Henry wrote:
> namesinorder =. (0&{"1 /: 1&{"1) -...@fisdir culling ifany 1!:0 filespec
>
> Is it tacit? Is it explicit?
With the exception of the filespec, which is explicitly an argument, and the
missing glue (@) between some verbs here, this is tacit.
Let's be clear. Tacit means "an unspoken understanding"; in J the
understanding is where the arguments are. Naming pieces of (tacit) code does
not make them any less tacit.
Tacit does _not_ mean a dense, unbroken, uncommented line of anonymous ASCII.
Whether tacit coding encourages this style is a different argument altogether.*
No one would argue that
namesinorder =. (0&{"1 /: 1&{"1) @ (-...@fisdir culling ifany) @ (1!:0) NB.
Monad takes a filespec as its argument
is tacit. Is that really much harder to read than your original?
-Dan
* The question of whether tacit code attracts weirdos is also a different
argument :)
----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm
