[Jprogramming] jhs example

2014-03-07 Thread Raul Miller
I want to set up a jhs server on localhost and a much more limited J http server listening on some other ip. Does anyone have an example of how to do this? Or do I need to "roll my own"? Thanks, -- Raul -- For information about

Re: [Jprogramming] jhs example

2014-03-07 Thread Joe Bogner
I've used nginx to reverse proxy specific URLs to the JHS server -- effectively acting as a more limited http server. nginx supports caching and logging and all the things you'd expect in a production grade web server. That's why I opted for it over a native J web proxy. I'd be interested in a J pr

Re: [Jprogramming] a good bar bet!

2014-03-07 Thread Jim Russell
The Aho of AWK fame I presume? > On Mar 6, 2014, at 8:10 PM, Roger Hui wrote: > > Probably not. It's not related to the problem of finding the minimum that > I know of. It's the general problem of how you avoid initializing a large > but sparse array. Recently, I did use the idea in a case of

Re: [Jprogramming] J in 5 minutes

2014-03-07 Thread Martin Saurer
Hello all Jers, My two cents (or 15 minutes) to show what J can do. Feedback is welcome. https://www.youtube.com/watch?v=VSJpJt3c11c -- For information about J forum

Re: [Jprogramming] J in 5 minutes

2014-03-07 Thread Tom Arneson
Very nice, Martin! -Original Message- From: [email protected] [mailto:[email protected]] On Behalf Of Martin Saurer Sent: Friday, March 07, 2014 07:12 To: [email protected] Subject: Re: [Jprogramming] J in 5 minutes Hello all Jers,

[Jprogramming] `: 6 binding to verb

2014-03-07 Thread Pascal Jasmin
+/`'' ( 4 : 'x ` y') %`# ┌───┬─┬─┐ │┌─┬───┐│%│#│ ││/│┌─┐││ │ │ ││ ││+│││ │ │ ││ │└─┘││ │ │ │└─┴───┘│ │ │ └───┴─┴─┘ I would like to be able to define a single function (verb) that produces    (+/`'' ( 4 : 'x ` y') %`#)`:6 +/ % # is that possible? my failed attempt: g2v =: 1 : ('( u y

Re: [Jprogramming] J in 5 minutes

2014-03-07 Thread Ni Bo
Impressive. At 0:22 seconds a small typing mistake: "J is a high-performance.." not "an high". Nick On Fri, Mar 7, 2014 at 5:00 PM, Tom Arneson wrote: > Very nice, Martin! > > -Original Message- > From: [email protected] > [mailto:[email protected]

Re: [Jprogramming] a good bar bet!

2014-03-07 Thread Roger Hui
Yes, it's that Aho. The other two authors are also well-known in the field. For example, Hopcroft won the Turing Award in 1986. On Fri, Mar 7, 2014 at 4:56 AM, Jim Russell wrote: > The Aho of AWK fame I presume? > > > On Mar 6, 2014, at 8:10 PM, Roger Hui wrote: > > > > Probably not. It's n

Re: [Jprogramming] J in 5 minutes

2014-03-07 Thread Yike Lu
Nick, that's not always a typo ;). In many languages, the "h" is silent. On Fri, Mar 7, 2014 at 9:18 AM, Ni Bo wrote: > Impressive. At 0:22 seconds a small typing mistake: "J is a > high-performance.." not "an high". > > Nick > > > On Fri, Mar 7, 2014 at 5:00 PM, Tom Arneson wrote: > > > Very

Re: [Jprogramming] a good bar bet!

2014-03-07 Thread Roger Hui
No, it's not fair. - You need to sort 1-byte characters to approximate operations on 1-byte integers. - You need to ditch the conversions to and from booleans. In J such conversions are relatively costly (compared to /:~, f'instance); in C, the conversion is free. On Thu, Mar 6, 2014 at 11:47 P

Re: [Jprogramming] jhs example

2014-03-07 Thread Raul Miller
I built some example J http servers before JHS - I was modelling the concepts and I wound up using them for user authentication support for usatoday.com, back before most of that got switched to facebook. They were pretty simple and you can find some of them up on the J wiki. But I have not decide

Re: [Jprogramming] J in 5 minutes

2014-03-07 Thread Joe Bogner
Very impressive. I was watching the clock in the upper right corner and it seemed like it all happened in real time. If that's the case, your typing is impeccable! Regardless, great demo. The collatz conjecture was very interesting. I had never seen qjide in action so I'll have to experiment with

Re: [Jprogramming] J in 5 minutes

2014-03-07 Thread Brian Schott
I agree with Joe, that is a fine demo of J. It is long enough that you might want to include a table of contents in the details. For example, this was how I did it with another video. Introduction: http://www.youtube.com/watch?v=wXjlDUGDXDw Install/Launch JHS on host computer: http://www.youtube.c

Re: [Jprogramming] J in 5 minutes

2014-03-07 Thread robert therriault
Well done Martin, Really high production values. I do think that you could afford to enlarge the workspace window to allow the text to appear larger as the background is not so necessary. You did a great job of setting up your context and letting the viewer know what you were going to cover.

Re: [Jprogramming] J in 5 minutes

2014-03-07 Thread Stefano Lanzavecchia
> -Original Message- > From: [email protected] [mailto:programming- > [email protected]] On Behalf Of Martin Saurer > Sent: Friday, March 7, 2014 2:12 PM > To: [email protected] > Subject: Re: [Jprogramming] J in 5 minutes > > Hello all Jers, > My

Re: [Jprogramming] J in 5 minutes

2014-03-07 Thread Vijay Lulla
Impressive, Martin! Like Joe, I'm impressed with your impeccable typing too. On Fri, Mar 7, 2014 at 11:43 AM, Stefano Lanzavecchia wrote: > > -Original Message- > > From: [email protected] [mailto:programming- > > [email protected]] On Behalf Of Martin

Re: [Jprogramming] `: 6 binding to verb

2014-03-07 Thread Jose Mario Quintana
Orthodox verbs, explicit verbs in particular, can only take nouns and produce nouns; in contrast, tacit wicked verbs can take words and produce words of any kind (use them at your own risk). For example, 9!:14'' j701/2011-01-10/11:25 o=. @: ar=. 5!:1@< Cloak=. (0:`)(,^:) Cloak=. (

Re: [Jprogramming] J in 5 minutes

2014-03-07 Thread Jim Russell
Very nice. Thanks! > On Mar 7, 2014, at 8:11 AM, "Martin Saurer" wrote: > > Hello all Jers, > > > > My two cents (or 15 minutes) to show what J can do. > > > > Feedback is welcome. > > > > > https://www.youtube.com/watch?v=VSJpJt3c11c > >

Re: [Jprogramming] J in 5 minutes

2014-03-07 Thread Michael Dykman
very nice. I hope I did not exceed the intent by reporting on twitter. On Fri, Mar 7, 2014 at 12:41 PM, Jim Russell wrote: > Very nice. Thanks! > >> On Mar 7, 2014, at 8:11 AM, "Martin Saurer" wrote: >> >> Hello all Jers, >> >> >> >> My two cents (or 15 minutes) to show what J can do. >> >> >>

Re: [Jprogramming] `: 6 binding to verb

2014-03-07 Thread Pascal Jasmin
Thank you Jose, I am liking Cloak very much, but was having difficulty applying it to `:6 . Here is what I was after, more generically (applying to verb that outputs gerunds) :    +/`'' (evoke&6 @:( 4 : 'x ` y') ) %`# +/ % # ( +/`'' (evoke&6 @:( 4 : 'x ` y') ) %`#) 1 2 3 2 - Original Me

Re: [Jprogramming] J in 5 minutes

2014-03-07 Thread Roger Hui
V. nice. I note that the video basically does not make use of sound in the sense the if you mute it, the information content is the same. The video https://www.youtube.com/watch?v=DmT80OseAGs (Sudoku in APL, by John Scholes) has a different approach wrt sound. From what I understand the narratio

Re: [Jprogramming] `: 6 binding to verb

2014-03-07 Thread Linda Alvord
f=: 13 :'x (+/%#) y' f +/ % # 4 f 3 2.3 2.3 2.3 2.3 f 4 4 Linda -Original Message- From: [email protected] [mailto:[email protected]] On Behalf Of Jose Mario Quintana Sent: Friday, March 07, 2014 12:27 PM To: Pro

Re: [Jprogramming] a good bar bet!

2014-03-07 Thread Peter B. Kessler
On 03/06/14 17:10, Roger Hui wrote: Probably not. It's not related to the problem of finding the minimum that I know of. It's the general problem of how you avoid initializing a large but sparse array. Recently, I did use the idea in a case of x i. y (you know, the problem I've been working on

Re: [Jprogramming] a good bar bet!

2014-03-07 Thread Roger Hui
Thanks for finding the typo. You are a friend if you have that book on your bookshelf. :-) Version 0.3 of an implementation in C: #define ASSERT(p,stmt) if(!(p)){stmt;} #define SPARSE_VALUE0 T sparse[M];// sparse array of values I4 sptr[M]; // sparse array of poin

Re: [Jprogramming] J in 5 minutes

2014-03-07 Thread Bo Jacoby
Yes, impressive. But as usual the part that I could not follow just told me that even if the author can do it easily, I can't. I have been a newbee on J for some years, but a lot of it I still don't master. I still use J602 because I didn't manage to make the later versions run. I love short

Re: [Jprogramming] J in 5 minutes

2014-03-07 Thread Murray Eisenberg
A comment on Roger's comment as well as some comments on other aspects of the video. (1) In fact, I find the sound background not just superfluous, but distracting. (Yes, I know that's why we have a mute function on our computers' sound controls.) (2) The background globe graphic is a bit dist

Re: [Jprogramming] J in 5 minutes

2014-03-07 Thread Roger Hui
(1) There is perhaps a generational difference. Young people might like some background music? (6) 10&#.^:_1 x converts a number to its decimal digits. Alternatively, "."0":x would be shorter than what's on the video. But we quibble. 10&#.^:_1 ]!100x 9 3 3 2 6 2 1 5 4 4 3 9 4 4 1 5 2 6 8 1

Re: [Jprogramming] J in 5 minutes

2014-03-07 Thread Murray Eisenberg
Thanks, Roger; I suspected Base (#.) was involved; I just didn't think of taking its inverse. Interestingly, although the solution using Base seems closer to the content of the problem -- keeping things in the realm of integers instead of going back and forth to and from strings -- on my system

Re: [Jprogramming] J in 5 minutes

2014-03-07 Thread John Baker
Very nice. If you've written even a simple macro you will be able to relate to this presentation. You may have a hit. On Fri, Mar 7, 2014 at 4:15 PM, Roger Hui wrote: > (1) There is perhaps a generational difference. Young people might like > some background music? > > (6) 10&#.^:_1 x converts

Re: [Jprogramming] J in 5 minutes

2014-03-07 Thread Roger Hui
Well, sometimes you'd use #: (aka encode) but you'd have to know the number of 10s to put in the left argument: n=: !100x (10&#.^:_1 n) -: (10$~1+<.10^.n) #: n 1 If you're going to resort to benchmarks, let me bring out http://www.jsoftware.com/help/release/digits10.htm , which introduced s

Re: [Jprogramming] `: 6 binding to verb

2014-03-07 Thread Jose Mario Quintana
I wrote: "Orthodox verbs, explicit verbs in particular, can only take nouns and produce nouns; in contrast, tacit wicked verbs can take words and " Actually, explicit verbs (even if they should not) can take any kind of words as arguments when the sentences in the verb's body are syntactically co

Re: [Jprogramming] J in 5 minutes

2014-03-07 Thread PMA
Maybe Stockhausen's _Klavierstuck IX_? It's based on the Fibonacci series. Roger Hui wrote: (1) There is perhaps a generational difference. Young people might like some background music? -- For information about J forums see

Re: [Jprogramming] J in 5 minutes

2014-03-07 Thread Raul Miller
Distractions are probably good, to a certain extent, in a promotional video. Not everyone that watches the video is going to want to take action immediately and start solving euler or investing in stocks or whatever else. Ideally, a promotional video should be fun enough that it can find its way th

Re: [Jprogramming] J in 5 minutes

2014-03-07 Thread Yike Lu
A few comments: I think that having music is better than no music. But I also feel that the music chosen is uhh... too upbeat. Usually music for coding videos is more ambient. Generally, I think that promotional materials need to answer the question "what's the killer app?" Hell, I don't even know

Re: [Jprogramming] J in 5 minutes

2014-03-07 Thread Raul Miller
I can think of worse problems to have, than being "too upbeat". ;) Not having killer apps might be an example of a worse problem to have... :/ So, switching back to "too upbeat" mode... let's start writing some killer apps? If we write enough of them, some of them might catch on? (But, of course,