A) I'm leaving the basic location alone, but probably changing the name of
it, because it's not the reference location
B) I'm adding an element to the offset, or adding some new element at the
same level as the offset, that can have some fields that are in a PIDF
civic, specifically, CAtypes, or a new geo location that is as general as
the original (point or shape, WGS84 CRS, ...)
C) This refinement of the reference point for a civic is constrained to be
more specific -- that is the CAtypes have to be more specific then the last
one in the initial location
D) With a geo, the only way to make any use of this is to specify a location
with a large uncertainty in the original location, and then specify the same
location with a smaller uncertainty in the extension. We would probably
want to constrain the implementation to that. This means that it's Option
A for geo.
E) I imagine there is no reason to require the second location -- that is,
if you just have the original location, and an offset, it's fine.
Brian
On 4/16/10 10:10 AM, "Alissa Cooper" <acooper@cdt.org> wrote:
> There are still some details to be worked out, but I think we are very
> near to consensus on "Option C." I'd like to suggest that Brian in his
> editorial role try to work this approach into the current draft and
> share it back with the design team and then the list.
>
> Alissa
>
>
> On Apr 11, 2010, at 8:20 PM, Thomson, Martin wrote:
>
>> James Polk writes:
>>>> In Geodetic:
>>>> -- Normal: outline of the building
>>>
>>> we should discuss this - as I think this (point "outline of the
>>> building") is problematic. What if you are at Yankees stadium or
>>> Soldier's Field. Where is the point "outline of the building"? that
>>> still means Yankees stadium or Soldier's Field?
>>
>> That depends. It might be reasonable to use a shape that encloses
>> either the stadium or the enclosing grounds.
>>
>>> I think for Geo, we
>>> have to agree if the geo points to a property, and then there can be
>>> either:
>>>
>>> - a reference point within that property aside from the macro geo
>>> coodinates, or
>>> - us the exact geo as the reference point to then do an offset from.
>>>
>>> I know James and Martin will balk at this latter choice, saying why
>>> not just recalculate the external geo -- but I think that isn't so
>>> obvious, but it may be.
>>
>> Not so much as you might think. While I always favour the inclusion
>> of more information, and Richard's presentation is what I would
>> consider ideal, this usage isn't that offensive.
>>
>> A recipient that receives the information is not presented with more
>> precision than there is accuracy, since there is no claim made with
>> respect to accuracy.
>>
>> --Martin
>>
>>> James
>>>
>> _______________________________________________
>> Geopriv mailing list
>> Geopriv@ietf.org
>> https://www.ietf.org/mailman/listinfo/geopriv
>>
>
>
>
>
>
>
>
>
>
>
>
>
>
> _______________________________________________
> Geopriv mailing list
> Geopriv@ietf.org
> https://www.ietf.org/mailman/listinfo/geopriv
_______________________________________________
Geopriv mailing list
Geopriv@ietf.org
https://www.ietf.org/mailman/listinfo/geopriv
