Well, you asked. So here's my thoughts:

1) I am not sure what your suggestion would mean for something like (1
+ (: (u x)+u y ) 2).

2) Also: the "existing syntax for non-tacit verbs" has already already
been extended above and beyond the syntax for J sentences.

3) And multi-line blocks are already an extension above and beyond the
syntax for J sentences.

4) Or rather, I suspect that your suggestion would mean a lack of
modularity in the underlying parser implementation, where the sentence
parser needs to be extended to handle multi-line non-sentence
constructs.

5) Also: saving a single character in explicit definitions doesn't
seem like a big win.

6) J also contains the wd sub-language, which also uses a different syntax.

I hope this helps,

-- 
Raul

On Tue, Oct 20, 2020 at 7:04 PM Elijah Stone <[email protected]> wrote:
>
> The addition of DD for verbs is a welcome one.  I have a suggestion
> regarding the syntax, however.  The use of {{ }} to delimit a DD verb
> seems to mirror APL, but it overloads existing syntax in j.  The previous
> form, (. )., was better in this regard, but I have an alternate proposal.
>
> Use (: to open a DD, and plain ) to close it.  Advantages:
>
> - shorter
>
> - : matches the existing syntax for non-tacit verbs (3 : et al)
>
> - suggests a syntax for ambivalent verbs (:monadic form:dyadic form)
>   which also matches the existing non-tacit syntax
>
> - suggests a syntax for DD adverbs and conjunctions, perhaps with (. )
>
> (Added bonus: the leading : matches expression syntax in k9.)
>
> Thoughts?
> ----------------------------------------------------------------------
> For information about J forums see http://www.jsoftware.com/forums.htm
----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm

Reply via email to