[QUANTITY] Requirements and apology

[Ray Plante]

> >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."]
> >
> I read Ray's values as only allowing a vector (as in the math sense of 
> an arrow that requires multiple components to describe it).  Hopefully, 
> Ray will clarify.

Hmm, I thought I had been more explicit about this, but now when I look 
back I see that I was indeed a bit vague on this point.  

My quantity definition implies a requirement:

>   A quantity is a measurable, countable, or comparable property of 
>   a physical phenomenon that can be represented as a set of numeric 
>   values.  

that is, a quantity can include multiple values.  This was to include 
vectors in simple cartesian space (e.g. magnetic field).  It could also 
handle other simple sets like ranges (although some perhaps "out-of-scope"
property would be necessary to indicate this).  Multi-dimensional arrays 
could possibly be included, too (e.g. tensors), but I can't think of a 
vital example.  For many cases of 1+D arrays, incorporating support 
directly in Quantity is not necessary, because of the following:

>  9.  It should be possible to use a quantity instance to refer to a
>      measurement of a physical phenomenon in the abstract without
>      actual values (i.e. values are optional).  This allows one to
>      use the instance to request specific quantity data ("give me
>      fluxes in this unit") or describe data stored outside the
>      model instance (e.g. in a table column, database, or image).  

What I'm saying here is that when dealing with a lot of quantities of the 
same type, it is possible to tag values "UCD-like" that are stored outside 
the Quantity model instance.  If your components are fairly separable, you 
can imagine reusing them to more efficiently describe, say, a 2-d array 
(image) of values with a common unit and phenomenon tag, but with varying 
error.  

BTW, Pat caught me with this AtomicQuantity model.  On first pass, this 
looks to fit my requirements pretty closely.  I don't think I had seen the 
UML diagram prior to my write-up (perhaps I had absorbed it subconciously 
from his and Gerard's ADASS poster), so for me it was a good confidence 
boost in the process.

>     We can adjust our diagrams in a few minutes and establish such a 
>model.  We can now claim that this quantity effort is completed and we 
>can move on.
>    Now, that it is done, it needs to be extended.

Not quite.  If we want to be serious about this, we need to put this 
through the full process.  That means,

  1.  A prose document describing the model.  It should include an 
      enumeration of all the concepts used in the model (e.g. unit, 
      error, atomic quantity) with a definition (a la the RM document), 
      along with the relationships between them.  The aim is to try to be 
      fairly free of a specific modeling framework (e.g. UML, XML schema).
      (I bet this could be made fairly short.)

  2.  A UML diagram.  The one I saw is most of the way there; however, I 
      think it needs more details (e.g. cardinality, any specific forms of 
      the classes, ...).

  3.  An XML Schema realization of the model.

cheers,
Ray