Proponents of reform struggle to avoid the patent office’s rubber-stamp treatment. One prospective bill would make tech patents harder to get and easier to challenge.
When patents were first formalized by England’s “statute of monopolies” in 1623, they were intended to stimulate the sharing of information and the spread of technology. Inventors were granted monopoly control over their developments in exchange for publishing them, thus discouraging the practice of keeping inventions secret.
Ironically, there are now those who argue that some patents have the opposite effect: that they ultimately discourage the efficient distribution of information. Software patents have been awarded to computer “inventions” as trivial as interface changes, the number of times a hyperlink must be clicked, and whether an auction bidder sets his or her own desired price or responds to a price request initiated by the seller.
The fact that a few holders of such trivial patents have pursued perceived violators through lawsuits–generating excessive legal expenses for the accused–has made online retailers and other Web-site developers more hesitant to add new content or functionality. Calls (both emotional and logical) to reform U.S. patent law have come from both software developers and legislators.
Until the 1980s, the United States Patent and Trademark Office (PTO) did not grant patents for software of any kind, regardless of an application’s worthiness (or lack thereof). In fact, the PTO avoided granting any patent if the invention utilized a calculation made by a computer. The rationale was that patents could only be granted to processes, machines, articles of manufacture, and compositions of matter. Patents could not be granted to scientific truths or mathematical expressions of them.
Software patents are still forbidden in Europe, but for the past few years the countries that operate the European Patent Office have been encouraged by large software companies and patent lawyers to allow patents covering mathematical computations. Last year the EPO drafted a proposal for “no limit” patents which, if approved, will allow the same patents permitted in the United States. Several user groups, such as Free Patents, are working to generate popular opposition to the European “no limit” patents proposal.
Debate over software patents has been fairly heated and well-publicized, but anyone unfamiliar with those arguments may be surprised to learn that the PTO is the subject of such controversy. In the mythology of grade-school history, after all, the patent office is virtually synonymous with Thomas Edison and the can-do American spirit of invention–not to mention a certain larger-than-life German physicist who changed our concept of the universe while working in a patent office. The PTO is supposed to be a good, noncontroversial institution.
Though its recent actions are controversial, the criteria that the PTO uses to determine whether a patent is warranted has not changed in the last century. A patent has to be both original and nonobvious. The controversy surrounds how the office applies the criteria. Hundreds of patents have been granted and later overthrown in court because the plaintiffs could easily demonstrate prior art–that the patented technology was not original. It’s a bit more difficult to prove that an invention is obvious; such a subjective notion will always lead to legal disputes. So most of the controversy surrounds prior art; and an industry has sprung up in developing technologies that do the due diligence that the PTO lacks the resources to do. Unfortunately, too often this happens after a patent has been granted, a situation that leads to unnecessary legal battles.
The news is full of reports of so-called frivolous patents. We need not rehash them here; readers need only look to Amazon.com and Priceline.com for examples of patents designed to give the dot-coms a business edge. Walker Digital, Priceline’s parent company, has gone so far as to say that obtaining and suing to enforce such patents is part of its business plan.
As mentioned earlier, the PTO did not willingly enter into software patents; as late as the 1970s the PTO avoided them on the basis that they were mathematical algorithms. The PTO’s change of position was actually forced upon it by the Supreme Court and other judicial actions.
In 1981, the case of Diamond v. Diehr provided the first instance in which the U.S. Supreme Court ordered the PTO to grant a patent on an invention that used computer software. In that case, the invention related to a method for determining how rubber should be heated in order to be best cured. The invention used a computer to calculate and control the heating times for the rubber. However, the invention (as defined by the claims) included not only the computer program, but also steps relating to heating rubber and removing the rubber from the heat. The Supreme Court ruled that in this case, the invention was not merely a mathematical algorithm, but also a process for molding rubber, and hence was patentable. This was true even though the only novel feature of this invention was the timing process controlled by the computer.
From that point, the PTO and inventors were left trying to determine when an invention was merely a mathematical algorithm, and when it was in fact a patentable invention that simply contained a mathematical algorithm. In the early 1990s, the Federal Circuit (the highest court for patent matters other than the Supreme Court) tried to clarify when a software-related invention was patentable. The court stated that the invention as a whole should be examined. Is the invention only a mathematical algorithm, such as a computer program designed to convert binary-coded decimal numbers into binary numbers? If so, then the invention is unpatentable. However, if the invention uses the computer to manipulate numbers that represent concrete, real-world values (such as a program that interprets electrocardiograph signals to predict arrhythmia, or one that analyzes seismic measurements), then the invention is a process relating to those real-world concepts and is thus patentable.
Although the PTO has developed internal guidelines for dealing with software patents, it seems that even those are not interpreted consistently. The PTO guidelines specify that software can be patented only if it “is used to control something external to the software routine (such as curing rubber), and those having precomputer process activity, meaning software programs that manipulate numbers representing concrete, real-world values (such as electrocardiograph signals and seismic measurements).” Under this standard, something like a Web page (let alone a hyperlink) should be either ignored or denied patent.
Time to reform?
There is a seemingly endless list of incidents suggesting that the PTO is slowly spiraling out of control, including non-software patents for such items as animals’ genetic sequences and the practice of using a laser pointer for cat play.
Support for the idea that PTO is operating effectively is drying up; its supporters seem invisible but for those, like Walker Digital, who have tried to profit from this misconception. Software developers and users remain vehemently opposed to current patent practices. “Imagine that each time you made a software design decision, and especially whenever you used an algorithm that you read in a journal or implemented a feature that users ask for, you took a risk of being sued,” suggests Richard Stallman, founder of the Free Software Foundation and a legendary Internet developer. “That’s how it is today in the United States, because of software patents.”
Even Congress, often maligned for its members’ lack of savvy in technical issues, has members who understand the problem. Last year, Rep. Dick Boucher, D-Va., commented that “something is fundamentally wrong with a system that allows individuals to get patents for doing the seemingly obvious. And it has led to a lot of unnecessary litigation.”
Activists opposed to software and business-method patents are currently working to quarantine the European Patent Office from PTO-style open patents, but are still hoping for a way to improve the situation here. There are several ways in which the United States’ patent procedures could be improved:
The U.S. Patent and Trademark Office could revise its own internal guidelines and procedures. There is some effort being applied to this type of activity in the form of the PTO Director’s Initiatives on Patent Business Methods for 2001, but the written objectives of this initiative seem relatively insignificant.
As in the early 1980s, the Supreme Court could direct the PTO to alter its operations. There are no high-profile cases approaching the court that could form a challenge to the patent system, however.
Developers and other corporate entities could voluntarily act in the best interests of an equitable patent system. In other words, they would need to police themselves to not submit patent applications for questionable technologies or business methods. Some of the companies that were scandalized by public outcry over software and business-method patents (namely Amazon) have reduced their patent-related litigation. However, the potential value of a patent, and the stated intent of companies like Walker Digital to utilize litigation, makes this unlikely.
The U.S. Legislature could mandate reform for the PTO.
Of these, the most concrete choice is the last–legislative mandate. Patent-reform legislation has been before the House and Senate in the past. In 1999, Rep. Howard Coble, R-N.C., introduced HR 1907, the American Inventors’ Protection Act, which proposed significant changes to patent law and procedure. Although HR 1907 was not exclusively directed at e-commerce, several of its provisions were designed to improve efficiency of the submission and review process, and one section, the Patent Litigation Reduction Act, could have reduced the patent-generated judicial demands. However, HR 1907 passed the House but not Senate (a similar bill suffered the same fate in 1997.
A bill with a bit more of a head start, as well as a technology-oriented pedigree (its sponsors are Howard Berman and Boucher, representatives from the technology-heavy areas of California and Virginia, respectively) is HR 5364, the Business Methods Patent Improvement Act of 2000. This bill would still permit patents to be issued for business-method inventions, and the term would not be affected. Such patents would be harder to obtain and easier to challenge.
We should note, however, that HR 5364 has not even been scheduled for introduction in the 107th Congress, but is liable to be sometime this summer, and would eventually need to be approved by the Senate. In short, legislative relief for technology patent problems has no guarantee of effectiveness in the near future, so users and developers need to be aware of companies abusing the currently open system in ways that could affect computer usage.
Contributing Editor Joe Rudich is a network administrator with the St. Paul Companies in St. Paul, Minn.