In 2009, I wrote for IPWatchdog about the case of In re Lister where the patent applicant barely avoided a potentially fatal “printed publication” bar based on his own copyrighted manuscript that was searchable in two commercial databases, but only after the critical “bar date.” See CAFC Rules Patent Applicant’s Own Copyrighted Manuscript Not Publicly Accessible. That was followed by a much more comprehensive article that I wrote which was published in 2011 on applying the “printed publication” bar to electronically posted documents. See Guttag, Applying the Printed Publication Bar in the Internet Age: Is It as Simple as Googling for Prior Art. That 2011 article included a discussion of the Lister case, but focused primarily on the 2008 case of SRI International, Inc. v. Internet Security Systems, Inc. where a majority of the Federal Circuit panel ruled that there were “genuine issues of material fact” sufficient to preclude a grant of summary judgment that an electronic document temporarily posted (but freely accessible) on the FTP server of the patentee (SRI International) was a “printed publication” bar.
One of the “revelations” I discovered in researching and writing that 2011 article is that the “printed publication” bar is really two distinct and parallel doctrines that the Federal Circuit has all too often unfortunately commingled together. The first, more active form is the “dissemination” doctrine. The second, more passive form is the “publicly accessible” doctrine (sometimes referred to as “constructive publication”).
The recent case of Voter Verified, Inc. v. Premier Election Solutions, Inc. (Nov. 5, 2012) is the latest example of the Federal Circuit being oblivious to the parallel existence of these two doctrines (and especially the requirements for each of these distinct doctrines) for applying the “printed publication” bar.
Under the “dissemination” doctrine, the courts focus primarily on the following three factors: (1) who received the document (i.e., who was the “relevant audience”); (2) how widely was the document circulated (i.e., how many copies of the document were disseminated to the relevant audience); and (3) how long (or temporarily) was the document in circulation. See, for example, the seminal 1928 case of Jockmus v. Leviton authored by Judge Learned Hand.
Under the “publicly accessible” doctrine, the courts focus on where the documents are stored and especially whether those documents are “catalogued or indexed in a meaningful way.” See, for example, the 1989 case of In re Cronyn. Sadly, in the 2006 case of Bruckelmyer v. Ground Heaters, Inc., the majority opinion of the Federal Circuit (authored by Judge Lourie) paid lip service to the requirement that the documents be “catalogued or indexed in a meaningful way” in ruling that two cancelled drawings figures remaining in the file wrapper of a Canadian patent constituted a “printed publication” bar. As dissenting Judge Linn correctly observed, those two cancelled drawings figures remained in file wrapper of that Canadian patent “by sheer happenstance,” and were otherwise “not indicated” as being present in that file wrapper, i.e., those cancelled drawings figures were not “catalogued or indexed in a meaningful way.”
One of the dangers I pointed out with respect to electronically posted documents being “publicly accessible” in my 2011 article was that of being unduly influenced by current advances in Internet search capability, even though the document in question was posted 15, 10, or even 5 years prior. That was particularly true of the SRI International case where the document in question was posted “in 1997 during the early stages of increased Internet usage, not 2006 when the patent infringement suit was filed.” In fact, at the end of my 2011 article, I cautioned that “[c]are should be to taken to assess Internet search capabilities as of the time” the document was allegedly electronically posted “without reference to those capabilities at the time the case was decided.”
In Voter Verified, the Federal Circuit panel ruled that an article (the “Benson” article) obtained from online periodical was a “printed publication” bar available as prior art under 35 U.S.C. § 103. But what was disconcerting was the discounting once again of the Cronyn requirement for electronically posted documents to be “meaningfully indexed” if it was to be “publicly accessible.” Indeed, while Judge Lourie’s opinion in Voter Verified is couched in terms of the “publicly accessible” doctrine, the facts much more strongly support a “printed publication” bar under the “dissemination” doctrine. (There was also an interesting question of whether the patentee in Voter Verified missed out on an indirect infringement argument based on the recent en banc ruling in Akamai Technologies, Inc. v. Limelight Networks, Inc. that was deemed “waived” by Lourie)
In Voter Verified, the asserted reissue patent (U.S. Reissue Pat. No. RE40,449, a reissue of U.S. Pat No. 6,769,613) which was originally filed December 7, 2000 disclosed and claimed automated systems and methods for voting in an election. The key feature of these claimed automated systems and methods was a self-verification procedure which permitted “machine and human error” to be detected before the ballot was submitted by the voter for tabulation. On summary judgment, the district court ruled that claim 49 of the asserted reissue patent was invalid under 35 U.S.C. § 103 in view of the Benson article which disclosed an electronic voting booth which automatically prompted the voter to confirm whether the filled out ballot card was accurate, and if not, permitted the voter to restart the ballot process (the incorrect ballot card then being shredded). The Benson article was obtainable from an online periodical known as the Risks Digest directed to computer safety and security. Here are potentially relevant findings of fact listed in Voter Verified with respect to the Benson article being a “printed publication” bar prior to the critical “bar date” (i.e., December 7, 1999):
1. The Benson article was posted to Risks Digest on March 4, 1986. At that time, Risks Digest was distributed online via a subscription mailing list and was also made available for download through an FTP site.
2. Starting in January 1995, all content published by Risks Digest (including the Benson article) became available worldwide on the Internet through the web site http://catless.ncl.ac.uk/Risks.
3. The Risks Digest was “well known to the community interested in the risks of computer automation, including those concerned with electronic voting technologies” and by 1999 “contained more than 100 articles relating to electronic voting.” All submissions for publication in the Risks Digest were “treated by the community as public disclosures.” Users could also “freely and easily copy Risks Digest content.”
4. Since September 1995, the Risks Digest web site included “a search tool” that would have retrieved the Benson article in response to search terms such as “vote,” “voting,” “ballot,” and/or “election.”
On appeal, Voter Verified argued that the Benson article must be “searchable by pertinent terms over the internet” to qualify as a prior art “printed publication” bar. Voter Verified also argued that the accused infringer (Premier Election) “provided no evidence of any indexing on any database” that would have allowed the interested public to locate the Risks Digest website, “much less the Benson article contained therein.” Voter Verified also emphasized “indexing” as being a “determinant of public accessibility.”
But Judge Lourie disagreed, stating that “indexing” was not a “necessary condition for a reference to be publicly accessible,” citing Lister. (That Lourie found “indexing” not to be an “absolute prerequisite” for a document to be “publicly accessible” is unsurprising, given that he was the author of the Bruckelmyer majority opinion which said as much.) Instead, Lourie ticked off findings of fact nos. 1 through 4 above to support “the district court’s conclusion that the Benson article was publicly accessible before the December 7, 1999” critical date. In addition, Lourie argued that the “uncontested evidence indicates that a person of ordinary skill interested in electronic voting would have been independently aware of the Risks Digest as a prominent forum for discussing such technologies” (whether or not that the Risks Digest web site had been “indexed” by 1999). Once the Risks Digest web site was accessed, Lourie further argued that “such an interested researcher would have found the Benson article using that website’s own search functions and applying reasonable diligence.”
But Lourie also acknowledged that, “although commercial internet search engines were available by 1999, the record is devoid of evidence indicating whether or not the Risks Digest website had been indexed by any such services as of the critical date [i.e., December 7, 1999].” In other words, whether search engines, as of December 7, 1999, could have found the Benson article via the Internet was in doubt, as even admitted by Lourie. And given that the accessibility of the Benson article via the Internet was being determined on a motion for summary judgment (not after a full trial on the merits), Lourie’s position that this article was “publicly accessible” as a matter of law so as to constitute a “printed publication” bar is problematic in my opinion. (Another of the “cautions” in my 2011 article is the need to avoid rushing to summary judgment on “printed publication” bars involving electronically posted documents “because of the factual intensity required to resolve such bars for such documents.”)
That Lourie’s opinion focused on the “publicly accessible” doctrine is also (unfortunately) largely due to the arguments asserted by the patentee in trying to avoid the “printed publication” bar posed by the Benson article. But what makes Judge Lourie’s reliance upon “publicly accessible” doctrine so irksome in view of such “debatable” facts is that it was unnecessary. Instead, Lourie had a much more direct and “ironclad” basis to support his reasoning that the Benson article was a “printed publication” bar under the “dissemination” doctrine. As noted in finding of facts above, the Benson article: (a) had been posted for at least 13 years prior to the critical date; (b) to the Risks Digest “well known to the community interested in the risks of computer automation, including those concerned with electronic voting technologies”; (c) distributed online via a subscription mailing list. In other words, the Benson article satisfied the three factors (who received the article, how widely was the article circulated, and for how long) required for a “printed publication” bar under the “dissemination” doctrine. See also the 1985 case of Massachusetts Institute of Technology v. AB Fortia, as well as Jockmus v. Leviton, all of which ruled that a “printed publication” bar had occurred based on “dissemination” of the invalidating document to the “relevant audience,” i.e., those of ordinary skill in that art.
In the end, Voter Verified reached the correct result that the Benson article was a “printed publication” bar. But the “publicly accessible” doctrine relied upon by Judge Lourie’s opinion was not the best (and most direct) pathway to get to that result. Even more distressing, the important Cronyn requirement of “meaningful indexing” of the electronically posted document was again discounted by Voter Verified. Put differently, if you can’t readily find the electronically posted document by doing a reasonable search (with appropriate search terms) via the Internet, how can you say that the document is “publicly accessible” to be a “printed publication” bar? (One of the questions I noted in my 2011 article that the Lister case left unanswered was when the number of Internet search results became sufficiently “manageable,” and thus not “overwhelming” so as to support a “printed publication” bar.) And because the “printed publication” bar remains part of the America Invents Act (AIA) when the “first to file” regime goes into effect March 16 of next year, the question of when an electronically posted document becomes “publicly accessible” will assume even greater significance.
*© 2012 Eric W. Guttag. Posted on IPWatchdog.com November 8, 2012.