[postgis-users] Computing a centroid on GEOGRAPHY objects

[Paul Ramsey]

It won't work very *well*. It'll work.

On Thu, Dec 6, 2012 at 2:15 PM, Ed Linde <edolinde at gmail.com> wrote:
> So does this mean that if the set of points span across 2 UTM zones, then
> its not possible to compute their centroid location?
> The ST_Buffer does say that it won't work if the points are across multiple
> UTM zones.
>
>
> On Thu, Dec 6, 2012 at 11:05 PM, Paul Ramsey <pramsey at opengeo.org> wrote:
>>
>> If the points of the features are truly spread out across the world,
>> the centroid is pretty meaningless. If the points of any given feature
>> are relatively compact, then using an approach with _ST_BestSRID would
>> work (see the def'n of ST_Buffer(Geography) for an example).
>>
>> P.
>>
>> On Thu, Dec 6, 2012 at 1:55 PM, Sandro Santilli <strk at keybit.net> wrote:
>> > On Thu, Dec 06, 2012 at 09:34:53PM +0100, Ed Linde wrote:
>> >
>> >> I know there is a centroid computation function in Postgis, but it
>> >> seems to
>> >> only work on a geometry type.
>> >> The lat/long coordinates are spread all across the globe and hence I
>> >> cannot
>> >> convert it to local UTM formats because there is a discrepancy when
>> >> going
>> >> across different zones!
>> >
>> > I wonder how the centroid of points spread across the globe would look
>> > like.
>> > Jules ?
>> >
>> > --strk;
>> > _______________________________________________
>> > postgis-users mailing list
>> > postgis-users at lists.osgeo.org
>> > http://lists.osgeo.org/cgi-bin/mailman/listinfo/postgis-users
>> _______________________________________________
>> postgis-users mailing list
>> postgis-users at lists.osgeo.org
>> http://lists.osgeo.org/cgi-bin/mailman/listinfo/postgis-users
>
>
>
> _______________________________________________
> postgis-users mailing list
> postgis-users at lists.osgeo.org
> http://lists.osgeo.org/cgi-bin/mailman/listinfo/postgis-users
>