Analysis of Only Everything Lasts Forever

Alright, this is just too deliciously nerdy to not post about.

I recently started writing some things in Processing. I will of course post about that at some point, but I found something that distracted me quite strongly a little while ago.

I have been looking at lots of Processing example programs, as well as open-source projects such as ones on Open Processing. There’s so much you can do with Processing in three dimensions, but the 2-D stuff is what really holds my attention. It must be the simplicity and the ability to understand the code without having to clutter my head with all the transformations necessary for pretty 3-D effects. I appreciate a program so much more when I can grasp what it’s doing.

Anyway, I stumbled upon Kyle McDonald‘s Shuffle Enumeration Diagram, which looks like this:

It might look super-complicated, but if you read the source code, you’ll see it’s a fairly uncomplicated graphical representation of a bit shuffling algorithm in motion.

From there, I was pretty interested, and was even moreso when I read that he had used something similar in a project poetically entitled Only Everything Lasts Forever. Quoted, with some added emphasis of my own:

Only Everything Lasts Forever is a very long sound composition for MP3.

It contains every sound we can distinguish as humans, as dictated by the MP3 specification (ISO/IEC 11172-3). It explores the social and political associations of sound representation, and the psychology and philosophy of noise and emptiness.

While the entire composition is approximately 10450 years long, the first month of the composition was streamed from a server room at EMPAC starting on Sunday March 28th 2010 at 7 PM EST.

The theoretical scale of this project struck me as something unusually large, and I decided to crunch some numbers. (Warning: Informal (though hopefully correct) math ahead.)

(Note: for the sake of my own sanity, I rounded the results of my calculations to three significant figures.)

Assuming the world population remains constant* at roughly 6.7 billion, and every person begins listening to unique parts of OELF 24/7, how long would it take for the entire composition to be heard?

10450 years of music / 6.7 x 109 years of music listened per year = 1.48 x 10440 years of listening if we all worked together.

Anyway, to give some perspective, assuming the universe is roughly 13.75 ± 0.17 billion years old, the amount of time it would take for us to listen to the whole composition would be about…

1.48 x 10440 years / 1.375 x 1010 years = 1.08 x 10430 times the current age of the universe.

More perspective: If you wanted to store the entire composition as an MP3 with a constant 64 kbps bitrate (remember, that’s kilobits per second, not kilobytes)…

64 kilobits/second = 8 kilobytes/second
1 year = 31,556,926 seconds
thus 8 kB/s x 31,556,926 s = 2,524,455,408 kB per year = 2.35108231 terabytes per year
10450 years x 2.35108231 terabytes/year = 2.35 x 10450 terabytes total
1012 terabytes = 1 yottabyte
thus 2.35 x 10450 terabytes / 1012 terabytes = 2.35 x 10438 yottabytes would be needed in order to store the whole thing.

To put that into perspective…

the information capacity of the observable universe = 1092 bits = 1.03397577 × 1067 yottabytes
2.35 x 10438 yottabytes required / 1.03 × 1067 yottabytes available = 2.28 x 10371 universes would be required in order to store the entire composition. Take that, Moore’s Law.

* I tried for a few hours to figure out how that number would look if I included population growth rate (either constant or gradually declining) in the equation. The whole effort is made cumbersome by the immensity of the numbers. I may return to this at some point, but it’s largely tangential.

My thanks to Kyle for the inspiration for this post.

Licensing information:

Shuffle Enumeration Diagram by Kyle McDonald, licensed under Creative Commons Attribution-Share Alike 3.0 and GNU GPL license.

Only Everything Lasts Forever by Kyle McDonald, licensed under Creative Commons Attribution 3.0 United States.

This entry was posted in CS, Thoughts and tagged , , , , , , . Bookmark the permalink. Both comments and trackbacks are currently closed.
  • Recent Posts

  • Topics

  • Archives