The Longest Straight-Line Ocean Voyage Is Pretty Damn Long

May 3, 2018 7:25 AM

Do you want to take a 20,000 mile ocean voyage but don't want to steer, like, ever? Good news! You can start in Pakistan, head southwest (aim for just a little off the coast of Africa), and end up on the Kamchatka Peninsula. It's only 19,939.6 miles, but what's 60.4 miles among friends?
posted by Etrigan (45 comments total)

Did they think about submarines? Submarines with drills? Submarines with laser drills? Way to science, guys.
posted by GenjiandProust at 7:40 AM on May 3, 2018 [10 favorites]

This is the best. I recently saw someone from the Metafilter Strava group who was running a pretty straight 10k and got curious about the longest straight line road (it is about 100 miles and there are tons of clickbait articles on the subject), so this definitely pertains to my interests.
posted by Literaryhero at 7:43 AM on May 3, 2018 [2 favorites]

Well that's my weekend plans sorted...
posted by instantidealism at 7:43 AM on May 3, 2018 [1 favorite]

Bummer about the AdBlocking policy on their site. Even when I excluded their site, they still threw that pop up at me telling me to disable AdBlock.
posted by NoMich at 7:45 AM on May 3, 2018 [1 favorite]

Good. Fuck steering.
posted by The otter lady at 7:47 AM on May 3, 2018 [19 favorites]

Atlas Obscura has a non-whiny page, NoMich.
posted by Johnny Wallflower at 7:51 AM on May 3, 2018 [1 favorite]

I can't help but think the flat earthers are going to get ahold of this and irritate the eff out of me with a malformed interpretation.
posted by missmobtown at 8:13 AM on May 3, 2018 [2 favorites]

I disagree that "not steering" has to equal "straight line" because my dream is to tie a rope to my ship's wheel so I am forever gently circling the Pacific garbage patch in my decommissioned aircraft carrier, playing shuffleboard and having pizzas delivered via drone from various earthquake-prone countries.
posted by lefty lucky cat at 8:18 AM on May 3, 2018 [12 favorites]

I was confused by the large gaps in Antarctica's landmass and in the center of Greenland on their map. Turns out large parts of Greenland and Antarctica are below sea level! Once you get to the bedrock below all the ice, that is. Didn't realize that. (Most (all?) of that is just because the ice weighs down the land, though, and it would rebound if the ice were to go away.)

Similarly on the article's maps, it looks like the great lakes mostly count as land. So at least the map used for display here looks like it's counting "land" as any rock/dirt surface that's above sea level, even if there's a big old lake above it, and "sea" is any rock/dirt surface that's below sea level, even if there's a big old ice cap above it. It seems like an odd choice.
posted by whatnotever at 8:26 AM on May 3, 2018 [2 favorites]

put me on this boat, i am ready
posted by poffin boffin at 8:28 AM on May 3, 2018 [1 favorite]

That last leg from Cape Horn to Kamchatka is a doooozy.
posted by stinkfoot at 8:30 AM on May 3, 2018 [2 favorites]

Submarines with laser drills?

ITYM "submarines with frickin' laser drills"
posted by Greg_Ace at 8:32 AM on May 3, 2018 [3 favorites]

and "sea" is any rock/dirt surface that's below sea level, even if there's a big old ice cap above it. It seems like an odd choice.

This is an uninformed guess, but maybe that's a side effect of the algorithm used, that needs "boundaries" set by what's above/below sea level?
posted by Greg_Ace at 8:35 AM on May 3, 2018 [1 favorite]

Oh, it's unlust and the one-dimensional boy.
posted by davebush at 8:35 AM on May 3, 2018 [1 favorite]

In the Bermuda Triangle, you can go in a straight line forever.
posted by clawsoon at 8:36 AM on May 3, 2018 [5 favorites]

I disagree that "not steering" has to equal "straight line"

Absolutely. Not steering is a pretty surefire way to end up not going on a straight line at sea, over a large enough distance.
posted by Dysk at 8:39 AM on May 3, 2018 [8 favorites]

AROUND THE HORN!?
posted by Krazor at 8:43 AM on May 3, 2018 [1 favorite]

The video showing that the path was indeed a straight line really messed me up after seeing how curvy the map made it look.
posted by A Bad Catholic at 8:55 AM on May 3, 2018 [2 favorites]

This opens up strategies for Risk that I had never imagined. India/Pakistan to Kamchatka in one move?
posted by Strange Interlude at 9:11 AM on May 3, 2018 [3 favorites]

I'm going to point my ship southwest, stand on the prow, and start swinging my fists. If you get hit, it's your own fault, landmasses!
posted by lefty lucky cat at 9:14 AM on May 3, 2018 [3 favorites]

Literaryhero, there's a Metafilter Strava Group?
posted by evilDoug at 9:34 AM on May 3, 2018 [2 favorites]

The voyage as described would be a real shitter, first riding the Agulhas current down Africa straight into the opposing polar currents coming around Antarctica, then approaching Cape Horn from the Southeast, and then a close reach bucking the Roaring Forties, followed by crossing the trades, only to head straight through the North Pacific High near Hawaii. No thank you.
posted by gyusan at 10:11 AM on May 3, 2018 [5 favorites]

Does this account for the imperfect sphericity of the earth? The tropical regions are basically uphill from the temperate ones because the earth bulges around the equator, so this would add some length, perhaps that problematic 60.4 miles.
posted by Rumple at 10:54 AM on May 3, 2018

I used Chrome developer tools to remove the ad blocker overlay (right click the ad blocker popup, inspect, walk up the dom, delete the parent div) and then read the article in print preview.
posted by jeffamaphone at 10:58 AM on May 3, 2018 [2 favorites]

Except that “straight line” is an abstraction based on what projection you are using for a map. The actual course in the initial figure of the article appears to be a great circle (projected) on a Mercator projection map. As far as steering, you would not fix a course and then go below and play solitaire. Steering a perfect circle requires constant change in heading. As an example, leaving Pakistan, your heading is ~SSW, passing Cape Horn nearly due W, in the mid-Pacific NW, and approaching Kamchatka its back to ~WSW. Luckily, modern nav systems can take care of the continuously varying course of a great circle route automatically.
posted by sudogeek at 11:36 AM on May 3, 2018 [1 favorite]

You don’t need to steer to drive/sail along a great circle. Your heading changes automatically, without you needing to turn the rudder. You *do* need to continually adjust if you want to follow a rhumb line (which looks like a straight line on a mercator projection), unless that rhumb line is the equator.
posted by aubilenon at 12:58 PM on May 3, 2018 [3 favorites]

Oh, and “Straight line” is pretty clear; it’s straight in spherical geometry, not on some dumb projection. Which is basically the definition of a great circle.
posted by aubilenon at 1:00 PM on May 3, 2018 [3 favorites]

It's a constant change in heading, but not a change in the position of the rudder, no?

(Of course you'd still have to use the rudder to handle currents and such.)

The Portugal-China route looks like a fun trip. The Beijing-to-Paris road race of 1907 is kind of an approximation (though longer, due to not being a straight line).
posted by zompist at 2:07 PM on May 3, 2018

I just like saying "Kamchatka."
posted by nebulawindphone at 4:08 PM on May 3, 2018 [1 favorite]

gyusan: The voyage as described would be a real shitter, first riding the Agulhas current down Africa straight into the opposing polar currents coming around Antarctica, then approaching Cape Horn from the Southeast, and then a close reach bucking the Roaring Forties, followed by crossing the trades, only to head straight through the North Pacific High near Hawaii. No thank you.

Would it be any better if you went the other way?
posted by clawsoon at 4:32 PM on May 3, 2018 [1 favorite]

I imagine maths probably has a definition of “straight line” on the surface of a sphere (or oblate spheroid) that works and is not some artifact of abstraction. I imagine it probably involves perpendicular lines converging at certain points maybe... or maybe comparing tangents relative to the center of the sphere... actually damn... other than being the shortest distance between two points how do you define straightness on a sphere?
posted by midmarch snowman at 4:55 PM on May 3, 2018

It's the shortest distance between two points on a sphere too, is my understanding. So one way to think about this is it's the longest sea route that couldn't be made more direct with the same endpoints.
posted by nebulawindphone at 5:09 PM on May 3, 2018 [2 favorites]

Wikipedia to the rescue!

The equivalents of lines are not defined in the usual sense of "straight line" in Euclidean geometry, but in the sense of "the shortest paths between points", which are called geodesics. On a sphere, the geodesics are the great circles.

So yeah, it's the shortest distance, which are great circles. The only way you could reduce the distance further would be by digging a tunnel
posted by aubilenon at 5:15 PM on May 3, 2018 [3 favorites]

evilDoug, yes there is. We are a friendly (albeit quiet) group.
posted by Literaryhero at 5:21 PM on May 3, 2018 [1 favorite]

The only way you could reduce the distance further would be by digging a tunnel

What on earth did I just read??

I want a burrito now
posted by Dysk at 5:40 PM on May 3, 2018

other than being the shortest distance between two points how do you define straightness on a sphere

As already stated, such a line is a geodesic or great circle.

Any line on the surface of a sphere will eventually meet itself if extended far enough, and thus form a circle.

A geodesic is a line that if extended as far as possible [i.e. so that one end meets the other] it forms a circle which will bisect the sphere.

Geodesics can have any orientation, but on a globe of the earth, the lines of Longitude are geodesics. The lines of Latitude are not.

Any two geodesics must intersect at two points.
posted by HiroProtagonist at 7:49 PM on May 3, 2018

the lines of Longitude are geodesics. The lines of Latitude are not.

One of the lines of latitude is!
posted by aubilenon at 8:53 PM on May 3, 2018 [3 favorites]

Laying a taut piece of string between two points on a globe gives you a geodesic. This also explains why you don’t need to steer when sailing a geodesic.
posted by monotreme at 8:55 PM on May 3, 2018

One of the lines of latitude is!

Good point! Musn't forget the equator.
posted by HiroProtagonist at 9:20 PM on May 3, 2018

(I think the "don't need to steer" thing is maybe a bit misleading, because in the real world you often do need to steer just to go in a straight line. Where you'd end up if you started off Kamchatka and didn't touch the steering wheel has a lot to do with wind and waves and currents, and not much to do with geometry.)
posted by nebulawindphone at 8:33 AM on May 4, 2018 [1 favorite]

Thanks to all for helpfully correcting my obviously very serious nautical plotting error and pointing out that you cannot actually “not steer” for twenty thousand miles. I cannot imagine how such a mistake made it through the editing process.
posted by Etrigan at 9:22 AM on May 4, 2018 [2 favorites]

Hang on I think you might even have to steer extra, to compensate for the Coriolis effect, when crossing from the northern hemisphere to the southern. I just hope nobody has already set out to try this, before we've worked out all the details! They could be lost at sea for significantly less time than they'd intended.
posted by aubilenon at 9:47 AM on May 4, 2018 [3 favorites]

Hey, sorry, I'm not trying to be a dick about the phrasing in the article -- I actually thought it might be helpful to the "what counts as a straight line" question.
posted by nebulawindphone at 9:50 AM on May 4, 2018

Is the "Cooke Passage" not correct, then? By eyeball it looks about the same length, but he claims its 2000 miles longer.
posted by ctmf at 10:36 PM on May 4, 2018 [1 favorite]

This is our Stack Overflow thread.
posted by clawsoon at 5:20 AM on May 5, 2018