RE: [huml-comment] PC-33 -Section 4.4.6-race

["Bullard, Claude L (Len)" <clbullar@ingr.com>]

I've been pointing to context usage for terms. 
The terms exist, are in wide use, and  have 
issues with the situatedness of their use varying 
by individual, group, culture and so on.  

In a real sense, the categories of the primary should 
frame such terms, or simpler, enable the user to 
create a context of metainformation for the use of 
the term.  Race is a good example of a term where 
the original contexts (historical/temporal) are 
changing and becoming less relevant to current 
contexts.  Yet we cannot state categorically that 
the current contexts are more right, only that they 
are different.   We can say simply and supportably 
that we don't need race because it adds no value 
to the primary where the primary serves as a means 
to create contextualizing metadata.

Removing race from the primary is supportable by 
argument that as a category, it can be created in 
a secondary by reference to physical descriptors 
and other observables.   Now HumanML has a clear 
use case.  Given a codelist in a secondary for a 
term not inherited from the primary, the secondary 
wrapper element or the codelist members could have 
an attribute or element that references a metadata 
document whose elements and attributes are derived 
from HumanML and the contents of which create the 
necessary framework for using that codelist.  HumanML 
types are thus a means of contextualizing the sign.

What I am resisting is removing a term because it 
offends an individual or group.   That is a path that
once taken, becomes irresistable to those who 
would impose their own individual and potentially 
narrow contexts.  Rationale is needed that is 
reapplicable or we will become a political debating 
list.  To drop it, we don't need to warn or otherwise 
comment beyond it being a term that while in common 
use in descriptors of individual humans and groups 
in a database, does not have a universal application and  
is derivable by use of contextual signs now available 
in the primary.   It can make a good example of 
a way to apply HumanML.  In short, a goal of HumanML 
is ameliorate miscommunication.   Race is a term 
that can cause miscommunication.  It is easy 
to discover such candidates by observing that their 
use provokes strong emotions.   Such terms can be 
framed using HumanML-derived context metadata and 
this is of great value because this metadata provides 
information for making correct selections from, for 
example, codelists.

len

-----Original Message-----
From: m batsis [mailto:mbatsis@netsmart.gr]

cognite@zianet.com wrote:

>Being non-objective as well as prejudicial, this term thus merits being
>dropped from huml Physical Descriptor primaries and  relegated to possible
>user-developed secondaries, with a warning that its usage may entail
>problems of vagueness and poor communication.
>
I also fail to see how the benefits of this term are enough to shadow 
the negative elements described above.

Since this is supposed to be against misscommunication, we should be 
exploring more objective models that actually help in isolating the 
legacy terms one can abuse, at least when viable workarounds exist. I 
haven't been convinced that race can justify it's position in the 
primaries or  the possible loss in case of it's omission from the primaries.

I'm wondering about the opinion of lurkers in this and any issues around 
the primaries, i mean "hello people, this is the public comment period"...