[Data-modeling] Rivalry ~ Just a base of views or type(s)?

[Robert Cook]

I took a stab at a generic rivalry type:
   http://rivalries.freebase.com/view/base/rivalries/rivalry

Instead of a CVT-mediator, I made it a full topic just to see if that  
feels better.  Indeed for many sports rivalries there seem to already  
be topics.

The alternative would be specific versions of this schema repeated,  
say, for people, companies and sports teams.

But this is as good as place as any to start collecting real data.

R


On May 20, 2009, at 3:10 PM, Gordon Mackenzie wrote:

> I was looking at a entry about the high school I attended when on e of
> the opening sentences sparked a thought; rivalry. It was noted that
> the Petaluma High School's long-time rival was the Casa Grande High
> School... Brought back memories.
>
> All sports seem to promote rivalry... Baseball, Tennis, Football, etc.
> All feature dramatic historical rivalry between teams and individuals.
>
> Then of course, as a type it could be very generic to be used across
> many domains, a CVT probably. It might include date range for start
> and end of rivalry (if such things ever truly end). Maybe a property
> to capture notable events?
>
> There are all sorts of rivals in life: Business, the Arts,
> Exploration, Military Branches, Countries.
>
> So my thought is to create a single type that has a property for type
> of rivalry (commerce, sport, artistic, national, educational,
> cultural, ideological, etc, or is it better to have separate types for
> the various kinds rivalries?
>
> =====
>
> Note: I see that Robert created a celebrity rival property:
>
> http://www.sandbox-freebase.com/view/celebrities/rivalry
>
> Here are topics about it:
>
> http://www.sandbox-freebase.com/search?limit=30&start=0&query=rivalry
>
> Note the topics about specific notable rivalries
>
> ~ Gordon
>
> <<< gordon at metaweb.com >>>
>
>
>
> _______________________________________________
> Data-modeling mailing list
> Data-modeling at freebase.com
> http://lists.freebase.com/mailman/listinfo/data-modeling