Researcher Runs IP Network Over Xylophones 83
joabj writes "Following up on experiments of running Internet Protocol(IP)-based networks with carrier pigeons or bongos, UofC grad student R. Stuart Geiger has demonstrated that it is possible to transmit simple ping requests across two computers using people playing xylophones. Throughput is roughly 1 baud, when the participants don't make any mistakes, or get bored and wander off. The OSI encapsulated model of networking makes this project doable, allowing humans to be inserted at Layer 1, the physical layer. Vint Cerf wasn't kidding when he used to say, 'IP on Everything.'"
this isn't even cool (Score:2, Interesting)
I remember doing "networking via scraps of paper" at school.
Writing silly notes since kindergarten and then actually implementing some real protocol in computing classes.
Then a couple of years later an LED, a bit of fibre and an LDR and we were building "fibre networks".
Mind you, this was two decades ago.
Looking even at the "cool" projects which come out of MIT undergrads, I get the impression that almost all children are exposed to absolutely nothing interesting whatsoever before the age of 18.
Re: (Score:3)
Looking even at the "cool" projects which come out of MIT undergrads, I get the impression that almost all children are exposed to absolutely nothing interesting whatsoever before the age of 18.
Welcome to education in the US.
Re: (Score:1)
It's parenting.
"We must keep the child safe from all danger, real or perceived."
It isn't *just* parenting. (Score:2)
In the public grade schools in Hawaii, the class will share about four textbooks on any given subject, and the state mandates and actually teaches toward state tests with state lesson plans and quizzes that are frequently wrong. They mark the kids as wrong when they get things right, and then tell them "you were right but I have to mark it lower because the state's answer is X."
And I don't mean normal smarter-than-the-test wrong, I mean things like singular v. plural.
Parenting is deficient in a lot of plac
Re: (Score:2)
I remember doing "networking via scraps of paper" at school.
When I visited the Miraikan Museum in Odaiba, Toyko, back in ~2005 they had a mechanical IP network that routed 2 byte messages of black and white wooden balls over rails and Archimedes screws. The first byte was the "IP" address (I think there were 4 or 5 nodes) and the second byte was the message. You set up your message in the staging area of your workstation, then pulled the release lever. The balls would roll down the track past an eye that would read the colors and various gates would be opened/clo
Re:they forgot to add parity notes (Score:4, Informative)
Re:they forgot to add parity notes (Score:4, Interesting)
Re: (Score:1)
Here's a mockup of what a noisy channel [youtube.com] would sound like.
Re: (Score:2)
Play major chords. If a note is incorrect then the chord is dissonant and you have an error condition.
or a sus4 chord :).
Re: (Score:1)
Except when a wrong note just yields a different major chord, like F major instead of C major.
Re: (Score:1)
That would require two notes to be wrong in order to end up with another major chord.
Re: (Score:3)
Error detection is a Layer 2 responsibility. TCP only does sequence numbers and resends lost packets.
Re: (Score:2)
Re: (Score:2)
That is not correct. The IP header does have a checksum, but it covers only the header itself. Corrupted data would not be detected by the IP checksum. TCP has a checksum that covers the TCP header, all the payload data, and a few important fields of the IP header (such as source and destination IP).
Re: (Score:2)
Also, there is something important to be said about link-layer error detection: it is not end-to-end. If corruption occurs while the packet is being handled by a router (as opposed to while
Re: (Score:2)
Very true. I have gotten into some verbal fights over that in the past with people insisting their error checking was good enough, and they didn't want to support the same kind of error checking other people were using. I kept pointing out, that the it w
Re: (Score:3)
Really they didn't implement IP-over-xylophone -- you cannot, because there is no provision in the IP standard for framing between packets. They implemented some as-yet-undisclosed link layer protocol, and then ran IP over it. They could just as easily have run DECNET. Since they gave no details of the link layer protocol, we don't know if it had checksum support.
Re: (Score:2)
Re: (Score:1)
Still uses IP. ICMP packets are delivered in IP packets.
Re: (Score:2)
Layer 3, IP, needs a minimum of 160 bits for the header... and the guys need to be good in arithmetic calculation to provide an accurate checksum!
And why not a router at that layer, two men, one listens to the last xylophone and the other one translates to a piano in the next room... while there could be some networks interferences
Re: (Score:3)
"Ping"
"Your tits look nice today."
Replied very inappropriately!
UofC? (Score:1)
Do we call Cal UofC? I had to check the article to make sure this very serious research project was coming out of California and not the University of Cambodia.
Re:UofC? (Score:4, Interesting)
In California, we call it UC Berkeley [wikipedia.org].
i.e. yoo see burr clee, and even that is incorrect, as the town and the school where named after Bishop Berkeley of "if a tree falls in the forest and no one is around to hear it, does it make a sound?" fame. His name is pronounced BAR clee.
/pedant
Re: (Score:2)
relevant: Observer effect [wikipedia.org]
just FYI Bishop Berkeley's solution was (paraphrasing): "yes, it makes a sound, because God hears everything."
Re: (Score:2)
as the town and the school where ??? named after Bishop Berkeley
Where what was named after Bishop Berkeley?
If you are going to be pedantic, spell your post correctly.
Re: (Score:2)
Most of my neighbors call it Cal or just Berkeley. They only use the UC when talking of one of the "lesser" campuses.
Anyone else think of CE3K (Score:4, Insightful)
Speilberg thought of communicating using musical notes 35 years ago
Re: (Score:1)
Speilberg made $337,700,000. This guy will die a virgin. Big difference.
Re: (Score:3)
Dude, why did you post anonymous. You would have had a great moderation war.
Those who see the humor (and fact) in wrote you wrote would mod you up.
Those see Spielberg as a deity and will do anything to appease him, and mod you down.
Those who the virgin comment hits just a little too close to home would mod you down.
My gut instinct is that you would have had at least a couple dozen moderations, and landed at a nice solid "+4 troll"
Re: (Score:1)
Speilberg pffft.
Try this http://en.wikipedia.org/wiki/Hedy_Lamarr [wikipedia.org] - secure communication using frequency hopping from an actress in 1941 ;-)
Edison patneted this idea (Score:2)
Upcoming Studies (Score:4, Funny)
Upcoming studies:
Improving Baud: Coffee vs Electroshocks
IP Backbone Implementation Using Michael J. Fox
River Rapids: High-Speed Internet With Riverdancers
Increasing Latency With Rube Goldberg Networks
Replacing LCD Screens With Bored College Students And Etch-a-Sketches
Stone-age Computing: Exploring Completely Inadequate Alternatives To Modern Technology
Re: (Score:2)
River Rapids: High-Speed Internet With Riverdancers
This I'd like to see - would it be implemented as a highly parallel channel?
Re: (Score:3)
only 1 baud? (Score:2)
Re:only 1 baud? (Score:5, Informative)
Your math is way off. With 8 notes, each encodes 3 bits; two notes allow 64 different combinations (not 1.6 million!), or 6 bits.
Re: (Score:1)
1 baud means one symbol per second, not one bit per second. If there are only two symbols in the alphabet (for example two notes), then 1 baud results in 1bps, but if there are more symbols, the bitrate can be higher.
Re: (Score:2)
8 notes is three bits each, I don't know how you got to 3 bytes from two consecutive notes...
Even two simultaneous notes would only be six bits.
Re: (Score:2)
Well, no.
Actually, 2 octal digits can do 64 different combinations.
Here's a better article.... (Score:4, Informative)
So a quick Google turns up this Black-boxing the User: Internet Protocol over Xylophone Players (IPoXP) [altchi.org]
It needs an acronym! (Score:2)
Re: (Score:2)
Re:It needs an acronym! (Score:5, Funny)
Next phase, IPoXP tunneling (Score:2)
Protocols (Score:1)
That's a Glockenspiel, not a Xylophone. (Score:2, Informative)
The article states that the musical instrument has "aluminum keys". From http://en.wikipedia.org/wiki/Glockenspiel :
"[A glockenspiel] is similar to the xylophone; however, the xylophone's bars are made of wood, while the glockenspiel's are metal plates or tubes, thus making it a metallophone."
Re:That's a Glockenspiel, not a Xylophone. (Score:5, Informative)
Re: (Score:1)
Originally xylophones were all wooden, and the expensive ones still are, but many are made of synthetic material now.
As a band director that teaches percussion it drives me nuts when people call a Glockenspiel (or bell set) a xylophone. I think it's caused by some of the toys that are available for little kids that are mislabeled.
Almost interesting (Score:1)
Stuart Geiger has demonstrated that it is possible to transmit simple ping requests across two computers using people playing xylophones.
Was there ever any doubt that this could be done? It's the same as that carrier pigeon IP thing - it was always going to work. Has it taught us anything new?
One baud? What does that mean? (Score:3)
Baud is a measure of symbols per second, so it's meaningless unless the amount of information per symbol is defined.
In this case, it turns out that a symbol is a hexadecimal value, so the data rate is about 4 bits per second.
Increased speed with solenoids & FFT (Score:2)
With today's DSP technology, FFT algorithms, and a bank of solenoids, two computers could, in theory, transmit data via xylophones a LOT faster than one baud!
FFT analysis on the receiving ends determines which notes are being played and when, even simultaneously. By using notes unique to each machine, both can be playing and receiving simultaneously. It would be quite noisy, but would definitely work.
It would also be a good idea to "damp" the chimes, to dramatically reduce the audio decay rate. This would a
Re: (Score:2)
Re: (Score:3)
Because it's fun?
Re: (Score:2)
Lame. Slashdot had IP over bongo drums in 2003! (Score:1)
Nearly 9 years ago, ./ reported about IP over bongo drums already, featuring double the data rate.
http://slashdot.org/story/03/09/27/175242/tcpip-over-bongo-drums [slashdot.org]
As the original page is offlne since years, here's archive.org:
http://web.archive.org/web/20031230015730/http://eagle.auc.ca/~dreid/ [archive.org]
VOIPOXP (Score:2)
What's next, VOIPOXP (Voice Over IP over xyloPhone? Latency will make satellite and lunar communication look really good. :-)
OSI never made anything doable (Score:1)
Somethings smells funny at Vint Cerf's place (Score:5, Funny)
Re: (Score:2)
"'IP on Everything." Sounds like his office is a really disgusting place.
LULZZZZ! [www.last.fm]
sorry, no online video or audio available... but trust me, its ridiculous
Re: (Score:1)
Sounds like my brother-in-law's hound dog.
Does anyone else have visions of... (Score:1)
RFC1149/2549 coupled to a keyboard under a line of birds?
Video demonstration of IP over Xylophone Players (Score:2)
Here is the video demonstration that didn't get posted in the original article:
Video: http://youtu.be/qCT7SisWh38 [youtu.be]
Internet Protocol over Xylophone Players (IPoXP) situates humans at the lowest layers of the Internet. Read the full paper at http://www.stuartgeiger.com/ipoxp.pdf [stuartgeiger.com]. A project by R. Stuart Geiger, Yoon Jeong, and Emily Manders at the University of California, Berkeley. Presented at alt.CHI 2012.
Video demonstration of Xylophone Players (Score:2)
You'll like this little gem [britishpathe.com] better.
Re: (Score:2)
that's just spam, has nothing to do with the original article IP over Xylophone.
Ping has been done using ducks! It's true (Score:2)
Computer Science (Score:2)
IP over black holes (Score:2)