On Wed, Jun 9, 2010 at 2:21 AM, Alexander Mikhailov <[email protected]> wrote:
> What would be an advice to J-ers? Clarity of idea - or,
> in extreme cases, soundness of algorithm? Or, rather,
> how to find a balance here?

First, I would write things the simple way.  If performance turns
out to not be an issue, I would move on.

If performance turns out to be an issue then... I might go
with an explicit definition, or a reduction, or I might try to
find relevant special code, or I might ask for help.

But, also, I might try manipulating the problem in its
abstract form, to see if I could express the same idea
differently.

(For example, relatively often, inductive approaches
turn out to  be equivalent to something like addition,
though perhaps in a different domain.)

-- 
Raul
----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm

Reply via email to