From: "Hal E. Fulton" <hal9000@...>
Date: 2003-05-05T05:07:59+09:00
Subject: Re: "Collage" of images -- more pychological randomness

----- Original Message -----
From: "ahoward" <ahoward@fsl.noaa.gov>
Newsgroups: comp.lang.ruby
To: "ruby-talk ML" <ruby-talk@ruby-lang.org>
Sent: Sunday, May 04, 2003 2:35 PM
Subject: Re: "Collage" of images -- more pychological randomness


> On Sun, 4 May 2003, Hal E. Fulton wrote:
>
> > Let image sizes be 4x4, 3x3, 2x2, and 1x1. (I assumed 64 pixels = 1
unit;
> > and yes, I hardcoded it, and yes, that's bad form). Let there be at
least
> > one 4x4; at least two 3x3 if they'll fit; at least 5 2x2 if they'll fit;
and
> > the rest 1x1.
>
>   "at least two 3x3 if they'll fit" seems to mean:
>
>     n_3x3 = 0 or n_3x3 >= 2
>
>   "at least 5 2x2 if they'll fit" seems to mean:
>
>     n_2x2 = 0 or n_2x2 >= 5
>
> is this correct?

That's probably what I said, but not what I meant.
My mistake. The words "at least" should be omitted
or replaced with "at most."

The code speaks more eloquently than I do: Aim for
two 3x3s, but fewer if they won't all fit.

Of course, that rule isn't sacred. It's tied to the
grid size I picked and it's related to the rules of
distribution that I mentioned in another email (a
smooth descending curve).

Hal