Oleg wrote: > 4 is in mtab, so "not e." is probably primes? Correct.
Raul wrote: > It rather inefficiently finds prime numbers in the sequence up > through 1 plus its right argument. Correct, and the most precise answer yet. > I think that i.&.(p:inv)@+&2 would perform > this same operation more efficiently. It would, but then p. essentially says "primes", and would be no better a litmus test of J's obscurity than asking (as an exaggerated example) "what does + mean?". Plus, the challenge as written is easier to translate to other languages and post in their communities, so we can compare "comprehensibility", as it were. Most other languages don't have an equivalent to p. (or inv for that matter), but almost all of them will be able to express the algorithm as written. Donn wrote: > [image: s_i=1/nsum_(j=i)^(i+n-1)a_j.] I don't know what this is. LaTEX? Can you post a challenge in J? The motivation of this thread is to ask "how comprehensible are non-trivial expressions in a language to the general constituency of that language?". Or are you asking how we'd express the above in J? (If so, I expect someone who already knows what language that is will post an answer before I can.) -Dan ---------------------------------------------------------------------- For information about J forums see http://www.jsoftware.com/forums.htm
