Re: Additional issues to address for the constraints_experiments branch?

[Elijah Newren <newren gmail com>]

On 11/14/05, Elijah Newren <newren gmail com> wrote:
<snip>

> There are 3 intuitive ways to break this into rectangles (which you
> listed), none of which satisfy the property I listed before ("a window
> fits onscreen if and only if it fits in _one_ of these rectangles").

I realized I was being a bit sloppy with my wording here and may have
misled for anyone reading closely enough.  The clarified wording would
be "a window fits in the region if and only if it entirely fits in at
least one of the rectangles".  (The difference in wording is due to
the possibility of overlap with these minimal spanning sets; it is
possible to have a window that can fit into either of two or more
different spanning rectangles.  The key is that if the window doesn't
fit into any of the spanning rectangles, then it couldn't possibly fit
into the region).

- Elijah