RE: [humanmarkup-comment] Re: Base Schema-geolocator

[Rex Brooks <rexb@starbourne.com>]

There are now several specific XML vocbulries to which we will be 
referring in the final draft of our base schema. Under OASIS we have 
specifications for name and address (separate but interoperable). 
Under HR-XML.org there are any number of specifications relating to 
employment, evaluations, medical information, insurance benefit 
plans, employee stock options, etc. I do not propose to work on 
harmonizing the HR-XML vocabularies in the first draft, but will 
include that organization in the announcement that goes out when we 
submit it to OASIS-Wide Approval. During that RFC period we will 
discover many things.

Note: geolocator is not likely to be treated as a derivation, but as 
a standalone element.

Otherwise, what Len said.

Ciao,
Rex

At 12:48 PM -0500 8/19/02, Bullard, Claude L (Len) wrote:
>From: Ranjeeth Kumar Thunga [mailto:rkthunga@interposting.com]
>
>>There isn't much information on the original Schema draft, but based on
>>my assumption this element is referring to an absolute point referring
>>to a specific location of a human being on a plane (or, rather, on a
>>globe). 
>
>Yes.  Geolocators abstract positions in a reference coordinate system
>from the address name.  This is basic mapping and there are plenty
>of reference systems for that.
>
>>Obviously, being part of the Base Schema, we would leave it up
>>to a higher level to define the actual units and framework involved.
>>These could be harmonized with some of the other metadata initiatives I
>>would assume, as Rex mentioned last and as James provided for.  
>
>That is exactly so.  Reference systems can vary (say GPS, say Lat/Long)
>
>1)>What would be the formal distinction between this and address?
>
>You have it.  An address is a name more or less (104 Honeybrook Drive)
>vs a coordinate position.  One needs this because address naming
>systems vary by culture (work out the British Postal Addressing
>system sometime).  It gets more obscure when talking about areas.
>For example, most American states have counties, but Louisiana
>has parishes.  Boundary conditions are also difficult so sometimes,
>a combination of locator types is required.  For example, at an
>intersection knowing N, NW, S, SW etc. or knowing things such as
>a liquor store cannot be located within 500 yards of a school zone.
>
>>Geolocator I could see as an absolute frame of reference based on a
>>series of coordinates, and address based on a postal convention
>
>Yes.
>
>>Could 'address' be a higher level definition that is one level
>>abstracted from geolocator?
>
>Yes but mapped, not abstracted.  Keeping such maps alive is a
>very big problem for municipal, state systems, etc.  Names of
>things change.  Coordinates don't.
>
>>It is theoretically possible to abstract an
>>address from a specific geolocation I would assume.
>
>One can map it through a lookup table.
>
>>I don't know if any of the metadata standards proposed by James takes
>?into account virtual or other worlds, but I guess it doesn't really
>>matter at this stage as they would be defined by a higher level schema.
>>Still, something to note.
>
>Yes.  Simply think of these as lookup maps.   If a virtual world is
>used, say a VRML world, this is basic stuff because it uses a nested
>coordinate system to organize space, so it is just a name.
>
>>Also, I am assuming that the geolocator probably describes the 'center
>>point' of a human as well.  We are not referring to the geolocation of
>>the appendages of a human being--would that correspondingly be defined
>>in bodyLocation I am assuming. 
>
>Probably not.  A geolocator for that would make the map very difficult
>to maintain.  A human is a local coordinate system within a global
>coordinate system.   Until one puts them in the ultimate destination,
>a box in the ground or a vase on a shelf, or spread to the wind on
>the Ganges, their center is ever changing. (hmmm... a place to drop
>into philosophy but no....).
>
>>Finally, we may want to consider, or allow for, a particular trajectory
>>to be accounted for as well of a human being.  This would go hand in
>>hand with location.
>
>Sure, but that is either a function emitting a set of positions or a
>position list such as a VRML interpolator object.
>
>len
>
>----------------------------------------------------------------
>To subscribe or unsubscribe from this elist use the subscription
>manager: <http://lists.oasis-open.org/ob/adm.pl>


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