[Data-modeling] Modeling numeric uncertainties

[Jeff Prucher]

There's an open task to modify the schemata for chemical elements to allow
for a data load, and one of the issues is that values like atomic mass are
almost always given with a degree of uncertainty. The question is how to
account for this in the model. The current suggestion is to use a CVT, with
two properties: one for the base value, and one for the degree of
uncertainty (+/-); both measured in the same unit. 

See the property "atomic weight (new style)" on sandbox:
https://www.sandbox-freebase.com/view/en/thorium

What do you think of this approach?

Thanks,
Jeff