Re: APL in Colorado, USA? Am I the last one?

On Apr 29, 1:28 pm, RHui...@[EMAIL PROTECTED]
 wrote:
> A solution to problem 2 in J.
>
> combsum=: 4 : 0 M.
>  if. (x>:y)+.0=x do. ,: 1 ,~ +/"1 i.(x<:y),x
>  else. (~.@[EMAIL PROTECTED]
 ,. +//.)/ |: (((x-1),0)+"1 x combsum&<: y),(x,0)+"1 x
> combsum y-1 end.
> )
>
>    6!:2 't=. 6 0 (+"1) 6 combsum 49'
> 0.0252536
>
> That is, 0.025 seconds.
>
>    $t
> 259 2
>    3{.t
> 21 1
> 22 1
> 23 2
>    _3{.t
> 277 2
> 278 1
> 279 1
>
> Details inhttp://www.jsoftware.com/jwiki/Essays/Combination_Sums
>
> On Apr 23, 10:51 pm, AAsk <AA2e...@[EMAIL PROTECTED]
> wrote:
>
> > It would be interesting to read a detailed account of what swayed the
> > Matlab fellow. Perhaps a book on APL and advanced mathematics is
> > needed.
>
> > A while ago,
>
> > 1. I was exploring the shortest path problem using APL: I wanted to
> > list all the paths. The shortest path is easily given by APL, see
> > Learning & Applying APL by B Legrand.  No APL text book delved into
> > this area ie. enumerate all the paths. Google highlighted several non-
> > APL solutions. Eventually I found an APL solution written by a US
> > academic (can't remember the URL).
>
> > 2. A colleague was doing an investifation. Take the UK national
> > lottery - it has 6!49 combinations where the first (1,2,3,4,5,6) and
> > the last (44,45,46,47,48,49) have each got a sum and the number of
> > combinations that have the same respective sum is exactly 1. The
> > objective was to list each sum between the 2 with the number of
> > combinations ... quickly. After a lot of experimentation, the solution
> > ran in less than 2 minutes on a 1.6 GHz Pentium processor.
>
> > I've heard that APL had something to do with the planning of the
> > (glass) roof of the British Library but not read about it.
>
> > More in print on the undeniable case for APL would be very welcome.
> > Would it not?

Wow.