How many inventors does it take to patent a light bulb?
One might hope that the answer is "only one," but in fact the right answer, or at any rate the historically correct answer, is more like three. While every schoolchild learns that Thomas Edison invented the light bulb, the history of the light bulb is fraught with more litigation than one would guess.
As a point of fact, he fought two other inventors in the patent courts for
years, ultimately winning more through perseverance than priority (it is
almost certain that an Englishman, J.W. Swan, had come up with the idea ahead
If you happen to recall, hazily, this encyclopedia tidbit of technological lore, it is likely to pop back into your mind as you read about Jay Walker. Walker is the Connecticut entrepreneur whom Forbes magazine in May called the "New Age Edison" -- comparing his intellectual property incubator, Walker Digital, to Edison's own invention factory. Nobody has been more assiduous in patenting Internet business ideas than Walker.
The first of these patented ideas that Walker has turned into a business is Priceline, the Web site that lets customers name the price they are willing to pay for everything from airline tickets to cars. That patent and the business built around it have already made Walker enormously wealthy -- Priceline is worth nearly $10 billion, and Walker's stock holdings are worth over $4 billion, if only on paper.
Walker is a pioneer in the game of staking a claim to intellectual property. He holds 18 patents on Internet business ideas and has about 250 more pending. This in itself makes him something of a controversial figure, because among the intellectual property cognoscenti "business process" patents and software patents are a very new knot in an otherwise well-worn fabric. From the moment that Priceline, with much hoopla, revealed its patented business, observers have been predicting that such patents, rather than fostering genuine new ideas, were more likely to result in litigation as companies tried to block each other's businesses.
Walker himself has been accused of patenting the merely obvious. Bill Gurley, then a high-profile securities analyst and now a venture capitalist, argued bluntly that seven months before Priceline launched, he'd outlined a business model almost identical to Priceline's in a Fortune magazine column. "How can an idea that so easily pops into the head of an Internet pundit be considered 'unobvious' to the general Internet practitioner?" wrote Gurley with evident sarcasm.
Now, Walker finds himself under attack in the courtroom. A San Francisco company called Marketel is suing Walker and Priceline, alleging that Walker stole its trade secrets. Meanwhile, Walker Digital is accumulating new patents, each seemingly broader than the last, that would theoretically give him ownership of the business models common in whole industries.
In looking closely at three of Walker's key patents, a pattern emerges. In two cases, the core Priceline patent and a patent for "expert based commerce," Walker's inventions sound remarkably similar to ideas whose existence Walker was aware of but whose inventors didn't bother to run to the patent office. In another, an almost stunningly broad patent for anonymous communication over a network, Net businesses were already acting on remarkably similar ideas not only before the time the patent was granted (that's a process that can take several years), but before Walker even applied for it.
Well, Edison didn't come up with all his inventions first, did he?
The Marketel lawsuit gives a good sense of the peculiarities of Walker's methods of invention.
Until a Fortune story last week, major publications did not give the suit much play. Walker's personal charm is considerable; he tends to shrug off jabs like Gurley's by patiently explaining that what his detractors come up with is no more than a vague idea -- the equivalent of writing an article about how biotechnology could lead to a cure for cancer and then taking credit for actually curing cancer. The air of invincibility that he has acquired in the business world as the reigning pacesetter in a whole new field of intellectual property is almost unpierceable.
Then, too, the litigants involved do not necessarily inspire much confidence. The company that is suing Walker, Marketel, does not operate a full-blown consumer business but is involved in the development of intellectual property. The principals will provide no details, claiming that to do so would jeopardize projects in development. The central figure in the lawsuit, Marketel president William Perell, already projects the almost too helpful air of a professional litigant as he politely declines to speak about the case on the advice of his lawyers. As evidenced in a letter that is included in court filings, Marketel's CEO, Byron-Eric Martinez, actually faxed Walker in 1998 looking for a job.
And yet the facts of the lawsuit are serious indeed. Marketel's lawsuit against Priceline has its origins in BookIt!, an airline ticket reservation business that Marketel launched in 1991. BookIt let customers send in requests for airline tickets by fax, guaranteeing their bid with a credit card. If an airline was willing to sell a ticket for the price a BookIt customer bid, the deal went through. The fax system is similar in concept to the airline reservation system that Priceline launched in 1997 (there are also significant differences; BookIt, for instance, charged users to submit a bid). It is sufficiently similar that articles about Marketel are cited as "prior art" -- the legal term for similar inventions -- in the patent that Walker acquired for the core of Priceline's business model (for a "Method and Apparatus For a Cryptographically Assisted Commercial Network System Designed to Facilitate Buyer-Driven Conditional Purchase Offers").
Court documents filed by Marketel say that in 1987 Perell hired a friend, Andre Jaeckle, to find investors for the BookIt business. Jaeckle, says Marketel, introduced Walker to Perell in the fall of 1988. Over the next 18 to 24 months, according to Marketel, Walker, Perell and Jaeckle talked about the BookIt business plan, and Perell revealed details of how it works. Marketel says Walker actually signed a non-disclosure agreement, though Priceline's filings say that Marketel has not produced it. Jaeckle, both sides agree, got stock in Priceline. Documents that Priceline filed with the Securities and Exchange Commission say that warrants to buy Priceline stock were part of a deal for a million-dollar loan that Jaeckle made to Walker Digital.
BookIt flamed out. Its launch was accompanied by a flurry of publicity in publications ranging from the Wall Street Journal to Consumer Reports, but within a year it was out of business. The death of a business, however, is not the same as the death of an idea.
To prevail in the lawsuit, Perell and Marketel might need to prove a whole host of facts that fall into the turgid realm of legal briefs. Was Walker privy to Marketel's trade secrets? Did he indeed sign a non-disclosure agreement? Did he use Marketel's trade secrets in developing Priceline? Did Walker have any substantial knowledge that wasn't revealed in press articles that described BookIt's business and are already cited in Priceline's patent application? All that and more will be for the courts to decide. But whatever the legal issues, the central question is simple: Did Walker talk to Perell, and then later resurrect the main ideas of BookIt, and pass them off as his own?
Priceline has provided a partial answer to this in court documents. Before revealing what it is, you could have some fun by looking at three possible answers and deciding which one Priceline gave:
(a) "Defendants ... admit that at some point Jaeckle introduced Perell and Walker."
(b) "Defendants ... admit that at some point Jaeckle took steps to try to put Perell and Walker together."
(c) "The Defendants have no recollection of any contact with individuals related to Marketel whatsoever ..."
Made your guess? Good. You probably figured out this was a trick question, and answered (d) all of the above. Because, of course, that's the answer.
Answer (a) comes from Priceline's initial answer to Marketel's complaint, filed on Feb. 5; (b) comes from an amended answer, filed on Feb. 16; (c) comes from a statement filed by Priceline's lawyers on March 11. Walker was traveling and could not be reached to comment.
Brian Ek, a spokesman for Priceline, says these responses are "consistent." In an e-mail responding to a request for clarification, he writes: "'Introduce' wasn't meant in the physical sense: rather, it was a suggestion that they might talk." And Priceline officials say that, in this case, "Defendants" doesn't include Jaeckle, who has known Perell for years.
Well, OK. But it is telling when people with prodigiously gifted memories start failing to recall things in the course of a lawsuit. When in the 1980s Ronald Reagan was called to testify about the U.S. government's role in funding the Nicaraguan Contras, the former president said, "I don't recall," "I don't remember" and all its variations so many times that satirists made something of a sport of tallying up the lapses of memory. So, too, with Bill Gates, whose startlingly poor memory for the details of Microsoft's business practices when faced with government antitrust lawyers has provided more than its share of grist for the press.
Judging from the early evidence, Priceline may have been hit in the head by the same amnesia-inducing brick.
Whether Walker had privileged information about Marketel is for the courts to decide. In the end, however, the issue comes down to this:
Does Walker start with business problems, come up with ideas to solve them, and in the process uncover the "prior art"? Or does Walker go looking for ideas already in the public domain and modify them to turn them into patentable processes? It's conceivable that he got his inspiration (legal) but not trade secrets (illegal) from a well-publicized and unsuccessful business idea.
Ek insists that is not the case. "Walker Digital always begins the invention process with a clearly identified business problem that can be solved for profit," he says. "Next, they work to come up with the best solution they can think of for the problem. After the solution is fleshed out, the prior art research begins."
That's the official line. But Priceline isn't the only case where Walker seems to have picked up a lot of know-how from a promising business idea.
U.S. patent No. 5,862,223 is titled "Method and Apparatus for a Cryptographically-Assisted Commercial Network System Designed To Facilitate and Support Expert-Based Commerce." Alas, that is not a title that would whet your appetite for breakfast reading, even if the only alternative were a cereal box. And that would be too bad, because patent 5,862,223 is an extraordinary document. If you were to pick up the document, and somehow manage to get past the first 35 pages of flow-charts, eventually you would come upon what is surely some of the best prose ever to cross the desks of the patent office. It is precise, it is hilarious, it is a little weird. Really.
In essence, the patent covers the creation of an electronic marketplace for the sale of expert advice and services. But that doesn't really do justice to it. Here is an example, taken from the text of the patent, of the kind of problem the invention is supposed to solve:
Imagine a professor of nanotechnology at the University of Makinsk, located in Kazakhstan, formerly part of the Soviet Union. Although the professor's teaching schedule keeps him fairly busy, there is not much to do in Makinsk. To relieve the boredom and to supplement his limited salary, he decides the time has come to embrace capitalism by using some of his time to offer freelance nanotechnology consulting. The professor begins by making a list of what needs to be done before being able to sell his nanotechnology expertise ...
And so it goes, enumerating the problems of finding clients, finding clients who will have work that the professor can do at his preferred Saturday evening time, finding clients who will pay on time (unlike the "many unscrupulous nanotechnology companies that have been known to prey on ex-communist scientists in the former USSR"), finding a way to get the information they want to the clients, hiding all this activity from his boss, and so on, almost, but not quite, to infinity. By the end of the passage, a two-column page of small type, the patent examiners must have been half-ready to imagine that if they rejected this patent application, they would be condemning a brilliant nanotechnologist to poverty and obscurity.
This is a pity, because the long story of the nanotechnologist serves to gloss over a pertinent fact about the invention: The idea at the heart of it simply isn't new.
The original idea of selling information on an electronic marketplace was tried first by an economist named Phil Salin. Salin started working on the software for an electronic marketplace as early as 1984, and launched AMIX, the American Information Exchange, in 1991. The company was well-publicized. Walker's own patent cites an analysis in the Economist magazine and a Forbes article by computer industry guru Esther Dyson discussing the system.
The system, as Dyson describes it, was fairly sophisticated even by contemporary Net standards. It involved both the posting of documents -- information that has already been created -- by experts as well as the solicitation of consulting services by buyers looking for advice. It included a mechanism for auctioning information and advice; Dyson said Salin also envisioned complex, automatic pricing systems that, for instance, could let information drop in price as it ages. Though high-profile names like Dyson and Lotus founder Mitch Kapor were involved, the company was unable to realize its ambitions and folded.
Walker's own patent recognizes AMIX as the most notable experiment with information marketplaces. "To understand the failure of AMIX, and all other prior attempts to create working expert-exchange marketplaces, it is necessary to understand that effective markets, whether they be physical or electronic, require a complete and highly specialized set of conditions in order to function and thrive," the patent notes, "A single missing ingredient ot feature might ... lead to the collapse of the Exchange."
But then the patent launches into a description of Walker's own system, and the reader, alas, never does find out exactly what feature was missing from AMIX that made it fail.
Observers, including Doc Searls, who worked on AMIX, say its greatest failure was being ahead of its time. AMIX existed before the Internet and struggled to build an online service that could handle its data and traffic. Now that the Internet exists, however, there are other businesses following in its footsteps. In fact, whereas Walker Digital simply has a patent, at least two companies, Advoco.com and Adeste.com, are working on full-fledged businesses that connect buyers and sellers of information and expert advice.
Ironically, Salin himself was a vocal opponent of patents on software and software-based systems. He likened creating software with writing, and said, in a public comment filed with the patent office as it was considering rules on software patents, that "patents on writings discourage trial and error [and] perfection of important ideas."
Salin's skepticism when it comes to software patents is shared by some of his successors. "Do software patents promote innovation?" queries Adeste founder Michael Stern. He leaves it as an open question. One senses that depending on whether the patent is his own or somebody else's, the answer can go either way. Like everyone else in the post-Walker era, however, he's hired a hotshot intellectual property law firm, in his case the New York firm of Kaye, Scholer, Fierman, Hays & Handler.
It might seem perplexing that an idea so obvious that several companies would have launched businesses around it even as the patent wound its way through the patent examiners' hands could be the subject of a patent at all. But if it seems perplexing in the case of a complex system for expert-based commerce, Walker's patent for a method for "establishing and maintaining user-controlled anonymous communication" seems downright awe-inspiring in its chutzpah.
That patent was granted to Walker and his associates in March (he had applied for it in September 1996) and bears the number 5,884,272. In essence, the patent is for a system by which people who do not want to reveal their identities -- corporate whistle-blowers, to use one example in the patent, or potential blind dates, to use another -- could send anonymous messages back and forth to each other. A central database would store information about all the participants, who could then search the database to find people who matched what they were looking for but wouldn't know precisely who they were communicating with.
It is worth quoting from the patent at some length:
A goal of the invention is to provide a communication system incorporating a central database of information supplied by one or more parties and managed by a central administrator, where all parties to the system can manage and control the release of any or all information about themselves or their identities, and where such a system allows for electronic communication between the two parties without the necessity of revealing their identities.
Another goal of the invention is to allow parties to submit criteria for searching a trusted agent's confidential database and receive a count of the number of records that satisfy the criteria, without revealing the identity of either party.
That's quite a mouthful. The patent could certainly apply to sophisticated business communications systems -- trusted third-parties for financial transactions and the like.
When Walker applied for his patent, however, the most obvious uses were more profane. Compare Walker's description of his invention to a press release from Match.com, an online dating service, sent out on July 12, 1995, more than a year before Walker applied for his patent:
Match.com members enjoy benefits over traditional print personals: enhanced privacy through anonymous e-mail exchange searching and matching based on self-selected criteria.
It's public relations happy talk, but the gist of it is clear. The database is there, the anonymous communication is there, the searching is there. And John Spottiswood, the president of Match.com, says even the ability to get a count of the number of people in the database who match your search criteria was there too (it certainly is now). Matchmaking, remember, was actually one of the example uses Walker's patent lists.
Anybody who applies for a patent must disclose to patent examiners any prior art -- effectively any similar work -- that can be relevant. But this yields a paradox:
Says one person involved with patent filings who did not want to risk alienating Jay Walker by speaking for attribution: "You have to disclose everything you know when you make the application -- anything else would be fraudulent -- but prior art can invalidate your application. So there is a big incentive to find out as little as possible about the prior art."
Presumably, if Walker did not disclose it, he did not know about Match.com. But that doesn't mean his patent will stand up to a court challenge. When contacted about the patent, Spottiswood said he had never seen it, but said that he would certainly be prepared to challenge it if it was close to what Match.com is already doing.
Walker takes care not to portray himself as a litigant in waiting.
When I asked him in a conversation about two months ago whether he plans to pursue a host of patent infringement cases, Walker was conciliatory.
"Our goal in life is not to go around chasing people doing things similar to what we're doing," said Walker. "We're focused entirely on using our property. Patents are part of a whole process of creating value. You can build a business around them yourself, or license the intellectual property. And you'd want to prevent others from copying what you're doing."
It all sounds so reasonable. Just a guy -- well, a guy with a team of professional brainstormers working for him -- trying to find a way to make sure that others don't take the good ideas he comes up with. And yet, behind the reasonableness is the scary possibility that an intellectual property samurai can go around patenting every idea that he can paste a serial number on.
You almost can't blame him. If Edison were alive today, and if indeed he were to invent the light bulb, it is almost certain that his patent application would read something like this:
The claimed invention is a method and apparatus for generating light in a dark room from an inorganic source. In one preferred embodiment, it encompasses a glass globule containing a filament of copper wire ...
Why just be the person who invented the light bulb? Why not be the person who invented inorganic light? (Yes, you would have to cite the candle as "prior art.")
All this reasonableness, however, could add up to a round robin of courtroom drama. Or not. It is hard to avoid the suspicion that in a lot of cases, companies faced with the choice of paying a fee to license the obvious or go mano-a-mano in a courtroom session with a billionaire will choose the first option.
In a conversation I had with Walker about two months ago, he said that he was not interested in pursuing "accidental" infringements on his patents. I found that strange, because the assumption that the ideas involved could be common enough to be routinely infringed on by accident struck me as perhaps diminishing a little of their luster. And yet, perhaps, we should be grateful, because the prospect of Walker going after all the people who could have come up with the same ideas sounds like it would demand a separate federal circuit court. And those cases would be especially grating to observe, because it's more than a little likely that some of the "accidental infringers" would be the same people who came up with the ideas in the first place.