Re: Common-lisp-stat & QuickLisp

[A.J. Rossini]

Ok folks, 2 things:

One, is that there is a tony2 branch, checked in, which eliminates
liblispstat, and some of the extraneous dependencies. Code looks
pretty raunchy, actually, and there is a good like of things to do,
and check, and fix. I'm focusing on data frames, mainly.

second, I'll start using the googlegroup that I set up, as a mailing
list/forum host, since I can't seem to find if github has such a
thing. It's Common Lisp Statistics, or lisp...@googlegroups.com

I'd prefer to use this if possible, when discussing development. And
I have :-).

If you don-t like that, please let me know.

On Wed, Oct 3, 2012 at 1:56 PM, A.J. Rossini <blind...@gmail.com> wrote:
> No, just visiting for the swiss school holidays. But perhaps the 3 of
> us can catch a bit about it while we are in the same time zone (though
> from here to switzerland isn't bad, better than from here to the usa
> west coast).
>
> Anyway, made a bit more progress this afternoon, and even more this
> evening I hope.
>
> Best,
> -tony
>
>
> --
> best,
> -tony
>
> blind...@gmail.com
> Muttenz, Switzerland.
> "Commit early,commit often, and commit in a repository from which we
> can easily roll-back your mistakes" (AJR, 4Jan05).
>
> Drink Coffee: Do stupid things faster with more energy!



--
best,
-tony

blind...@gmail.com
Muttenz, Switzerland.
"Commit early,commit often, and commit in a repository from which we
can easily roll-back your mistakes" (AJR, 4Jan05).

Drink Coffee: Do stupid things faster with more energy!

[A.J. Rossini]

---------- Forwarded message ----------
From: Steven Núñez
Date: Mon, Oct 8, 2012 at 2:48 PM
Subject: Re: Common-lisp-stat & QuickLisp



The gmail list is probably good. Github tools have never impressed me.


Quickly on the topic of graphics, I was looking for an XML <-> CLOS
solution for some PMML work I'm doing and found this:

http://code.google.com/p/xmlisp/

Looks like it might help with the graphics. Worth investigation anyway;
has a MS Windows version. Sadly it doesn't look like it does (in an
obvious way) XML/CLOS, like it did here:
http://www.agentsheets.com/lisp/XMLisp/

The author is quite active on the CCL list.

Regards,
- Steve

On 2012-10-08 9:44 , "A.J. Rossini" <blind...@gmail.com> wrote:

>Ok folks, 2 things:
>
>One, is that there is a tony2 branch, checked in, which eliminates
>liblispstat, and some of the extraneous dependencies. Code looks
>pretty raunchy, actually, and there is a good like of things to do,
>and check, and fix. I'm focusing on data frames, mainly.
>
>second, I'll start using the googlegroup that I set up, as a mailing
>list/forum host, since I can't seem to find if github has such a
>thing. It's Common Lisp Statistics, or lisp...@googlegroups.com
>
>I'd prefer to use this if possible, when discussing development. And
>I have :-).
>
>If you don-t like that, please let me know.
>

[A.J. Rossini]

It's not off topic. I spent some time with Hadley during his first
round of implementing ggplot (and discussing some fo the thesis issues
with an interactive GoG, which I'd like to see.

The idea is a good one -- would need to build a general infrastructure
similar to what Paul Murrell did for static graphics -- but perhaps
there is a general API approach that would work for a mix of
static/dynamic, interactive and fixed (one can have fixed dynamic
visuals, and static interactive visuals, the first being whether data
"moves" or is re-projected, the second being whether the human can
modify (directly or programmatically -- but anything directly doable
MUST be programmatically doable).

I should add a section on visualisation in the TODO.org list.

Peter, my recommendation would be to read the extensive, sometimes
silly, documentation in the code itself, in the sections you might be
interested in. FOr example, sticking the javascript stuff somewhere
in the TODO section of visualization would be good, since javascript
shares the XLispStat v1 prototyping object system, so design could be
jump-started from that. Also, I'd like to emulate the XLS dynamic
graphics capability, but just like Paul broke the mold when he did
grid graphics for R, I think that a next generation to that system
should and could be implemented, especially since we have minimal
backwards compatibility to worry about. XLS still runs, but it's
incredibly primitive, 22 years later...

Clone, read and edit, and when you have questions, stick them in the code.

Since eventually some pure statisticians will be reading this, I see
no reason to be comment-sparse. verbosity will be useful, soon
enough.



On Tue, Oct 9, 2012 at 5:00 PM, Peter Schmiedeskamp
<pe...@thoughtspot.net> wrote:
> It's exciting to see some new movement on the Common Lisp Statistics front.
>
> This may be off topic, but since you mentioned graphics, I thought I'd make
> mention of the Grammar of Graphics as a way of thinking about
> visualizations. In R, I appreciate the flexibility and consistency of
> ggplot2. I also see that notable lisper, Peter Seibel, has a working
> implementation of a grammar of graphics (in Javascript, not CL):
> https://github.com/gigamonkey/gg
>
> Is the readme here the right place to go to get started poking around?
> https://github.com/blindglobe/common-lisp-stat/tree/tony2