I was also thinking the hover area might be too small, …however:

I’ll correct that. I think it’s correct for triggers and curves. But cursors might be difficult to select.

Might there also be a Z-order issue? The reason I ask is because I noticed that while trying to get cursors in focus, my curves were often lighting up instead. Are the curves higher in the Z-order than the cursors?

Well, the algorithm is not as simple as I wished; it is affected by the way I store objects in memory (it’s not a simple List of objects, because it is very CPU-consumming with a lot of objects). So if two objects are at the same position, the priority is :

1. group name

2. object’s activity (inactive first then active)

3. object’s type (curves, triggers then cursors)

and finaly, random! Because I use a Hash-table, the order is not predictable.

(I made these observations while working with the quite dense-with-curves Metastasis score)

Yes, curve selection in Metastasis is horrible… I know 🙂