Comments on: Robinson Map Projection

Robinson Map Projection

KirkJobSluder — Tue, 16 Nov 2004 05:53:09 -0800

Arthur Robinson died last week. He is famous for the Robinson Projection which compromised on the Greenland problem of while being (IMO) more pleasing than the Peters Map. The map was widely used by the National Geographic Society in the 70s and 80s making it one of the most familiar to a generation of adults. The wikipedia has a summary of the advantages and disadvantages of the Robinson map as a compromise between equal area and spatial distortion. The map library at the University of Wisconsin is named for him.

By: ZachsMind

ZachsMind — Tue, 16 Nov 2004 06:28:50 -0800

I can't believe nonazimuthal pseudocylindrical orthophanic projection never caught on. Well, Robinson Projection never caught on either. When I was a school kid it was just called "that map on the wall." Arthur was a true visionary who saw the poles as lines instead of points. ..Far out, man.

By: piskycritter

piskycritter — Tue, 16 Nov 2004 06:32:00 -0800

I heard about it on NPR.

By: nthdegx

nthdegx — Tue, 16 Nov 2004 06:52:34 -0800

What does "pleasing" have to do with how a map should be laid out? Seriously.

By: KirkJobSluder

KirkJobSluder — Tue, 16 Nov 2004 07:25:18 -0800

What does "pleasing" have to do with how a map should be laid out? Seriously. There is no way to project a spherical surface onto a flat piece of paper that does not involve some form of distortion. You cannot show both shape and size accurately. Given that you have to compromise between distoring shape and/or size, aesthetics plays a big role in determining how much you are willing to compromise for which latitudes.

By: KirkJobSluder

KirkJobSluder — Tue, 16 Nov 2004 07:31:10 -0800

Aww, heck. Neither the Peters nor the Mercator really look right to me. The Peters makes South America and Africa look too skinny compared to what I expect from looking at a globe, while the Mecator has that big problem with the British Isles and Greenland. Projections like the Robinson have their distortions, but those distortions don't stick out like a sore thumb to me.

By: rocketman

rocketman — Tue, 16 Nov 2004 07:50:37 -0800

If you're ever in Madison, I recommend checking out the Robinson Map Library. It's got great atmosphere - a giant hall in an old, giant red block building like some German fortress, 20-foot ceiling, and rows and rows and rows of map cabinets topped with tables. It's like something out of Indiana Jones. It's a miracle they can find anything in there, but when I needed a copy of a 1848 political map of Europe, they knew right where it was.

By: leapfrog

leapfrog — Tue, 16 Nov 2004 08:39:25 -0800

My favorite has always been the Dymaxion projection, just because it makes the most sense to me. Start with a regular polyhedra which approximates a sphere, then unwrap it. I'm also a big fan of good ol' Bucky. (may he rest) One of these days, I'd like to work on software to compute a moving Mecca map; something where the central point changes based on user input in real time. It would make a neat screensaver. Or, if you had live GPS input, it would be a huge ego boost. (I am the center of the earth!)

By: ChasFile

ChasFile — Tue, 16 Nov 2004 08:48:40 -0800

What does "pleasing" have to do with how a map should be laid out? Seriously. Um, wow. By that logic, then, what does 'pleasing' have to do with how a car is laid out? or a menu? or a computer screen? Because "usefulness" in this case is about much more than "information presentation" it is about "effective information transmission," which includes little things called "perception" and "interaction" and "limpidity" as well. The Robinson projection has always been the best, IMO, because it attempted to minimize distortions while maximizing visual integrity and appeal. The Peters projection, on the other hand, sacrifices all precepts of asthetics in the interest of "accuracy." And despite sacrificing ALOT of asthetics (it really looks like crap) and uselfulness (much of Russia, Northern Europe, and Canada are completely unintelligible in the Peters) it still suffers from a great deal of distortion, as all 2-D maps do. Your belief that mapmakers should not be concerned with their maps being "pleasing" is premised on the belief that there is a perfectly correct way to lay out a map and that any deviation from that in the name of asthetic liscense is improper. There are many things worng with that, but most obviously is the belief that the transition from geography to cartography is a direct and precise one. This simply isn't the case. All mapmaking requires value judgements, trade-offs, and asthetic decisions.

By: Songdog

Songdog — Tue, 16 Nov 2004 09:31:24 -0800

There is at least one damned good way of laying out an accurate map: a globe. I'm just sayin'.