In the late1990s, when the federal courts began to show interest in protecting software with patents, it caused a stampede at the U.S. Patent Office. "Microsoft (Corp.) five to six years ago had very few patents, and now it has roughly 2,000," says Suzanne Harrison, senior vice president of ICMG, an intellectual property management company in Palo Alto, Calif. IBM has 20,000 patents and gets US$1.7 billion per year for licensing those patents to other vendors, Harrison says.
With intellectual property, one of the biggest untapped sources of value in companies today, she says, the software industry has begun to look more like the pharmaceutical industry, where companies zealously protect drug patents. "CEOs would be remiss if they didn't go after that value," Harrison says.
But it's not clear whether what works for pharmaceutical companies will work for software. There is a threat for 2003 and beyond that patents could hurt software innovation as much as or more than they help it. Critics' complaints that software is poorly understood by the Patent Office aren't helped by the fact that vendors don't release their source codes for scrutiny. (Pharmaceutical companies, in contrast, release development documentation by the truckload.) That's meant that the overwhelmed Patent Office has approved several dubious patents because it could not find evidence that someone else had already come up with the idea (during its so-called prior art research). Worse, software changes so rapidly that patents have to be broad, covering the concepts behind the software so that they can remain relevant for years. That may make the patents too vague for other inventors to know whether they can work around them. It also means the patents are less likely to hold up in court. For example, British Telecom (BT) patented hyperlinking before the Web came along, and then tried to enforce the patent against companies that used hyperlinking on the Web. A federal court ruled that BT was trying to cover too much ground, and in late September, BT was forced to abandon its efforts to uphold the patent.
Thomas Edison, who held 1,093 patents, is shown here with his favorite invention, the phonograph. Could patent law inhibit software innovation?
A big company like BT can afford a long, bitter legal battle to test the merits of its patent. Smaller vendors and independent developers don't have that luxury. Indeed, the darkest view of the patent frenzy is that poorly written, overly broad patents owned by big companies with deep pockets will be used as blunt weapons to discourage those small companies from trying to invent new software. The little guy simply won't want to risk getting dragged into court, no matter how questionable the patents may seem. "Taking out a patent doesn't cost that much, but challenging it in court does," says Mark Webbink, senior vice president and general counsel for Red Hat, the Raleigh, N.C.-based distributor of the open-source GNU/Linux operating system.
Of course, patent proponents have an equally forceful argument: If you don't allow inventors to protect their ideas from copycats, they will be less likely to invent in the first place.
These arguments have come up before. When a new technology comes along, there's always a lot of noise and litigation, says Mark F. Radcliffe, a partner at Gray, Cary, Ware & Freidenrich, a law firm in Palo Alto. "In the long run, patents have always proven to be good for innovation," he says.
But as inventors increasingly move behind the walls of corporations -- which have goals to build and protect revenue as well as to innovate -- it's not clear at the start of 2003 whether the software patent frenzy will cause innovation to flower or be trampled.
Join the CIO New Zealand group on LinkedIn. The group is open to CIOs, IT Directors, COOs, CTOs and senior IT managers.