RE: [huml-comment] Request for a motion on PC-33 -Section 4.4.6-r ace

[Rex Brooks <rexb@starbourne.com>]

Title: RE: [huml-comment] Request for a motion on PC-33 -Sect

Thanks, Len,

I understand this a little better now. I'm afraid I'm more concerned with a compelling reason to remove the attribute race from the attributeGroup phyicalDescriptors. I am satsified that we can get to the current usages through this set of properties method. As of now, for me, the argument that it is not objective is sufficient, and if were not a term used in current practices, it would be compelling in an of itself, but with current usages, however one views those usages, it doesn't rise to the level of compelling for me and I would be happier with a clearly compelling reason the completely trumps current usages, which can nevertheless still be accommodated. That does not mean I favor retaining it. I don't.

Ciao,

Rex

At 9:01 AM -0600 12/5/02, Bullard, Claude L (Len) wrote:

> I'm not sure "derived" is exactly
> right.   If in fact, one can't say it is a
> physical
> characteristic native to an instance, one can say it
> can be created as a sign
> whose prototypical properties (in the Sowa sense of a
> prototype as a set
> of properties that taken together, determine set
> membership) are derivable
> from HumanML primary types.   Using the
> example I provided earlier,
>  
> Race:
>  
> A = Asian
> B = Black
> I = Indian
> W = White
> U = Unknown
>  
> one sees a term followed by a codelist.  
> What HumanML could provide
> would be a set of properties for the term, race, which
> could then be used
> to make a selection from that set given an
> instance.   That set might
> include rules or might be a simple prototype such as
> slots with values for physical
> characteristics common to members of the set,
> historical origins, and so on.
> (In practice, this is hard to do, but I think that
> difficulty itself is valuable
> in focusing the community of interest's use of the
> term; that in fact,
> the real value of the term will
> decline.)
>  
> Note:  the same approach would apply to declaring
> prototypes for
> races of trolls, woodland, plains, mountain, or
> otherwise.   So this
> is not special pleading, but precisely how HumanML is
> intended
> to be used.
>  
> Thanks to Dennis for pointing this one out.  It
> provides a good
> example of how HumanML can be applied to a term of
> contention
> such that typical users of the term can clarify
> precisely how
> this term is defined below the level of the initial
> codelist.
>  
> len
>  
>  
> > -----Original Message-----
> > From: Rex Brooks [mailto:rexb@starbourne.com]
> >  I would prefer a stronger reason than that it simply
> > can be shown to be non-objective despite current usages, and it can be
> > derived from the existing PBS in a secondary schema, and it can be
> > imported or declared from other interoperable resources. Those are
> > sufficient, but not compelling reasons, whether reapplicable or
> > not.
> >

-- 

Rex Brooks
Starbourne Communications Design
1361-A Addison, Berkeley, CA 94702 *510-849-2309
http://www.starbourne.com * rexb@starbourne.com