I don't see any real need for these changes.
When developing functions in a script, isn't it simpler to make
comments in comment lines?
For Jan-Pieter's example, I would write
NB. sum divideby count
avg=: +/ % #
For Elijah's multiline definitions, of course these can include
comments, either as full comment lines, or at the end of code lines.
The only example that makes sense to me is the statement separator in
a single line {{ }} definition, but here [ works as the statement
separator.
On Sat, Apr 1, 2023 at 7:48 AM Jan-Pieter Jacobs
<janpieter.jac...@gmail.com> wrote:
>
> I'd be in favour of both .. and ... .
> Regarding ... , I'd like to note that Matlab uses the same as line joiner
> (see
> https://www.mathworks.com/help/matlab/matlab_prog/continue-long-statements-on-multiple-lines.html),
> and also turns everything after that into a comment (which is very handy).
> This would let one also comment longer tacit verbs well, without any
> extraneous NB.'s . Trivial example would be:
> avg =: ...
> +/ ... sum
> % ... divided by
> # ... count
>
> Your first comment would be incompatible with the easy commenting (without
> using NB.).
>
> I don't like the .:. proposal, since it is more difficult to type different
> characters instead of the same; I'd rather have ::: (could imagine them
> looking like the holes for shoelaces, that can be used to tie together
> lines)...
>
> Jan-Pieter
>
>
>
> On Sat, 1 Apr 2023, 11:26 Elijah Stone, <elro...@elronnd.net> wrote:
>
> > Some time ago, Michal proposed that a line separator be added. I want to
> > rekindle that discussion. The proposal was that .. behave like a line
> > break
> > when placed on a single line, such that e.g. {{ a=. y+y .. a }} 2 would do
> > the
> > obvious thing. Short, distinctive, and to the point.
> >
> > I also want to propose a line _joiner_, analogous to \ in shell or c: ...,
> > placed at the beginning or end of a line should join it with the previous
> > or
> > next. Joining lines happens _after_ stripping comments, unlike the other
> > languages I cited; the goal is to enable large, multi-line definitions
> > with
> > commentary for intermediate terms, without the need for pointless
> > intermediate
> > definitions.
> >
> > It might be objectionable to use such similar symbols for separators and
> > joiners. But maybe it's not such a big deal. Two more ideas:
> >
> > 1. Could use the _same_ symbol for both, with its sense depending on where
> > it's placed in a line.
> >
> > 2. Separator could be .:.; metaphor: a hill stops the interpreter in its
> > tracks.
> >
> > I don't like 2 because the larger the separator is, the more annoying it
> > is to
> > use. Joiner can afford to be large, since it only comes into play if the
> > rest
> > of the line is sufficiently large.
> >
> > Anyway--thoughts? Comments? Suggestions?
> >
> > -E
> > ----------------------------------------------------------------------
> > For information about J forums see http://www.jsoftware.com/forums.htm
> >
> ----------------------------------------------------------------------
> For information about J forums see http://www.jsoftware.com/forums.htm
----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm
