[
Lists Home |
Date Index |
Thread Index
]
"Jonathan Borden" <jonathan@openhealth.org> writes:
> Joe English wrote:
> >
> > What are the real-world use cases of minOccurs and maxOccurs?
> > In my experience, occurrence constraints specifying anything
> > other than zero, one, or many are almost always an indication
> > of a bad design decision somewhere in the system.
>
> Number of living parents,
> grandparents,
> Number of (one's) kidneys
> Digits on left hand
> Chambers in heart...
> Human chromosomes...
Jonathan, no-one is arguing that one-to-many relationships exist. What
Joe was asking about was if you ever want to specifically quantify the
"many", IOW when "*" or "+" is not precise enough.
Note that your first 5 examples don't even describe interchangeable
things. Presumably, if you want a schema to describe digits on one's
left hand (perhaps to describe scars on them, or fingerprints?), you
don't want to treat them as "hand (digit{0-5})", where the digits are
only differentiated by position. What if a person had 4 digits? Which
one is missing? Similarly, "parents (mother? father?)" might be a
better representation than "parents (person{0-2})". Parents are not,
as a rule, interchangeable. Neither are heart chambers, but I'm sure
you know more about that than I do.
So back to Joe's question - a real world case where a list of
indistinguishable items is bound by some specific and precise number
other than 0 or 1.
Ari.
|
