Jesse,

> Jim, have you looked at my CursorsForResizing changeset?

Yes

> It provides
> resize cursors instead of visible yellow dots for both window resizing
> and subpane resizing, using the old mouseEnter/mouseLeave event logic.
> I have had it in mind that this old logic should be replaced with the
> simpler sort of logic you're using, with transparent resize handle
> border morphs, but haven't got around to it.  Perhaps we should look
> into folding this work together.
>

Sure. The subpane resizing should be a straight drop in.

You have to remember that the current example, is just that, an example.
It's tries to take some of the concepts that I was talking about, and make a
simple working example to play around with. In this simple case, it doesn't
really handle any of the cursor functions properly. I'll probably make
several versions that explore different avenues of implementation. In a more
complete implementation with the cursor, there are a couple of design
decisions you need to make.

The simplest way, which probably is not be completely correct, is to
squirrel away whatever cursor is current when the mouse enters the window
frame, and restore it when the cursor transition outside of the window
frame, outside of the window. In other words, if the cursor goes into the
window itself, it needs to generate a mouseEnter: for the window to handle,
if it leaves the window and the window frame all together, it restores the
saved cursor. The 'ideal' way is to have each morph have a cursor associated
with it and let everyone take care of the cursor themselves, but I don't
think this is a reasonable expectation at this point.

Jim