Improper to take Official Notice of legal conclusion (Ex parte Yardley)

Ex parte Yardley
Decided July 21, 2009
(Appeal 2009-001146; Appl. No. 10/689,379; Tech. Center 3700)

Ex parte Yardley reminds us that Official Notice is limited to facts, and taking Official Notice OF a legal conclusion is improper.

Some of the claims at issue here were directed to a method of making a paper napkin, and included limitations for dimensions and weight. The Examiner admitted that the reference did not teach ranges of dimensions, nor the weight of the product, but took Official Notice that "it would have been obvious to one of ordinary skill in the art to apply the folding method of [the reference] to various sizes and weights of webs including those of claims 82 and 87." (Decision, pp. 15-16.)

The Board held that Official Notice of obviousness is improper:

[Here] the assertion of official notice is not directed to establishing evidentiary facts, but is rather an attempt to establish the ultimate legal conclusion of obviousness as to the rejected claims. The procedures setting forth the circumstances in which official notice may properly be taken do not recognize any basis for the taking of official notice of a legal conclusion (MPEP § 2144.03), nor do we believe that this would be proper under any circumstances.
(Decision, p. 16.)

The way I understand this, taking Official Notice of a fact is acceptable because a fact is a Premise from which a Conclusion is drawn. Thus, taking Official Notice is a merely a short cut to proving the fact with real evidence — a shortcut that is permissible in limited circumstances. On the other hand, short-cutting the entire analysis by taking Official Notice of the Conclusion itself is not acceptable. 

Based on my own experience, and the BPAI cases I've read, it's pretty hard to traverse an assertion of Official Notice in a way that convinces the Board that you've complied with the law (see MPEP 2144.03(c), for starters). But if you ever see the Examiner taking Official Notice of a legal conclusion, argue that the MPEP § 2144.03 does not provide for this.

Obviousness is probably the strongest form of an "ultimate legal conclusion". But I think other aspects of obviousness analysis could be described as legal conclusions.

For example, I think you could traverse the following statement on the same grounds: "Official Notice is taken that it would be an obvious design choice to use an LCD display instead of a CRT display".

Or even this one: "Official Notice is taken that it's merely a matter of design choice to use an LCD display instead of a CRT display".

New Patent Blog: Gray on Claims

I've discovered a relatively new patent blog which I've added to my list of favorites: Gray on Claims. It's a blog about claim construction case law, authored by an IP Litigation Attorney at Foley & Lardner. Though the initially blog focused on Federal Circuit decisions, more recent posts have included BPAI decisions.

Non-Patent Prior Art: Web Pages

Recently I posted (here) about the use of software screen shots of as prior art, under the "known or used by others" prong of 102(a). Web pages are another kind of prior art which I see a lot. Pages from a website are considered publications and thus also available under 102(b). What sort of evidence of a publication date is an Examiner required to provide for a web page?

In most of the cases I've seen involving a web page as prior art, the Examiner used an Internet archive website (such as www.archive.org) to obtain evidence of the publication date. An archive site works as follows: type in a website URL, and the archive provides a list of dates for which past pages from that site are available; choose a date, and the archive serves up the website as it existed on that particular date; thus, you can browse to a copy of a particular page as it appeared on the website in the past.

The MPEP does not mention Internet archives, but does state that the critical date is "the date the item was publicly posted:"

Prior art disclosures on the Internet or on an on-line database are considered to be publicly available as of the date the item was publicly posted. Absent evidence of the date that the disclosure was publicly posted, if the publication itself does not include a publication date (or retrieval date), it cannot be relied upon as prior art under 35 U.S.C. 102(a) or (b). 
(MPEP 2128.)

The BPAI appears to approve of the use of archived web pages as evidence of an electronic publication date. I found two decisions in which the Board found an archive date to be sufficient evidence, and I found no decisions to the contrary.

In Ex parte Shaouy, the Examiner indicated that the web page was available in "internet archive 'wayback machine' " and provided a copy of the web page as retrieved from the archive. The copy contained the following URL at the top of the page: "http://web.archive.org/web/20001209085500/http://www.forecastpro.com/". The Board noted that the numbers encoded in this URL correspond to the date the Examiner used as a publication date (December 9, 2000). The Board found that the Examiner's reference to "internet archive 'wayback machine' " in the Advisory Action, in combination with the date stamp encoded in the URL, was sufficient evidence of a publication date. 

Ex parte Molander contained a similar fact pattern. The Examiner referred to the archive site and the archive date in the rejection. The print out of the archived web page included the URL of the archive site, and that URL encoded the date relied on by the Examiner. The Board once again found that this was sufficient evidence of a publication date for the web page.

So if the rejection isn't very clear about the date relied on for a web page reference, it's probably a good idea to look at the print out of the archived web page, and to check the encoded archive date for yourself.

Non-Patent Prior Art: Screen Shots of Software

I prosecute a lot of software patent applications, and it's not uncommon in my cases for Examiners to use a screen shot of software as prior art. A screen shot isn't a publication, but it is evidence of "known or used by others" under 102(a).

Software typically has a copyright notice with a date, so it's not surprising that an Examiner might rely on that copyright date as the date of public use. But a given version of software has a range of copyright dates. Which date(s) can the Examiner use?

The BPAI discussed this issue in Ex Parte Martinez. The application in that case had a filing date in 2001, and the Examiner presented screen shots of Microsoft Word 2000. One of these screen shots include a copyright message "© 1983-1999". The Board ruled that this copyright notice "provides prima facie evidence that the features of [Microsoft Word] were at least known in the United States in 1999." (Decision, p. 7.)

Note that the Board used the latest date in the copyright range, and not the earliest. Though the Board didn't make this point explicit, it's important. The earliest copyright date tells us nothing, since a copyright date range provides no information  about which particular features of the software were present in which version/at which date.

So if an Examiner uses a screen shot as a reference, the first thing to check is that he/she is relying on the latest date in the copyright range. But you may want to dig further, because sometimes software is released without updating the copyright date. For example, it's common for intermediate releases of software to be updated via a download, and that update doesn't necessarily include a revision of the copyright string. In that case, it's possible that the feature relied upon by the Examiner was added (via the update) after the latest copyright date.

If you have reason to believe that an update was applied after the latest copyright date, does the burden shift back to the Examiner to provide evidence that the feature was present before the update? Another BPAI decision, Ex parte Chen, appears to say Yes.

In Ex parte Chen, the software at issue was Microsoft Word, and instead of deciding on the merits the Board remanded the application to the Examiner to find corroborating evidence of the relevant date of the Word feature at issue. 

We note that the figure 1 includes both copyright information and an indication of what “Service Pack” was present on the computer at the time of the screen shots. While difficult to read it appears that SP-3 (Service Pack 3) was in use at the time of the recordation of the screen shot. From our brief Internet search, Service Pack 1a (SR1a) was available on or about May 12, 2000, Service Pack 2 (SP-2) was available on or about November 14, 2000, and Service Pack 3 (SP-3) was available on or about October 21, 2002. Therefore, if the functionality was not added by Service Pack 3, then the date of the reference should make Microsoft Word 2000 prior art, and if Service Pack 3 added the disputed functionality described/shown, then the date is after the critical date in the instant application and is not prior art.
(Decision, p. 1.)

The Board appeared to say that because the version information indicated that an update had occurred, the burden shifts back to the Examiner. This position is consistent with the idea that a copyright date is prima facie evidence.

What if the screen shot did not indicate that an update had been applied? Could the Applicant still shift the burden back to the Examiner by simply providing evidence that updates were available for that software?

This appears to be the Appellant's position in this case, since the Appellant never mentioned the presence of the Service Pack 3 string in the screen shot. Instead, the Appellant basically argued that online updates to software are common, that the Examiner had not "fully documented the history" of the software relied on, and that the the Examiner bore the burden of establishing that a reference is prior art. (Reply Brief, pp. 15-16.)

The Board didn't comment on the Appellant's arguments, so we don't know the minimum showing required to shift the burden back to the Examiner for better evidence of a "known or used" date. Therefore, it's probably best to do a little digging on your own and see if you can find evidence that an update was applied after the copyright date. For example, the About box in my Mozilla Firefox® browser includes a version number, and the Mozilla® website has a list of all versions and their release dates.