Given that the layeredimage already knows which attributes correspond to arm and torso movements, it would be nice to be able to give a with statement to those layeredimage groups (which if I understand correctly, really means being able to give one to all attributes in that group) that would kick in if an attribute change within that group is triggered as part of a show or say statement (only if no other transition is attached directly).
The level of flexibility this would add above and beyond what is already provided by config.say_attribute_transition would enable:
- the afore mentioned distinction between groups such as facial expressions and body parts
- expanded scope to impact not just say but also show calls
- characters to move at different speeds
(I love the idea that some characters could have faster or slower transitions based upon their age, fitness and/or personalities!)