>I'm not entirely clear on what you're doing here, but
you may want to
>consider using custom attributes for some of this - for
instance, the
>units that the class uses could be implemented as a type-
level
>attribute.

>Right. I would have attributes such as:
>
>[LengthUnit(LengthUnits.Metre)]
>[TimeUnit(TimeUnits.Second)]
>
>either at a type or method level. (I'm not comfortable
with the names
>above, as LengthUnits suggests it's a flag-based enum,

Ranier Dunno <ebuffer@hotmail.com> wrote:
> I am wondering if there is some construct to force a
> class to implement a set of static methods? Sort of like
> a "class interface" as opposed to the common "object
> interface"?

Ranier Dunno <ebuffer@hotmail.com> wrote:
> Ok, that's what I thought, thanks. The reason I was
> asking, is actually that I want to add some "reflection-
> like" properties to a set of objects I have. My objects
> have properties which have values in a given scientific
> unit, and I want to provide the ability to query a class
> about its properties and the scientific units of those
> properties. Since all objects of the same class will have
> identical properties and units, it made sense to make the
> methods static (plus I don't want larger objects).

Ranier Dunno <ebuffer@hotmail.com> wrote:
> OK, brilliant! I'll look into that - I am definitely
> looking for a more elegant way of doing things. In
> essence, I want to do this: say, I have a "Sprint" class
> representing short runs made by a single person on a
> track field. A "Sprint" will of course have some non-
> static properties, like the length of the sprint, name of
> the runner, time spent, average velocity and so forth.
> What I would like to do, is to add some meta-data for
> those properties. For instance, I would like the ability
> to specify that length is measured in meters, not in
> feet, and that velocity is meters/second, not mph.