The long-running patent dispute over browser plug-ins is due for another round, thanks to a ruling delivered on Wednesday in Microsoft’s favor in the six-year old patent infringement case brought by Eolas Technologies.
A federal appeals court upheld the lower court’s interpretation of the Eolas patent on browser plug-ins, but said the jury in Eolas’ patent infringement suit against Microsoft should have considered whether the patent itself is valid.
Judge James Zagel of the U.S. Court of Appeals for the Federal Circuit ruled that a lower court was wrong to reject Microsoft’s arguments that browser plug-ins were an obvious enhancement that had been done before and therefore shouldn’t have been patented.
Moreover, it said the lower court erred in throwing out Microsoft’s defense that Eolas had engaged in inequitable conduct by withholding information about an earlier browser that also allowed embedded interactive programs in a Web page.
Therefore, it erased the district court’s ruling and sent the case back for a new trial on those specific issues.
Eolas executives Michael Doyle, David Martin and Cheong Ang received U.S. Patent 5,838,906 on November 17, 1998, assigning it to the University of California because they had developed the technology at UC San Francisco. They applied for the patent in October 1994, saying they first had demonstrated the technique in 1993.
Eolas claims the patents cover the embedding of small interactive programs, such as plug-ins, Java applets, scriptlets or Microsoft’s ActiveX Controls, into online documents that are triggered when the page loads or when a user clicks. Eolas sued Microsoft in February 1999 for infringing on the patent, saying its Internet Explorer browser took advantage of the Eolas invention.
In the decision delivered today, the Federal Appeals Court declined to narrow the patent, agreeing with the way the district court interpreted the patent. That is, the appeals court agreed that, if the patent is valid, Microsoft and presumably, anyone else creating interactive Web pages, infringes on it.
Patrick Ross, a research fellow and communications director for The Progress & Freedom Foundation’s Center for the Study of Digital Property, said the decision was promising for Web developers and Internet users. “We are all in favor of property protection when a patent is done correctly,” Ross said. “Patents themselves aren’t bad, but poorly issued patents are.”
Predictably, both sides claimed the ruling was a victory.
“We won on all the issues that would have been most important for Microsoft,” said UC spokesperson Trey Davis, “namely on the question of patent infringement and the damages that followed from that infringement.”
“We have maintained throughout this process that the Eolas patent is not valid and today’s ruling is a clear affirmation of our position,” Microsoft spokesperson Jim Desler said in a statement.
“The infringement stuff stuck,” said John Rabena, a partner at the law firm Sughrue Mion, which specializes in software patent litigation.
“But it’s probably to be viewed as a victory for Microsoft, in that the federal court has opened the door to a significant batch of evidence dealing with the invalidity of the patent.”
Microsoft’s Desler said the company welcomed the opportunity to try to prove that Eolas had engaged in inequitable conduct in its patent application. According to U.S. law, a patent is not valid if the patentee took unethical actions such as not informing the patent examiner of possible prior art. “The ruling also gives Microsoft the opportunity to present evidence that Eolas knowingly withheld information about Pei-Wei’s invention to the patent office,” Desler’s statement said.
The new trial will hinge on whether Viola, an early browser created by Perry Pei-Yuan Wei, qualifies as prior art that should have disqualified Eolas from patenting its technology. Wei did not respond to requests for an interview, but his personal Web site contains his history of Viola.
“The Viola browser was the very first Web browser to support interactive embedded objects, and other features such as tables, input forms, stylesheet (sic),” Wei claims. The site includes a screenshot of Viola.”
According to Wei, he contacted Michael Doyle, the former UC researcher and CEO of Eolas, informing him that the Viola browser qualified as prior art and included a paper that contained the source code. Yet, Wei wrote, Doyle mentioned only Cello and Mosaic in his patent application; those browsers did not include embedded interactivity.
According to the court document, Wei testified at the trial that he had written the Viola code in May 1993 and demonstrated it to engineers at Sun Microsystems
soon after that. But Wei attempted to demonstrate a different version of Viola for the jury — and couldn’t get it to function. That flub may have led the lower court to incorrectly decide that the second version of Viola didn’t support plug-ins.
The district court ruled that Wei had abandoned the first version of Viola, and therefore it didn’t qualify as prior art. Wednesday’s appeals court ruling said that was a mistake.
UC’s Davis said it has maintained that the ordered reconsideration of the Viola browser is not a major issue. “We’ve made the claim repeatedly that it’s not a prior art,” he said. “If we have to do it again, we will.”
Next page: Case Law Brightened, and A Battle on Two Fronts
Continued from page 1
The decision brightens an area of the law that was somewhat murky, Rabena said. “The federal court clarified the law to say that there are only two categories of abandoning an invention. This didn’t fall into either category. I think a number of people didn’t recognize that in the law before.”
Rabena said there are two ways an inventor can abandon an invention: by an unreasonable delay in filing for a patent, perhaps because the inventor had other irons in the fire, or by not disclosing it because he wanted to keep it secret so he can make money on it. The district court ruled that Wei’s actions fell into the latter category; the higher court disagreed.
Rabena said the most significant part of today’s ruling is that the new jury will have to consider Microsoft’s defense that the Eolas patent is invalid because Eolas engaged in inequitable conduct — or withholding information — an argument that can be more powerful in front of a jury than quibbles about prior art.
“To [be able] to argue that [this company] is trying to get money from us, and they’re sleazy, that’s an emotional argument that will turn a jury a lot faster than boring facts about who did it first,” Rabena said.
A battle on Two Fronts
The battle over plug-ins is being fought on another front, as well.
When an appeals court upheld a ruling against Microsoft, in October 2003, the World Wide Web Consortium weighed in, asking the Patent Office to reconsider. When the USPTO agreed in November, the W3C presented prior art in the form of early drafts of the HTML specification, e-mails from the W3C’s Dave Raggett and a paper by George Toye, referring to NoteMail, which was part of a collaborative Internet communication project called SHARE.
Based on that information, patent examiner Andrew Caldwell rejected the claims of the Eolas patent in an initial review. But that was only the first step in a process that could take another year.
Dan Ravicher, executive director of the Public Patent Foundation, said the ruling wasn’t a clear victory for patent reformers. “[The appeals court] could have changed the claim construction to interpret the patent more narrowly, in a way that Microsoft doesn’t infringe.” Instead, the higher court upheld the broad reading.
The Public Patent Foundation is dedicated to fighting overly broad patents — and has instituted a Microsoft patent watch. Although it’s not participating in the fight to overturn the Eolas patent, it finds itself on Redmond’s side in this one. “This patent is one that could cause the public significant harm,” Ravicher said. “At this time, Microsoft’s interests are aligned with the publics interest against the patent.”