A plehtora of Quantities

[Patrick Dowler]

On Friday 14 May 2004 02:57, Guy Rixon wrote:
> This list-valued thing is what breaks Q for me.  I would like to see the
> Q's
> inside the list, not the list inside the Q. _That's_ intuitive, and it
> also
> means the listing structure can express the relationship between the
> Q's; the
> standard Q structures don't have to deal with relationship.

This was exactly the concept that Gerard and I expressed in the domain model, 
where we had only AtomicQuantity (single, indivisible value) and 
CompositeQuantity (structured value aka tuple). Logically one can create a 
list/set/array of Q and the collection expresses some aspects of the 
relationship between the members, but doesn't have to be a Q - it could be, 
but that is a separate issue maybe better left for design or even 
implementations. Clearly an implementation would be very heavyweight if it 
was actually done this way - putting a copy of a unit string in every Q in 
the list, for example - but **logically** the units applky to the members of 
the list, not to the list. Plenty of room for optimisations, but they do not 
belong in the design and certainly do not belong in the domain model.

Now, the reason we had included CompositeQuantity is that the object-property 
thing Ed referred to looks to us like:

Subject
	list of Property

Property
	Phenomenon name
	Quantity value

ie we explicitly pulled the name/ucd/whatever out of Q because property is 
easily understood and mostly used as a name-value pair. Some phenomenon 
implicitly correspond to non-atomic/compsite/structured values, e.g.

"this galaxy has has shape (phenomenon) of ellipse(a=2, b=1, e=0.5)"

We toyed with using structured data types like geometry inside AtomicQ, but 
the problem is that the various things that describe something like an 
ellipse do not have the same units, coord systems, reference points, etc.
so it is kind of ugly.

So, I completely agree with Guy that "Q is a list" is not as clean or 
intuitive as "list of Qs". 

-- 
Patrick Dowler
Tel/Tél: (250) 363-6914                  | fax/télécopieur: (250) 363-0045
Canadian Astronomy Data Centre   | Centre canadien de donnees astronomiques
National Research Council Canada | Conseil national de recherches Canada
Government of Canada                  | Gouvernement du Canada
5071 West Saanich Road               | 5071, chemin West Saanich
Victoria, BC                                  | Victoria (C.-B.)