Piotr wrote:RogerOwens is more sensitive to vertical stretch/horizontal squish (or just "vertical"), while I'm more sensitive to horizontal stretch/vertical squish (or just horizontal).
I consider both to be equally undesirable. The measure consisting of min/max scale counts both directions of compression equally.
But yes, it's true that the shape-distortion of Africa stands out particularly much, because of Africa's size and location.
This is why I think 45° equirectangular looks much better than 30° or 0° equirectangular. Or maybe it's the bias of me living in Poland. This, that or both.
Sure, and likewise Equirectangular with standard parallel at lat 45 likewise portrays the U.S. more accurately than does the one with standard parallel at lat 30.
One thing I like about Behrmann equal-area is that its shape-compromise is between the equator and lat 41.4 ...resulting in the U.S. being about as accurately-portrayed as Africa.
With Cylindrical Equal-Area, lat 35. 4 has cos^2(lat) that's equal to the average cos^2(lat), averaged over the entire surface of the Earth.
Half of the Earth's surface is farther from the equator than lat 30, and half of the Earth's surface is closer to the equator than lat 30.
To me, there are boring projections (like Robinson; this one seems to mirror the RogerOwens' dislike of Winkel Tripel) that I have no reason to like or use, but I don't really hate any map projection.
I feel that Robinson achieves its design purpose of overall good appearance. I like it because it's pseudocylindrical. I understand and respect the preference that many people have for maps that minimize the worst magnification and shape-distortion, without being conformal or equal area.
I just personally prefer a map that gives the exact information of a conformal or equal-area map.
I, too, like all map-projections (...but prefer ones with useful properties and without appearance-spoiling distortion that doesn't bring any usefulness).
Why do I think Robinson is boring? Because it just is to me. It's pseudocylindrical, so there is extreme shearing
It's just a matter of what a person wants from a map. To me, usefulness is important. Pseudocylindrical maps have some useful properties. For pseudocylindrrical maps, a latitude-ruler can be marked with scales that tell latitude, horizontal scale, and area-magnification at any particular Y-value on the map. They more meaningfully portray latitude-related information. Longitudes can be easily judged or measured too, on a pseudocylindrical.
, and makes the map more biased (less compromise).
Well, it's a subjective individual matter, what kind of compromise one prefers.
Yes, Robinson must have valued reducing Arctic magnification more than accurate Arctic shape-portrayal. Well, which is more important: The area of the Arctic, or island-shapes in the Arctic Ocean?
Relatively good shapes, and areas not so bad, in the non-polar Arctic, instead of more accurate shapes very near to the pole.
It's impossible to say which kind of compromise is better.
As I said, I personally prefer maps that are exactly equal-area or conformal, but I consider Robinson to be a good compromise, if a compromise is desired--and many people like a compromise map whose area and shape errors aren't so bad, over most of the map.
It's not simple, compared to Apian II.
I like the simplicity of Apianus II. I've suggested it as an early map to display in school-rooms for the early school-years.
It [Robinson] has way too much flattening on higher latitudes (and I seriously don't just mean Arctic or Antarctic).
I don't find it so. The U.S. is portrayed with shape that I don't notice anything wrong with*. Canada isn't seriously magnified, in comparison to the U.S., compared to their relative sizes on a globe. Greenland is, of course, flattened, in comparison to a globe, but not as badly as it is in cylindrical projections.
*Sure, Robinson's U.S. is
barely east-west compressed, compared to its shape in Mercator. But that's barely noticeable, and certainly not problematic. It doesn't stand out as an error. Poland is farther north, and farther from Robinson's standard parallel, and I can understand that residents of Northern Europe might prefer Winkel's North-European shapes to those of Robinson. Winkel probably has a more northern standard parallel than Robinson does.
But i want to repeat here that half of the Earth's surface is within 30 degrees of the equator, and that about 75% of the Earth's surface is within 49 degrees of the equator.
Sure, there's great flattening in the high arctic, but not so outrageously bad in northern land-masses, other than arctic islands.
Winkel Tripel is just better and more "compromise" to me.
Africa is unrealistically skinny in Winkel-Tripel. That could be justified in an equal-area map, but there's no excuse for it in a map that has no exact useful properties. The U.S., too, is greatly horizontally-compressed in Winkel.
Half of the Earth's surface is in the +30 to -30 latitude band, a region of outrageous horizontal compression in Winkel.
About 75% of the Earth's surface is in the +49 to -49 latitude band (a parallel that coincides with the northern U.S. border), which, as I said, is highly horizontally compressed.
There's just no excuse for that in a "compromise" map that has no exact useful properties.
It seems that I'm completely different than RogerOwens.
Undeniably different people often prefer different advantages for maps. That doesn't mean that anyone is wrong.
Michael Ossipoff
