A letter to David Lowery at Trichordist, Carbon-Copied to Emily White

Dear Mr. Lowery:

Thank you for your well-thought-out reply to Ms. White’s blog at All Things Considered.  As you correctly point out, “fairness for musicians is a problem that requires each of us to individually look at our own actions, values and choices and try to anticipate the consequences of our choices.”  As an amateur musician of little repute and a listener of great passion, this question is a double-edged sword.


It really sucks to be on the “destruction” end of creative destruction, but I think that’s where artists are finding ourselves.  We’re just on the same end of the stick as opera singers and vaudeville performers were when records and movies came out—and I can assure you, it’s not the clean end.  Opera stages and vaudeville theatres made money because the number of seats in the house was, by definition, limited.  They could sell a scarce product, just like fair-trade coffee is scarce.


Their business model was killed off because recording technology allowed the song to be carried outside the walls of the theatre.  And it wasn’t just the people on the stage losing jobs—my grandmother’s nonagenarian friend Polly lost her sweet gig playing piano for the silent movies, too.  The recording industry made money off of recording and distributing tracks, assembled into albums.  But the track is the reification of the song.  The track is not the song.  And while you’re correct that “file sharing sites could get the same license if they wanted to, at least for the songs,” they don’t want to bother because they know that they’re paying for vapor.  File sharing has rendered tracks about as scarce as water.  


Your reasoning is strong, and you are making an important plea for us to recognize the plight of the artist.  But yes, I am asking us “to change our morality and principles to fit the technological change,” because the proverbial cat is out of the bag.  However, you jump right back on the correct-train when you point out the Cloud not being the solution.  The Cloud is siphoning off profits while offering very little, other than convenience.  I say we cut them out of the picture as well.


We as a listening community should share and share alike, burning CDs, e-mailing .torrents, and popularizing our favorite musicians to the best of our abilities.  And then we and our peers must bear the responsibility of supporting the artists.  Instead of relying on corporations to make the bets on musicians, we need to be doing so ourselves, through crowdsourcing platforms (which, although they do skim profits off the top, also provide a credit-card processing feature that I don’t have the knowledge myself to emulate) and through attending shows and tipping the band.  I’m excited to learn about SweetRelief, and that’s another great way to help out. 


As for musicians, we’ve got to be honest with ourselves: the idea that we can buoy ourselves off our past recordings is a thing of the past.  We’ll need to practice music im derech eretz and find some other things to do to get by in the world.  But the best songs we’ve ever written are the ones we haven’t written yet, and as long as we can convince our audiences this is true, I think we’ve got a shot.

40GB and a "Data Mule"

I finally stumbled upon the preferred term for the sort of transfer I'm looking for in this network, and it's not exactly elegant: a "data mule."  Coined in an article about underwater environmental research, the problem of rapid radio frequency die-off underwater was solved by a robotic "mule" traveling to sensor nodes and waking them up briefly to transmit their data before moving on down the line.  The mule could be recharged and sent back and forth, but power on the independent nodes was then conserved until it was needed.  The model closely resembles the opportunity-passing ring topology discussed earlier on this blog, so I'll take the term if it's already in use.

But let's face it: "data mule" carries with it some unpleasant connotations of rubber balloons full of heroin or cocaine jammed in uncomfortable places.  And it appears that the Flame virus may be capable of doing the same sort of thing on airgapped machines, riding along invisibly on a host USB until it can reach its target.  Seems pretty clear that if you have highly-secure data, you should basically just weld shut the USB ports on that machine... but then again, don't forget that "the most secure computer system is one encased in five feet of concrete, powered off, disconnected, and at the bottom of the ocean."  

The mule idea, though, also lends itself to a hardy animal that can reach the most remote of locations.  Through a service called DakNet, villages far off the beaten path in India and Cambodia are able to get access to Internet communications via a data mule.  Still, the implemented solution was on a village-wide level, and required mobile stations located on motorcycles or buses as well as local data kiosks.  That level of implementation, though far cheaper and negligence-tolerant than stringing up telephone wire hundreds of miles, is still a bit pricey for the attempt I'd like to make with our project.

So now we at least have a term for our vehicle of transmission, regardless of the connotations.  And I don't think I'll blush too much about muling data with painfully-slow connections-- I'll just be hiding the USB in my pocket.

Levinas and the Face-to-Face Network

To answer a few questions about the rationale for schlepping data from place to place by foot, I'm going to go off in a radically different direction than just talking over tech specs.  Part of the philosophical underpinning of the entire programme is rooted in Emmanuel Levinas's notion of the "face-to-face" encounter.  Peer-to-peer and friend-to-friend networks seem to have some of the same elements, but that actual encounter of the human being on the other end-- the Other, so to speak-- is a profoundly alienating experience.  Not only do I not know from whom I'm getting the music or movie that is being shared, but in most cases it is in my best interest not to know.

The first batch of mp3s I ever received came before I'd ever installed Napster.  A friend of mine was something of a whiz when it came to magically finding music, and he handed me a CD he'd burned.  I believe most of the White Zombie music I currently may or may not have on my computer is descended from that CD.  There is a moment of vulnerability in sharing files via the Sneakernet, but that's part of the point: I know that the person sharing with me actually values me enough to see my face and not kick my ass.  Indeed, Levinas states that the face protests to the observer, "Do not kill me."

The face-to-face is an ethic that comes before morality or taste.  It is the agreement that we will have a discussion, that we will form a network, that we will agree to at least keep talking.  I might share files p2p, but face-to-face I'm sharing an experience.  So the point of this network is not so much its efficiency (though it can be far more efficient than using the Internet) but instead the connections it forms with data.  I still know where all my White Zombie came from, and that's a layer of fun that I don't have with so many other files.  In the same way, I should also thank Aunt Linda for that 3.5" diskette of Tetris clones way back in the day.  Tetris will always be connected with you in my brain because of that.

Sharing should mean caring, no?

So You Want to Start a Sneakernet

I'm hoping some of you have followed this blog because you actually want to put this whole Sneakernet idea into practice.  So what will you need?  Well, at a bare minimum, you'll need a USB drive and a friend for a point-to-point network.  But let's look in a bit more detail into the two configurations I mentioned in an earlier post, the RING and the STAR.

SETTING UP A RING SNEAKERNET (Interval passing)

You'll need:

• a small group of friends (say 3-6)
• one USB drive per person
• a time to meet up regularly (say, Friday at the pub)

On Friday at the pub, everyone in the group meets up and passes the USB drive to the next person.  You can make a list if you like to keep things straight:

1. Mike
2. Steve
3. Lexi
4. Chris

And then Mike always passes to Steve, Steve to Lexi, Lexi to Chris, and Chris to Mike.

The week or so each person has with each drive should give enough time to find or create some interesting content for other members of the network, and since you won't get the drive back for a month, there should be some surprises waiting there for you each time.

SETTING UP A RING SNEAKERNET (Opportunity passing)


You'll need:

• a small group of friends (say 3-6)
• one USB, period

With just one USB drive, the meetup time is not quite the same incentive.  Instead, the USB drive travels in the same pre-determined order, but whenever we see each other to pass it off.  So if I'm seeing Steve this evening, I'll just give it to him then, and he can pass it on to Lexi when he gets the chance.  Latency is likely to be much higher on this sort of network, especially if someone spaces for a few weeks that they were the one left holding the drive.  Still, if you've only got one USB between you, this might be the way to go.

SETTING UP A STAR SNEAKERNET

You'll need: 

• a group of friends of most any size
• anywhere from a few to a handful of USB drives
• a mug, bowl, box, or bin located in a safe but open place

This form of the Sneakernet is likely to be the most welcoming to outsiders.  Anyone who wants to can come grab a USB drive from the mug or can drop one off there; we browse whatever content we pick up and share whatever we feel is worth sharing with anyone who decides to pick it up next.

The key to this topology is finding a good place for that mug.  Suggestions include the counter at your favorite coffee shop, the desk in the dorm room of wherever everyone ends up hanging out all the time (you know there's always one room), or behind some books in the library in a section no one ever browses (the Westlaw journals can provide an ironically amusing location, and I've never seen anyone take one off a shelf).  You could put them all in a box and store them out by the woodpile.  You could even register them as a GeoCache and wait for the fun to start rolling in.

These are just a few ideas, but I'm sure you all will have others!

Money on the Sneakernet

Monetization. 

It's kind of an ugly word, and it makes me think of those Gordon Gekko types sitting around, cigar in mouth and feet up on desk-- you know, the ones who greenlight productions like Teenage Mutant Ninja Turtles III, Troll II, and any movie starring Jean-Claude G-d Damn I Can't Act.

But let's be real: money really does make the world go 'round.  That's not saying I'm an anarchocapitalist, but we need an abstract representation of the value of labor and commodities.  And rather than reinvent the wheel, why not use money?

Here's the problem, though: money is scarce (for most of us... if it's not for you, please, hook me up!), but the types of things we can schlep around the Sneakernet aren't: movies, books, music, etc.  These things were scarce when you had to have a physical copy of the book, or the physical piece of vinyl record, or the actual copy of the VHS tape.  But now that everything can be copied a bazillion times... how do we determine the value of something?

Perhaps "scarcity" can be replaced with "difficulty of acquisition."  It's pretty easy to get a Kim Kardashian sex tape anymore, but much harder to acquire, say, a Moldovan passport in 48 hours.  (Not that you'd need one.)  But the same thing holds true for, say, people looking for music on the Sneakernet-- while something can be infinitely copied, we still have to get a copy of it first.

That's actually a boon for indie artists, and let me tell you why: if you're a talented person, I want your music.  And I want it badly enough that I want to be one of the first people to get ahold of it and listen to it. So when one of my favorite bands, The Shondes, needed help getting a new van after theirs was stolen, I was more than happy to give them money.  Why?  It's not just because I want the music that they've already produced.  It's also because I am investing in the music that they will produce in the future.  Because that's the scarcest commodity of all!

However, it's a bit tough to support every Kickstarter project that I like.  (Believe me, THE WOMAN would for some reason rather eat than have a new copy of random indie albums.  I don't get it.)  So what if, rather than basing a system of exchange on sharing scarce resources (like markets in gold and BitCoins), we based exchanges on obligations to other people?

Enter RipplePay.  It's a system built on IOUs, but you can trade the debt of people to be able to pay other people.  You know that scenario: I owe you $10, but you owe our friend Bill $10.  The most direct way to pay that off is for me to hand Bill $10.  Ripple automates this, and allows us to, in a way, create currency out of nothing.  So perhaps I'm broke as heck and want to thank you for the USB drive full of folk music you gave me... instead of just promising you some housecleaning services, I put an IOU out, and then you can trade that IOU to other people just like it was cash!  It has to be a trusted network, but if you're already in the Sneakernet, hopefully you're meeting people who you can trust.

Because you're creating a currency out of nothingness, you can actually reward your friend for creating new music by giving him an IOU that he can spend with anyone else who would value your IOU.  If he publishes enough music, you might be willing to give an item you might have put on eBay to someone else that he owes something to.  Savvy?

Anyway, just thoughts.  Hopefully they're slightly-liberating ones.

How Networks Are Born

lo

--Charley Kline

10:30 pm, 29 October 1969

Boelter Hall 3420, UCLA

If there were an "In the beginning," for the Internet, this would be it-- or to be more precise the first pair of characters transmitted on ARPAnet, one of the main predecessors to the Internet.  Mr Kline was actually attempting to send the phrase "login" to a waiting computer at Stanford, but the system crashed two characters in and they had to try again an hour later.  By November 21, a persistent connection was up and running between the two schools, and by December 5, a four-node network was running between UCLA, Stanford, UC Santa Barbara, and the University of Utah.

Before the Internet was this monstrously large network, it was four computers strung together.  If you're anything like me, you are totally excited by this concept and want to read all about it on your own, which you can in the collection of all Request for Comment (RFC) memos ever published.  Some of my favorites include the very first one, the first cut at Telnet protocol, and a curious one discussing some undisclosed criticism that the contractor, BBN, didn't want to publish for fear of embarrassing the researchers involved.  Much later, of course, we get RFC 1149, which details the transmission of IP datagrams on avian carriers-- i.e. sending data via carrier pigeons.

ARPAnet grew and grew, reaching 213 nodes by 1981, but it was eventually pared back down, and was finally decommissioned on February 28, 1990.  Pioneer Vint Cerf wrote the following poem in memorium, entitled "Requiem of the ARPAnet":

---


It was the first, and being first, was best,
but now we lay it down to ever rest.
Now pause with me a moment, shed some tears.
For auld lang syne, for love, for years and years
of faithful service, duty done, I weep.
Lay down thy packet, now, O friend, and sleep.


---

A touching moment, after which Al Gore invented the Internet.

But back to lo and that infant network.  In the beginning, ARPAnet really just consisted of those four network nodes passing information back and forth.  That's about all it would take to establish one in your hometown, too-- just find three or four other people who are interested in file sharing and maybe browse through a couple of network topologies to plan out how you're going to pass the drives around, and what you're all interested in sharing.  My suggestion is music you've recorded yourself!

Lo and behold, you'll have your own Sneakernet!

Facebook is Not The Network

I wanted to tackle one of the more basic issues with the Sneakernet pretty early on: how do I route a message to someone?  I mean, normally sending e-mail, I know the persons's e-mail address and host name, so I just send a message to steve@hostname.com, and presto!  All the little automatic switches and routers magically make the message appear at his end of the Internet.  (Okay, it's not quite that simple, but to most of us the Internet is a Black Box.)  On the Sneakernet, it's a little more complicated.  And because of that, we're going to go old school on this one.

Forget Web 2.0-- this is more like Web 0.2.

See, the thing is, computers used to have to connect directly to each other to share information over UUCP.  That's fine if I'm sending something directly to someone else, but sometimes I want them to PSST!  PASS IT ON.  And the way I would explain how to pass it on was by sending the mail with a BANG PATH.  Essentially the notation inserts a ! (bang) in between each node on the network that the message has to go through.

Let's say I wanted to send a song I recorded to my friend Steve.  Since I know how to dial his home, I'll hook up my dial-up modem, set the headset in the cradle, and we're off at a rocket speed of 150 baud.  Like I said, old school.  And at 150 baud, grab some coffee, because this is going to take a while.  There's really no reason to construct a bang path for this scenario, but here it is:

mike!steve

Steve shares the song with a friend.  Because my music is so awesome, now Steve's friend Chris wants to send something back to me for critique.  The bang path now becomes the following:

chris!steve!mike

Pretty simple still.  But let's say I have Lexi over to my house, and she has some comments for Chris?  For me to contact Chris, I just needed to be able to pass something by way of Steve.  But for Lexi to do so, she's got to go through two people.  Even more problematic, where both Chris and I knew Steve, Lexi doesn't know Steve or Chris, and Chris doesn't know me or Lexi.  But if we document a path between each other, the feedback can get there.

lexi!mike!steve!chris

Now let's say Lexi's father-in-law Abe has a friend named Dan whose sister Karen knows a guy in Hollywood named Kevin who wants to use Chris's song in an upcoming movie.  Oy gevalt.  It's a pretty tenuous connection, but with the bang path, we at least have a chance of the message getting from Chris to Kevin:

chris!steve!mike!lexi!abe!dan!karen!kevin

Now let's assume that the Kevin in question is none other than the illustrious Kevin Bacon.  Can we count the degrees of separation here?  While bang paths were intended to pass messages between computers, they also are remarkably useful for describing a particular route through a social network.  And as described here, the Six Degrees of Kevin Bacon is essentially a graph problem, using the social network as a big graph.

It also turns out that people are already adapting mathematical graph theory into analysis of the social graph.  I'll give you one guess as to who is making a killing off of this analysis...  FACEBOOK!  And good for them-- they jumped on a neat idea and are getting rewarded with the moneys.  No hard feelings.  But Facebook is not The Network.  Facebook analyzes and displays The Network for easy observation and use, but Facebook is not The Network.

We are.

And that's the key of the Folk Sneakernet-- just finding out ways to utilize our own person-to-person connections (Yes, P2P) to begin sharing things with a handshake rather than with a pulse of electrons.

So my message me!you!yourFriends is this: the social network is ours already and always has been.  We just need to adopt practices that let us use The Network in any way we choose.