Federal Circuit Finds Mere Uploading Of a Paper to an FTP Site Does Not Make the Paper Publicly Available
In SRI International, Inc., v. Internet Security Systems, Inc., 511 F3d 1186, 85 USPQ2d 1489 (Fed. Cir. 2008) the Federal Circuit affirmed in part, and vacated and remanded in part the District Court’s decision finding Internet Security Systems (ISS) was liable for infringement of U.S. Patent Nos. 6,484,203 (“the ‘203 patent”), 6,708,212 (“the ‘212 patent”), 6,321,338 (“the ‘338 patent”), and 6,711,615 (“the ‘615 patent”) (collectively the SRI patents).
The SRI patents relate to cyber security and intrusion detection. Specifically, the SRI patents describe “[a] computer-automated method of hierarchical event monitoring and analysis within an enterprise network including deploying network monitors in the enterprise network, detecting, by the network monitors, suspicious network activity based on analysis of network traffic data.” All four patents originated from a November 9, 1998 application by inventors Phillip Porras and Alfonso Valdes.
The EMERALD 1997 paper
In June 1997, SRI posted on its SRI file transfer protocol (“FTP”) server a paper entitled “EMERALD: Event Monitoring Enabling Responses To Anomalous Live Disturbances” (“the EMERALD 1997 paper”). The EMERALD 1997 paper contains a detailed description of a tool for tracking malicious activity across large networks. The EMERALD 1997 paper and the ‘212 specification contain some overlapping material.
During prosecution of the ‘212 patent, SRI disclosed the EMERALD 1997 paper in its Information Disclosure Statement, and the paper was listed in the patent’s Other Publications section. However, the application issued as the ‘212 patent despite this disclosure.
The District Court granted summary judgment in favor of ISS finding that the EMERALD 1997 paper anticipated the ‘212 patent. SRI challenged the District Court’s grant of summary judgment contending that the EMERALD 1997 paper is not an enabling disclosure with respect to the ‘212 patent.
On appeal, the Federal Circuit agreed with the District Court finding that the EMERALD 1997 paper did enable one of ordinary skill in the art to practice the claimed invention. As evidence of enablement, the Federal Circuit noted that the ‘212 patent did issue with a similar disclosure as the EMERALD 1997 paper, including similar figures and descriptions for those figures. As such, the enablement of the EMERALD 1997 paper is tied to the enablement of the same description in the ‘212 patent. The Federal Circuit also found that the testimony of SRI’s expert witness, Dr. Kesidis, regarding the differences between the ‘212 patent disclosure and the EMERALD 1997 paper lacked the detailed analysis needed to create a genuine issue of material fact. As such, the Federal Circuit affirmed the District Court’s finding that the EMERALD 1997 paper was enabling for purposes of 35 U.S.C. §102.
The Live Traffic Paper
The District Court also granted summary judgment in favor of ISS, ruling that all asserted claims in the patents were anticipated due to the publication of “Live Traffic Analysis of TCP/IP Gateways” (“the Live Traffic paper”). Mr. Porras and Mr. Valdes authored the Live Traffic in 1997. SRI displayed the Live Traffic paper on its web site on November 10, 1997 for seven days. On August 1, 1997, Mr. Porras sent an email to Dr. Bishop, the Program Chair for SNDSS, in response to the SNDSS call for papers. He included the specific FTP address, ftp://ftp.csl.sri.com/pub/emerald/ndss98.ps, in the email. In four instances, Mr. Porras provided the full path and filename of the Live Traffic paper. The District Court ruled that the Live Traffic paper was a printed publication that anticipated all asserted claims of the four patents-in-suit. The District Court thus determined that SRI’s FTP server’s directory structure gave access to the article to a person of ordinary skill in the art. In the District Court’s view, one of ordinary skill would know that the SRI FTP server contained information on the EMERALD 1997 project and therefore would navigate through the folders to find the Live Traffic paper.
On appeal, the Federal Circuit did not find enough evidence in the record to show that the Live Traffic paper was publicly accessible and thus a printed publication under 35 U.S.C. § 102 (b). The court found that SRI’s case fell somewhere between Application of Bayer, 568 F.2d 1357, 1358-59 (C.C.P.A. 1978) and In re Klopfenstein, 380 F.3d 1345, 1347-50 (Fed. Cir. 2004), two cases dealing with public access.
In Application of Bayer, a graduate thesis in a university library was cited as prior art. The library had not catalogued or placed the thesis on the shelves and only three faculty members knew about the thesis. In that case, the court found that the thesis did not constitute a printed publication for purposes of 35 U.S.C. §102 because a customary search would not have rendered the work reasonably accessible even to a person informed of its existence.
A similar result occurred in In re Cronyn, 890 F.2d 1158, 1161 (Fed. Cir. 1989), where the thesis document was in a library, but was only indexed by the author’s name. In this instance, the Federal Circuit found that there was no public accessibility because “the only research aid in finding the theses was the student’s name, which of course, bears no relationship to the subject of the student’s thesis.” 890 F.2d at 1161.
In contrast, in In re Klopfenstein, two professional conferences displayed posters. These posters were printed publications for purposes of 35 U.S.C. §102 because their entire purpose was public communication of the relevant information.
Using the Klopfenstein scenario, the Federal Circuit found that posting the paper to the FTP site was most similar to placing posters at an unpublicized virtual conference with no attendees. Like posters at a vacant and unpublicized conference, only a person who wandered into the conference by happenstance or knew about the conference via unpublicized means would have access. As such, the Federal Circuit found the paper even less available to the public than the thesis in Bayer. This is largely due to the fact that the Live Traffic paper was not finished and posted on the FTP site solely to facilitate peer-review in preparation for later publication. The court further found that the relatively obscure filename would not lead a person to find the unpublicized paper such that the paper was not indexed in a meaningful way as is required to allow a member of the public to find the paper. As summarized by the Federal Circuit,
The record on summary judgment does not show that an anonymous user skilled in the art in 1997 would have gained access to the FTP server and would have freely navigated through the directory structure to find the Live Traffic paper. To the contrary, the paper’s author, Mr. Porras, thought it necessary to provide Dr. Bishop with the full FTP address for the file. Surely Dr. Bishop, the Program Chair for SNDSS, would have qualified as one of ordinary skill in the art in 1997. Yet, despite his knowledge of the field, FTP servers, and the paper, Dr. Bishop apparently would not have found the reference without Mr. Porras’s precise directions.
As such, the Federal Circuit rejected the District Court’s holding that the Live Traffic paper was a printed publication for purposes of 35 U.S.C. §102.
Significance for Patent Owners
Given the prevalence of online collaborative research, inventors are tempted to post drafts of papers for comment by coinventors and for limited peer review. Such sites, when publicly accessible, do present a risk that the drafts have been published for purposes of 35 U.S.C. §102. However, as demonstrated in SRI International, the mere fact that a paper can be accessed publically is insufficient to find that the paper was published for purposes of 35 U.S.C. §102(b). Instead, the access must be through a mechanism which is meaningful for someone, without prior knowledge of the paper or its location, to locate that paper. Thus, for purposes of collaborative research, the use of online postings for drafts does not necessarily represent a patentability bar under 35 U.S.C. §102(b) as long as the posted draft is not well indexed (and is not at a location normally indexed by such search engines as Google or Yahoo).