Government, Industry Agree: Something Must Be Done to Improve Software IP Rules
We conclude our examination of intellectual property with a brief history of software patents and a look at what the future may hold for this space. Because the pace and product of innovation are so heavily influenced by IP regulation, this area has important resonance on all technology development, with digital media high on the list of affected sectors.
Software patent history
Prior to 1980, the U.S. Patent and Trademark Office did not issue patents for software, or any computer-related process, under the rationale that computer operations were simply mathematical expressions of scientific truths, and thus could not be claimed as original or inventive. The PTO only issued patents to devices, processes or tangible manufactured items, and the purely algorithmic operation of computer programs specifically were not included.
In 1980, however, the U.S. Supreme Court forced the PTO to change its tune, by ordering the grant of a patent to a process for curing rubber that used computer control of temperature. The only unique element of the patent was use of a computer and its heating-control programming, but the court felt that the computer program was only a part of an otherwise patentable process.
Thereafter, the PTO and the courts attempted to set guidelines on determining whether a filing included computer software used as a component of a patent-worthy process, or if it attempted to claim originality for software per se (which continued to be considered generally non-patentable).
This process was confused and misleading, and often acted to extend the already lengthy period of patent review. (It is not uncommon for five years to elapse between filing and final patent grant.)
In the rapidly paced software world, this meant that patents often were granted so long after application that they often seemed inappropriate when finally granted, because the environment had changed so much in the interim. To counteract this problem, in the 1990s the PTO attempted to accelerate its processing work, and also began to include patents on software-related business methods.
These techniques occasionally resulted in inadequate diligence of review, again resulting in problematic patent grants. So in attempting to correct the problem of software patents being too difficult to procure, the PTO may have overcompensated and made such patents too easy to obtain in some cases.
This conundrum gave fuel to the Open Source software movement, elevating it from the province of marginal “true believers,” and giving it some traction in the corporate mainstream. Supporters felt that the Open Source precept of essentially eschewing the pursuit of IP royalties in the software space was the only solution to this complex problem. Meanwhile, others continued to believe that the Open Source process did not provide sufficient incentive for innovation, which was the fundamental purpose of patent protection in the first place.
Of course, the abuse of the current patent system by IP terrorists, discussed in a previous column, has a detrimental influence on innovation, as well.
The risk of huge expense to implementers in defending themselves against patent infringement suits brought by non-implementers seems to be a serious misappropriation of the process intended by the founding fathers, when they authorized Congress to “promote the progress of science and useful arts, by securing for limited times to… inventors the exclusive right to their respective… discoveries” (U.S. Constitution, Article 1, Section 8).
Today, government and industry agree that the patent process inadequately addresses the current environment, although there is not yet consensus on how things should change. Providing some thought leadership on such amendment is the Federal Trade Commission, which recently issued the first of two reports on the subject.
FTC recommends PTO changes
In 2002, the U.S. Federal Trade Commission and the Department of Justice held nearly a month’s worth of hearings on intellectual property law, delving into antitrust, standards and patent issues. In late 2003, the FTC published a 300+ page report on the commercial aspects of its findings, including recommendations for reform. (The FTC and DOJ soon will publish a companion report focusing on antitrust matters.)
Reforms proposed by the FTC center almost exclusively on improving and updating the U.S. patent system, and making it more in balance with competition law and policy, which the FTC administers. The report cites improvements required in the quality of patents, and it recommends changes to the patent process.
In terms of improving patent quality, the FTC proposes institution of new patent-examination criteria (which the PTO is already developing, and which will extend the breadth and depth of patent examiners’ work), and recommends a substantial increase in funding for the PTO.
Regarding process changes, the FTC proposes a new administrative procedure that would allow post-grant review of (and opposition to) problematic patents, if the patent can be shown to be questionable under certain specific thresholds. Currently, any review of a granted patent requires a special petition to the PTO (as in the Eolas case, discussed in a previous column), and such review is rarely granted.
Other fundamental changes in points of patent law also are proposed, in which the burden of proof on any patent challenges is reduced, and the criteria for declaring a patent claim as “obvious” (and therefore invalid) are more tightly specified.
The FTC’s recommendations also call for the publishing of all patent applications 18 months after filing (which has become standard practice for patents filed outside this country).
This would alleviate a fairly common and potentially expensive problem in today’s business world, as follows: Company A spends several years investing in development and planning for a product. After the product comes to market, Company B is issued a patent that reads on the product, and Company A must now pay Company B substantial, unanticipated royalties. Company B had indeed filed for the patent before Company A’s development, so it is legitimately owed these royalties, yet Company A’s equally legitimate due diligence on prior art did not discover Company B’s intellectual property claim, because it spent those years in the nether world between patent filing and issuance. The new statute for publishing patent applications would not allow this period to extend longer than 18 months.
Do the numbers
Although this all may seem an arcane undertaking, the patent process has real impact to the everyday lives of consumers and technology companies.
Consider that a typical digital consumer device produced today may read on hundreds if not thousands of patents, and simply the process of securing all necessary rights for the device’s production (let alone paying the royalties) can add to its cost. As more patents are filed and devices become more complex, this problem will only increase.
This is further complicated by the continuing trend toward software-based invention, which generally shortens product development cycles at the same time that patent processing is taking longer than ever. The FTC’s research indicates that the PTO now receives more than 1,000 new patent applications each day, and that patent examiners average only 8 to 25 hours on each before making their judgments. Clearly something must change.
Most of the recommendations made by the FTC so far require some kind of congressional action, and given the surfeit of other issues on the national agenda (and the lack of any champions rising to sponsor IP reform to date), it is unlikely that these changes will take any sort of fast track toward enactment. Nevertheless, the issues have been raised, and will likely occupy a growing measure of debate on the national stage over the next several years.
Thanks to numerous contributors for help with this series of articles, particularly Jim Burger of Dow, Lohnes & Albertson, PLLC, and Andy Updegrove of Lucash, Gesmer and Updegrove, LLP.