In case you haven't heard of BountyQuest, let me summarize. BountyQuest is a Web-based service that connects those who seek prior art (earlier publications covering the same subject) to invalidate a patent with those who might be able to find it. "Posters" (people or companies who want to invalidate a patent or otherwise establish prior art in an area of technology) post a minimum bounty of $10,000 on a piece of technology, normally a patent. Posters usually list both the patent number and a description of the art they want to invalidate. "Bounty Hunters" who have registered with BountyQuest for free can submit anything in the public domain that they think describes the same technology. The first Hunter to submit prior art that meets the Poster's criteria wins the bounty. [Editor's Note: For more information, see Nancy Lambert's Better Mousetrap column in the January issue of Searcher magazine or at http://www.infotoday.com/searcher/jan01/lambert.htm]

CEO Charles Cella launched BountyQuest in October 2000, with a substantial portfolio of initial bounties, most of them in the areas of computers/Internet/business methods and biotech/pharmaceuticals. A bounty's minimum posting time is 2 months, so late December was the soonest that BountyQuest could announce the first winners. In fact, the first announcements came on January 30; as of mid-March, BountyQuest had awarded five bounties. Details appear at http://www.bountyquest.com/winner/winnermain.htm.

I found it interesting that, although both Posters and Hunters are guaranteed anonymity unless they give written permission to reveal their identities, all but one of the initial bounty winners have gone public. Also interesting, and tending to bear out Cella's opinion on where the prior art would come from, is that none of the first five winners is a patent-information specialist. All five work in the areas in which the bounties were posted, and three of them submitted their own or their companies' publications. Another consistent trend: All the bounties paid so far are for Internet, computer, and business method patents.

But the big news is the conclusion to one of the first bounties posted: US5960411, Amazon.com's patent for 1-Click online shopping. Tim O'Reilly, head of O'Reilly & Associates (a publisher of books on software and the Internet) and a vocal critic of Internet patents, posted the bounty. Interestingly, Jeff Bezos, founder and CEO of Amazon.com and one of the inventors listed on the 1-Click patent—as well as an investor in BountyQuest—had no objections to the posting. He felt his patent was strong enough to survive the test and would, in fact, be strengthened if no one found close enough prior art to invalidate it.

BountyQuest deemed the end of this posting worthy of a telephone press conference. The teleconference on March 14 started with Cella giving a summary of BountyQuest's activities and declaring them a "tremendous success." Of the first 25 bounties posted, five have produced winners. However, he said, the bounty on the Amazon.com 1-Click shopping patent closed without a clear winner. Thirty submissions came in against it, but not one of them did what the bounty required—that is, invalidate every claim of that patent.

Cella did state, however, that the submissions included "very significant prior art" that will have an effect on the "patent going forward." The art made it clear that others had long ago come up with one-click shopping in different contexts than the Internet. For instance, one of the submissions included marketing materials from a bank in Montreal that allowed single-click ordering of checks. Another involved one-click ordering of items in TV ads via a TV remote control. Less serious submissions also came in: one citing a "Doonesbury" cartoon series on virtual-reality one-click shopping, another citing an episode of Cheers in which Norm orders his beer with one gesture, and one referring to Captain Jean-Luc Picard ordering Earl Grey tea from his replicator on board the starship Enterprise. Details of all the submitted art also appear on the BountyQuest Web site.

BountyQuest's stated policy is to award bounties only when a single piece of prior art fulfills the Poster's requirements. In this case, the requirement was to invalidate all the claims in the 1-Click patent. Members of the patent-information community have objected to this policy from the start, and in the teleconference, several call-in questioners pointed out that it's not realistic. A single piece of art can invalidate a patent for lack of novelty, but non-validity is often established with multiple pieces of art which, taken together, present an argument for obviousness. In fact, it would be almost impossible for a single piece of art to invalidate all of a patent's claims. Nor is this policy fair to the Hunters, who would receive no payment for submitting art that could be of great use to the Poster even if it didn't meet the exact criteria of the posting.

Pressed on this issue, O'Reilly agreed that he would award the $10,000 to those who submitted the nearest prior art. BountyQuest later announced that it would split the bounty among three Hunters. The winning art includes US4734858, which describes the use of a remote data terminal to place orders; US5303393, which describes a simplified way of placing an order via a two-way radio system; and European patent EP680185 (U.S. equivalent US5819034), which actually describes one-click shopping within the patent specification. The relevant text of this patent, compared to a section of the summary of the invention in the Amazon.com patent, appears at http://www.oreillynet.com/pub/a/patents/2001/03/14/patent.html.

EP680185 was considered the closest of the art submitted. But what does that signify? What impact will this and the other art have on the future of the 1-Click patent? Some of O'Reilly's statements showed that he considered the art significant. During the teleconference, he actually said that he felt the European patent did knock out the Amazon.com patent's broad claim. Afterward, in a letter on the BountyQuest Web site dated March 15, he said: "This was a tough one to win because the Amazon 1-Click patent is so specific to the Web. But these three people submitted terrific prior art that narrows the scope of what Amazon can now claim. In my opinion, it will be much more difficult for them to enforce their patent from here on out…. Although none exactly matched all the elements necessary to outright win the Bounty, the three runners-up came very close. Their submissions should go a long way toward making the 1-Click patent virtually unenforceable. (Not a legal opinion.)"

On the other side of the coin, in an open letter again available from the BountyQuest site (full text at http://www.oreillynet.com/pub/a/patents/2001/03/14/bounty.html) dated March 14, O'Reilly said: "[B]ecause of the complexity of Amazon's complete patent, and because it is specific to the Web rather than to these earlier media, we weren't able to match every claim. And, in fact, all the prior art that was submitted specifically for the Web [my emphasis—NL] confirms Amazon's belief that they were doing something original…. So, I want to offer Jeff something of an apology. Amazon may not have deserved a patent on 1-Click … but they appear to have been staking out new territory in ease of use for Web shopping." Then he flipped the coin again, saying, "However, we do think that the prior art that we found sufficiently narrows the scope of what Amazon can claim [so] that it will be much more difficult for them to enforce their patent."

In other words, although O'Reilly admitted that the art submitted for this bounty hunt did not invalidate all the Amazon.com 1-Click patent's claims, he still felt it may have a significant impact on the broad claim. Amazon.com spokespeople, not surprisingly, say that because the hunt failed to turn up prior art that invalidated all the patent's claims, "We remain confident that the 1-Click patent is valid."

So, there is some confusion as to whether the art submitted will impact the 1-Click patent enough to deserve the bounty, or even whether O'Reilly believes that. The courts will decide on the fate of Amazon.com's patent, of course. But it occurs to me that awarding this bounty, despite the uncertainty, serves BountyQuest's best interests. First, BountyQuest wants to publicize how effective its system is in finding art that professional searchers might miss. Second, it probably realizes that it has to get away from its one-piece-of-art-must-do-it-all policy, and this is a good start.

Most interesting of all, at least for professional searchers, was the way different news services treated the story. In some cases, incomplete and hurried reporting produced stories that seemed almost biased. A Reuters article, entitled "Amazon Patent Naysayers Come Up Empty," mentioned that none of the art submitted completely matched Amazon's 1-Click technology and quoted O'Reilly as saying, "There was no winner, mainly because the 1-Click patent is specific to the Web…." It failed to quote O'Reilly's statements that he felt the art submitted could impact Amazon's enforcement of the patent. A New York Times article, entitled "Patents: No Claims on Amazon's ‘One-Click' Shopping Device," was similarly pro-Amazon. It, too, stated that none of the art submitted addressed all the claims in the patent. It quoted the Amazon spokeswoman's statement that the BountyQuest contest vindicated Amazon's 1-Click patent. It quoted only the apology section of O'Reilly's open letter. Nowhere did it show the flip side of the coin.

On the other hand, Newsbytes and InternetNews (services more specific to the computer industry) presented much more balanced stories, quoting both pro-Amazon and pro-BountyQuest statements. This is an interesting lesson for those of us in the business of evaluating sources: On a technical story, you're better off going to the selected technical sources than to the more general ones. Even that advice bears a caveat, however. For example, the coverage by CNET, a prominent computer-industry portal service, carried the less-than-complete Reuters story.

The issue remains whether software/business methods should be patentable at all. During the teleconference, O'Reilly talked about his concern with the "chilling effect on innovation of patents in general." Although he realizes that good arguments exist for patents in areas where research requires large investments, O'Reilly does not feel that the same arguments hold when applied to software, business methods, and the Internet. He asked if we want to have a freewheeling technical culture or one in which programmers must work with lawyers and other programmers looking over their shoulders. In his open letter, he said: "The fact that the idea of 1-Click shopping was even lampooned in cartoons years before Amazon's ‘invention' suggests that the bar on obviousness has been set far too low. The simple transposition of existing or obvious ideas onto a new medium such as the Web should not be treated as an invention worthy of government-backed monopoly protection. Amazon's innovation was not an invention so much as it was a bold business move, which we should all applaud—and be free to imitate!"

In the teleconference, Cella pointed out that software patent claims are usually ambiguous, even to experienced patent lawyers. However, patents identify what you may not do, while the public domain tells you what you may do. So, Cella said, if you find prior art that describes what you want to do (and is not itself an active patent), then go ahead and do it. He feels that BountyQuest has torn down some fences with art submitted that won this and the other bounties, and tearing down fences is what BountyQuest is all about.

An interesting side note: The United States Patent and Trademark Office (USPTO), no doubt aware of the controversy in this area, has announced a considerable slowdown in issuing business-method patents. In March, it instituted a new policy requiring a second review for these patents. During the last quarter of 2000, the USPTO issued only 36 percent of the applications it examined in this area (compared to 72 percent of applications across all technologies). The USPTO is also developing extensive training, including field trips to appropriate industries, to help patent examiners learn which processes are routine and obvious and therefore ineligible for patenting.