Forgent Networks has announced that it has a patent to one of the most commonly used digital image formats in the world and that it intends to enforce its rights by charging royalties to those who use it.
There's no question that people should be allowed to patent something (like a technology) and then reap the financial rewards should that something become ubiquitous.
But morally (unfortunately, not legally), there's a right and a wrong way to go about staking your claim. In particular, the patent claim needs to come before the technology achieves ubiquity, not after.
Whether you have a patent or not, you shouldn't be able sit around for years, or even decades, watching a technology get adopted en masse by the entire world only to suddenly speak up and say "Excuse me. We have a patent on that. Pay up."
Whatever happened to honor and integrity? That was a rhetorical question. Please don't answer me directly. In response to my column about how Microsoft sowed its own seeds of mistrust, I get hundreds of e-mails every week telling me to take my head out of the sand and wake up to reality that this is the way business is done these days. If that's the current state of affairs and we continue to endorse such behavior by patronizing the companies that engage in it, then we're a pathetic bunch indeed.
This isn't the first time that technology scoundrels have attempted to cash in on their patents long after the relevant technologies achieved ubiquity. JPEG's closest cousin GIF went through precisely the same thing when, in 1999, Unisys decided it was going to enforce its patent on LZW, Gift's compression algorithm. Unisys waited more than 10 years before deciding to enforce that patent and eventually started notifying the operators of non-commercial Web sites that made use of GIF-based images that they would have to pay a $5,000 royalty to Unisys.
One year later, British Telecom decided it would try to enforce one of its patents by asking ISPs that supported hyperlinking (the key to the Web's usefulness) to pay BT a licensing fee. BT felt that a patent it held since 1989 was enough to make such a demand. But when asked why the company didn't come forward sooner, it claimed that it had only recently discovered the patent during a routine trawl through its own patents.
IBM, Microsoft patented standards
This isn't the only way that vendors try to cash in on patents. More recently, IBM and Microsoft tried something a bit more insidious. Whereas GIF and JPEG became de facto standards for the deployment of images on the Web, IBM and Microsoft started to turn their patented Web services technologies into real standards that had the imprimatur of the World Wide Web consortium (W3C).
When the W3C started to stiffen its patent policy to ensure royalty-free access to the standards it sets, Microsoft and IBM formed their own pseudo-standards organization called the Web Services Interoperability Organization (the WS-I). (Officials from both companies deny a connection between the two events.) Publicly, the WS-I's charter was to create test suites that ensured interoperability. But an official from the organization later admitted that the WS-I's intellectual property release agreement also covered specifications (like the ones the W3C standardizes on) and that those specifications would not necessarily be royalty-free. The admission made the organization look more like a patent shelter than an interoperability group.
Fortunately, since those revelations, most of the Web services specifications--including the more recently announced WS-Security spec--have had some royalty-free language attached to them. Even so, the patent issue has fractured the standard-setting community to the point that the W3C may no longer be the central coordinating point for Web standards. As a result, it will become more difficult for those considering the adoption and deployment of "standards" to determine if there's a risk that some vendor might come back 10 or 20 years from now to demand a royalty payment.
Unambiguous licensing needed
So, when and how should patent holders go about asserting their patent rights? Ethical business practice should be a prerequisite to cashing in. Patent holders should be on the hook to notify the Internet community with unambiguous licensing terms long before their technologies reach any level of ubiquity. The unambiguous part is essential.
All too often, vendors say they will license their technology on reasonable and non-discriminatory terms (better known as RAND) but rarely elaborate on what the actual terms are. That could mean royalty-free. Or maybe not. Some cross-licensing of the licensee's technology on behalf of the licensor could be required. Or maybe not. While the non-discriminatory part is pretty straightforward, the word "reasonable" might as well be changed to "ambiguous." To eliminate any confusion over the ambiguity, the Free Software Foundation's Richard Stallman has been insisting that I stop using the term RAND and universally replace it in my columns with UFO--for Uniform Fee Only. More often than not, a RAND license results in a uniform fee since any non-uniformity would be considered discriminatory.
Regardless of what the license is called, the ambiguities need to be eliminated. Vendors like to hide behind the term RAND. It's a way for them to quickly proliferate their technologies without raising any red flags. If Forgent had come forward at the beginning of the Web and told the world that it was going to charge royalties for the right to use JPEG, do you really think any of us would be using it? Talk about a red flag! Vendors that want to see their technologies achieve ubiquity hate red flags. But they should have no other choice. Otherwise, it's just Tom Foolery. And if we're suckers enough to be fooled, then we deserve to get slapped with the royalties.
What do you think? When and how should patent holders go about asserting their patent rights? Join the discussion in our TalkBack forum or e-mail me at david.berlind@cnet.com.
