Patent Office chief aims to be 'technology neutral'
The 200-year-old system wasn't meant to rule out patents on software and business methods, and it's also pretty well built for weeding out the "trolls," says Jon Dudas.
HOLLYWOOD, Calif.--Open-source and free software fans who despise software patents shouldn't look for an ally in the head of the U.S. Patent and Trademark Office.
Patent Commissioner Jon Dudas says his office is all about staying "technology neutral," so long as the invention meets certain standards.
"The system we've had has worked to promote technology for 200 years, and it can do that in the software industry, so long as you follow the principles that (a technology) is useful, new, and nonobvious," Dudas said during a wide-ranging interview with CNET News.com at a technology policy conference here this week.
Read on to hear more about what Dudas, who has also held the post of Commerce Department undersecretary since June 2004, had to say about things like business method patents (think Amazon.com'sand eBay's ), so-called Congress' attempts to overhaul the system he oversees, and the Patent Office's own efforts to curb bad patents from the start. (Editor's note: Responses have been edited for length and clarity.)
On software patents: "Software, biotechnology, business methods--In the United States, the Supreme Court has consistently held that those are areas where there should be patents, and those industries have flourished.
"Specifically with open source, I think the two should coexist very well. If someone gets a patent, then that intellectual property has to be respected, but so long as that patent isn't used, open source can be as open as it needs to be. You can license some (patents) and not license others. There are some who feel by definition you should only have open source or only have a patented model. The administration's position has always been that...both open source and patents help innovation thrive."
On business method patents: "We essentially said, we shouldn't base whether or not something is patentable on what type of technology it is or even what kind of method it is...so long as it's new, useful, and nonobvious.
"Probably in the last three years, of all the (business method patent) applications that have come in the door, the office has said 85 percent of these are not allowable. The patent system, I believe, is working very well... We're starting to get a higher percentage of business method patents being approved, probably in the 20 percent range."
On "so-called" patent trolls and whether they're overhyped: "It depends on how someone defines a patent troll. 'Patent troll' is meant to be a pejorative term. Some have defined it as someone who doesn't develop a project commercially. Certainly, someone who comes up with a great idea and licenses it, that's a very efficient way to do it.
"I think the concern there is more a concern with the judicial system as a whole than it is with the patent system."
On gripes by large technology companies that patent infringement fines are too high: "The administration has said we do think judges could give better guidance to juries. We definitely feel that in the case of damages, it would be good if judges could give very specific guidance to juries. With the provisions in the existing bills right now, the administration has said, 'Webased on that because they favor one type of damage model over another.'
"There are a variety of different factors you can apply to determine what the right compensation is, and you, as best as possible, want to mimic the market. But the bills direct you to a certain couple of factors that may or may not mimic the market. We're saying the judge should tell the jury; don't just say, 'Here's 15 factors; figure it out for yourself.' The judge should tell jury, 'Here are a number of factors. These 3 seem critical; look at factors 2, 5, and 6."
"When they talk about damages awards and things like that, people say there's a lot of bad patents. When you get to the point of a damages award, there's a judge and jury that already determined this is a valid patent... By the time you look at damages awards, you're talking about good patents and you're talking about actual infringement of intellectual property."
On weeding out "bad" patents from the get-go: "Making sure applicants can give good information up front...will lead to a much better patent system--more than anything else. (The Patent Office is proposing that applicants) do a basic search (for "prior art," which is previously published technology that relates to an invention) and short report about why their invention should be patented.
"In 27 percent of cases, nothing is submitted in terms of prior art. In another 17 or 18 percent of cases, people just give us a long list of references. There really is a responsibility for people to come in and say, 'Here's why I think I deserve a patent; here's some of the areas to look at.' The decision will always rest with the patent examiner. The patent examiner will always do the full search and review and analysis."
On reducing in-court patent litigation: "The administration proposed post-grant review (a process within the Patent Office where people would be able to challenge patents just after they're issued). It has to be a true alternative to litigation. We don't think the idea is to have another way to question a patent, and then you can go to court and you can go to post-grant. The standard we've held is that there should be a threshold in order to get a post-grant review, an actual threat of litigation.
"The second thing--this is the critical point--if you choose to go to post-grant review, any issue you raise or could've raised, you have to raise there, or you can't raise it again. You don't want an innovator to have hurdle after hurdle after hurdle after they've gotten a patent."
On whether Congress and the Bush administration can reach an accord on new patent system rules: "We feel like senators and members are cognizant and care about concerns of the administration. Ideally we'll have a bill with a letter from the administration saying we support this bill as it is. That's our goal."
On handling a flood of patent applications: "We've hired more examiners in terms of raw numbers and percentages in the history of the office. We've said very clearly, hiring more examiners in and of itself won't be the solution. Even hiring 1,200 examiners, even increasing a lot of different flexibility initiatives that increase productivity...we still get more applications in the door than we're able to examine.
"What we really need to do is get higher quality applications. The allowance rate, those applications that are approved, used to hover between 62 and 72 percent, back 35 to 40 years ago. In the last 6 years, it went up to 72 percent in 2000, then dropped steadily down to a 43 percent allowance rate. So what we see is 57 percent of what comes in the door doesn't lead to a patent. "
On the long line to get a patent--and how some may game the system: "Most of the time when people say it takes me 33 months to get a patent, people think, "What's taking so long?" The examiner only takes a few weeks. That time is all waiting in line. That line is getting longer and longer and longer--filled with applications that never get approved. What's, in lots of ways, more disturbing is in over half the cases where we say this isn't patentable, people just file again and get back in line.
"We put out a rule (that was supposed to take effect last November) saying that you can only do that three times. But we were sued. (A judge issued a temporary injunction against the rule.) Right now, the way the law has been interpreted is people have unlimited opportunities to refile their application. You have more opportunities to have us look at your patent application than you do to appeal your death sentence.
"We want to make certain that people can't apply with a very broad patent application, which they know will get rejected. And then they get back in line, and meanwhile, they're looking out and seeing what's happening in the market. Sometimes they see that if they focused that broad claim, it could cover an existing technology... Then, (going by) the date of first filing, they can then say, 'I own that technology'... That's a very real concern. That gets more in line with concerns of troll behavior--someone who is literally watching the technology...so they can rise up out of the bridge and sue people."