Paul Allen’s lawsuit against Google and other search, advertising and e-commerce companies is attracting a lot of attention, including explanations from folks like me about how the lawsuit demonstrates one way to profit from privately funded research, to how the lawsuit demonstrates horrific inefficiencies in patent licensing mechanisms, contributing to valuation of even economically simple inventions. Another thing the news coverage of this lawsuit brings to light is the fundamental misunderstanding of patents. Consider the following from a recent article covering Interval’s amended filing:
Finally, the last patent (the ‘682 patent), he appears to be claiming that it covers making recommendations to users on things they might like.
Leaving aside, for the moment, the fact that Interval’s legal team was able to re-file this case with relatively little effort (as I previously predicted), as well as the fact (as I also predicted) that the new complaint provides less information than what was provided in this report (see link on page 2), Masnick’s cursory treatment of the patent coverage highlights the reason why media coverage of cases like these ignite passionate responses in the minds of inventors. Certainly, the thought that someone owns the right to make recommendations is preposterous, so, the reader is inevitably led to the conclusion that a patent system that allows such patents is unacceptable.
However, a closer examination of US Patent 6,757,682 reveals claims reciting a specific series of steps, rather than a simple idea. Consider Claim 3, for example:
3. A method of disseminating to a participant an indication that an item accessible by the participant via a network is of current interest, comprising:
receiving in real time from a source other than the participant an indication that the item is of current interest; processing the indication;
determining an intensity value to be associated with the indication and an intensity weight value, and adjusting the intensity value based on a characteristic for the item provided by the source; and
informing the participant that the item is of current interest.
Masnick’s explanation focuses on the first (a method of disseminating … an indication) and the last (informing the participant) portions of this claim when suggesting that it “covers making recommendations” while skipping the real focus of the claimed invention, involving receiving “indications” about relevance of information, and determining an “intensity value” and “intensity weight value” based on information learned from the source and the circumstances of the indication. Much of the argument in Interval’s case will center around what the inventor’s meant by “intensity value” and what Google and the other defendants do (or don’t do, as the case may be) that are identical or equivalent to the claimed “intensity value.”
As a practical matter, is this a dramatic technical limitation on Interval’s patent? Not really. Any information processing system in modern times would most likely want to pass along only the “best” information, rather than all of the information. However, for those who make their living by studying patents, as well as educating and counseling clients about patents, there is a world of difference between the suggestion that a patent covers the idea of making recommendations, and the reality that the patent covers a process by which the recommendations are chosen. Questions still remain, such as whether the determining step is too broadly described to have been novel in the first place; whether phrases like “intensity value” are so vague or devoid of meaning as to render the claim indecipherable; or whether contemporary needs to filter “noise” in favor of valuable information render the use of intensity values to be obvious.
Another difference between Masnick’s take and my own is that one leads readers to a snap judgment about the effectiveness of the US patent system, while the other encourages readers to stop, think, and come to their own conclusions.