Something old, something new Part 1
A few weeks ago, the Federal Circuit decided an interesting case, Bruckelmyer v. Ground Heaters, construing the printed publication prong of 35 USC 102(b). The case garnered a bit of attention because the court went a bit further than it had before in holding foreign patent filing materials to be publicly accessible “printed publications.” Patently-O covered it, as did Patent Prospector.
The specific invalidating materials at issue in Bruckelmyer are two drawings. These two drawings were included in the originally filed version of a Canada patent application, but they were excluded from the final, issued version of that Canada patent.
Bruckelmyer is a tough case, as Judge Linn’s dissent in the case indicates. The majority frames the decision as a matter of applying controlling precedent — namely, In re Wyer, 655 F.2d 221 (CCPA 1981). Having read Wyer, I’m not convinced it’s as tightly “controlling” as the Bruckelmyer majority states. On the other hand, Bruckelmyer may be the appropriate result. In any event, I admit that my sense of the cases comes from a policy rationale that none of the opinions brings up. That policy rationale is what I want to talk about here. I’ll take up Wyer and Bruckelmyer in future posts. More below the fold …
So, how should we decide whether something is sufficiently publicly accessible to count as a “printed publication” under 102(a) or (b)? Well, we should, if we can, relate our criterion for “printed publication” to the reasons we have a novelty requirement. One of those reasons is to avoid encouraging people to (re)invent existing solutions in the hopes of getting patent protection.
Consider the following thought experiment. (I’m adapting this from the cogent discussion of the point in Robert P. Merges & John F. Duffy, Patent Law & Policy: Cases & Materials 419-20 (3d ed. 2002).) Inventor Joe confronts a practical problem. Would we rather have Joe invent a solution to the problem or use a known solution to the problem? The answer, from an efficiency perspective, is plainly, “whichever approach is likely cheaper.” Now, imagine our patent system has no novelty requirement at all yet; instead, it merely has a no-patent-for-locating-a-known-solution rule. What will Inventor Joe do in response to his problem? And is that what we want him to do, from an efficiency perspective?
Suppose Inventor Joe correctly estimates that the cost of locating the known solution is much higher than the cost of inventing the solution on his own. Joe will invent on his own, which is the cheaper course of action. Society is better off, and Joe is better off (he saved his own resources), and he can also receive patent protection.
Suppose, instead, that Joe correctly estimates that the cost of inventing on his own is much higher than the cost of locating the known solution. Society is better off if Joe locates the known solution. Joe’s private interest, however, may be different. If Joe invents (which, under the hypothetical patent system, means he can obtain a patent) and can generate enough profit from exploiting his patent right, he can make inventing cheaper for him than locating the known solution. Society is worse off, because Joe takes the higher-cost route.
Can we prevent Joe’s private interest from getting out of alignment with the social interest? The novelty rule can certainly help: If we add a novelty rule (i.e., “no patents for things that are known”), and thus deny Joe patent protection if what he invented could likely have been found more cheaply by searching existing information, we eliminate Joe’s “I’ll just invent it myself” strategy. The trick, then, is to define the category of “existing information” to include the material that is likely to be cheaper to search for a known solution, relative to the likely cost of (re)inventing a solution. (As Merges & Duffy put it, “[t]he best novelty rule might exclude from consideration categories of prior art if the courts can be fairly confident that the cost of finding the information will be higher than the cost of recreating the information through independent research.” Id. at 420.)
The distinctions sections 102(a) and (b) make are a little easier to understand against this backdrop. For example, although intra-U.S. public use and knowledge can be invalidating, extra-U.S. public use and knowledge can’t be invalidating … unless that extra-U.S. information has been reduced to a “printed publication.” This roughly tracks the common sense predictions one might make about the relative cost of locating known solutions to a problem in different circumstances (inside v. outside the U.S., written down v. not written down). It is also easier to understand the contours of the “printed publication” concept itself. Well-indexed, readily searchable text information is less costly to find than non-indexed, isolated text information.
To sum up, we should interpret section 102’s “printed publication” requirement such that we confer “printed publication” status on information that is strongly likely cheaper to locate in response to a problem than to recreate from scratch in response to that problem.
Interesting thoughts on the rationale of what should or should not be a printed publication. Your policy analysis seems consistent with the other recent and somewhat controversial CAFC decision on what constitutes a “printed publication,” although the court did not decide the case on the basis of a policy analysis. In re Klopfenstein, 380 F.3d 1345, 1352, 72 USPQ2d 1117, 1121 (Fed. Cir. 2004) (affirming Board’s ruling that a fourteen-slide presentation, made by named inventors and a colleague more than one year before the critical date, was a printed publication that invalidated the application claims, and rejecting applicants’ argument that the slides were not a “printed publicationâ€? because they were never distributed or indexed and catalogued, and ruling that the slides had sufficient public accessibility to be a “printed publicationâ€? because they were displayed continuously for almost a three-day period at two technical conferences, with out any reservation or prohibiting of copying, the attendees were of those of skill in the art, and only a handful of slides with two or three bullet points fully conveyed the claimed invention).
From a pratical prespective, there are questions on just how far the Bruckelmyer decision goes. The court appears to rely heavily on the fact that the text of the sepcification of the foreign application contained a reference to the specific allegedly invalidating subject matter, and therefore, it could be presumed that one of skill seeing the issued foreign patent would be motivated to obtain and review the entire file history and thereby find the canceled drawings that showed the anticipatory subject matter. In fn. 3 the court warned that it was not “express[ing] an opinion regarding the unusual situation where the prior art reference at issue is contained in a foreign file wrapper, but otherwise bears little relation to the subject matter of the issued and published patent.” Denying printed publication status to subject matter bearing little relation to the issued patent seems consistent with your policy analysis since if the subject matter bears little resembleance to the issued patent, then the costs of searching for that subject matter likely exceed the costs of just searching the information in the issued patent.
While these policy thoughts are interesting and helpful to understanding the legal doctrines, Judge Lourie, in some recent remarks to a bar association (and printed in this week’s patent BNA weekly newsletter/journal), noted that the CAFC is not a policy-driven court. Rather, in his view, the court is duty bound to apply the statutes and precedents that exist as best it can to the situtation it faces without regard to whether the precedents and statutes achieve the best policy goals. He made these remarks in response to suggestions by the court’s critics that the CAFC should have a closer relationship with acedemia. Accordingly, while policy concerns can sometimes impact how a specific statute or precedent should be interpreted, they may not carry the day for a litigant at the Federal Circuit in the face of contrary precedent. Indeed, long ago the Sup. Ct. instructed in the context of section 102 that Congress, and not policy, sets the boundaries on prior art status since “the patent or publication can only have the effect as evidence [of prior art] that is given to the same by the act of Congress.” Bates v. Coe, 98 U.S. 31, 33-34 (1878). Not all members of the CAFC may agree with Judge Lourie. Several years ago, another member of the court in public comments to the bar noted that litigants often failed to address policy arguments in their briefs to the court, and that such arguments should be made. The FTC in its famous 2003 report “To Promote Innovation” noted the CAFC’s apparent lack of interest in addressing policy issues for patent cases, and suggested that the court should rethink its position. See Ch. 6 of the 2003 rpt at pp 7-9.
Given this flux, the policy discussions like the one you applied to the Bruckelmyer should continue. In time they’ll find a proper home; although it may not be in the body of a CAFC opinion. But nonetheless, these discussions should help people more fully understand the issues. I for one enjoy reading the policy discussions, as they force me to think of the patent law beyond the application of a series of judicial precedent and extension of those preceents.
Comment by Bob Matthews — May 14, 2006 @ 10:58 am
[...] Bob Matthews, in a comment to my recent post on the “printed publication” portion of section 102, offers some important observations on the role of patent policy in deciding patent cases, and on the way some Federal Circuit judges think about that role. [...]
Pingback by The Fire of Genius » Patent law is patent policy — May 14, 2006 @ 1:38 pm
[...] As I explained in my prior post on the “printed publication” aspect of section 102, “we should interpret section 102’s ‘printed publication’ requirement such that we confer ‘printed publication’ status on information that is strongly likely cheaper to locate in response to a problem than to recreate from scratch in response to that problem.” This approach is an effort to link our interpretion of “printed publication” to our basic reasons for having novelty requirement in the first place. [...]
Pingback by The Fire of Genius » Something old, something new Part 2 — May 14, 2006 @ 4:05 pm
[...] As I explained in my first post on the “printed publication” aspect of section 102, “we should interpret section 102’s ‘printed publication’ requirement such that we confer ‘printed publication’ status on information that is strongly likely cheaper to locate in response to a problem than to recreate from scratch in response to that problem.” This approach is an effort to link our interpretion of “printed publication” to our basic reasons for having novelty requirement in the first place. [...]
Pingback by The Fire of Genius » Something old, something new Part 3 — May 16, 2006 @ 11:06 pm
[...] Early last month, I blogged a bit about the Federal Circuit’s decision in an interesting “printed publication” case, Bruckelmyer v. Ground Heaters, Inc. Those posts are here, here, and here. [...]
Pingback by The Fire of Genius » Bruckelmyer Redux — June 28, 2006 @ 8:41 am
[...] Des figures supprimées lors de la poursuite d’une demande de brevet canadienne et contenues dans le dossier de poursuite sont utilisées valablement comme de l’art antérieur invalidant un brevet dans l’affaire Bruckelmyer v. Ground Heaters (Fed. Cir. 2006) -Texte du jugement original -Texte de l’appel refusé -Commentaire de Patently-O, et 2eme commentaire -Commentaire de Fire of Genius [...]
Pingback by “C’est é-patent!” » Blog Archive » Collection de faits divers du USPTO et des États-Unis au cours de cet été 2006 — August 22, 2006 @ 7:03 am