[Data-modeling] Professional sports teams

Dan Milbrath dmilbrath at metaweb.com
Sat May 24 00:08:49 UTC 2008


I'd probably vote for having two separate types -- though it seems we could live with it either way.
- danm


----- Original Message -----
From: "Jeff Prucher" <jeff at metaweb.com>
To: "Freebase data modeling mailing list" <data-modeling at freebase.com>
Sent: Friday, May 23, 2008 3:02:20 PM (GMT-0800) America/Los_Angeles
Subject: [Data-modeling] Professional sports teams

I recently added a type for school sports teams
(http://www.freebase.com/view/sports/school_sports_team) to the commons
sports domain. This type has "sports team" as an included type
(http://www.freebase.com/view/sports/sports_team), but some of the
properties on Sports Team are not applicable to non-professional sports
teams (either school-based or amateur). These properties are Owner(s) and
Draft Picks. My question is, should we move these two properties to a new
type called Professional Sports Team, or leave them where they are with the
understanding that they will not be applicable to many topics?  

As I see it, abstracting them to a new type makes the model a bit more
complicated, since there's one more type to add. But it also provides extra
semantic value -- by typing something Professional Sports Team, you would be
asserting something about the team, even if the two properties were blank
(similar to the way the type Deceased Person can assert that someone is
dead, even without knowing the date, place, or cause of death). 

I also don't know how important this distiction between professional and
non-professional teams is. It matters somewhat for the major sports in the
US, but it seems less important in association football, if I understand
things correctly.

What do you think?

Jeff Prucher
Type Librarian & Ontologist
Metaweb Technologies, Inc. 


_______________________________________________
Data-modeling mailing list
Data-modeling at freebase.com
http://lists.freebase.com/mailman/listinfo/data-modeling



More information about the Data-modeling mailing list