[Bf-taskforce25] New tools mockups

[Jason van Gumster]

I'm just not sure that this is a workflow improvement. Sticky keys
introduce multiple meanings to a keypress that aren't easily hinted to
the user, obfuscating the experience. Also, Blender has generally
employed a minimal amount of the press-and-hold paradigm, which is
better for prolonged use.

I'm not completely opposed to the concept of Select->Act->Tweak, but it
does add an additional step to the flow because you have to confirm
after tweaking. And I certainly don't want to rely on holding down a
key and applying on key-up... especially if that's the more common
use-case. In the case of extrusion, the current moving mouse method is
faster (and more likely to be used) than the tap method you propose...
but the tap method has the quicker access. It seems backwards.

I just feel that there's a more elegant solution here. Parametric
editing is great, but the point here is that it's a *tweaking* step. By
their definition, tweaks are optional or after-the-fact and shouldn't be
a forced step in every operation.

Of course, I could clearly be overstepping my bounds here, and missing
some critical point. If so, please let me know.

  -Jason

William Reynish <william at reynish.com> wrote:

> One solution to this problem is to use sticky keys.
> 
> Let's use the extrude tool as an example. You'll be able to invoke  
> this tool in two ways: either by holding the e key for a period of  
> time, or by simply tapping the e keys quickly. These different ways
> of invoking the tool produces two different results.
> 
> Press and hold: extrudes out the selection and lets the user
> position the extrusion using the mouse. Letting go of the e key
> applies the tool, but the tool settings can still be tweaked
> afterwards. Tap: performs the extrude, but applies no translation.
> This can be numerically tweaked using the tool tweak settings, along
> with steps and other tool settings.
> 
> Selecting an tool from the menus would work like the tap scenario,
> so that mouse positions don't produce problems.
> 
> Same system could work for transforms, bevels, or whatever.