I'll prime the pump here. Over several centuries a global location referencing system developed organized around meridians of longitude and parallels of latitude. It works for mariners, cruise missiles and atlas publishers, and doesn't seem to have any serious bugs (unless you consider the fact that the earth isn't a perfect sphere as a bug).

So, why try to improve on this system by triangulating the earth into some strange polyhedron? What's to be gained by messing with lat and lon, and how might it be achieved? Where do such pursuits lead us?

These are not idle questions. Quite a few researchers in diverse disciplines are beginning to ask and answer them. We feel it's time to start to collect these threads and weave a whole cloth that might wear comfortably on more than one body. What do you think?

-- Geoffrey Dutton (dutton@spatial-effects.com), April 28, 1999

Answers

That's a neat thing to do. Is the globe transluscent, with geography painted on it? How do you construct a solid globe using only the support of oct edges? How is it presented to visitors?

-- Geoff Dutton (dutton@spatial-effects.com), August 22, 2000.

For the sake of stirring the pot, maybe global grids are overrated.

For applications that are not truly global, and for some that are, a well-chosen or designed map projection may be sufficient. In the plane, everybody can be happy with whatever tessellation they like. Projections can minimize distortions most unwanted.

Some global grid advocates, at least I would include myself, are fascinated by the glories of geometry. Maybe we exalt these fascinations sometimes over getting the job done for a specific purpose.

Heretical enough?

-- denis white (denis@mail.cor.epa.gov), April 30, 1999.

Heretical only for an unreconstructed gridder. Most users are happy with their local coordinate systems, even though these don't knit together with their neighbors' systems without some effort. But I think there is value in appreciating that one's geodata is part of a larger picture, and thus one might want to foster a holistic sense by making the data more "interoperable" rather than just a private project.

Global grids, when they are hierarchical, enable the encoding of location to express the resolution and scale of measurements, in fact to indicate the precise region across which a geocode has dominion. Standard coordinates (plane and spherical both) don't do this. As a result, a hierarchical planetary address carries its own positional metadata; this means it needs less external documentation to be useful to someone else. And that is something that GIS has implicitly promised but has never really delivered.

Yes, geometry is fascinating, but it's just part of the implementation.

-GD

-- Geoff Dutton (dutton@channel1.com), May 03, 1999.

One basic global grid is important to me. I use the octahedron when working with grade school children and some times adults. I make a globe that you can go inside of to become the earth. It too is a basic octahedron (spherical) with a few more great circles depending on the circumstance. I havent made one for a few years but will again as Earth School gets cranked up in this 12 acre parcel we are restoring to native vegetation in Poway CA.

-- Randolph Howell (songeagle@netzero.net), August 07, 2000.