Patents on software: Hot and ugly

April 22, 1990

THE ISSUE of software patents is getting hot, heavy and potentially ugly. In an increasingly lawsuit-happy world, some of the industry’s best long-range thinkers are beginning to question the morality, if not the business sense, of the iron-bars kind of “protection” that patenting software introduces into what’s widely perceived as the most creative, free-wheeling, innovative industry in the world.

How to protect the rights of software creators is a thorny problem. That’s because a software program, at least compared to the familiar kinds of intellectual property like books and mechanical processes, is a kind of hybrid beast. It’s written much like poetry is written — in a stylized, potent form of language — but also acts like a machine, since it makes a computer perform certain specific operations.

This is not a concept that patent offices seem to grasp. In the past few years, software developers have been shocked by increasingly frequent announcements of patent awards for programming processes which they consider, in many cases, to be the obvious results of computerizing tasks which were once done mechanically.

Their biggest fear, and one that is mostly likely to be borne out in practice, is that the granting of patents for such obvious tasks will put them under an enormous economic burden. For them, the spectre of lengthy, expensive patent battles has already produced the dreaded chilling effect on their creative output.

One particularly poignant example is being tested in the British courts. A British video production firm, Quantel Ltd., applied for — and was granted — U.K. and U.S. patents on what everyone in the computer graphics world considers a no-brainer technology which, among other things, allowed for the creation of the digital “airbrush,” a now-ubiquitous tool in computer paint programs.

Despite its patents, Quantel’s was not the first digital airbrush. Graphics pioneers at New York Tech — including legendary names such as Alvy Ray Smith (now at Pixar), Jim Blinn (Cal Tech) and Lance Williams (Apple Computer) — claim they invented it in 1978.

“We considered the airbrush too trivial to patent,” says Smith. “I wrote the program in 1977, and documented it in 1978. Quantel took out its patent in 1980 or later.”

But two years ago, Quantel brought suit against a small U.K.-based video production firm, Spaceward Microsystems, to enforce its patents. Spaceward lost, to the tune of 1.6 million British pounds (more than $2.5 million), and is now essentially out of business.

Barbara Robertson, west-coast editor for Computer Graphics World magazine, wrote recently that the Spaceward suit is already making customers think twice about buying systems from anyone other than Quantel. Many predict that Quantel, without competition, will have less reason to develop its technology. “The question then becomes, at least in the U.K., who can move the technology forward?” she says.

The problem, says Pixar’s Smith, is that judges don’t have the expertise to make good decisions about such matters. As a result of testifying at the Quantel trial, he says, “I now think it’s extremely difficult to argue a simple technical matter in front of a non-technical judge and a highly paid team of attorneys who are highly motivated to make sure the judge doesn’t “get it.’”

The Quantel trial sent shock waves through the graphics software industry in the U.S. Most developers don’t want to talk about it, for fear of being targeted next by the company’s attorneys.

Cases like Quantel are drawing attention to the gaping holes in the U.S. patent process. The April 1990 issue of Technology Review carried a comprehensive story called “The Software Patent Crisis,” by attorney Brian Kahin, an adjunct research fellow in the Science, Technology and Public Policy Program at Harvard University’s Kennedy School of Government.

Kahin believes, as do others, that the “old” way of protecting software — via copyright, as opposed to patent — was working just fine. In short, a patent prohibits anyone else from making, selling, or using an invention for 17 years, no matter what the circumstances of its creation. Copyright protects the independent creation of products — in other words, as long as your invention doesn’t look just like mine, both of us are protected, even if our programs do the same tasks.

That’s why there can be more than one spreadsheet on the market at once — although, as Kahin points out, the U.S. patent office actually granted a patent for “natural order re-calc,” a common feature of spreadsheet programs that allows a change in one calculation to change throughout a document.

By the way, the company that holds that patent, Refac International Ltd., sued six major spreadsheet publishers last August, including Lotus, Microsoft and Ashton-Tate. You haven’t heard of Refac because it’s not in the software industry. It’s business, Kahin points out, is acquiring, licensing and litigating patents.

This kind of nonsense is demoralizing the software community. One very visible member of that community, Mitch Kapor, founder of Lotus Development — most recently returned to entrepreneur status with his new company, ON Technology — is trying to talk some sense into the U.S. government on the issue before it’s too late.

Kapor recently testified before a House Judiciary Committee subcommittee that’s exploring the issue of intellectual property rights. He is rightfully appalled at companies who seem more and more to be using litigation as a “business tactic — not a practice of last resort.”

Kapor believes that any intellectual-property law about the software industry should err on the side of innovation rather than protection. “It’s the nature of software for ideas to slosh and flow back and forth between competitors, companies and industries,” he said.