Protection of Software via Patents and Trade Secrets

| 0 TrackBacks

This past Monday, December 3, 2007, I gave a talk in Cartagena, Colombia on the Patent/Trade Secret interface in the protection of software. I was invited to do so by ASIPI (Interamerican Association of Intellectual Property) at their XIII Work Sessions and Administrative Council Meeting, held in Cartagena on November 30 to December 5. This meeting featured seven panel discussions, one of which (Panel 4) dealt with the controversial topic of “Software Protection via Patents.” It was to explore the controversy from business, academic and governmental perspectives and also “discuss alternative ways to protect innovation in the software industry.” With James Pooley (AIPLA President) as moderator, Dick Wilder of Microsoft represented business and Giancarlo Marcenaro of the Colombian Superintendence of Industry & Commerce spoke for government. Protection of alternative ways was my topic.

The paper, which I prepared for the panel discussion and which will be translated into Spanish and published in ASIPI’s Derechos Intellectuales series, dealt first, because of the title, with the development of software patents, the current USPTO regulations, current cases shaping software patent protection, the differences between US, European, Japanese, and Korean software patent protection. It also gives an example of software patent enforcement, examines the impact of software patents on innovation and entry into the software market, and briefly outlines how Open Source Software Licenses can affect software patents.
The crux of my presentation, however, was that all IP categories can be enlisted in a complementary, rather than alternative, way to protect software. I advocated reliance on trade secrets while obtaining copyright as well as patent protection on software. Finally, my paper postulates that sui generis protection would be the most suitable form of protection, as was proposed when the issue first arose in the mid-sixties.

The first point I made related to the importance of integrating all IP categories for synergistic multiple protection via exploitation of the overlap because this can cover additional subject matter, strengthen exclusivity, afford additional remedies and provide backup in case of invalidation of an IPR.

Then I focused on how trade secrets are compatible with and complement copyright protection, which is true only for software, inasmuch as the Copyright Statute and a regulation of the Copyright Office (Copyright Circular 61) permits the blocking out of trade secrets by, e.g., submitting only the first and last 25 pages of the copyright application. This recognizes officially the existence of trade secrets in software development and the need to protect them without foregoing copyright protection.

Importantly, trade secrets are also compatible and complementary with patents to protect software. Patent applications need only disclose the best mode and enablement of the software innovation and the code itself and collateral know-how need not be disclosed, but can be maintained as trade secrets. The glue holding the patentable and trade secret ideas together, which transforms these ideas into expression, is often the technical know-how, the grist for trade secrets. Software libraries, compilers, implementation schemes, and code optimizations techniques constitute just the tip of the iceberg in the average software company’s know-how.

Any contention that trade secrets cannot coexist with patents on a given invention because of the best mode requirement, overlooks the simple truths that this requirement applies:
• Only at the time of filing,
• Only to the knowledge of the inventors, and
• Only to the claimed invention.

In fact, the best mode requirement is actually no impediment to the coexistence of patents and trade secrets for almost any invention. Patent applications are normally filed very early, providing only rudimentary R&D data. Tom Arnold asserted that it is “flat wrong” to assume, as “many courts and even patent lawyer seem prone” to do, that “because the patent statute requires a best mode disclosure, patents necessarily disclose or preempt all the trade secrets that are useful in the practice of the invention.” (1988 Licensing Law Handbook).

Almost every software program consists of a combination of patentable material, copyrightable material and proprietary information and know-how, i.e. trade secrets. Copyright protects against literal copying of the code, in whole or part. Trade secrets, allowed to be redacted out of the copyrighted code, protect important innovations that, for one reason or another are not disclosed in a patent application, as well as collateral know-how. The patented inventions, of course, fall under the protection of the Patent Act.

Thus, it is clear that software protection is possible via trade secrets, copyrights or patents individually or simultaneously under the Copyright Statute together and side by side with trade secrets as well as under the Patent Law again in conjunction with trade secrets. The best of all worlds! In no other area of artistic or technological innovation is such total integration and exploitation of overlaps across the IP spectrum possible.

However, I submit that such an all-inclusive protection scheme would be rather convoluted and expensive and because of this and because each applicable IP category has shortcomings as means for effective software protection, given the special nature of software, a sui generis type of protection system should have been fashioned.

Many believed that copyright protection was inappropriate for being an “artificial construct,” inasmuch as the aims of copyright law and computer programming are diametrically opposed, the former stressing subjective, individualistic, creative elements and the latter, objective, technical and scientific systematization. Software is functional, non-literal by nature as it performs a task or generates output. Also a copyright term for 100 years is incongruous for protection of ephemeral software programs.

However, there are also significant problems with patenting software, in particular the statutory subject matter (algorithm) issue and the non-obviousness requirement.

With patent and copyright forms of protection thus being “Procrustean beds,” the notion of a sui generis form of protection for software had considerable appeal.

Indeed, I remember well that the first impulse by the IP profession back in 1965, when the issue first arose, was to provide a sui generis form of protection and sui generis IP protection is not uncommon at all. It exists for the protection of plants and artistic products. Moreover and very significantly, Congress, in 1984, fashioned the Semiconductor Chip Protection Act (17 U.S.C. § 901) to protect mask works for ten years in a copyright-like manner. This was done hastily and improvidently according to some practitioners and this statute is apparently rarely used.

While the Intellectual Property Owners Association (IPO) and others are strongly in favor of a “unitary patent system,” it can be argued equally or more forcefully that a “frozen in time one size fits all patent system” (per former PTO Commissioner Bruce Lehman) is not well suited for protection of software, given its special, i.e. sui generis nature. Also, Congress had no hesitation to pass the Vessel Hull Design Protection Act of 1998 as sui generis protection (17 USC 1301 et seq.) and recently even passed broadening amendments. And protection for databases via a sui generis protection regime has also been under consideration.

In conclusion, it is incontrovertible that both copyrights and patents in conjunction with trade secrets are available for synergistic multiple protection of software. However, from a puristic point of view, properly drafted sui generis protection for software could provide more appropriate and less controversial protection for software.

Karl F. Jorda

No TrackBacks

TrackBack URL:

About this Entry

This page contains a single entry by Jon Cavicchi published on December 7, 2007 10:42 AM.

Jorda on : Initial Patent/Trade Secret Evaluation Guide was the previous entry in this blog.

Are trade secrets nothing but “baby patents”? is the next entry in this blog.

Find recent content on the main index or look in the archives to find all content.