> Numbers x*y such that x + y or |x - y| is a square.
Perhaps I'm stating the obvious here: this defines a set not a sequence.
If you want a comment perhaps that's the place to over-explain the set
is assumed to have increasing order instead of weird order.
Cross-referencing to squares is also a good suggestion from the reviewer.
Another good idea for crossreference is if you know of or can find other
sequences including or included by your new suggestion.
Composite numbers would not work as a super set since you are allowing
1's to appear as factors. Your submission should be a superset of this:
My usual first question about a sequence like this is how many crossrefs
would we need to define it in terms of set operations on those seqs?
You could also cross reference the primes as a lot of them are missing.
Best wishes,
--Brad