On Wed, Jul 29, 2009 at 5:18 AM, neville holmes<[email protected]> wrote:
> Thank you Dan; have a good trip.  Your rephrasing displays
> an understanding of J far beyond mine.  I've never used
> $: for example.  But your rephrasing encompasses my suggestion,
> as far as I can make out.

Well..

To be fair, this will be different from $.

t=  $.

t is a verb -- albeit, a verb without a domain.

This means that the cost of implementation would be significant.

If he were to implement this, I think Roger would have to introduce
up to six new classes of parsing tokens.

Here is a list of them with some not particularly good examples:

pseudo verbs that will be come adverbs
   + [. *

pseudo adverbs that will become adverbs
   (+ [. *) "

pseudo nouns that will become adverbs
   (+ [. *) 0

pseudo verbs that will become conjunctions
   + ]. *

pseudo adverbs that will become conjunctions
  (+ ]. *) "

pseudo nouns that will become conjunctions
   (+ ]. *) 0

I do not think we can have tacitly defined conjunctions
or I would also have to worry about some other
possibilities.

It's also possible that some of these cases would
be arbitrarily ruled "illegal" (to make the implementation
easier).

Before I can even begin to think about what
all this means, I would want to come up with
a variety of useful or at least plausible examples
of each.

But it seems to me that implementing this mechanism
would involve quite a bit of work, and would have
non-obvious implications, both for the user and for
the implementation.

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

Reply via email to