3(***(***)(***))4
35831808
3((***))4
144
3(***(***)144"_)4
35831808
3(*)4
12
3(**12"_(***)144"_)4
35831808
12(***)144
2985984
3(**2985984"_)4
35831808
3(12"_*2985984"_)4
35831808
3((***)(***)***)4
429981696
3(144"_(***)144"_)4
429981696
/Erling
On 2014-07-20 02:49, Linda Alvord wrote:
I'm still looking for the phrase or concept that explains the difference
between these two. Why are they producing different results?
3(***(***)(***))4
35831808
3((***)(***)***)4
429981696
If you grew up in the early years using APL, an idea might come to mind.
Linda
-----Original Message-----
From: [email protected]
[mailto:[email protected]] On Behalf Of 'Pascal
Jasmin' via Programming
Sent: Saturday, July 19, 2014 8:34 PM
To: [email protected]
Subject: Re: [Jprogramming] High Speed Train Challenge
3 * utu utu 4
3 * 1 : '[: u~ u' 1 : '[: u~ u' 4
429981696
----- Original Message -----
From: Kip Murray <[email protected]>
To: "[email protected]" <[email protected]>
Cc:
Sent: Saturday, July 19, 2014 6:55:02 PM
Subject: Re: [Jprogramming] High Speed Train Challenge
utu =: 1 : '[: u~ u'
3 *:@* utu 4
429981696
uses the ideas that *** is equivalent to *:@* and u u u is equivalent to [:
u~ u
On Saturday, July 19, 2014, 'Pascal Jasmin' via Programming <
[email protected]> wrote:
I'm not sure I'm arguing for the idea. I've disliked ~ in the past
because I've read it as one of the other 2 meanings that was written. I
wonder if making reflexive more common would help overcome forgetting it
exists quicker.
from your examples, it would appear that good candidates for monadic + and
* would be +~ and *~. Monadic * and + could have been chosen with +: and
*: symbols. We can (fortunately) implement such bivalence ourselves:
area =: *~
area 8
6 area 8
----- Original Message -----
From: robert therriault <[email protected] <javascript:;>>
To: [email protected] <javascript:;>
Cc:
Sent: Saturday, July 19, 2014 6:16:58 PM
Subject: Re: [Jprogramming] High Speed Train Challenge
Hi Pascal,
Not arguing against the idea but they are only functionally the same for
monadic.
4 *: 4
|domain error
| 4 *:4
4 +: 4
|domain error
| 4 +:4
5 *: 4
|domain error
| 5 *:4
5 *~ 4
20
5 +: 4
|domain error
| 5 +:4
5 +~ 4
9
Cheers, bob
On Jul 19, 2014, at 2:59 PM, 'Pascal Jasmin' via Programming <
[email protected] <javascript:;>> wrote:
kindof the same as your idea
*~(^:3) 12
429981696
completely off topic, but would it be a good or bad thing if, assuming
there was a shortage of ascii mnemonics, and some need, if monadic +: and
*: were redefined considering that +~ and *~ do the same?
----- Original Message -----
From: Erling Hellenäs <[email protected] <javascript:;>>
To: [email protected] <javascript:;>
Cc:
Sent: Saturday, July 19, 2014 3:05:03 PM
Subject: Re: [Jprogramming] High Speed Train Challenge
Another way to do the same thing, but not a solution, is this
expression:
3 (*(*(***)*)*) 4 NB. Funny way
429981696
Anyone can find a nice recursive way to write it? My best shot:
12 1:`([ * [ $: [: <: ])@.([: * ]) 8 NB. Complicated way
429981696
It's a recursion? * $: *
/Erling
On 2014-07-19 20:48, Raul Miller wrote:
Probably, yes.
And I was sort of provocative by not going with the implied
limitations.
But there's can be quite a bit of ambiguity when key issues are
implied, rather than addressed or illustrated.
This is a problem I face myself, quite often: How can I be aware of
important issues which matter to other people, when I am incredibly
focused on my own point of view?
That said:
(1) Erling Hellenäs had already posted some solutions which satisfied
the "one verb" constraint using * as that verb (at the time I made my
42981696"_ post).
(2) Realizing that derived verbs are J verbs is an important lesson
which beginning J programmers often overlook.
You can't really be a good J progammer if you don't understand the
grammar of the language. And it's not that the grammar is hard to
understand - it's extremely simple. But it's so simple that it's also
easy to sometimes get by with false generalizations about its rules.
This leads into the almost inevitable "no that's not what I meant"
sorts of social issues.
So yes, my post was - in a sense - somewhat bratty. But I felt that
the underlying issue was important enough to raise the point and stick
with it at least until someone called me on it.
Thanks,
----------------------------------------------------------------------
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
----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm
