Ah, yes.

On Thu, Mar 24, 2016 at 6:28 PM, tim Rowledge <tim at rowledge.org> wrote:
>
>> On 24-03-2016, at 4:21 PM, Chris Muller <asqueaker at gmail.com> wrote:
>> [snip]
>> It’s a side-effect of a too-simple test for whether the morph has the cursor and we have exactly the same problem on the bottom-left corner. And indeed, with less obvious effects on the top two corners.
>>
>> Lovely picture illustrating the problem perfectly.  Would it be sufficient to move the grip morphs behind the morphs within the bounds of the window?
>>
>> Isn't the problem simply that its extent is too large?  It should simply match the width/height of the window border widths.
>>
> Nah; notice how the extent of the grip morph is (unsurprisingly) the whole square shown, overlapping considerably with the bottom part of the scrollbar. The top-left one almost completely overlaps the close button by the way but morph precedence combined with whatever makes it not a problem. What we want is for the effective region to be an L shape not a square. Or perhaps for it to be behind the scrollbar?
>
> If we shrunk it to the border width/height it would be only couple of pixels in size and we’d probably never be able to hit it. Especially on a high resolution display.
>
> tim
> --
> tim Rowledge; tim at rowledge.org; http://www.rowledge.org/tim
> Try not to let implementation details sneak into design documents.
>
>