In a previous life, I developed Digital Rights Management (DRM) software. With the rise (and decline) of Napster and other forms of P2P file sharing, DRM has suddenly become in vogue again. It seems that DRM, along with its cousin copy protection, is on a five year cycle. About every five years digital information copying becomes a real industry worry, companies implement a DRM or copy protection solution, that solution fails, and then DRM goes away for another five years.

This issue first became important while the PC was a new idea on the market. I recall purchasing software that had some form of copy protection included. One mechanism required a hardware “dongle” to be plugged into the serial or parallel port of the computer. Others required the original disk (with hidden files) to be inserted to start the software. Yet another scheme used a laser hole in the distribution medium to assure that a particular sector of the floppy couldn’t be written to or read from. Overall, these schemes generally failed. Why? Because five minutes after the scheme was invented, someone developed a means to bypass it. Again, these schemes ranged from the obvious to the sublime — I even recall traipsing through the assembly code of some application to remove the calls to the routine which talked to the laser hole on a diskette.

Later, copy protection schemes required answering some question from the installation guide (these even exist today). I’m not sure that anyone could develop a more perverse mechanism than that. I actually learned all of the capitals of the republics of the then-Soviet Union so that I could play my copy of Welltris without having to drag the book around.

Today this sort of thing has become fashionable again as the recording industry seeks to protect its profits by preventing people from copying music files. This was a big enough issue that it took down the king of MP3 sharing systems — Napster. (Sometime in the future I’ll post a retrospective on the legal defense of Napster.) Napster was easy with their big central database that was used for searching, but other tools based on completely decentralized searches (Gnutella, LimeWire, etc.) will be harder to take down.

And so we come upon the latest installment of Digital Rights Management. The recording industry, seeking to ensure that music doesn’t get shared, is going to put anti-copying alterations on CDs and is attempting to release music to the net by selling licenses to it. Even Windows Media Player includes license management to protect the redistribution of music and videos.

There’s a real problem with all of this however. First of all it’s a losing battle. The minute one scheme comes out, someone comes up with a way to break it. Generally speaking, it’s not that complicated to attack DRM because ultimately the unprotected goods have to be sent to your computer. From there, it’s only a matter of capturing the unprotected data. The five year cycle of DRM occurs because of this.

Second, it’s a bad thing in general. Today most content is protected by copyright. The beauty of copyright is that after a certain period of time, the content passes into the public domain. The founding fathers of the US did this (both for copyright as well as for patents) so that our culture would be enriched over time. However, DRM could actually undo virtually all of that.

DRM offers the opportunity to specify the terms and conditions for use of the content — a license. This is how software is distributed today — under license. If music and books were distributed this way, the works would never pass into the public domain. The terms and conditions would be spelled out and the buyer would have to agree. Only those who agreed to the license would ever be able to use those works. Imagine if the license also included a prohibition on criticizing the content (for example, you would never see a bad movie review). That may seem unrealistic to you now, but it’s wholly possible.

While the debate rages over MP3 and music sharing, the momentum is still against DRM schemes. Recently the recording industry was criticized for attempting to copy protect a recently released CD. In fact, I believe that we are at the point where the recording industry, represented by the RIAA, needs to recognize that a certain amount of music is going to be shared. To account for this, the RIAA membership should lower the price of music CDs. I believe that the high price of CDs will continue to drive the music lovers away from the record stores and onto the Internet. While bandwidth is still at a premium for many people, this is the time to get people used to paying 10 to 25 cents for a song that they can download. For a quarter, it’s hardly worth waiting hours to download via LimeWire, yet high quality audio could be made available via high performance servers that would counter the low performance and unpredictable quality of the peer networks

Clayton Christensen, who developed the seminal theory on disruptive technology, examines why companies fail and why theory trumps data. From the article:

Businesses get blindsided because they focus on their best, most profitable customers and ignore other potential markets or customers seeking lower-cost products. This narrow view, Christensen says, ignores the fact that every market is characterized by three distinct change trajectories:

• Performance improvement that customers can readily use (that is, it matches their own changing needs).
• Technology advances driven by sustaining technological improvements.
• New performance introduced by a disruptive technology, which typically begins at a lower level of performance, but rapidly improves until it meets the majority of customers’ needs.

I think this is an interesting market market observation. Markets change by improving performance of existing solutions, improving existing technology or introducing disruptive technology. If you don’t change with them, then you stand a good chance of tanking. This is what lead to the end of the mini-computing era — PC’s were disruptive technology that ended the life of mini-computing companies. There were no longer significant performance improvements or sustained technological improvements that could prevent the PCs from taking hold of the market.

You can find the full article on CNet News.

I admit that I’m no different than the rest of the population of this country when it comes to trying to understand the events of 11 September 2001. It’s almost impossible to fathom why the attacks would occur and what the effect will be on the future. One of the core questions that we want to answer is the one of cause. What was the chain of events that occurred that caused the terrorists to take the actions that they did? As it happens, this discussion relates closely to one of my areas of interest — Complexity Theory.

Complexity theory, rooted on Chaos theory, basically draws from the idea that things are interconnected in such a way that the smallest, seemingly insignificant events can create significant emergent behaviors. Edward Lorenz, a meteorologist, described what is considered the classic example of this — a butterfly’s wings flapping in Brazil can create a tornado in Texas [1]. While that may seem implausible, consider that weather patterns are profoundly complex and events surrounding them have a tremendous degree of interconnectedness that makes weather prediction almost impossible.

Complexity talks about collections of independent, goal-oriented individuals self-organizing into self-consistent systems with emergent behavior. Emergent behavior is the idea that new results can occur while the individuals are doing their own best efforts. In a sense, the whole becomes greater than the sum of the parts. If you think about human beings emerged from dividing and differentiating cells, it becomes quite apparent. If cells divided and differentiated and combined to form a human being from a complete, detailed set of instructions (as most theories held), the incidence of birth defects would be astronomical. That’s because detailed, continuous sets of instructions tend to produce drastic outcomes if anything in the process gets upset. (Think about assembly lines in a factory where one misshapen part of the car destroys the entire car itself, e.g. bad tires.) However, almost all human babies are normal and the incidence of defects is pretty rare. That’s evidence that in fact the system of producing life is about very simple organisms (cells and genes) following very simple instructions that, when brought together, creates that system from which a human emerges. [2] The fact that all humans are constructed in virtually identical fashion, yet we are all distinct, backs up that theory further. This theory is essentially fractal — it works at all levels, from atoms to genes to cells to people.

People in society organize in the same fashion — each of us, with our individual objectives and goals, is connected to each other and the actions that we take affect those around us and the emergent behavior of all of us together. That means that we are all interrelated — our actions affect one another, regardless of how significant that action may seem.

Complex systems self-organize — in effect, you get order for free. Order naturally rises out of complexity and chaos. That’s not to say that bad things can’t happen as an effect of this self-organization, because they certainly can. But it also means that we are actually perfectly at home in this world — the things we do in life are patterned up and down the cosmic ladder. Our society acts in similar fashion to collections of cells or perhaps even collections of stars. The ability for self-consistent, self-organizing systems with emergent behavior from independent actors is a powerful and pervasive concept. The independent actors in the system (cells, genes, people) likely don’t work directly towards the overall goals of the system, if they understand them at all, yet as a collective achieve powerful results.

It’s hard to know if the incidents on 11 September are significant in the overall scheme of things. Even if those incidents are significant, it’s likely that as individuals within the greater complex system, we will fail to understand them because we don’t understand the behavior that is supposed to emerge. But one thing is certain about these events — our individual actions, no matter how large or small, in some way caused those incidents to occur. Just like tornados in Texas, the smallest, seemingly insignificant things that we do in our day-to-day lives have profound effects because of our tremendous degree of interconnectedness. That interconnectedness, however, creates the complexity that makes it impossible to identify the particular causes of these events and, even more so, makes these events impossible to predict.

And that may be as close to understanding 11 September 2001 as we ever get.

[1] Sardar, Ziauddin and Abrams, Iwona, “Introducing Chaos”, Totem Books, 1999, p54.
[2] Waldrop, M. Mitchell, “Complexity: The Emerging Science at the Edge of Order and Chaos”, Touchstone, 1992, p.106.

p.s. Most entries won’t be this long — I may convert this into an essay and just post the link here. (Is there a rule that says that weblogs can’t be edited once posted? :-)

Just getting the web log on-line today, finally. I’ve been putting it off because because frankly I haven’t had much time. But it’s probably worth taking the time to jot down my observations, both technologically and otherwise.

By way of introduction, I’m Jeffrey Kay. I’ve been developing distributed agent-based systems for the past several years and along the way have been promoting peer-to-peer software. I’ve worked with the World Wide Web Consortium as an advisory committee representative. Recently I spoke at the O’Reilly P2P and Web Services Conference on the technology I’ve been developing. Look for more rants here on P2P, Web Services, and technology in general.

I recently took a weekend motorcycle journey, which included a trip through Skyline drive. As you can see, the weather was poor — I encountered rain, sleet, pea soup fog and very cold weather. But you know what they say — the worst day riding is still better than the best day at work. In case you are wondering, I ride a ’96 Honda Shadow VT1100C — the last of the four speeds.