Tacit programming was sure brought to the foreground by J, but function trains apparently precede it by a couple of years, and have since been implemented in Dyalog APL and NARS 2000.
The idea of functions having rank, and modifying rank, was also made central to J, but has been included in Dyalog APL more recently - as much as they can without breaking backward compatibility, I think.
But it's the "notation" change I'm struck by - the linked article says "discuss the future prospects of “the notation” in its various forms." as if APL and J and K are the same notation, when they casually, visually, aren't, and in the sense that J differs from APL of 1990 with rank and tacit functions and linguistic naming and behaviours (gerunds, conjunctions, et al), they aren't the same notation semantically either.
