Yeah, I think that many of us wouldn't mind to use a crashing beta, but I understand that releasing a beta depends on how much the code is broken right now, and besides at this early stage I think there's no need to rush. I'm crossing my fingers though...
About the automated process, yeah I imagined that much and, as I said, you could go either ways (easy automated / pro with tweakable values) but considering that if you want to work on a complete model for any kind of production you will probably have to retopologize it anyway, and in that case the UV unwrapping becomes easier to deal with (at least if you know what you're doing), I think that for now an automated solution could be enough.
After all, usually the real purpose in having UVs in the first place is to have textures usable with other painting programs like photoshop, but I think that in this case you just want to export the texture maps done in meshmixer and import them in any other rendering program so I wouldn't waste much time in tools for fine tuning of UV-islands for now.
But how will the texturing process work? Will meshmixer create automatically unwrapped UV's like UVmaster in zbrush (that is basically a pelting process where the UVs are open on a plane and relaxed using tresholds)? Will it accept photograps, maybe via planar projection (it could be the case looking at the buddha image)? Or we will just be able to paint like before but with the chance to bake the texture without the need to go to meshlab to do so?
Will meshmixer regenerate the UVs just when asked to? In that case it could be useful to remesh the model in order to have an even distribution of the UVs before painting.
On the other hand, that could lead to another problem: unless you can store informations from one format to the other (vertex/UVs), you could loose all the previous work because an automatic process of UVs regeneration could recreate UVs in a completely different disposition, same thing if I want to separate or stitch/boolean parts and I want/need to regenerate the UVs. You need to save the color informations so that you don't lose them after a topology tranformation of any kind (at least to a certain degree).
About vertex and UV painting I think that you have two solutions
- you could have both of them at the same time, like on two layers that you can merge when you feel like it, or
- you could just have one at a time, bake the informations from one method and discard them after those informations are transferred to the other, be it UVs or vertex.
In meshlab you need to import two objects (one as origin and the other with UVs as target); it would be great to have an automated process that does all that for you with a single click (unless you can find a method that doesn't involve the creation of a copy of the same object to transfer the informations from one to the other).
This is going to be fun, thanks and keep up the good work.