The patent case against those chains—as well as Microsoft, Cabela's, Dell, Newegg, Hewlett-Packard and Audible—was brought by Kelora, which said it had patented a way to let shoppers isolate search results by color, price, date, brand, size, etc.
The arguments discussed in the opinion from U.S. District Court Judge Claudia Wilken, sitting in the Northern District of California, primarily focused on the elements of the patent. Kelora argued that, although the elements of such response-filtering were well-known at the time of the filing, the idea of piecing them together in that particular way was new.
Part of the discussion involved moving to a stateful server—one that remembers earlier queries and can use that information to make future answers much more accurate—and whether that was an obvious move at the time.
"Such a design had critical advantages over a stateful server, such as to reduce the amount of disk storage and memory space required and to simplify crash recovery," the judge wrote. "The inventors themselves acknowledged disadvantages of stateful servers on the World Wide Web. If the Web server has to identify and track individual sessions with each such user in order to know what criteria the user previously searched in earlier search iterations, there is clearly a tremendous overhead load on the Web server to service such individual search sessions."
Kelora argued that there existed at the time "several different approaches to keeping previous search results, only one of which is resubmission of the original search terms along with the new search terms and that this would not have been the single solution necessarily chosen by a person of ordinary skill in the art."
That's the legal standard. Would this have been obvious to a person of ordinary skill in that area at the time?
One interesting discussion involved a Kelora argument that it had tested the obviousness with a group of computer science and engineering students.
"In its supplemental opposition, Kelora refers to anecdotal evidence from one of its experts about her first-hand experience reviewing solutions generated by computer science and engineering students in response to an assignment to adapt a computer application with search capability to a networked application. Kelora argues that, because none of these students adapted the application to the Web or to use resubmission, such steps were not obvious to a person of ordinary skill in the art. Kelora's other expert also refers to the fact that the inventors themselves were unfamiliar with Web servers until the second quarter of 1994 and could not have adapted the invention to the Web before that time."
The judge drew a different conclusion. "This evidence is not determinative. The issue of obviousness is determined entirely with reference to a hypothetical 'person having ordinary skill in the art.' Further, even if the students were of ordinary skill in the art, the fact that the students did not choose to go beyond the basic requirements of a school assignment to adapt their applications to the Web or use resubmission does not establish or suggest that they lacked the skills to do so with the appropriate motivation. Similarly, there is no showing that these students had access to all prior art references at the time, or that they viewed the collective teachings of the prior art 'as a whole,' as is presumed of the hypothetical person of ordinary skill."