[racket] Racket community exceed expectations.

From: Neil Toronto (neil.toronto at gmail.com)
Date: Tue Dec 11 19:02:44 EST 2012

On 12/11/2012 12:36 PM, daniel rupis wrote:
>   To balance my (a little trolling post), I must say that racket community is
> very nice. Thanks to all that respond to the post.

For my part, you're welcome. :)

>   I am interested in math, and I follow for example maxima, julia and R.
>
>   It should be wonderful if the new math library could be integrated in any way
> with the symbolical capabilities of maxima.  In a recent post there was a
> question or problem  about the  GPL license.  I don't know if that is a real
> obstacle, but I think that union is a power tool.  Perhaps some form of union
> between maxima and the library in typed racket would be wonderful.

It could be wonderful. I love using Maxima, and I find it to be stable, 
consistent, and very useful.

Licensing is a real obstacle, though.

I don't know whether Maxima can be used as a library. If it can, because 
it's GPL, we can't include Racket bindings to it in our decidedly LGPL 
standard library. If anybody does distribute Racket bindings, they must 
be GPL, and any project that uses those bindings must also be GPL.

One option is to talk to Maxima by squirting text through pipes, but 
that sort of thing has always been brittle. I wouldn't try it, and I 
don't think it would pass code review with the other Racket devs.

Since starting the math library, licensing has been a constant thorn in 
my side. The projects that I would love to just translate 
proven-to-be-stable numerical code from generally fall into four camps:

  1. Closed-source.

  2. GPL, as backlash against #1. The Gnu Scientific Library (GSL) is
     very vocally in this camp.

  3. GPL, because they wanted to reuse code from or link to #2 projects,
     usually the GSL. I think R is in this camp.

  4. Don't care about licensing, so terms are unclear.

I get the impression that licensing for mathematical code in general, 
not just numerical code, is messy and annoying.

So maybe this is an opportunity. If the only option for getting good 
math support with liberal licensing terms is Racket, we could see a lot 
more people using Racket. I've been trying to outdo R in particular in 
plotting, distributions, and statistical functions. We're not yet to 
where (require math plot) is an R substitute, but we should be close - 
perhaps with a few extra Planet2 packages - in 1-2 years.

I think we have the expertise around here to do computer algebra the 
right way. Any takers?

Neil ⊥


Posted on the users mailing list.