On Thu, Jun 4, 2009 at 6:38 AM, Don Watson <[email protected]> wrote:
>        When longer trains of verbs (and other language components) are
>        created in Tacit J, why is the fork/hook grammar better than
>        right to left grammar?

Answering this question is awkward, because the question itself
has some awkward elements.

First, there is no universal "better".  Forks allow the literal use of
a set of not-uncommon mathematical expressions, but I am
uncomfortable with trying to express this as some kind of
universal "better" mechanism.

Second, fork/hook grammar is right-to-left.  For example,
   f g h i j
is equivalent to
   f g (h i j)

That said, I think what you are really asking is for a contrast
between J's current grammar and some of your proposals.
I might be thinking wrong, but I can try providing some of
these contrasts.

One important issue is that J's grammar has been implemented.
This means that any inconsistencies which would prevent
implementation have already been resolved.

Another potential issue is expressiveness.  J's forks
let me express some ideas rather concisely.  I suspect
that I would have problems with this kind of thing in an
implementation of monadic chaining.  I can do a lot
with forks which I can not do with sequences of verbs
separated by @: (which seems to be equivalent to what
I think you were asking about).  The reverse is not
the case, because of capped forks.

Does this help?

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

Reply via email to