Hi all,

I've issued a pull request for an early pass at defining how to handle unstructured mesh data.


I would really, really appreciate any feedback.  If the YTEP as written is too opaque or poorly organized, feel free to leave suggestions.  If there are spots where I've completely missed something or sidestepped it without digging int, that would be extremely useful, too.  As I note in the alternative implementation area, I'm not sure I got the namespacing right, and I'm not sure that the way I've thought about the data and how individuals would want to interact with it is right, either, but I did try to think though how *I'd* want to interact with it.

Also, as I mentioned in the PR, I think this could be a gateway to improving support for particle/discrete data too.