Wrinkle in ‘Seamless’ Feel of the Web

Last year, well after Microsoft Corp. negotiated a slap on the wrist with the federal government to end its antitrust problems, the University of California helped deliver a roundhouse blow to the big software company’s gut.

The punch was in the form of a $521-million judgment against Microsoft for violating a 1998 UC patent on Web browser technology. After legal fees and payments to the former UC researchers who invented the technology and to its sole licensee, the university stands to clear more than $50 million. That would be one of the largest payoffs from an invention in its history.

But UC would be wise not to start counting its money just yet. Microsoft, not surprisingly, is appealing the ruling. More important, the consortium that oversees the Web’s technical protocols, the World Wide Web Consortium, was so shocked to discover that a key element of its technology was subject to royalties that it asked the U.S. Patent Office to overturn the patent outright. The agency has agreed to take another look, a process that could take at least a year.

The university’s victory has set off a round of soul-searching in cyberspace and academia over the implications of commercial control of a key part of a public standard relied on by millions of people around the globe.


As Tim Berners-Lee, the founder of the Web and the consortium’s director, wrote to the Patent Office, enforcing the patent could cause “substantial economic and technical damage to operation of the World Wide Web.”

He outlined a scenario in which millions of innocent website designers would have to spend millions of dollars and countless hours redesigning their sites to avoid infringing the patent, and predicted “several years of disruption.”

The consortium, which goes by the shorthand W3C, seems particularly irked that after Berners-Lee and other pioneers spent their careers producing public technical standards through an open academic collaboration and without cashing in, three researchers from UC San Francisco are claiming a piece of the action -- a piece, incidentally, that gives the Web much of its pizazz.

UC’s position is that back in 1994, its researchers developed a new way for Web browsers to display certain kinds of multimedia material, including video streams, simple animation formats such as Macromedia’s Flash, and Java applets. Previously, such programs, known as plug-ins, launched in separate browser windows, a clunky and slow process. The researchers say they developed a way for plug-ins to launch automatically within the original page, providing what Web developers like to call a “seamless” user experience.


The patent in hand, UC assigned its rights to a private contractor as an exclusive licensee. This was Chicago-based Eolas Technologies Inc., which had been founded by Michael Doyle, one of the original inventors. (As a result, the patent is commonly known as the Eolas patent, or sometimes the 906 patent, which derives from its formal designation as U.S. Patent No. 5,838,906.) UC was to receive a portion of any royalties Eolas managed to raise.

Because the patent technically covers the technology of Web browsers, rather than multimedia programs, Eolas reasoned that the place to start enforcing its rights was with the developer of the browser that ran on 95% of all the computers in the world: Microsoft.

After several years of litigation, an Illinois federal judge and jury ruled last year that Microsoft’s Internet Explorer browser does indeed infringe the patent, and that Microsoft owes Eolas and UC $1.47 for each of the more than 350 million Explorers that it shipped as an integrated element of its Windows operating system between November 1998 and September 2001. (The bill for later shipments is yet to be calculated.)

Most computer users are probably unaware of the verdict’s effect on the professional Web community, but it might help to imagine a hunter loosing a shotgun blast in the direction of a flock of slumbering geese. That was especially so after Microsoft mentioned that it might redesign Explorer to circumvent the infringement -- which meant that pages built around Flash, Java and a host of other such programs might have to be rewritten to make them compatible with Microsoft’s tweaks.


Others wondered how a protocol that had been widely and publicly used for years ended up as someone’s patent. Indeed, the core of the consortium’s appeal to the Patent Office is the contention that it had been invented by others at least a year before UC’s researchers came on the scene.

As an example of such “prior art” the Patent Office may have overlooked, the W3C cites two papers published as early as 1993 by Dave Raggett, a Hewlett-Packard scientist, describing a process very similar to UC’s.

The problem with this argument is that, while the Patent Office didn’t look at the Raggett papers, they were cited by Microsoft in its legal defense -- in fact, Raggett was called as a witness -- but for some reason they didn’t figure in the verdict. Eolas and UC, for their part, contend that the patent examiners did a thorough job of looking at all the relevant prior art, and concluded that Doyle and his colleagues had invented something new.

The patent holders, meanwhile, sound mystified at having been cast as the villains of the affair. They figure they deserve panegyrics, not brickbats, for skinning a company so routinely depicted as the Sauron of the software world.


“It’s quite misguided and not real fair for the W3C to be blaming Eolas for this, when it’s UC and Eolas that contributed the technology to the Web in the first place,” says Marty Lueck, Eolas’ lead attorney in the case. “W3C should be telling Microsoft to pay up for the very substantial profit they’ve made.”

UC also is sensitive about the idea that as a public institution, it should be donating its research for the public good, as if to strike a blow for academic openness.

Trey Davis, a spokesman for the UC system, argues that the principle it is upholding is at the center of the innovation process. “An enormous amount of time or money or intellectual property goes into new products,” he told me. “The core principle is that these parties should be compensated for their time and money.”

There also is some fear in the software community that Eolas and UC unwittingly have played into Microsoft’s hands by giving the company the incentive and means to tighten its control over the Web.


The patent, if it’s upheld, would cover browser displays of third-party plug-ins -- but not the display of programs integrated into the browser or operating system itself.

“This is ideal for Microsoft,” says Matt Liotta, an Atlanta software developer who has posted a “workaround” allowing Web designers to avoid pages that might infringe the patent. According to this reasoning, Microsoft wouldn’t allow users of its browser to play video or audio clips through RealNetworks Inc.'s Real Player because that would infringe the Eolas patent. But Microsoft could embed its Windows Media Player in its browser without infringing.

Web designers would face the choice of designing non-infringing Web pages to display Flash, Java and other programs, all at great effort, or simply writing for Microsoft’s home-grown plug-ins.

Their pages would no longer display properly on other company’s browsers, but if they did so on the program used by more than 95% of the market, what would be the loss? Soon enough, Microsoft would command 100% of the market and undisputed control over the programming language of the World Wide Web.


Eolas and the university might say that such suspicions present all the more reason to hold Microsoft’s feet to the fire now -- particularly because some of the Microsoft profit being demanded by the patent holders would be plowed back into academic programs and research at UC.

“You’re talking about Microsoft making a profit of 80, 90% on its sales and not wanting to fairly compensate Eolas and the university,” Davis says. The amount at stake is not a small consideration in this age of financial famine, he adds. “The university does have bills to pay too.”

Golden State appears every Monday and Thursday. You can reach Michael Hiltzik at and read his previous columns at