CONCURRENT PATENTS WOULD RESOLVE DILEMMA

CONCURRENT PATENTS WOULD RESOLVE DILEMMA

As experts and news sources have repeatedly observed in recent weeks, the software patent dilemma continues to reach new heights of frenzy.

On March 9, the founder of Amazon.com - itself widely criticized for abusing the patent system - proclaimed to the world that software and business-method patents are ''fundamentally different'' and should be recognized as such. In fact, he suggested, they should be recognized as so fundamentally different that they might only be awarded a lifespan of three to five years, as opposed to the usual 20.Jeff Bezos' proposal is novel and constructive. Immediately, however, it came under fire from those who said special treatment for certain categories of patent is inappropriate, and that all patents must be treated the same.

Perhaps this is so. In either case, a simple possibility for profound patent system reformation lurks right under our collective noses, and has yet to be recognized: concurrent patents.

Under the patent system as it now stands under U.S. law, the first party to devise an invention receives a patent and holds the exclusive rights to the technology it covers for 20 years from the date of application (in effect, given the examination period, this gives the holder about 17 years).

The monopoly conferred by a patent - monopoly otherwise condemned in our free market system - was supported in theory by a persuasive social argument: The short-term benefit to the inventor gives him incentive; the long-term benefit that accrues to society from use of the invention after the patent expires outweighs the costs imposed upon society from the initial monopoly. A perfectly reasonable theory.

Today, however, the infrastructure designed to produce innovation is a core component in every major institution. As a result of this advanced research-and-development infrastructure, conferences, journals, listserves and other information-sharing tools tend to lead inventors in similar directions at roughly the same time.

The U.S. patent system cloaks applied-for patents in secrecy. Doing so protects an inventor's interests in his proprietary knowledge during the examination period.

Doing so also means that it is not possible for patent applicants to know whether they were the very first to invent - and thus, whether they will receive that profitable right to exclude others - until some party receives a patent, the patent is published, and the guessing game is over.

Nowadays, the confluence of ideas in public forums, as well as the unprecedented pace of technical development, means that a virtually identical idea may be presented to the patent office by multiple parties within the ''blind'' period of examination secrecy.

The problem is especially acute in the constantly changing information-technology arena, where everyone knows there is a sizeable backlog in the software-examination division of the U.S. Patent and Trademark Office.

Patents for materially identical processes and procedures may sit, unretrieved and untended, for two years or more. In the interim, numerous successful businesses may be built on these same concepts.

Yet if a patent is eventually awarded, its owner has the right to force any other company using the protected technology to revamp it or to pay hefty licensing fees to the de jure monopolist.

As Kenneth Nichols recently observed in Computer magazine, the adoption of software-related technology tends to take place with astonishing rapidity. If the first to market - and only the first to market - receives patent protection, then the existing patent system rewards only one of many deserving parties.

The upshot is that other, similarly innovative parties are in effect punished, rather than rewarded, for investing resources in technical innovation. Fair competition degenerates into a winner-take-all lottery.

One doubts that the authors of the Constitution, which recognized the need for patents in Article I, Section 8, intended such a result.

And neither is it the intent of the patent system to create an effective monopoly over the Internet, which is the most fundamental of public networks. Yet by designating a single owner for a potentially powerful technology, the patent system narrows the foundation upon which our technological future moves forward.

The clearest theoretical resolution to this problem is to reward similar inventors similarly.

If materially identical concepts are presented to the PTO for patenting within, say, six months of one another, the statutorily required preservation of secrecy means that none of the applicants is aware that their idea is covered by another patent application. All of the applicants should be entitled to some portion of the reward associated with the innovation. They should all receive patents - concurrent patents.

The concurrent patent would allow any of its holders to do business using the patented technology, and would allow any of its holders to license out the technology to other parties seeking to deploy the technology in their own businesses.

Providing for concurrent patents would help balance the range of interests seeking a technology lead in the fast-moving software arena. It would allocate economic rewards more fairly in this marketplace than is currently possible.

The concurrent patent concept could be applied equally across all categories of patents, avoiding the objections associated with applying special treatment to particular varieties of patent. In addition, this approach would not create special classes of patents that are difficult to distinguish and that are a continuing administrative burden as trends in patenting evolve over time.

Concurrent patents could restore the patent system's effects to those intended by the founders: strong incentives, fair rewards.