Editor’s Note: The U.S. Patent Trademark Office (USPTO) is seeking input on software-related patents beginning with its first-ever public roundtable in Silicon Valley next week, so Wired Opinion is closing its special series on “The Patent Fix” with specific solutions like these that don’t require changing the law. This expert’s opinion is one of the proposals under consideration by the USPTO.
The issues and fallout around software patent protection have manifested in many ways — from patent trolls to patent thickets. But at the heart of the problem are fuzzy boundaries in the patent claims.
Any incentive system that’s based on protecting exclusive rights depends on the ability of competitors to determine where the rights of others end and where theirs begin: where they can productively pursue their projects. This is true in developing land, and it’s true in patenting inventions.
But where land boundaries are clear and searchable, the boundaries surrounding software patents are notoriously fuzzy.
Unfortunately, wholesale reshaping of the IP protection system for software appears to be out of reach as there are political and pragmatic constraints on such reform. So I have advocated putting more onus on the patent applicants to clarify the boundaries of their claimed inventions at the outset of the process.
This approach improves the quality of patent examination, enabling competitors to more easily navigate patent boundaries. It reduces the tremendous confusion surrounding the interpretation of patent claims by judges. And it can do these things without any new legislation.
Why are software patents so vague? Many software inventions are inherently abstract. But the PTO also doesn’t place a lot of emphasis on ensuring applicants clearly delineate their inventions; examiners get little or no reward from scrutinizing claim scope, and it takes time to fine-tune the many claims set forth in so many applications. Since the inventor (or prosecuting attorney) gets to define the scope of the patent, there is an incentive to seek broad and malleable claims. Together, these factors cost the technology sector, investors, courts … all of us.
So the patent system needs to take patent claim clarity more seriously.
Patent claims should be presented through a standardized, detailed format, with check-boxes and related explanatory fields for clarifying the meaning and boundaries of claims. Such a structure should:
Separate claim restrictions. Currently, the only constraints on claim format are the “one sentence” rule (an antiquated rule requiring patent claims to be drafted as single sentences) and other provisions governing dependent claims. The frequent result, especially in software-related inventions, is amorphous claims. But requiring claim restrictions to be set forth in separate fields would help examiners, the public, and courts distinguish the claim components accurately and easily. Furthermore, search engines would be able to parse patent claims by claim restrictions — the most useful search unit for locating relevant prior art.
Define ambiguous terms, and include a glossary of specialized terms. Any potentially ambiguous claim terms — including terms of degree (“approximately,” “substantially) — should be defined. Applicants would include a glossary of terms that are prone to confusion, and designate a default dictionary for guiding the interpretation of any undefined claim terms.
Distinguish between limitative and illustrative embodiments. The form should force applicants to expressly indicate whether examples in the specification are intended to limit — or merely illustrate — the patent claims. This is a critical distinction in determining the scope of patent claims, yet applicants currently have no duty to clarify their intentions here. The current approach allows applicants to pick a position based on litigation strategy … which leaves their competitors, and the courts, in the dark.
Identifying means-plus-function elements. Patent law only presumes that the phrase “means for” invokes the “means-plus-function” interpretive rule [section 112(f)]]. It is not often clear which examples set forth in the specification correspond to the “means” clause. As a result, competitors and courts struggle to interpret such claim elements. Much of this confusion can be eliminated by having a check box for applicants to designate their intention (to invoke section 112f), along with hyperlinking or other transparent techniques for specifying the corresponding “structure, material, or acts.” This would turn an often muddy claim element into a crystalline disclosure.
Record interviews. The PTO permits phone or in-person interviews with applicants to better understand what’s being claimed and the basis for a patent’s validity. Since these discussions form a basis for granted patents, they ought to become part of the public record so others can better understand why a patent was granted. Such statements help limit the scope of patent claims by clarifying what the examiner and the applicant understood up-front.
Make the form electronic. All of the above should be implemented through an electronic application form, resulting in a hyperlinked, red-lined, layered, online-accessible form. That way, examiners could easily highlight ambiguous claim terms in their requests for clarifications, and applicants could annotate their responses. Capturing this interchange would enable competitors, subsequent inventors, litigants, and courts to more easily decipher how and why patent claims were crafted so in the first place.
This proposal provides simple, cost-effective ways of delineating claims at the front-end of the patent life cycle. It would also move the arcane, 19th century paper patent file system into the 21st century, not to mention enable a new generation of search and evaluation tools.