If you look at this site about map projections you can learn more why it's a problem about the poles:
Try to wrap a piece of paper around an orange. Maybe you can figure out the problem Google has at the poles
Software is like entropy. It is difficult to grasp, weighs nothing, and obeys the second law of thermodynamics; i.e., it always increases.
Norman R. Augustine
Ask Gore! Global Warming is the answer, It's all gone.
Actually there is a relatively simple solution to this problem: construct the Earth as a geodesic sphere and use dymaxion projection for the map file.
Or without the jargon: construct the sphere out of lots of triangles, think Spaceship Earth at Epcot. In fact when you strip out the bezier curves and other fancy stuff this is exactly how you construct a sphere as a basic 3D model.
Now each triangle gets its own overhead slice of the map. There are no poles with such a model or map projection so whether you're looking at the Congo or Antarctica things would be stretched out or displayed any differently.
To be honest it's a shame GE doesn't use a system like this. It might make some things a bit more complicated but it would allow for a far more even display of the maps. It would also allow you more freedom in how to use image overlays, even if the actual mechanics would be a bit trickier. For example, putting an image overlay near the poles does all sorts of wacky stuff to it - but this wouldn't be an issue with my suggested set up.
So I suppose now I need to learn how to program so I can create this amazing piece of software and learn about the programming pitfalls it would stumble into
This was extensively discussed in this link.
Google Earth Hacks Forums > Google Earth > Great things you've found >
really really weird?
Quoting from that link :
Hungry Donner is right provided geometric accuracy is what you are looking for. But then the 'feel' of a spherical earth will not be there for most users who are laymen and use it only as a visual display and not for geodesy applications.
The point in the placemark is exact North Pole where all the Longitudes merge into a single point.
The terrain at north pole is sea.
For the sea area GE uses only numeric data indicating the depth of sea at a particular place ( bathymetric data, not imagery acquired by camera. ) . This data is coarse resolution data and therefore has very poor resolution and each pixel may span about a kilometer by kilometer at equator.
E.g. You can see this by zooming in the sea water ( till the scale legend at bottom left shows about 50 kms )near 13░32'42"S 15░21'29"W .... there are squares of varying intensities of blue and black. They are the actual pixels.
Now imagine what will happen if you try to bring closer the northern corners of a pixel .. it will become a triangle.
That is what happens at the poles.. the pixels become triangular ( spherical triangular actually ) and so each pixel at the pole is triangular and the vertex of each triangular pixel is at pole, giving rise to the flower like appearance for the area around pole and each 'triangle' represents the color corresponding to average color shade in that area
From your bookmark if you go up in altitude slowly then you start seeing next line of pixels surrounding the first set and then next ... and so on to ultimately the limit where the pixels become very small and you get a smooth picture of sea bottom.
a perfect sphere!
So the feel of a spherical Earth would be better perserved if GE used a DP map instead of the standard Mercator.
From the user's perspective there really aren't any drawbacks to this. So why didn't Keyhole do this? Most likely because the Mercator route is simpler, especially with the data set they were using. However it's worth pointing out that the DP route I suggest is actually how 3D objects are normally skinned by a computer. This is why the skin of a human figure looks all weird.
Is it my imagination, or does it look like the earth is expanding? If you study the faultlines, it kinda looks like a balloon filling with air.