A month or two ago I watched the keynote to ChefConf 2013 by Adam Jacob. In the beginning of Adam’s talk, he lays out the groundwork for what he calls (unsurprisingly) the continuous delivery economy. Adam talks about the personal driver service Uber, and how companies like it are becoming more prevalent. In case you haven’t used it before, Uber is a slick application for your phone that dispatches a car to your location and takes care of all of the payment details. Open the app, ask to be picked up, get in the car, get out at the destination. It’s really that easy.
Adam goes on to talk about a hypothetical company UberNana that is just about ready for the prime time. Due to increased connectivity in rural areas, UberNana can connect banana sellers and buyers to a mutually beneficial end. Instead of being limited to known seller areas, buyers can now connect with all the sellers — wherever they might be. Sellers can wait until their bananas are ripe, and will receive a fair market value for their goods because the flow of knowledge is symmetrical.
Buyers get a larger market, sellers get an equitable price, and consumers get ripe bananas.
The real take away is that — at least in the service industry for now — software has become the middleman. I don’t need to call a taxi to go from one place to another. I don’t need to have a wholesaler to buy bananas. I just fire up the app, make a request, and it’s done. Continuous delivery in action; tight feedback loops, goods delivered right when you need them — right now.
I definitely think Adam is on to something, but I think it would be a mistake to label this as a recent trend. Take this 2000 Senate Judiciary Committee meeting, ‘The Future of Digital Music‘. I’ve appended Gene Kan’s written statement to this article, as it is only available on the Internet Archive, and the more copies of it, in my opinion, the better. At about time index 1:05:30, Gene delivers an excellent account of this continuous delivery economy, and how individuals and corporations might use it to their advantage.
In brief, I think his message can be summarized as follows: The internet is a near-zero marginal cost distribution network. An increasing number of goods are becoming digitized. Preventing the digitization and dissemination of those goods is almost impossible. How are companies and individuals leveraging these new paradigms and distribution channels to their advantage?
Napster was the Uber of its time. I want a song. I request the song. The software handles all the details, and I get my song. I didn’t have to go to a music store. I didn’t have to wait in line. I got my song, right now.
I think the extension of this paradigm to big-box retail is obvious. An increasing amount of floor space that was once dedicated to physical goods has been digitized — books, movies, and music to name a few. If I can get my product through the internet — right now, at a cheaper price, in greater fidelity, and so forth — why would I bother going into the store?
I think niche retailers like Best Buy and Petco have or soon will begin to feel these pressures. If my entire product selection becomes digitally distributed, my customers (all else being equal) are going to go where the price is cheapest. But what about physical goods? Food? Crafts? I say, behold the 3D printer, same-day delivery, and Etsy.
If I were a large retailer I’d be worrying about what I could be doing to differentiate my products and services. What makes me special when a half a dozen online stores sell the same good, at a cheaper price, and I can get it within the hour? What am I doing to enable an infrastructure that supports this consumption model? What am I doing to create a platform that facilitates these transactions — not only for first parties, but for the banana sellers and banana buyers of the world. How do I handle a world where I don’t necessarily control the distribution, consumption, packaging, or usage of my goods?
With tremendous disruption comes tremendous opportunity.
9 July 2000
Senate Judiciary Committee
Hearing regarding the future of intellectual property in the digital age
using music as an analog for all types of intellectual property
Utilitarian view of intellectual property and music on the Internet
Statement of Gene Kan
Founder InfraSearch, Inc.
- 1. INTRODUCTION 3
- 2. CONSUMER DEMAND4
- Environmental considerations5
- Quality control6
- 3. BENEFITS FOR THE RECORDING INDUSTRY6
- Recording companies7
- 4. RETOOLING MEDIA DISTRIBUTION7
- Turning pirates into paying distributors8
- Do or die8
- 5.TECHNOLOGY IS JUST TECHNOLOGY9
- Copy protection doesn’t work9
- End-to-end encryption10
- Napster and Gnutella are just the first wave10
- 7. WHAT SHOULD WE DO?111. IntroductionThe future of intellectual property in the face of broadband is uncertain.Intellectual property control on computer networks has long been a problem for the software industry1. Most recently, old-economy enterprises such as the Recording Industry Association of America (RIAA) have taken notice. Music is the topic of the day, so let’s focus there.Technologies enabling the simple and fast exchange of music have existed for decades. In the physical world, audio cassettes led the charge. Remember recording songs off the radio? Minidisc and recordable compact discs made it easy to swap and pirate music en masse with near-original quality. In the virtual world, swapping MIDI, tracker, and other formats of digitized audio were already commonplace when I began BBSing in 1994.Other media have had similar histories. Movies, images, and text have had their equivalent of the audio cassette: that first easy-to-use duplication instrument. Video cassettes, video compact discs, photocopiers, scanners, email, optical character recognition…. The list goes on for some time, and has been going for some time now.So what is it about the Internet that has made the long-lived problem of media duplication such a pressing issue? Recording industry executives claim that one main factor is ease of use. Anyone, particularly university students2, can easily download whatever music they want in minutes.For typical home Internet users (56 kbps modem), it takes approximately eleven minutes to download a high-quality MP3-encoded song3. For a typical university student it takes about thirty seconds. The recording industry would argue that the downloaded file, when played back using an MP3 player4, is similar in sound quality to the original compact disc.The cost to download? In the US, most Internet Service Providers (ISPs) charge a flat rate, as does the telephone company for local calls. That makes the incremental download cost approximately zero. It’s mainly a matter of patience and hard disk capacity.With the impending globalization of inexpensive broadband access, what is becoming a headache for the music industry will become a headache for the movie industry.2. Consumer demandDownloadable music attracts all manners of people. Recording industry employees, doctors, lawyers, students, adolescents…the list goes on. The people who use Napster are not criminals. They are not the thugs you see on the evening news. The people who use Napster are your family and friends.There are numerous theories on why MP3 is so popular, and I’ll cover a few of them below. But the simple fact is that digital downloadable music is hugely popular. So popular, in fact, that the recording industry recently capitalized on Napster’s success by using it to “leak” new songs by Madonna and Dr. Dre, among others. No, they’ll probably never admit to it, but it did happen, and it’s well known throughout the music industry that these songs were leaked from within the record companies to generate hype the same way movie trailers do.ConvenienceThe convenience of MP3 is undeniable. For less than 100 USD one can buy a 20 gigabyte hard disk, which stores approximately 5000 near-CD-quality songs. The hard disk fits in your shirt pocket, or neatly into your computer, where it will provide about 20000 minutes (fourteen days) of continuous listening. Portable MP3 players are about the size of audio cassettes. Compare that to compact discs which store about seventy-four minutes of audio and don’t fit into standard pockets.
Finding music is typically a very tough process. These are the steps I take: Drive to the record store. Scour the shelves. Buy, if I’m lucky enough to have found something worth trying. Drive home. Listen. Remember the tracks that I like.
The steps for Internet music purchases are much more simple. Let’s take a brief look at the buyer experience at Amazon.com. Click over to Amazon.com. Click “popular music”. Type in “John Denver”. Get a screenful of results. Click. Buy. Moreover, when I clicked on “John Denver’s Greatest Hits”, Amazon recommended several other albums I might be interested in. Record stores don’t do that, and if they did, I’d have to scour the shelves again for the recommended disc.
Unfortunately, at this time, Amazon doesn’t sell MP3s. Buyers must wait for the physical CD to be delivered. And MP3.com, which does sell MP3s, doesn’t sell an MP3 version of “John Denver’s Greatest Hits”. So while the Internet makes a step in the right direction, there is currently no well-known non-infringing method of downloading “John Denver’s Greatest Hits”.
The ultimate, really, is found in information-sharing communities such as Napster and Gnutella5. Users are able to search for exactly the information they want, and download it instantly. As it relates to music, that means users can search for exactly the artist and song title they are interested in. Forget scouring the store shelves only to find that what you wanted is not in stock6.
It’s my personal guess that if people could pay a reasonable price for the music they download, they would. At this time the only well-known pay-for-download services do not carry downloadable versions of popular albums. People have no choice. If they want the convenience of downloading, they unfortunately have no way to compensate the copyright holder.
The marginal cost of an MP3 is zero. Even for consumers. When the telephone company comes to your house to install your DSL, they might charge 150 USD for installation and 50 USD per month. Using that line, an infinite amount of music and be downloaded with little hassle.
Compare that to compact discs. Mine, transport raw materials, acquire raw materials, manufacture, box, transport, unbox, stock, inventory, sell, re-order, etc. All that costs a lot of money, and the Internet eliminates the whole cycle. On the Internet, it’s: record the music, sell the music.
A popular argument for MP3 is that it makes music easily portable. Flying on airplanes imposes interesting space constraints. When I fly, I generally have a laptop computer with me so I can work while I’m in the air. Before MP3, I carried a portable compact disc player and a handful of the CDs I liked most. Now, I can carry MP3 versions of all of my CDs on my laptop’s hard disk. The best part: a full hard disk weighs no more than an empty one, and I don’t have to find a place to stow my fragile CD player and CDs.
Compact discs, unfortunately, are made of matter. Matter which must be mined, manufactured, and delivered. Each step in that process holds numerous environmental disasters, and in the end the thing consumers are really after is the music carried on the compact disc, not necessarily the compact disc and its associated packaging.
The Internet makes near-zero-marginal-impact music ownership possible. Downloading an MP3 does not require the manufacture and delivery of a compact disc, box, liner notes, etc. To summarize it sweetly: no environments were harmed in the download of this MP3.
One thing consumers demand above all others is product quality. Currently, Internet bootlegs of music vary greatly in quality. Some are ripped7 from CDs which have scratches. Others are encoded at horribly low bitrates, diminishing the audio quality to unbearable levels. Yet others are encoded using low-quality encoding software, leading to diminished quality.
The RIAA and its constituent record companies have an opportunity to exploit their own brand name and quality control procedures to produce digital downloadable music with consistent high quality. Surely music downloaders would pay a small amount to ensure they aren’t wasting their valuable download resources.
When you buy toothpaste, do you buy Colgate, or do you buy Brand X Tartar Control?
3. Benefits for the recording industry
Artists and recording companies alike can benefit from digital music. The reasons are numerous. For artists, it is a chance to reach directly to their audience: a global audience. They have the opportunity to capture nearly 100% of the gross sales of their product. For recording companies, there is an opportunity to reduce the marginal cost of distribution to nearly zero, and to expand the scope of distribution to the entire Internet.
Whether or not it actually happens, artists have complained for a long time that they don’t get a fair deal in their record contracts8. Even Metallica, a band rumoured to have struck an outstanding contract with their recording company, sees the pot of gold:
“Yes, of course, the scenario that the gentleman asked in the question is very, very possible, and we’ve been looking at that for a long time. And when we are done with our record contract, I would say that something in that direction is somehwere (sic) between a real possibility and a certainty.”9
The Internet allows artists to reach their listeners directly. Listeners can provide feedback at the click of a mouse button. They can use the same mouse button to purchase music directly from the artist. In effect, the Internet combines the best of the patron and performance systems. The entire audience can function collectively as the artist’s patron: each individual listener funding the particular aspects of the artist that he likes, each listener encouraging the type of performance he funds.
Finally, artists will return to a situation where they can net much of the gross.
The benefit for recording companies is very simple: near-zero marginal cost of distribution.
Typically, this would be a business’s dream. Why it has apparently become a nightmare is the subject of rampant speculation.
Perhaps it is because recording companies want to protect their current mode of operation. Lars Ulrich summarized the role of recording companies very simply:
“Because what really, essentially, is a record company? A record company is really essentially a bank, a bank that funds a bunch of money to make records, and videos and promotion, publicity appearances and so on, and they take that shot that one day the artist is going to be so successful that they’re going to first of all get all their money back, second of all make a profit.”10
Recording companies are in fact exactly like venture capitalists. They fund, promote, and advise their portfolio artists. What they mainly do is everything necessary to distribute music in the physical world. They maintain relationships with radio stations, record stores, and they arrange for manufacture and shipment of CDs.
Of course, the Internet changes that last bit. If communities such as Napster and Gnutella are really successful, then perhaps the market for music on physical media will shrink. But is that a bad thing? Environmental considerations aside, a near-zero marginal cost distribution system should be a boon to the companies which own the copyrights to popular music. They would be able to sell the intellectual property and net the gross.
4.Retooling media distribution
Exploiting new technologies often requires retooling. Automobile manufacturers, computer chip manufacturers, the United States Postal Service…. All have had to continuously change their methods of doing business in order to remain competitive. In the efforts to retool have proven themselves time and again to be worthwhile. Costs of production and distribution decline, profits increase. The music industry is not exceptional. Nor is the motion picture industry.11
The only thing missing is a method for the record companies to financially exploit the Internet for what it is: the best intellectual property distribution mechanism yet seen on Earth. So here is an idea: turn pirates into paying distributors.
Turning pirates into paying distributors
Napster and Gnutella are giant music distribution networks. Every file-sharer is in fact a voluntary distributor. These communities are like potlucks. Everyone brings what he or she wants to share, and perhaps partakes in what others brought. Right now it all happens for free, mainly because there is no infrastructure in place to do it for profit.
Installing that infrastructure is a small matter, and its effects are huge. Tracy Scott’s method of turning pirates into paying distributors is elegantly simple, and its success is predicated on only two assumptions. First, Internet connections have finite capacity. That’s not just an assumption; it’s a rule. Second, people have pecuniary motivation.
Suppose I have a finite-capacity Internet connection and lots of music to share. Lots of people like to download the music I share. Now suppose I could charge each downloader in a way that made it easy for them to compensate me for the music I am sharing with them. I would, of course, also compensate the record company for the resale of their intellectual property.
Now, because I have a finite-capacity Internet connection I would seldom, if ever, allow someone to download from me for free when I could charge for the privilege. If I charged 1.50 USD for each track, and split the revenues evenly with the record companies, the record companies would make 0.75 USD for each download! They would never have to promote or distribute their products themselves again.
You and I would be doing our very best to promote and distribute their product for them, as we would have a profit motive. Multiply this over the thirteen million you and I’s on Napster alone, and the profits for distributors and copyright holders are staggering.
Do or die
Technology has interesting effects. Those who realize how to integrate and exploit new technologies gain potentially huge advantages over their competitors. Those who do not will acquire an equal disadvantage with respect to their competitors.
Intellectual property profiteers are now at a crossroads. If they adapt quickly, as they have in the past, they can leverage the Internet to revolutionize their industries. If not, they may find grassroots efforts invading their bailiwicks.
It’s happened before: phonographs ultimately led to the demise of the player-piano music industry. The forward movement of technology always leaves behind the stragglers.
5.Technology is just technology
Technology is neither good nor evil. Individuals choose how to employ technologies. Some choose for good, others for evil. Some use automobiles as a conveyance. Others use automobiles to rob banks and kill people. To boil it down into a popular refrain: guns don’t kill people, people kill people.
To adapt the saying for the discussion at hand: Napster doesn’t pirate music, people pirate music. In fact people have pirated music in large scale for decades. What’s happening on the Internet is nothing new.
Napster and Gnutella are on the one hand user communities. These users are the ones who are possibly pirating music. On the other hand, Napster and particularly Gnutella are notable technologies which are changing the way we look at the Internet.
Certainly these technologies are not running rampant on computer networks seeking out the latest Eminem tracks and pirating them. Humans are doing that.
People routinely use other means to distribute music over the Internet as well. Email, FTP, Usenet, IRC, ICQ, etc. All are used to distribute music on the Internet.
Keeping in mind that music is only a current-day analog for all types of intellectual property, we will spend a little time analysing how duplication of intellectual property is inevitable, and why it makes sense to use the carrot instead of the stick.
When my family got its first IBM PC/XT in the mid-1980’s, the first thing on my mind was games. In those days copy protections on games were very creative. They involved placing magic bits of data at secret parts of the floppy disks on which the games were distributed. For some time it was impossible to pirate those games. That is, until more advanced disk copying software was developed. After that, there were secret codes and product registration keys. Even physical keys (dongles). All have been defeated, generally on the day of the software’s release.
Copy protection doesn’t work
Certain cable television channels are scrambled. You can either pay fairly high recurring fees to watch those channels, or you can purchase a relatively inexpensive descrambler to watch those channels. More recently, DSS satellite television signals are encoded. So savvy consumers reprogram the DSS card to circumvent the encoding.
Protection schemes seldom work. Encryption for Microsoft WMA format files was broken almost immediately after its release. The process was incredibly simple. One would purchase the right to listen to the encrypted audio file. Play it back through special software which records the decrypted audio file, and mission accomplished12. You now have on your hard disk a permanently decrypted audio file.
SDMI, or the Secure Digital Music Initiative, is what many record companies are betting on to preserve their control over music distribution.
My suspicion is that people in significant numbers will choose to re-encode SDMI music into MP3 (or some other freely distributable format). In fact, if even a few people re-encode and distribute a song, the fluidity of information-sharing communities will ensure its rapid and extensive duplication.
One onerous tactic I have heard is posited end-to-end encryption. The idea is that at no point is the media unencrypted. The file is encrypted. The data travelling from your hard disk to your digital speakers or your digital monitor is encrypted. Currently even encrypted media formats are decrypted long before the data makes it out of your computer. In end-to-end encryption the data is encrypted right up until it is presented visually, audially, or otherwise.
It’s an interesting idea that at first appears to put a stop to piracy. But in my mind there are few if any panaceas that are predicated on authoritarian control, and end-to-end encryption is not among them.
End-to-end encryption is easily defeated. Currently, movies are often copied using a method commonly termed “telesyncing”13. If end-to-end encryption became a reality, audio and video would surely be telesynced. The encryption would be rendered meaningless.
The end-to-end part of the end-to-end encryption idea is misleading. Since humans don’t have decryption systems built into their anatomy, information must be deciphered before we experience it. And that is the failing. The only way to make music that cannot be copied is to make music that cannot be heard. The only way to make movies that cannot be copied is to make movies that cannot be viewed.
Napster and Gnutella are just the first wave
Napster and Gnutella are but the first of a succession of technologies which will make it increasingly difficult to control the distribution of intellectual property.
Napster was the first. It involved a central server, which has demonstrated that it can be a point at which controls can be applied14. Gnutella was second. It involves no central server, eliminating the possibility of easily controlling the habits of Gnutella users by strictly legal means15. Gnutella is only pseudo-anonymous. FreeNet corrects that. It, like Gnutella, is fully distributed with no central server, and it is completely anonymous.
If laws are enacted against these technologies, the ensuing replacements for these technologies would only be more difficult, if not entirely unfeasible, to police16. This is only the beginning. 7.What should we do?
A zero-marginal-cost means of distribution is a rare opportunity. It should be seized and exploited. Tracy Scott’s viral marketing method is a clear and simple way to give incentive to Internet music sharers to promote the legitimate sale of music. The current crop of technologies should be encouraged and adopted, not restricted or abolished, lest lawmakers and industry leaders wish to bring forth truly intransigent technologies.
We’re on the precipice of a slippery slope. The toothpaste is already out of the tube. It can be exploited nicely, or be turned into a huge mess.
- Software and Information Industry Association (http://www.siia.net), 9 July 2000. The SIIA claims a loss of 59.2 billion USD in the past five years with 12.2 billion USD lost in 1999 alone.
- American universities are commonly blessed with extremely fast Internet connections. They are also populated by young, technologically savvy people with few financial means.
- Popularly encoded songs (128 kbps at 44 KHz) are approximately one megabyte per minute.
- MP3 players are numerous and varied. Software-based players include the popular Winamp (http://www.winamp.com). Hardware players include the Walkman-like Diamond Rio and the in-car empeg (http://www.empeg.com).
- Gnutella is a fully distributed network comprising individuals and computers which actually build the strength of the network as they join it. Without users, Gnutella is nothing but a definition of how computers might communicate over a network to exchange information. Gnutella is developed by hundreds of individuals around the world. Numerous Gnutella-compatible softwares are released with complete source code, increasing its appeal to technologists.
- I have searched for two years for John Denver and Johnny Cash at San Francisco Bay Area music outlets (Tower Records, Wherehouse, Fry’s, etc.) and have been unsuccessful. I suppose my tastes are a bit unusual, but I heard their songs as a child, and wanted to hear them again. Strangely, even immediately after Denver’s death, no record stores carried his music. No stores even had a slot for John Denver! On Gnutella, I found hundreds of Johnny Cash and John Denver tracks. Similarly on Napster. If nothing else, this demonstrates the power of the Internet to assist consumers.
- Ripping is the process of extracting the data from a compact disc into a manipulatable file. After a CD is ripped, it is typically compressed and encoded into a space-saving format such as MP3.
- Courtney Love has recently made statements to this effect at the Digital Hollywood Conference. A summary can be found at http://rollingstone.lycos.com/news/newsarticle.asp?ID=10847&Artist=23 (9 July 2000).
- Metallica drummer Lars Ulrich on Slashdot.org (http://slashdot.org/interviews/00/05/26/1251220.shtml), 26 May 2000.
- Historically, both the music and motion picture industries have faced challenges from recordable media. One example is the video cassette recorder, which found motion picture companies suing Sony over Betamax. Video cassettes have since proven to become a huge profit centre for the motion picture industry by way of sales and home video rentals. After all, what would moviemakers do with straight-to-video productions if there were no video cassettes? Perhaps some day we will have straight-to-Internet music.
- This piece of software was interestingly called UNFUCK.EXE and has made its way around the Internet like a brushfire. Copies of it are largely found outside US borders, where the road to stamping it out is rife with jurisdictional hurdles.
- Telesyncing is crude, at best. The process is simple: a person enters a movie theatre with a video camera and tapes the movie. Later he digitises his tape and releases the results on the Internet. A large network of pirate movie FTP (File Transfer Protocol) sites propagate the digital media worldwide.
- Both Metallica and Dr. Dre have exploited Napster’s centralization in their campaigns to thwart piracy on the service.
- The idea that the RIAA or its designates may “spam”, or purposely overburden the network to cripple it, has crossed my email inbox numerous times. Fortunately for the legitimate users of Gnutella, spam and its relatives have been outlawed.
- Perhaps when Vice President Al Gore created the Internet he never thought it would come to this, but then, Senator Trent Lott probably didn’t think of the myriad uses for paper clips either.