Nick Papior

Hmm.. Ok, what about allowing 0 orbitals, and when creating `SparseGeometries` with geometries having 0 orbitals an error will be raised? I can't immediately see all implications, but at least...

I think there wouldn't be too much of a problem, I am just very worried that the 0-orbital atom will cause havoc in the SparseGeometry classes. The workflow for 0-orbital...

> What if such a geometry passed to `SparseGeometry` is internally filtered to move zero-orbital atoms into a separate placeholder? Then it could be simple flag if they should be...

Thanks for this. What exactly is it you want? :+1: I.e. I am a bit reluctant to do this since one can easily do: ```python aorb = AtomicOrbital(name) aorb.n aorb.l...

Additionally there are some problems with this logic, you will force a passed name to superseede explicit arguments: `"5px", n=2` will force `n=5`, while I would argue that explicit arguments...

I like this idea of `sel`, or at least a variant of it. One thing that really requires some careful thought is how it should be implemented. Here are some...

Jonas has also asked me about this named stuff which seems quite handy. However, I am not happy about its current usage, it is clumsy and not very easy to...

Thanks for the long code ;) I am not worried about generality, I am more worried about maintainability of what you have. One would have to *remember* this for all...

Yes, I plan on reworking the names at some point, I need to add an attribute to classes to enable simple handling of these things. But for now I would...

The branch #99 introduces the changes noted above. I would like some comments regarding the last thing mentioned above @tanocalogero @tfrederiksen @sleith-dk @sofiasanz @brandimarte @jonaslb @juijan @jmllorens Sorry for pinging...