YouTube: https://youtube.com/watch?v=kH9S9jTfUIU
Previous: Question Tuesday: The Goat Edition
Next: The Road Less Traveled By

Categories

Statistics

View count:363,182
Likes:17,854
Dislikes:219
Comments:1,508
Duration:03:55
Uploaded:2015-08-28
Last sync:2019-06-13 14:40
In which Hank just gets really nerdy for a while trying to figure out whether it's possible to encode and decode all 24 TB of Wikipedia onto and from a single QR code. Please feel free to check my math and let me know if I got anything wrong.

Minute Physics: How Far Can Legolas See: https://www.youtube.com/watch?v=Rk2izv-c_ts

Decoding a QR Code by hand: https://www.youtube.com/watch?v=KA8hDldvfv0 (which does a great job of explaining exactly how QR codes work.)

I know I didn't correct for the distortion at the edge of the wide-angle shot, but Henry and I went through the calculation and found that it didn't affect the overall result.

----
Subscribe to our newsletter! http://nerdfighteria.com/newsletter/
And join the community at http://nerdfighteria.com http://effyeahnerdfighters.com
Help transcribe videos - http://nerdfighteria.info
John's twitter - http://twitter.com/johngreen
John's tumblr - http://fishingboatproceeds.tumblr.com
Hank's twitter - http://twitter.com/hankgreen
Hank's tumblr - http://edwardspoonhands.tumblr.com
Good morning John. This is a QR code. Its a way of visually representing data, just like barcodes, or snapcodes, or facial expressions, or the written word. And I was looking at one of these things recently and I thought to myself: how much data could you really fit on a QR code?

Well, QR codes are complicated and I don't have enough time to explain them in total. But, suffice it to say that each block of 2 by 4 squares codes for one byte of data. One character of text. The question that I had was could you print out all of Wikipedia on one QR code and have a camera that could scan that code and you could get all of Wikipedia by scanning one QR code?

I'm aware that this is a dumb question, but it very much is the type of dumb question that I enjoy figuring out the answer to. Just the text portion of just the English portion of Wikipedia is about 24 terabytes. That's 24 trillion bytes. So in order to get the number of squares we need, we just multiply that number by 8 because it's eight squares per byte. 

I don't want to know how many squares this thing is gonna be, I wanna know how many squares it's gonna be on each side. So, we take the square root of that number. To figure out how much space that will actually take up, we have to figure out how big each one of those tiny little blocks are.

I'm gonna say about 2 millimeters because that's pretty easy to print and to read by a QR reader. So you multiply that number by two and then convert from millimeters to kilometers and there you have it, you have a QR code that's just 27 kilometers on a side that has ALL of Wikipedia on it.

Now, of course, this is just half the battle. We've printed out Wikipedia on a 27 kilometer by 27 kilometer sheet of paper, but we need a camera that can actually do that. Well, if you have a camera with a 90 degree field of view, then you can take a picture of this whole thing, you just have to be 13.5 kilometers away from it.

However, there's some bad news that comes along with this. First, at that distance, the atmosphere is going to distort this image way too much, so we've got to take it out of the atmosphere and into space, put the QR code on the surface of the moon and put the camera in orbit around the moon at around 13.5 km up. No big.

The second problem is actually a bigger problem cuz it deals with you know, just sort of the fundamental nature of the universe. From so far away, and with blocks so tiny to try and resolve, you run into what's called the defraction limit. Basically, when passing through the aperture of our camera, the difference between the size of our aperture, that's the hole that the light comes through, and the wavelength of light is significant enough that it blurs anything below a certain size.

So to take a picture of something so tiny from so far away, you need to do one of two things, or both of them. One, make the aperture of the camera really huge or two, make the wavelength of the light smaller. But if you want to take the picture of this thing with a cell phone camera, you can as long as your cell phone camera just happens to be able to detect gamma rays.

Gamma rays are radiation with a very small wavelength, also problematically with very high energy, so it's difficult to print with ink that would absorb gamma rays because they would normally just fly right through it, so you've gotta find some kind of really special gamma ray absorbing ink.

The other, in my opinion better, solution to this problem is to make your aperture just huge and one way to do that is to just use the James Webb space telescope which we're already developing and has a pretty massive aperture. Now, you'd have to modify it so it could be a wider-angle camera cuz it's a telescope right, it's got telescopic, not a wide angle lens.

But the James Webb space telescope does have an aperture big enough that it could conceivably resolve these two millimeter pixel things from 13.5 kilometers away. However, it does not have a sensor sensitive enough to take a picture of all of them at once, which I would like for it to do so that we can just take a picture and real quick have all of Wikipedia.

So we are going to have to develop a new camera sensor that has like 10,000 times more pixel resolution than the current highest in military spy cameras that we use, but I have confidence that we can do that in the intervening time between when we are able to actually print out all these pieces of paper, which is gonna take some time, and get them to the moon and lay them out in the proper order.

So the answer to the question "can you print out Wikipedia and have it be a scannable QR code?" The answer is yes, not really with current technology but it is possible. John, I'll see you on Tuesday.

I want to end this video by saying you know, I'm not entirely sure. I could have got something wrong, and I want to say thanks to Henry from MinutePhysics for coming over and helping me with the math of this. That was a really enjoyable project. That's my life.