arith.diff, arith.ratio

[Markus Demleitner]

Dear Semantics community,

Since we're puzzling about P and S UCDs a bit already, I'd like to
bring up two atoms the status of which I've never quite understood:
arith.diff and arith.ratio.

Suppose a table contains a difference between two RA measurements.
Going by the UCD spec's:

  While it is possible to build a UCD as a combination of several
  simple words, the primary word carries most of the meaning as to
  “what the quantity is”.

I would make this "arith.diff;pos.eq.ra", since this is a
*difference* (of RAs).  But given that arith.diff is S, I can't do
that, and it seems I'll have to go for pos.eq.ra;arith.diff.  That,
again, seems very wrong to me, as the column doesn't contain anything
like an RA.

A similar reasononing would apply to, say,
phys.temperature;arith.ratio.

So... does anyone remember why arith.diff and arith.ratio were
labeled S?  Am I perhaps using these atoms against their intended
meaning?

Thanks,

              Markus