In SRI International, Inc. v. Internet Security Systems, Inc., 511 F.3d 1186 (Fed. Cir. Jan 8, 2008), the Federal Circuit recently reviewed its previous case law on the printed publication requirement of 35 U.S.C. § 102(b). The SRI decision indicates that making an otherwise relevant prior art reference openly accessible to individuals via the Internet does not necessarily qualify the reference as a printed publication. This holding is notable because it suggests that intent may now be a factor in determining whether a publication is publicly accessible, while providing pointed commentary from the Federal Circuit on its own printed publication precedent.
Under 35 U.S.C. § 102(b), an inventor is not entitled to a patent if a printed publication dated more than one year before the filing of the application enables and discloses every element of a claimed invention; in other words, if each element of a claim can be found in a single prior art reference, the invention was already in the possession of the public, and cannot be patented. Congress introduced this provision in 1836. As technology advanced and new methods of printing were developed, public accessibility became the touchstone for assessing whether a reference is prior art under Section 102(b). The related concept of public dissemination and “availability and accessibility to persons skilled in the subject matter or art” has also come to play a role in this determination. Whether a reference is a printed publication is highly fact-specific and is determined on a case-by-case basis. It is within this historical context that the Federal Circuit addressed the invalidity arguments in the SRI case.
SRI, a network intrusion detection research company, sued Internet Security Systems for patent infringement. The patents at issue claimed priority to an application filed November 9, 1998. Notably, the patents incorporated by reference a technical paper entitled “Live Traffic Analysis of TCP/IP Gateways” (the “Live Traffic paper”) that SRI posted on its website on November 10, 1997. The article’s two authors submitted the Live Traffic paper for peer review for the 1998 Symposium on Network and Distributed Systems Security ("SNDSS"). As part of the peer-review process, SRI had emailed a version of the paper to a SNDSS committee in August 1997, and at the same time, placed the paper on SRI’s FTP server. The email directed the SNDSS committee chairman to the specific FTP address, which was not password protected. The Live Traffic paper was posted on the FTP server for a week. During that time, seven known individuals, other than SNDSS committee members, were directed to the specific FTP subdirectory to review documents related to network intrusion detection.
At the trial court level, the parties had agreed that the Live Traffic paper would be invalidating if it qualified as prior art, and Judge Robinson of the District of Delaware found that the paper was indeed a “printed publication” Under 102(b). On appeal, in a 2-1 decision, the Federal Circuit disagreed and remanded the case for further consideration, suggesting that the Live Traffic paper was not a printed publication because SRI did not intend it to be publicly accessible. Judge Moore dissented, finding that the paper was accessible and, therefore, a “printed publication.”
In reaching its decision, the Federal Circuit analyzed five prior cases covering a range of public accessibility scenarios. The panels in Cronyn and Bayer had held that the publications in question were not publicly accessible and did not therefore qualify as printed publications under Section 102(b). By contrast, the publications considered in Wyer, Klopfenstein, and Bruckelmyer were found to be invalidating references under 102(b), in view of their accessibility to the interested public.
The Bayer and Klopfenstein cases were discussed at length in SRI. In Bayer, a master’s thesis was determined not to be a printed publication. The key facts were that the university library took a long time to process the thesis once it was deposited, a period during which it was not catalogued or shelved. With the exception of faculty members to whom the thesis had been presented, no one had access to the thesis during this period. It was not publicly accessible. On the other end of the spectrum, in Klopfenstein, a technical poster was presented at a conference and displayed for three days. The displayed information had not been put into a hard copy format to be distributed at that time. However, the poster was held to be a printed publication, as it was accessible to an interested public that was capable of processing and retaining the information.
The majority and the dissent in SRI disagreed on where the Live Traffic paper fell within the spectrum of printed publication cases. The majority analogized the SRI facts to the uncatalogued thesis in Bayer, focusing on its impression that SRI did not intend to have the paper disseminated publicly. The majority perceived the Live Traffic paper’s FTP address to be particular to SRI’s organization, while not being classified, indexed, or abstracted in a manner that would permit outsiders to understand what was located at that address. Only one non-SRI person, the Program Chair for SNDSS, specifically knew about the availability of the paper on the FTP server. By comparison, the poster presented at the conference in Klopfenstein was publicly accessible. According to the SRI majority, because SRI was not promoting its FTP server to the relevant audience, and, therefore, Bayer was more analogous than Klopfenstein.
In dissent, Judge Moore found Klopfenstein highly relevant. After all, SRI had directed individuals, interested in intrusion detection, to the FTP subdirectory containing the Live Traffic paper. Like the district court, the dissent found it difficult to believe that a company involved in intrusion detection would argue that one skilled in the art of intrusion detection would not detect information purposefully posted on the internet.
The dissent set forth a list of factors to show why Klopfenstein was analogous to SRI. The first factor was the duration of display. The Live Traffic paper was displayed for a week more than double the time the Klopfenstein poster was displayed — which allowed ample opportunity for the public to capture, process, and retain the information. The second factor was the expertise of the audience, which in SRI would have been people interested in intrusion detection, individuals more than capable of understanding the paper. Another factor was how reasonable it would be to expect that the reference would not be copied by the public. In SRI, there were no protective passwords, disclaimers, anti-copying software, or other indications of privacy for the paper on the FTP server. The FTP server allowed for easy copying and retrieval of its contents. Having given the FTP subdirectory address to individuals with interest and expertise in this field, SRI created an expectation that the paper would be copied. Taking these factors together, the dissent concluded the Live Traffic paper was publicly accessible, and the reference was therefore a printed publication. However, these arguments did not prevail and the majority remanded the case for further analysis by the district court.
In addition to the specific analysis by the Federal Circuit about FTP servers, one lesson from SRI is that intent to make a reference publicly accessible is now a factor to consider in determining whether the reference constitutes a printed publication under Section 102(b). It remains to be seen how significant the intent factor will be in future 102(b) cases.