[Data-modeling] State of the commons (A-B)

[Kirrily Robert]

On Jun 30, 2009, at 9:53 AM, Robert Cook wrote:

> It would be very helpful if people could list the domains they think
> need serious work and what they'd like to be done and we'll work
> together to fix them.  I think this mailing list is a good place to
> air pent up concerns about domains and if it becomes too detailed,
> we'll move it to a different venue.


Here are my current thoughts, alphabetically by commons domain:

American Football: schema is fairly solid, I think.  However, there  
are no very active admins who actually know football.

Anime/Manga: no active admins, no schema development in years, and  
duplicates information already found in the TV, Film, and Comics  
Commons.  If we'd had bases when this was created, it probably should  
have been built as a base.  I don't believe anyone really uses this  
much, and I'm very tempted to suggest we delete it and refactor all  
the data into other commons.

Architecture: Brendan, Iain, and Spencer are working on integrating  
stuff from structure2 into Architecture, which is great.  We also need  
to move out the "Museum" type and model it properly.  I started  
working on that in http://mladraft.freebase.com/

Astronomy: fairly good, has active community, I think it's doing OK.   
However, there are ongoing issues of "Location" as applied to extra- 
terrestrial locations.

Aviation: Could use some work in splitting "Aircraft" into "Aircraft  
class" and "Aircraft instance" as we already have in eg. boats and  
rail.  Some discussion around aircraft accidents / disasters that  
never got resolved.  Basically there are no active admins with expert  
knowledge in this field, and there's bit rot afoot.

Awards: pretty solid.

Baseball: People are requesting properties, but there are no active/ 
responsive admins with a knowledge of the field.  Needs work.

Basketball: Very little schema here, very little activity, looks kinda  
moribund.  Many of these sports domains exist only because "position"  
varies between sports.  Have we learnt anything about modeling in the  
last year or so to know a better way to describe these things, so we  
don't need a different domain for every sport?

Bicycles: 3 types, very skeletonish.  Needs data import work to make  
it useful for base-builders (which is why it was originally made a  
Commons, because there were so many bike-related bases), then some  
build-out for bicycle specs and the like, which will vary considerably  
between eg. MTB and racing.

Biology: so-so, quality-wise.  I think what we really need is some  
data gardening work on organism classifications, not schema work per se.

Boats: needs major reworking.  Draft in progress at http://www.freebase.com/view/user/skud/boats

Broadcast: Needs work.  A lot of the schema design here, especially  
around Broadcast Content and Broadcast Artist, was done to support a  
particular application, but it's hard to understand outside of that  
context.  Luckily, the major types are fairly self-explanatory, eg.  
"Radio station" and "TV station".  But some are kind of bewildering.

K.

-- 
Kirrily Robert
Freebase Community Director
kirrily at metaweb.com
http://freebase.com/