putting to most people and it makes it a hard sell at work or to a research group. I suspect the only reason it found a niche in quant finance is that those guys are geniuses and can kinda do whatever the hell they want. They also don't have to share their code all that often AFAIK.

There's also no reason for it. Heck, you could even give each function an alias, write a "verbose language", and have it compile to the short form before being processed further. Then you could even write both long-form and short-form, and interleave the two if you really wanted.

[beagle3]

Q is exactly that for K. Almost everyone switched from Q to K sooner or later, because once you grok the ideas, the verbosity gets in the way of both reading and writing.

There's also kerf, which is verbose; AFAIK it was not met was the same success that APL, J or K have -- though likely for unrelated reasons.