<div><p>Is there any value in allowing particles to be classified as both (say) "quark" and "fermion"?  Adding "fermion" and "boson" to the enumeration, and making the "group" property non-unique would work for that, if we want to include them in the model.</p>
<p>Re spin: interesting.  We could refactor the particle spin type if need be, perhaps by turning it into a CVT. It would be an ugly one, though -- it might need properties for "spin numerator", "spin denominator", as well as "parity", since those pesky spins are often in thirds, and even if not, a spin of "0.5" would probably seem pretty weird. (The other alternative would be to treat the "spin" value as a machine-readable string, which would display better but can't be operated on like an actual number.)</p></div>