Join 3,512 readers in helping fund MetaFilter (Hide)


See Different
June 11, 2011 4:10 PM   Subscribe

The world is not as you think it is. While every map system has its faults, the Mercator we all know was designed for ship navigation five centuries ago, and introduces significant geographical distortion. Alternative projection systems, including perspective-cylindrical, pseudo-cylindrical and conic, attempt to portray correct relative size, accuracy of features, and position. Inverted maps diminish natural tendencies to see countries at the top as "superior".

The USGS has an excellent overview and comparison of cartographic systems. Maps are Territories, an online book exploring the cultural use of maps. A further explanation of the Gall-Peters projection. Many more "unusual" map projections available from ODT.

Upside-down maps and Progonos previously.
posted by Bora Horza Gobuchul (61 comments total) 102 users marked this as a favorite

 
Great post!
posted by tbonicus at 4:16 PM on June 11, 2011 [3 favorites]


But how big is Antarctica?
posted by TwelveTwo at 4:23 PM on June 11, 2011 [2 favorites]


I Love Turnbull's work with maps. I've actually written an article using his ideas as a basis arguing about the use of animals on maps and how their usage changed. Trying to get it cleaned up to publish atm.

Thanks so much for all these links.
posted by strixus at 4:26 PM on June 11, 2011


What about all of the swimming and burrowing creatures? Don't they deserve a map that's backwards? Or from their perspective, correct?
posted by StickyCarpet at 4:27 PM on June 11, 2011 [1 favorite]


This is something I've long been curious about:

In this era, where digital modeling can generate globe projections at will, why can't the Mercator Projection be redrawn based on the globe's circumference described by the arc made by the shortest-route path between any two points on the globe?

For many purposes (but obviously not all), scale and accuracy are most important along the relatively narrow territory nearest to a few points, and projections don't have to be bound to the Equator, poles, or key geographical elements.
posted by ardgedee at 4:28 PM on June 11, 2011


(Great post, btw. Thanks for the links I'm reading right now.)
posted by ardgedee at 4:30 PM on June 11, 2011


The "natural tendencies to see countries at the top as "superior""?

At last the answer to why everyone thinks Greenland is more important than France.
posted by joannemullen at 4:32 PM on June 11, 2011 [12 favorites]


This explains Tasmania in so many ways. Small, insignificant and at the very bum of Australia.
posted by greenhornet at 4:35 PM on June 11, 2011 [1 favorite]


Oi!
posted by Jimbob at 4:38 PM on June 11, 2011


My favorite mind-blowing map fact: Africa is really fucking big.
posted by Civil_Disobedient at 4:39 PM on June 11, 2011 [27 favorites]


> why can't the Mercator Projection be redrawn based on the globe's circumference described by the arc made by the shortest-route path between any two points on the globe?

Looks like this is partially answered in the Oblique Projections page of the site linked from "every map systems", above. An example is generated, using Campinas, BR as the southern polar coordinate.

Still and all, it would be interesting to, say, enter "London, UK to Seattle, WA, USA" to describe an arc to base the map's equator, and generate a map. Or enter a single location as a (northern/southern) pole and generate a map.
posted by ardgedee at 4:40 PM on June 11, 2011


Buckminster Fuller invented the Dymaxion Map to address these issues his own way.

Granted it's just the natural result of mapping a geodesic shape over a sphere and Mr. Fuller sure loved re-inventing his inventions, and it's arguable that he even actually invented the geodesic dome/sphere in the first place, but there's some good analysis and infoporn to be mined there.
posted by loquacious at 4:49 PM on June 11, 2011 [6 favorites]


Azimuthal projections came pretty close to that in practice, ardgeedee. ISTR it was not unknown for companies to have maps drawn up so that, say, directions and distances from their headquarters or shipping hub were shown true. Sometimes as a practical matter, sometimes as a "Hey, we're the center of the world!" PR gimmick. I can't find the example I'm thinking of right now though.


(And unrelatedly on preview, loquacious brings up Fuller's map so I don't have to!)
posted by hattifattener at 4:54 PM on June 11, 2011


from the upsidedown link China (Ancient): The Chinese were the first to invent the compass, which they always thought of as pointing south. South was a sacred direction, in ceremony the king would always face south. [ Source: email from Andy ]

This doesn't seem to make much sense (from the Ancient Chinese perspective even)?

also, not really buying the claimed "natural" tendency to think things nearer the top of the map are "superior".
posted by Bwithh at 4:59 PM on June 11, 2011


Something like this was the basis for one of my favorite parts of The West Wing.
"You can't do that."
"Why not?"
"Because it's freaking me out!"
posted by fifteen schnitzengruben is my limit at 4:59 PM on June 11, 2011 [12 favorites]


China (Ancient): The Chinese were the first to invent the compass, which they always thought of as pointing south.

Well.. opposite poles attract, and the same poles repel. People tend to think of the "North needle" of a compass pointing north. In fact, it is attracted to the south magnetic pole. Ponder that for a moment.
posted by Jimbob at 5:06 PM on June 11, 2011


Something like this was the basis for one of my favorite parts of The West Wing.

It's the first link in the post. I don't even watch the West Wing but I love that video clip. According to FireFox I've viewed that page about a dozen times in the last year.
posted by loquacious at 5:09 PM on June 11, 2011


Yeah but no matter how you draw a map, it still takes a long freaking time to fly from North America to Australia.
posted by bwg at 5:24 PM on June 11, 2011 [2 favorites]


Yeah but no matter how you draw a map, it still takes a long freaking time to fly from North America to Australia.

Nonsense, I draw my map with Montana renamed Australia, and now it doesn't take long at all.
posted by TwelveTwo at 5:41 PM on June 11, 2011 [1 favorite]


Why is everyone so fascinated with projections? Just get a globe (you know, a map on a sphere).
posted by rmorell at 5:54 PM on June 11, 2011 [6 favorites]


Because there's no correct way or "best way" to map a sphere onto a flat page, so picking the right technique for the right job is a sort of art form.
posted by crapmatic at 6:00 PM on June 11, 2011 [2 favorites]


why stop at inverted? why not orient the south pole left? i mean, if we're going to be illogical, why not be really illogical?
posted by nathancaswell at 6:03 PM on June 11, 2011


what if THE COLOR I SEE AS GREEN ISN'T THE SAME COLOR YOU SEE AS GREEN!?!??!?!

Did I just blow your fucking mind or what??!?!?!?
posted by nathancaswell at 6:09 PM on June 11, 2011 [2 favorites]


why stop at inverted? why not orient the south pole left? i mean, if we're going to be illogical, why not be really illogical?

Actually, it was fairly common in medieval times to put east at the top. Which has a logic to it: when traveling across open terrain, the one consistent thing you had to orient yourself by when you broke camp in the morning was the sunrise. In fact, that's the source of the term "orient yourself": it literally means to face east. And if you're facing east, you want a map designed to be read while facing in that direction.
posted by baf at 6:21 PM on June 11, 2011 [29 favorites]


Because there's no correct way or "best way" to map a sphere onto a flat page, so picking the right technique for the right job is a sort of art form.

There may be no "best way", but there are certainly worse ways.
posted by TwelveTwo at 6:23 PM on June 11, 2011 [1 favorite]


I love that inverted map. NZ...front and center.
posted by hal_c_on at 6:34 PM on June 11, 2011


The "natural tendencies to see countries at the top as "superior""?
At last the answer to why everyone thinks Greenland is more important than France.
posted by joannemullen


Even in France, there are people who think that Haute Medoc is a higher-ranked wine appellation than Medoc, when it's really just more Northerly.
posted by StickyCarpet at 6:53 PM on June 11, 2011


ardgedee: In this era, where digital modeling can generate globe projections at will, why can't the Mercator Projection be redrawn based on the globe's circumference described by the arc made by the shortest-route path between any two points on the globe?

Pretty easy to do yourself. Grab Hugin (an open source panorama stitching program, but we can mis-use it to "stitch" a single image of the world using different projections), add an equirectangular image of the world (lens: equirectangular, HFOV: 360), then in the preview select the Mercator projection, and drag the image (left-click move, right-click rotate) to put your chosen points along the "equator". In the stitcher tab, select "Stitch now...", and the map will be warped to your very own (for example) Seattle-to-London "equator" Mercator projection.

Hugin supports lots of other projections as well.
posted by JiBB at 7:33 PM on June 11, 2011 [9 favorites]


Self-link: interactive auto-reorganizing version of Fuller’s Dymaxion projection (I’m calling it “Faumaxion” because Fuller chose a project that had no inverse, so I used the similar gnomonic projection). Also check the first, more squirrellier version.
posted by migurski at 7:38 PM on June 11, 2011 [34 favorites]


These articles all discuss how each map is an "assumption," a different way of projecting the globe. Without even tearing this to shreds with PoMo theory, I'll just point out the most ridiculous assumption in the articles, "People feel pride in their country when its relative size is shown accurately."
posted by charlie don't surf at 8:05 PM on June 11, 2011


Of note, distortion happens on many levels with maps not only with the chosen projection. Essentially projections are a compromise used to highlight specific portions of the map that are important, such as scale or area. The Mercator projection is actually really useful since it displays Rhumb lines as straight lines. Thus, even if a particular projection is distorted in some form it still might have significant uses.

If you are like me and find distortion in maps and visual media extremely interesting, I would strongly suggest the wonderful How to Lie with Maps by Mark Monmonier. Its cheap, a quick read and very informative.
posted by graxe at 8:11 PM on June 11, 2011 [1 favorite]


Holy shit migurski. That's amazing. I'm speechless.
posted by notsnot at 8:17 PM on June 11, 2011 [1 favorite]


Mmmmm, tasty, tasty post, BHG.

Very interesting info in the upside down map link. Now I want. Want. Want.


...also, not really buying the claimed "natural" tendency to think things nearer the top of the map are "superior".

I'm with you on that one, bwithh. I tend to break the continents up into containing first, second, and third world countries--so by economics and standard of living. That means that Australia and NZ are 'superior' even though they're at the 'bottom' of the map.
posted by BlueHorse at 8:21 PM on June 11, 2011


The reason you put the north on top is because the northern hemisphere has twice as much land and 90% of the population. Jeez.
posted by kafziel at 8:28 PM on June 11, 2011 [1 favorite]


Actually, it was fairly common in medieval times to put east at the top.

I grew up in a place that has a large mountain range to the east--that is, a north/south range, several hundred miles long, and just to the east of the populated areas.

In short, you could always tell which way was east simply by looking towards the mountains.

So naturally, from the youngest age, I did all my navigation in relation to east as the primary direction.

I lived at various spots along that range for better than 25 years, so the habit was fairly ingrained.

Every other place I've live, I've oriented myself with the help of maps, and so naturally consider north to be the primary direction, and think of the other directions and navigation in relation to north.

It's amazingly confusing to go back to the childhood home. I can never quite figure out how, in that last few miles coming down the canyon, everything has suddenly shifted by ninety degrees.
posted by flug at 8:30 PM on June 11, 2011 [7 favorites]


flug, that’s a common thing in river towns as well. Many traditionally oriented their cartographic self-image up from the water, like this 1880 sewer map of Memphis (Perry-Castañeda collection).
posted by migurski at 8:35 PM on June 11, 2011


and the reason we don't have the top of the map as east is because compasses point north/south
posted by Bonzai at 8:40 PM on June 11, 2011


I habitually orient myself by large bodies of water and the like (probably because my mom does). It makes a lot of sense in cities that've grown out from a waterfront/riverfront/harbor over time. This sometimes makes it hard to give directions, but then, I find that people I'm giving directions to often have no idea what direction north is, either.
posted by hattifattener at 9:29 PM on June 11, 2011


Map lover here. People have already sufficiently pimped Fuller's map, so I'll just thank you for the great links.
posted by Meatbomb at 9:29 PM on June 11, 2011


Obligatory the map is not the territory reference.
posted by Splunge at 9:34 PM on June 11, 2011


I am so glad someone made this mefi post because when I first watched that scene from The West Wing years ago, I promised myself I would follow up on it but never did.
posted by OddlySurreal at 9:54 PM on June 11, 2011


migursky, you can see that too in Montreal, where "north" is actually almost 45 degrees off. That really messed me up when I figured that out.

Also being from Montreal (me) and Toronto (wife), where the water is south and the hills go north, messes us up whenever we're anywhere where that context is defferent.
posted by sauril at 10:49 PM on June 11, 2011 [1 favorite]


My personal preference is for either the Winkel tripel or Wagner IX projections.
posted by alby at 2:36 AM on June 12, 2011


Civil_Disobedient, yes, that's a lovely map and my current favourite too!
posted by infini at 4:16 AM on June 12, 2011


Oh, what a fantastic post! The world needs more exposure to non-UTM (or Robinson - don't even get me started on Robinson) maps.

I used to work for a GIS software company, small but growing with some rather large clients. They once sent me out of state to train a bunch of guys that would be using our software and I had developed a severe love of maps by that point in my life, so I was looking forward to being around like-minded people. It was my first work-related trip, and I was not entirely sure I was the right person for the job.

Apparently I was, because they actually asked the right technical questions and grokked by the end of the trip and because as I was leaving I was given a copy of a beautiful polar map of most of the northern hemisphere and an inverse world map. We had talked earlier in the day when the training concluded about our favorite maps - and yes, it is possible to have a favorite projection - and we all agreed that the world needed to see more non-transverse mercator, north-oriented topographic maps that challenge their way of thinking and their sense of their own world, and so they gave me two of the most disorienting-for-laymen maps they could think of.

They're both hanging on my wall alongside a bathymetric chart of the Savannah River and a Fuller map of the oceans. They remain two of my most treasured gifts.
posted by neewom at 4:21 AM on June 12, 2011


Since the 'significant geographical distortion' link in the FPP is a visual post, here's the relevant text explanation I found via another link for the Greenland/Africa problem.
posted by infini at 4:36 AM on June 12, 2011


Inverted maps diminish natural tendencies to see countries at the top as "superior".

Is there research about this? I see it come up occasionally with maps about the projection, but haven't seen anything that's very convincing. Seems to me people have a natural tendency to see the country are from superior no matter what it's size.
posted by cj_ at 7:10 AM on June 12, 2011 [1 favorite]


Ten years ago, a girl I knew came back from Australia with an upside-down map T-shirt. I desperately wanted one then, and now I really, really want one.

Excellent post.
posted by devinemissk at 7:26 AM on June 12, 2011


The reason you put the north on top is because the northern hemisphere has twice as much land and 90% of the population. Jeez.

That's a post hoc justification if ever I heard one. You think before Australia had been discovered the map makers said "this has the more land, we'll start here". No, they started where they were.

And even ignoring that, it doesn't make sense: why should the larger land be at the top? Don't maps fold? Do they get narrower as you go further down?

Anyway, I'd like to hear this pomo destruction of cartographic assumptions. They always go well here.
posted by bonaldi at 8:41 AM on June 12, 2011


We don't have to care very much about this stuff anymore. We don't need paper maps.

With a digital map, you can focus on the country you are talking about, minimizing the distortion for that country, and you can zoom out to get a broader view from space. If you like, you can have that view projected in any flat way you want and make the transformation from one to the other as slowly as you like (to show you the warping created by the projection you selected). You can have selected items on the map also displayed in their own windows to show relative size, distance, etc. If you are looking at two countries on either side of the globe, you can show each country in its own window with distortion minimized and have a third country show the relative view (where each is on the globe, how large each is, how to get from one to the other, plus non-geographic views such as wealth and population). With screens getting larger and larger, plenty of people will have screens the size and resolution of paper wall maps.
posted by pracowity at 9:50 AM on June 12, 2011 [2 favorites]


Pracowity, what you say is true but really only for a tiny minority of the population at large. For a growing number of people—possibly a majority, if not now then soon—the default screen used to interact with maps is getting smaller. Behind that screen is an economy of processing and storage scale which means that for the most part, every popular map you come into contact with from providers large and small alike uses an identical projection optimized for locally-conformal images and people who live in the 20°-60° latitudes. Paper maps may no longer be necessary, but the digital maps we’ve mostly replaced them with carry many of the same limitations.

What you describe could happen, but for a variety of reasons it mostly doesn’t.
posted by migurski at 11:56 AM on June 12, 2011 [3 favorites]


We don't need paper maps.

We don't need paper maps as much as we used to, certainly, but the need depends on the type of map and each individual user's intended use of his map.

We had debates about this in the office where much of our work was dedicated to scanning paper maps and engineering drawings and other location-dependent charts, georeferencing and, as is natural for a GIS company, playing with geographic databases and making digital maps with an insane amount of detail and customization.
There are people without the easy ability to use digital maps despite the relatively new technologies built specifically to store and view a large number of digital maps. Although our need for paper maps has diminished dramatically with the advent of new technologies, it has not eliminated the need entirely, and the world would be worse off in any case if we no longer made old-fashioned meatspace maps.

Distortion can, in any case, be diminished simply by changing scale. Again, depending on the intended use for the map, a larger-scale map is called for (links go to description of large/small scale maps, one of my most annoying peeves, having had to explain it time and again to a number of people).
Depending on the needed use, in other words, distortion can simply be diminished by using a large-scale map, as most road maps would arguably be, and not necessarily dependent on the coordinate system used. Transverse Mercator, for example, is useful in very large-scale maps and less useful for the small-scale (depending again on use), and others would be better for large-scale, for example Lambert.

Zooming in and out does still, if accuracy is desired, create the need for redrawing of the map so that distortion is minimized, and so projection and other geographic data will need to change dependent on desired scale just as it would with a paper map. In GIS, this is easily accomplished on the back-end of the database (I say easily, but GIS software is actually rather amazingly and beautifully complicated; I guess 'easily' is more aimed at the user and not the compiler). There is still, however, a need to take into consideration the horizontal and vertical data of the desired mapping area, the intended uses of the product and so on, and thus a need for someone, somewhere, to consider what sort of data the map (and thus the user) needs. There is someone, somewhere, that will print that resulting map out and use it as they would a regular old paper map.

I love maps that challenge people's view of their own world. Maps that, for some reason or another, are not north-oriented, are amazing works that people aren't used to seeing. Using traditional north-oriented maps, we think about cardinal directions more than we do features and usefulness sometimes, but with maps that are not north-oriented we suddenly begin to consider the features more. With polar maps, for example this one (large JPG, 812K), north and south becomes a different concept for the average user.

Basically, the uses for maps are varied and not all users are able to do away with paper maps entirely. Also, the more we challenge our view of the world with maps that aren't simple rectangular grids laid out the way we're used to is a great thing and can actually challenge one's sense of their place in the world.

(long-winded way of saying, again, hell of a post!)
posted by neewom at 12:14 PM on June 12, 2011 [2 favorites]


Damn it, I meant small scale in relation to lambert. *sigh* I just peeved myself.
posted by neewom at 1:55 PM on June 12, 2011


In the classroom, you can use nice big digital maps and you can flip the poles to try to reeducate students.

But most people want maps so they can get from A to B and they care mainly about exactly where they are now. They can view digital maps on tiny screens that announce when the next exit is coming up. They don't want to trundle down the highway in a dymaxion car while the person in the passenger seat tries to make sense of a dymaxion map, and if you try to change their world view they're going to give you welts on your Weltanschauung.
posted by pracowity at 2:05 PM on June 12, 2011 [1 favorite]


While chasing links from this post I came across this blog post of Accidental Map Projections. I like the warped polyconic.
posted by hattifattener at 3:56 PM on June 12, 2011


I like upside down maps because they support the Continental Drip Theory.
posted by plinth at 4:57 PM on June 12, 2011


If you're interested in being challenged further about mapping, you might want to check out Denis Wood's Everything Sings, an atlas of a small neighborhood that challenges notions of what gets mapped, not just how things get mapped.
posted by BlooPen at 6:51 PM on June 12, 2011 [1 favorite]


The Dymaxion map gives me serious fuzzy-wuzzies for depicting earth's land masses as an archipelago. We're all in this together!
posted by Scoo at 8:11 AM on June 13, 2011


Why is everyone so fascinated with projections? Just get a globe (you know, a map on a sphere).

It's kind of hard to fold up a globe and stuff it in your glove box.
posted by EmpressCallipygos at 10:16 AM on June 13, 2011 [1 favorite]


The concept of "north is up" is arbitrary both on a map and also in the solar system or even the universe. Does the Earth rotate clockwise or counterclockwise? Yes!
posted by JJ86 at 2:04 PM on June 13, 2011


Sorry for over-posting in this thread, but...

Maps not easily-viewable in digital form for their intended use (from UT map room, all are very large files, and I think they're pretty cool, YMMV):

Irkutsk, Russia. 1:250,000 JOG, UTM, 6.2MB jpg
Middle East. 1:1,000,000 ONC, Lambert Conformal Conic, 17.1MB jpg
San Antonio del Norte, Honduras & El Salvador. 1:50,000, 10.5MB GeoPDF (location-aware features do not work on any non-windows machine, requires a download to view location-aware features; disclosure: I worked at the company that developed GeoPDF and had a hand in its creation and in the creation of many spatially-aware PDF's)


Interesting/strange things being done with digital maps:

Bizarre Map Challenge, an annual challenge for students' bizarre map creations (that is sadly on hiatus due to funding issues), including The United Constellations of America, cities in the US named after other countries, and The Earth in Reverse.

OpenStreetMap, an open, remix-able mapping tech online (very, very, very previously).

Ushahidi, an open-source platform for mapping crises (also previously)

Why it's probably not a good idea to rely on Google Maps / Earth too much


Unusual maps:
Europe as seen from Moscow, a more cut-up and less-distorted Myriahedral Projection (second link is a PDF of a Cartographic Journal paper), lots previously
posted by neewom at 2:47 PM on June 13, 2011 [2 favorites]


« Older Attorney General Holder mandates, at a minimum, an...  |  Leicester, 1982. A school rock... Newer »


This thread has been archived and is closed to new comments