Amazon's Patent Reform ProposalJeff's open letter to Amazon customers [ The letter is no longer available on amazon.com. ] about the patent issues is a great outcome for the discussion we started this past week. After all, if Time Magazine's Man of the Year says that the patent system has gotten out of hand, that may well have more impact on the powers that be than if 10,000 relatively anonymous programmers do so. So I'm really excited to participate with Jeff in his call for a closer look at how we can update the patent system for the Internet age. We're both complete beginners at getting the attention of Washington, and even if we were experts, there is no promise that we'll get results, but I do agree that the reforms that Jeff has suggested would be a great first step towards rethinking the system.
"Rethinking" is of course the key word. Jeff's suggestions are one way to address the problem. Others who know more about the patent system may have even better ideas. We must start by recognizing that we have a problem. Once we do that, we can engage technical, legal, and regulatory experts in an effort to solve that problem. As Jeff points out, that may well be a long process--but not one that is more difficult or more contentious than many of the technical infrastructure issues of the Internet that have been solved by groups such as the IETF or W3C.
I do want to commend Jeff for listening to his customers. He didn't just hide behind his lawyers, or a PR spokesperson, but engaged directly, demonstrating that he does understand the new rules of the Internet economy. I first spoke with him less than 36 hours after I posted my open letter, and we've spoken at least a half dozen times since. Many of those conversations have not been during traditional business hours--he's keeping hacker's hours, like many of the rest of us! He's read a lot of the commentary left on oreilly.com, as well as letters sent directly to Amazon. I can't imagine any other high profile CEO who would be as forthcoming.
While Jeff hasn't done what I originally asked for--to rescind his patent claims--he has most definitely engaged with the problems I was raising, thought seriously about them, and proposed an answer that works for him and his business.
That being said, I don't want to let Jeff entirely off the hook. One thing about a call for action in Washington is that it could be seen as just a way of shifting the focus away from Amazon and onto the PTO. I don't think that is what Jeff is trying to do--but one way to tell that for sure is by the amount of followup we see over the next few months. But in any event, it is most certainly true that the software patent problem is far larger than Amazon's patents. Large as they loom in e-commerce, Amazon is a very minor player in the patent game. Any solution that simply involves Amazon and does not spark larger reforms will be of little importance.
I also do want to point out that I've learned a lot about patents in the past ten days. In talking with Jeff about the details of his patents and why he thought they were original, I was struck by how different his sense of what he had "invented" was from the sense I got by reading the patents themselves, and from commentary I'd read on the net (including my own :-).
In the case of 1-click, the patent claims look to a casual reader as if they broadly cover the use of saved state to make it possible to conduct an e-commerce transaction without forcing the user to identify him or herself. In fact, they cover only the single "point and click" aspect, such that the sale is made without any confirmation step. In short, this patent is far more narrow than it might at first appear. And in fact, Amazon did an incredibly successful job of making 1-click an easy-to-use feature. Most examples of "prior art" that readers sent in to me had various confirmation steps, and were not "1-click" approaches with anything of the slickness that Amazon brought to the table. One e-commerce pioneer claimed that he'd tried a 1-click type of approach several years before Amazon, but had given it up because customers found it confusing. He admitted that the way Amazon implemented it was a significant advance over the way he'd done it. As Jeff has claimed, in hindsight it looks easier to get this right than it did at the time.
In the case of the Associates patent, what is being patented is not the broad idea of referral marketing, but instead a mechanism that allows individuals on the net to establish a little virtual bookstore on their site, with fulfillment by Amazon, entirely on their own, without having to negotiate a business deal with Amazon. In effect, Amazon created an API that allowed others to use Amazon as a service. This was in fact a really nice piece of engineering, and a foretaste of things to come as next-generation web sites start to publish XML-based APIs for content syndication or for business transactions. Jeff points out that at the time the Associates program was introduced, there was plenty of commentary by folks like Esther Dyson hailing it as a real innovation.
But the more I became convinced that Amazon's patents might actually hold more water as original inventions than I thought, the more I became convinced that they were still a bad idea.
I banged this drum fairly loudly in my original postings, so I won't repeat myself here. Instead, I'll quote Harvard Law school professor Lawrence Lessig, author of the recent bestseller Code and other Laws of Cyberspace.
In an April 1999 article in The Standard, Lessig wrote:
No doubt we are better off with a patent system than without one. Lots of research and invention wouldn't occur without the government's protection. But just because some protection is good, more isn't necessarily better. Especially in cyberspace.
This last point is really key. We've got a good thing going. The free-wheeling development environment that has given us both the Internet and open source tools like Linux, Perl, Apache, MySql, Sendmail, and so on, has demonstrated convincingly that there is enormous power in an open, shared platform where ideas are given away as the foundation for further innovations. So far, entrepreneurs, stock market investors, and computer users have all benefited enormously from the open, non-proprietary approach historically used for developing internet technologies.
In this second article, Lessig wrote about the ever-more slippery slope of business method patents:
You don't have to be against all patents to believe that there is a reason to be concerned here. For contrary to the it's-all-OK-sayers, there is something new about patents in cyberspace. While the old problems with patents remain--for example, can the patent office keep up with the pace of change--there is a new kind of patent that is sweeping technologists in the Valley. Yet we know nothing about the effect this patent will have.
That is a sentiment with which I believe most of the 10,000 people who replied to my original call to action, as well as the many hundreds of thousands who have discussed this issue on the net over the past week, would heartily agree.
Return to: Ask Tim Archive