On 14/06/2007, at 2:17 AM, brian grant wrote:
my proposal for a recursive geospatial reference is this: a modified
Hierarchical Triangular Mesh (HTM) - please bear with me on this -
I need to
OK, I'm answering because you asked. I looked into that kind of
addressing when I was wondering how to locate my data, but concluded
that it was a case of premature optimisation. It is indeed more
efficient to store co-ordinates in that form of representation, but
it is also a form of compression where it is more difficult to work
with the numbers.
What concluded the argument for me was the observation that any form
of representation that is inherently based around a globe is not
going to be suitable when we're trying to locate points that are not
on the globe. How do you represent satellites in space? How do we
locate other planets when we start colonising the moon or Mars?
Despite the problems, latitude and longitude is "good enough" - there
would have to be a significant advantage to make it worthwhile to
overcome the default inertia. What specific improvement does HTM (or
Geotude) bring to the table that latitude/longitude cannot be
modified to also handle?
So yes, I would have to agree with Richard Abas that it is a solution
in search of a problem. :)
Steve.
--
[EMAIL PROTECTED]
_______________________________________________
Geowanking mailing list
[email protected]
http://lists.burri.to/mailman/listinfo/geowanking
