Getting screwed by your ISP

[avram]

My long-ago contributions to the Everything website weren’t anywhere near this interesting — Penises have higher bandwidth than cable modems:

Putting these together, the average amount of information per ejaculation is 1.560*10⁹ * 2 bits * 2.00*10⁸, which comes out to be 6.24*10¹⁷ bits. That's about 78,000 terabytes of data! As a basis of comparison, were the entire text content of the Library of Congress to be scanned and stored, it would only take up about 20 terabytes. If you figure that a male orgasm lasts five seconds, you get a transmission rate of 15,600 tb/s. In comparison, an OC-96 line (like the ones that make up much of the backbone of the internet) can move .005 tb/s. Cable modems generally transmit somewhere around 1/5000th of that.

(Via Boing Boing)

Comments

[supergee.livejournal.com]

Many people do not know whether they are producing information or just jerking off.

[batyatoon]

For some reason I am reminded simultaneously of Snow Crash and Neuromancer, but for completely different reasons.

My brain hurts.

Whee splat!

[autopope.livejournal.com]

The logical fallacy in this is, of course, the fact that the OC-96 is always on -- whereas your average human male is unlikely to be able to ejaculate more than ten times a day and quite possibly only manages two or three before flaking out. So we've got a burst rate or 15,600 tb/s, for five seconds, for maybe two or three bursts in 24 hours -- call it 250,000 tb/day -- against three minutes per terabyte over the OC-96, which works out at 480 tb/day. Now, when you consider that with wavelength dimension multiplexing we can cram over 2.5 tb/s down a single fibre already -- although this technology isn't fully deployed yet -- we begin to see that we can almost ship data around at that speed already.

Which begs the question of why.

Quick back of the envelope calculation: human eyeballs resolve 4K x 4K pixels in 32-bit colour, and the same again in 8-bit grayscale, twice over for two eyes, with a refresh rate of about 18 times/sec. We can approximate this to 1,152,000,000 bytes/sec without compression. If all other sensory inputs amount to the same volume of data, we get 2 Gb/sec (without compression) for the full afferent sensory feed of a human being. Multiply by population, and we see that the bandwidth we're discussing is enough to encode the entire sensory input of about 8 million human nervous systems in real time, with no compression.

This is, in case you hadn't noticed, a wee bit ridiculous :)

[theaceofspades.livejournal.com]

...but then we get into discussions about uptime, and that could just get messy.