> I can verify that that script runs under j807.
>
> I was disappointed with the lack of documentation,
>
> Still, it took me only a few seconds to crash my J session:
>
> ( A=. 3 3 $ [: (1&o.) (2&o.) ]sb )
> >A
From the wicked point of view, the behavior of the interpreters started to
deteriorate after j804, the Jx interpreter which I use is based on a
version of j806 which still is acceptable to me. I remember that I had to
modify the toolkit because j807 was too picky about certain things. I am
afraid you might have experienced a j807 issue; you might consider using a
j806 interpreter instead.
Unfortunately, I do not have an official j806 DLL handy; but, this is what
Jx does after running the toolkit,
( A=. 3 3 $ [: (1&o.) (2&o.) ]sb )
┌────┬────┬────┐
│1&o.│2&o.│1&o.│
├────┼────┼────┤
│2&o.│1&o.│2&o.│
├────┼────┼────┤
│1&o.│2&o.│1&o.│
└────┴────┴────┘
type 'A'
┌────┐
│noun│
└────┘
B=. >A
type'B'
┌────┐
│verb│
└────┘
$ o > A
3 3
>A
1&o.
b=. <@:[ { ]
(1 1 b >) A
1&o.
(2 1 b >) A
2&o.
((2 1 b >) A) 0 0.5 1
1 0.877583 0.540302
At any rate, I would advise staying away from unboxed arrays of verbs,
adverbs, and conjunctions at least until one has a pretty good idea of what
the expected results should be. In my experience, often one gets nonsense
when one asks a computer for nonsense and sometimes even a slap on the
wrist.
> For now, I will repeat a paragraph from
> http://jsoftware.com/pipermail/chat/2021-December/009020.html
>
> Boxed verbs come with costs: Documentation costs, implementation
> costs, maintenance costs (in the J implementation), debugging costs
> (in code which did not intend to use the feature but erroneously used
> it), opportunity costs (time that could have been spent on other
> things). To make a decent decision here one has to be aware of both
> the value and the costs of that decision (and someone has to step up
> to cover those costs).
>
> ....
I do not see much if any, disagreement there. What follows is an excerpt
from
[Jchat] Was [Jprogramming] Tacit completeness (jsoftware.com)
<http://jsoftware.com/pipermail/chat/2021-December/009020.html>
"
>* Accepting low value for high cost is only very rarely the right move.
*>Even if there was a desire to go that route (a huge if) you might be quite
right and probably it is already too late for J. BQN is heading that way,
as I said time will tell if the pioneer of tacit programming has been
left behind.
"
__________________________________________________________________________
On Thu, Dec 30, 2021 at 9:18 PM Raul Miller <[email protected]> wrote:
>
> I can verify that that script runs under j807.
>
> I was disappointed with the lack of documentation,
>
> Still, it took me only a few seconds to crash my J session:
>
> ( A=. 3 3 $ [: (1&o.) (2&o.) ]sb )
> >A
>
> ... which, perhaps, has something to do with why this glitch has been
> removed from more recent versions of the interpreter.
>
> For now, I will repeat a paragraph from
> http://jsoftware.com/pipermail/chat/2021-December/009020.html
>
> Boxed verbs come with costs: Documentation costs, implementation
> costs, maintenance costs (in the J implementation), debugging costs
> (in code which did not intend to use the feature but erroneously used
> it), opportunity costs (time that could have been spent on other
> things). To make a decent decision here one has to be aware of both
> the value and the costs of that decision (and someone has to step up
> to cover those costs).
>
> ....
>
> It might very well be that there's a useful approach here. But it
> looks like adequately specifying and documenting that approach would
> not be easy.
>
> Thanks,
>
> --
> Raul
>
> On Thu, Dec 30, 2021 at 7:09 PM Jose Mario Quintana
> <[email protected]> wrote:
> >
> > > Or possibly I made a mistake the last time I tried that exercise, as
> > > my copy of J crashed.
> > >
> > > Without a good specification, it's rather difficult to distinguish
> > > between machine problems and implementation problems.
> > >
> >
> > The attached J Wicked Toolkit.txt works for me...
> >
> > NB.
> >
----------------------------------------------------------------------------
> > NB. J Wicked Tacit Toolkit...
> > NB.
> >
----------------------------------------------------------------------------
> >
> > NB. DO NOT TRY TO LOAD THIS SCRIPT!!!
> > NB. Instead, run it using 0!:0, or similar, or paste
> > NB. it on an J editing window and use Crtl-A Crtl-E, or
> > NB. of course, run the script using a hot key
> >
> > NB. (0!:0)<'/.../J Wicked Toolkit.js_'
> >
> > NB. WARNING: This tookit relies unorthodox (theoretically illegal)
> > constructions
> >
> > ...
> >
> > mne=. (k e fetch o #) o ('()' ;: o -.~ ":) f.adv
> > assert (A B C D)mne -: box (0&({::))`(1&({::))`(2&({::))`(3&({::))
> > assert (A f. 0&{:: Match) , (B f. 1&{:: Match) , (C f. 2&{::
Match) ,
> > (D f. 3&{:: Match)
> >
> > erase e o <"0'ABCDEFGHIJKLMNOPQRSTUVWXYZ'
> > ┌─┬─┬─┬─┬─┬─┬─┬─┬─┬─┬─┬─┬─┬─┬─┬─┬─┬─┬─┬─┬─┬─┬─┬─┬─┬─┐
> > │1│1│1│1│1│1│1│1│1│1│1│1│1│1│1│1│1│1│1│1│1│1│1│1│1│1│
> > └─┴─┴─┴─┴─┴─┴─┴─┴─┴─┴─┴─┴─┴─┴─┴─┴─┴─┴─┴─┴─┴─┴─┴─┴─┴─┘
> >
> >
> > (9!:14)''
> > j807/j64nonavx/windows/release/commercial/
> > www.jsoftware.com/2018-10-05T10:48:34
> >
> > ( A=. 3 3 $ [: 1&o. (0) (0) (0) ]sb )
> > ┌────┬────┬────┐
> > │1&o.│0 │0 │
> > ├────┼────┼────┤
> > │0 │1&o.│0 │
> > ├────┼────┼────┤
> > │0 │0 │1&o.│
> > └────┴────┴────┘
> >
> > ;A
> > |domain error
> > | ;A
> >
> >
> > Again, I am just playing with the j807 DLL without a full installation.
I
> > hope it works for you.
> >
> > Good luck!
> >
> >
> >
> >
> >
> >
> > On Thu, Dec 30, 2021 at 4:08 PM Raul Miller <[email protected]>
wrote:
> > >
> > > On Thu, Dec 30, 2021 at 2:54 PM Jose Mario Quintana
> > > <[email protected]> wrote:
> > > > This is what Jx does which in most cases is what official j8xx
> > > > interpreters do. You could find out what the latter interpreters
> > illegally
> > > > do after running the wicked tacit toolkit holding your nose if
> > necessary.
> > >
> > > Possibly,
> > >
> > > Or possibly I made a mistake the last time I tried that exercise, as
> > > my copy of J crashed.
> > >
> > > Without a good specification, it's rather difficult to distinguish
> > > between machine problems and implementation problems.
> > >
> > > Thanks,
> > >
> > > --
> > > Raul
> > > ----------------------------------------------------------------------
> > > 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
----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm
