[QUANTITY] Requirements and apology

[David Berry]

Ed,

> Astronomers rarely query for a single number.

Do you mean a single number, or a single quantity? I read Ray's
message as implying that a Quantity can contain multiple values,
not just a single value. [He didn't say this explicitly but it is
implicit in his use of plurals in statements such as "It is *not* a
requirement that the values that make up the quantity may have
different units."]


> Not only is it inefficient to create many small objects, but
> one immediately desires a new class that can keep track of what is
> different about each of the individual atomicQuantities.

Why is this? There are two ways in which you can keep track of
the differences between two of Ray's Quantities without needing
sub-classes; firstly by including a UCD (or some such thing) as a
component of the Quantity [this would be a way of implementing Ray's
item 10: "It should be possible to associate a quantity with the
physical phenomenon it measures. This association is not required to
appear in model instances."]; secondly, by the context in which the
Quantity is used - if a quantity was used as a component of a higher
level object, it's place within the higher level structure could
determine the content of the Quantity.




David