On Aug 3, 2011, at 9:55 AM, V. Balaji wrote:
> I would still opt to have an unspecified datum take the value
> "unspecified" rather than "unknown". Despite Upendra Dadi's objections,
> I still feel my principal objection to specifying the datum is the
> implication of precision that does not exist. For instance, if I told
> you my model of the Earth was a sphere of radius R, and yours was a
> geoid of some kind, you might choose to apply a transformation of some
> kind, which in my opinion is completely uncalled for.
By uncalled for you mean "unnecessary" or "over-kill"? Seems to me the only risk here is wasted CPU cycles...
It seems the primary objection for some is that the datum is unnecessary because of the low precision in coordinates in some datasets (i.e. GCM data) and that assigning a datum inherently implies higher precision.
Would those with this argument be more comfortable if there was a convention for indicating the precision of spatial coordinates?
One specifies a datum (or unspecified) and spatial precision. win-win.
Tom Kunicki
Center for Integrated Data Analytics
U.S. Geological Survey
8505 Research Way
Middleton, WI 53562
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <
http://mailman.cgd.ucar.edu/pipermail/cf-metadata/attachments/20110803/fd0815bd/attachment.html>
Received on Wed Aug 03 2011 - 09:09:36 BST
