`_____________________
`
`BEFORE THE PATENT TRIAL AND APPEAL BOARD
`_____________________
`
`
`
`GOOGLE INC.
`Petitioner
`
`v.
`
`AT HOME BONDHOLDERS' LIQUIDATING TRUST,
`Patent Owner
`
`_____________________
`
`Case CBM: Unassigned
`U.S. Patent 6,286,045
`_____________________
`
`DECLARATION OF HENRY HOUH
`
`
`
`
`
`
`
`
`Google Inc.
`GOOG 1003
`CBM of U.S. Patent No. 6,286,045
`
`i
`
`
`
`
`
`
`
`I.
`
`CBM Review of USPN 6,286,045
`Declaration of Henry Houh (Exhibit 1003)
`
`I, Henry Houh, hereby declare as follows.
`
`Introduction
`1.
`
`I have been retained as an expert witness on behalf of Google Inc. for
`
`the above-captioned covered business method (CBM) review.
`
`2.
`
`I am over the age of eighteen (18) and otherwise competent to make
`
`this declaration. I am being compensated for my time in connection with this CBM
`
`at my standard consulting rate. I understand that the petition for covered business
`
`method (CBM) patent review involves U.S. Patent No. 6,286,045 ("the '045
`
`patent"), GOOG 1001, which resulted from U.S. Application No. 08/858,650 ("the
`
`'650 application"), filed on May 19, 1997, naming Michael John Griffiths and
`
`James David McElhiney as the inventors. The '045 patent issued on September 4,
`
`2001, from the '650 application. I further understand that, according to USPTO
`
`records, the '045 patent is currently assigned to the At Home Bondholders'
`
`Liquidating Trust ("Patent Owner").
`
`3.
`
`In preparing this Declaration, I have reviewed the '045 patent and
`
`considered each of the documents cited herein, in light of general knowledge in the
`
`art. In formulating my opinions, I have relied upon my experience in the relevant
`
`art and have also considered the viewpoint of a person of ordinary skill in the art.
`
`4.
`
`I am familiar with the technology at issue as of the May 19, 1997
`
`filing date of the '045 patent. I am also familiar with a person of ordinary skill in
`
`
`
`- 1 -
`
`
`
`
`
`
`the art with respect to the technology at issue as of the May 19, 1997 filing date of
`
`CBM Review of USPN 6,286,045
`Declaration of Henry Houh (Exhibit 1003)
`
`the '045 patent.
`
`II. My Background and Qualifications
`5.
`I have multiple undergraduate and graduate degrees from the
`
`Massachusetts Institute of Technology (MIT). I have a Bachelor of Science Degree
`
`in Electrical Engineering and Computer Science from MIT which I received in
`
`June 1989. I also have a Bachelor of Science Degree in Physics from MIT which I
`
`received in February 1990. I have a Master of Science Degree in Electrical
`
`Engineering and Computer Science from MIT which I received in February 1991. I
`
`have a PhD in Electrical Engineering and Computer Science from MIT which I
`
`received in February 1998. My thesis was directed to “Designing Networks for
`
`Tomorrow's Traffic," and was supervised by Professor David Tennenhouse and
`
`Professor John Guttag.
`
`6.
`
`I have extensive experience in creating web pages and commercially
`
`setting up websites during the 1993-1997 time frame, and have continued this
`
`development since then. As part of this experience, I regularly programmed using
`
`HTML and Perl programming languages. I also have experience with CGI scripts
`
`and Perl scripts as used for dynamic web pages.
`
`7.
`
`I have experience with the operations of web servers starting in 1993,
`
`including setting up one of the first hundred or so web servers, and have
`
`
`
`- 2 -
`
`
`
`
`
`
`configured and set up scores of web servers since. In 1994, I co-authored papers on
`
`CBM Review of USPN 6,286,045
`Declaration of Henry Houh (Exhibit 1003)
`
`our web server video system and on database-backed web sites which I presented
`
`at the first World Wide Web conference held at CERN in Geneva, Switzerland.
`
`8.
`
`From 2003-2004, I also was the architect for a next-generation web
`
`testing platform developed by Empirix, known as e-Test Suite. e-Test Suite is now
`
`owned by Oracle Corporation. e-Test provided functional and load testing for web
`
`sites. e-Test emulated a user's interaction with a web site and provided web
`
`developers with a method of creating various scripts and providing both functional
`
`testing (i.e., did the web site provide the correct response) and load testing (i.e.,
`
`could the web site handle 5000 users simultaneously). As part of this project, I co-
`
`wrote a javascript interpreter for e-Test so that it could understand the javascript
`
`code present on advanced web sites.
`
`9.
`
`I have web advertising experience as well, including starting a
`
`company in 1994 that was funded, in part, through web advertisements. Later in
`
`2008-2009, as the CTO of a venture-funded social network start-up company, I
`
`also evaluated different ad network tools.
`
`10.
`
`I am familiar with patents and have numerous provisional and non-
`
`provisional patent applications on which I am an inventor, including 4 provisional
`
`patent applications, and 20 non-provisional applications. These patent applications
`
`deal with various areas of technology including, but not limited to: utilizing
`
`
`
`- 3 -
`
`
`
`
`
`
`network processors, network and packet delivery systems, tagging media content,
`
`CBM Review of USPN 6,286,045
`Declaration of Henry Houh (Exhibit 1003)
`
`dynamic presentation of advertising and other content using enhanced metadata,
`
`methods for web site testing, and organizing and managing files that are
`
`transmitted over a network.
`
`11.
`
`In formulating my opinions, I have relied upon my training,
`
`knowledge, and experience in the relevant art. A copy of my current curriculum
`
`vitae is provided as GOOG 1004, and it provides a comprehensive description of
`
`my academic, employment, research, and professional history.
`
`12. With my extensive experience in the field of web based systems prior
`
`to and up until May 1997, I am qualified to provide an opinion as to what a person
`
`of ordinary skill in the art would have understood, known or concluded as of 1997.
`
`I have been asked to opine on the state of the art as of the May 19, 1997 filing date
`
`of the '045 patent.
`
`III. List of Documents Considered in Formulating My Opinion
`
`13.
`
`In formulating my opinion, I have considered the following
`
`documents referenced herein:
`
`Exhibit #
`
`1001
`
`Description
`
`Griffiths et al., U.S. Patent No. 6,286,045 (filed May 19, 1997;
`issued September 4, 2001).
`
`1002
`
`Prosecution history of U.S. Patent No. 6,286,045.
`
`
`
`- 4 -
`
`
`
`
`
`
`
`
`
`1005
`
`1006
`
`1007
`
`1008
`
`1009
`
`1010
`
`1011
`
`1012
`
`1013
`
`1014
`
`1015
`
`1016
`
`CBM Review of USPN 6,286,045
`Declaration of Henry Houh (Exhibit 1003)
`
`Novak et al., "New Metrics for New Media: Toward the
`Development of Web Measurement Standards"(September 26,
`1996, published in the World Wide Web Journal, April 1997,
`213-246)
`
`Pitkow et al., "In Search of Reliable Usage Data on the
`WWW"(Proceedings of the Sixth International WWW
`Conference, April 1997)
`
`J. Mogul and P. Leach, "Simple Hit-Metering for HTTP" (dated
`January 21, 1997; published January 22, 1997)
`
`C. Brown and S. Benford, "Tracking WWW Users: Experience
`from the Design of HyperVisVR," Webnet 96, Oct. 15-19, 1996
`
`H. Skardal, "A Trip Report and some reflections," W3C Meeting
`on Web Efficiency and Robustness (Apr. 22, 1996)
`
`"WRQ Express PageMeter delivers breakthrough accuracy in
`Web metering," Business Wire, Apr. 28, 1997
`
`U.S. Patent No. 5,751,956 to Kirsch (filed Feb. 21, 1996),
`"Method and Apparatus for Redirection of Server External
`Hyper-link References."
`
`Lopez-Ortiz, et al., "A Multicollaborative Push-Caching HTTP
`Protocol for the WWW," December 28, 1995
`
`Fielding et al., "HTTP Working Group Internet Draft Hypertext
`Transfer Protocol – HTTP/1.0"(dated February 19, 1996;
`published February 20, 1996)
`
`Fielding et al., "HTTP Working Group Internet Draft Hypertext
`Transfer Protocol – HTTP/1.1"(dated May 2, 1996; published
`May 3, 1996)
`
`"Top 20 Network Shows, Rated 4 Ways, Announced by Nielsen,"
`Broadcasting-Telecasting (Broadcasting Publications, Inc.), p. 93.
`December 8, 1947. Retrieved December 1, 2015.
`
`Meeker, Mary, "Technology: Internet/New Media The Internet
`Advertising Report," Morgan Stanley, U.S. Investment Research
`
`- 5 -
`
`
`
`
`
`
`
`1017
`
`1018
`
`CBM Review of USPN 6,286,045
`Declaration of Henry Houh (Exhibit 1003)
`
`(January 1997)
`Meeker, Mary, "Technology: New Media The Internet Report,"
`Morgan Stanley, U.S. Investment Research (February 1996)
`Stephen L. Manley, “An Analysis of Issues Facing World Wide
`Web Servers,” Bachelor’s Thesis, Harvard College (dated April 7,
`1997)
`
`
`IV. Relevant Legal Standards
`14. Counsel has provided me with various legal standards that I
`
`understand apply to my analysis.
`
`A. Definiteness
`15.
`I understand the Supreme Court has stated that a claim is indefinite
`
`when it fails to inform, with reasonable certainty, those skilled in the art about the
`
`scope of the invention. I also understand that the Patent Office may employ a lower
`
`“plausible” indefiniteness standard. In particular, I understand that if a claim is
`
`amenable to two or more plausible claim constructions, the USPTO requires the
`
`applicant to more precisely define the metes and bounds of the claimed invention
`
`by holding the claim unpatentable for being indefinite. These two different
`
`standards would not change my opinion set out below.
`
`B.
`16.
`
`Patent Eligible Subject Matter
`
`I understand that the patent statute defines four categories of
`
`inventions that are eligible for protection: processes, machines, manufactures, and
`
`compositions of matter. I also understand that courts have found ineligible a
`
`
`
`- 6 -
`
`
`
`
`
`
`process or a system directed to an abstract idea with no practical application or
`
`CBM Review of USPN 6,286,045
`Declaration of Henry Houh (Exhibit 1003)
`
`which preempts substantially all practical applications. I also understand that
`
`courts require that meaningful limitations beyond conventional or routine steps or
`
`components such as general purpose computer hardware be added to the abstract
`
`idea to avoid preempting all practical applications of the idea.
`
`C.
`17.
`
`CBM Eligibility
`
`It is my understanding that, to be eligible for this proceeding, the
`
`claims cannot be considered a "technological invention." One part of the
`
`technological invention test is whether the claims recite technological features that
`
`are novel and unobvious.
`
`18.
`
`It is my understanding that a claimed invention is not novel if a single
`
`prior art reference teaches each and every limitation recited in the claim. The
`
`disclosure may be explicit, implicit, or inherent. I understand that a reference is
`
`read from the perspective of a person of ordinary skill in the art ("POSA") at the
`
`time of the invention.
`
`19.
`
`It is my understanding that if an invention is not the same as prior
`
`products or processes, then it may be considered novel. However, for the invention
`
`to be patentable, it must include a difference from the prior art that would not have
`
`been obvious to a POSA. If the invention only includes routine, conventional, or
`
`otherwise well-understood functionality or improvement, then it is obvious.
`
`
`
`- 7 -
`
`
`
`
`
`
`
`CBM Review of USPN 6,286,045
`Declaration of Henry Houh (Exhibit 1003)
`
`20.
`
`I also understand that rationales that may support a conclusion of
`
`obviousness include: (a) combining prior art elements according to known methods
`
`to yield predictable results; (b) choosing from a finite number of identified,
`
`predictable solutions, with a reasonable expectation of success; (c) simple
`
`substitution of one known element for another to obtain predictable results; (d) use
`
`of known technique to improve similar devices (methods, or products) in the same
`
`way; (e) applying a known technique to a known device (method, or product)
`
`ready for improvement to yield predictable results.
`
`21.
`
`I
`
`also understand
`
`that
`
`any
`
`secondary
`
`considerations of
`
`nonobviousness must be considered. I understand that secondary considerations
`
`must have a nexus to the claim and that even substantial evidence of secondary
`
`considerations may not overcome a strong prima facie showing of obviousness. I
`
`am not aware of any secondary considerations having a nexus to the claim that
`
`have any bearing on the opinions I provide herein.
`
`V.
`
`Person of Ordinary Skill in the Art
`22.
`
`I understand that a person of ordinary skill in the art is one who is
`
`presumed to be aware of all pertinent art, thinks along conventional wisdom in the
`
`art, and is a person of ordinary creativity. A POSA would have had knowledge of
`
`web pages, Internet advertisements including knowledge of banner display
`
`counting, cache usage, and various HTTP-related technologies as of 1997.
`
`
`
`- 8 -
`
`
`
`
`
`
`
`CBM Review of USPN 6,286,045
`Declaration of Henry Houh (Exhibit 1003)
`
`23.
`
`I understand that in a related inter partes review of the '045 patent
`
`(IPR2015-00657), the PTAB found that a POSA would have had (a) a Bachelor of
`
`Science degree in computer science or a similar field or (b) at least 3 years of
`
`programming web-based information management and delivery systems. But given
`
`that the claimed invention uses common and well-known HTTP functions in a
`
`routine manner, it would have been sufficient for a POSA to have at least (a) a B.S.
`
`degree in computer science or a similar field, or (b) at least 3 years of experience in
`
`web-based information management and delivery systems.
`
`24. Based on my education and experience, I would have easily exceeded
`
`the criteria for a POSA proposed in the inter partes review in 1997, and I still
`
`exceed it today.
`
`VI. Overview of the Claimed Technology
`25. The '045 patent describes a "system for storing and delivering
`
`information," such as advertisements, across a computer network, "wherein
`
`accurate counts of the number of times the information is displayed or served to
`
`users or terminals can be made…[and] the operation of the computer network is
`
`not significantly affected." (GOOG 1001, 3:31-45.) The technology of the '045
`
`patent is best described by referring to Figures 3 and 4 of the patent. These figures
`
`are diagrams showing, respectively, "a prior method" and a "preferred method of
`
`
`
`- 9 -
`
`
`
`
`
`
`the present invention" that illustrate what the patent was intended to capture. (Id.,
`
`CBM Review of USPN 6,286,045
`Declaration of Henry Houh (Exhibit 1003)
`
`4:62-67.)
`
`26. The '045 patent describes method 72 of Figure 3 (provided below) as
`
`being prior art. (Id., Figure 3.) By the patent's own admission this method is
`
`"conventional," and is used, at least in part, to illustrate the problem with the prior
`
`art. (Id., 9:25.) Method 72 of Figure 3 begins with a user requesting a web page in
`
`step 74, which transmits a message from a user's terminal to a computer network
`
`via a proxy server. (Id., 9:25-36.) However, according to the '045 patent, the
`
`request in step 74 may not reach a server "[if] the user at the terminal 36 had
`
`previously requested the same page from the computer site 46" and the page is
`
`stored at the user's terminal or proxy server. (Id., 9:37-67.) Otherwise the computer
`
`site serves the page to the terminal, and the page may be cached. (Id., 9:67-10:9.)
`
`
`
`- 10 -
`
`
`
`
`
`
`CBM Revview of USSPN 6,2866,045
`003)
`
`
`
`Declaratiion of Hennry Houh ((Exhibit 1
`
`
`
`- 11 -
`
`
`
`
`
`
`
`
`
`227. Figurre 4 of thee '045 pateent is also
`
`
`
`
`
`
`CBM Revview of USSPN 6,2866,045
`003)
`
`
`
`Declaratiion of Hennry Houh ((Exhibit 1
`
`
`
`
`
`
`
`reproduceed below. TThe differrence
`
`
`
`
`
`
`
`
`
`betweenn the two
`
`
`
`figures (thhat is, the
`
`
`
`only diffeerence betwween the cclaimed "nnew"
`
`
`
`
`
`
`
`
`
`process and the prrior art) has been highhlighted.
`
`
`
`
`
`
`
`
`
`- 12 -
`
`
`
`
`
`
`
`CBM Review of USPN 6,286,045
`Declaration of Henry Houh (Exhibit 1003)
`
`28. The '045 patent alleges that "[t]he method 110 of the present invention
`
`[as shown in Figure 4] solves the initial problem of how to create accurate counts
`
`of banner information displays [sic] on user terminals while avoiding the problems
`
`created by requiring the banner information to be retransmitted across the network
`
`each time the banner information is requested by a user." (Id., 14:33-38.)
`
`29. To do this, as described by the specification, "the initial banner
`
`request signal sent by the terminal 36 during the step 112 is preferably a
`
`mandatory signal to be transmitted across the computer network 30 and that cannot
`
`be blocked or terminated by either the terminal 36 or the proxy server 50, even if
`
`the banner to be served to the terminal 36 is already stored in either the terminal 36
`
`or the proxy server 50." (Id., 14:57-63, emphasis added). Assuming the request
`
`signal is not blocked, it can then be used to count the number of advertisement
`
`displays. (Id., 15:33-35.) But instead of returning the banner to the terminal, the
`
`method 110 in step 114 returns a banner location address, such as a redirect signal,
`
`specifying the location address of the banner that was requested. (Id., 15:35-44;
`
`17:29-37.) The remainder of the steps of Figure 4 are essentially the same as the
`
`steps described with regard to the admitted prior art of Figure 3. (Id., 14:46-48.)
`
`30. The problem with the '045 patent's characterization of the differences
`
`between the "invention" and the prior art (e.g., steps 112 and 114 of Figure 4) is
`
`that those particular steps were already well-understood functions used in a routine
`
`
`
`- 13 -
`
`
`
`
`
`
`and conventional way. As discussed in greater detail below, these were known
`
`CBM Review of USPN 6,286,045
`Declaration of Henry Houh (Exhibit 1003)
`
`functions of HTTP.
`
`VII. State of the Art
`31. At the time the '045 patent was filed, Internet advertising was going
`
`through a massive growth phase. Internet advertising exploded from a $13MM
`
`industry in the 4th quarter of 1995, to a $66MM industry by the 3rd quarter of 1996,
`
`and was expected to continue ramping rapidly. (GOOG 1016, pp. iii, viii.) But the
`
`industry also recognized that accurate counting of advertisement exposure was key
`
`to the industry's success. (Id., p. xi ("Advertisers will not be totally comfortable
`
`advertising on the Web until confidence builds that Web advertising measurement
`
`is accurate and auditable by a reliable third party in a 'Nielson-like' way."))
`
`32. Advertisers typically paid a fee based on the number of times their
`
`advertisements were displayed on websites or accessed by users. (GOOG 1005, p.
`
`5.) As discussed in a September 1996 research paper, the "measurement process
`
`involves counting and summarizing the visitor transactions on a Web site," where
`
`"[m]easurement and analysis products tell managers who is accessing their site,
`
`when, and what is being accessed." (GOOG 1005, p. 7.) So ensuring an accurate
`
`count was necessary—overcounting would mean that the advertiser was charged
`
`too much, and undercounting would mean that the advertiser was not paying
`
`enough given the number of banner ad displays.
`
`
`
`- 14 -
`
`
`
`
`
`
`
`CBM Review of USPN 6,286,045
`Declaration of Henry Houh (Exhibit 1003)
`
`33. At the same time (and one of the reasons advertisers were so
`
`interested in using the Web as an advertising medium), the World Wide Web itself
`
`was experiencing rapid growth. In early 1996, the Internet was estimated to then
`
`have about 9 million interactive users and grow to over 150 million users by the
`
`year 2000. (GOOG 1017, p. 3-1.) The nascent infrastructure at the time was
`
`overwhelmed by the amount of traffic, causing increased bandwidth and speed
`
`delay. Forces were at work to try to reduce the strain on the infrastructure while
`
`maintaining a high quality of use.
`
`34. One well-known and conventional way to reduce traffic across the
`
`network was to cache copies of resources (documents, banners, other items) on
`
`intermediate servers, such as proxy servers. HTTP (HyperText Transfer Protocol)
`
`was then (and still is) the fundamental protocol on which the World Wide Web
`
`was built. (GOOG 1013, p. 1; GOOG 1001, 1:43-47.) The original HTTP
`
`specification recognized the benefits of using cache: "A cache stores cachable
`
`responses in order to reduce the response time and network bandwidth
`
`consumption on future, equivalent requests. Any client or server may include a
`
`cache." (GOOG 1013, p. 5.) The HTTP specification also described a "proxy" as
`
`an intermediary between clients and servers. (Id.) Systems using such proxy
`
`servers offered several advantages over systems without proxy servers. Among the
`
`advantages were that a proxy server would store cached information, preventing
`
`
`
`- 15 -
`
`
`
`
`
`
`the need for one or more client devices from having to locally store the cached
`
`CBM Review of USPN 6,286,045
`Declaration of Henry Houh (Exhibit 1003)
`
`information. (GOOG 1014, pp. 11-13, 43; ‘045, 9:59-67, 10:63-11:12.) The proxy
`
`server would also prevent additional requests to a server or original source of
`
`cached information, which could save time, bandwidth, or other resources, and aid
`
`in scalability. For example, the time to retrieve the cached information from the
`
`proxy server would be shorter than the time to retrieve the same information from
`
`a server over the Internet. (GOOG 1014, pp. 12-13; GOOG 1001, 9:25-59, Figure
`
`3.)
`
`35.
`
`It was routine, default functionality of HTTP to interoperate with a
`
`local cache in order to check for a resource prior to retrieving or requesting the
`
`resource from another server. For example, the HTTP specification stated that
`
`"Any party . . . may employ an internal cache for handling requests. The effect of a
`
`cache is that the request/response chain is shortened if one of the participants has a
`
`cached response applicable to that request." (GOOG 1013, pp. 6-7).
`
`36. However, the use of caching to decrease traffic across the network
`
`hampered accurate counting of advertisements—an effect that was well-known as
`
`early as 1995/1996. Due to caching by online service providers (OSPs), OSP
`
`caching distorted Internet usage counts "because the cached files are duplicates that
`
`reside in locations other than the original Web site . . . . This creates problems for
`
`advertisers on the Web, which can't be sure how many people they're actually
`
`
`
`- 16 -
`
`
`
`
`
`
`reaching." (GOOG 1017, p. 3-6.) One reason that ad displays were undercounted
`
`CBM Review of USPN 6,286,045
`Declaration of Henry Houh (Exhibit 1003)
`
`was that "browser software and many Internet gateways intercept some requests
`
`before reaching the server, and these cached requests are never logged." (GOOG
`
`1005, p. 13.)
`
`37. To counteract the effects caused by caching on advertisement counts,
`
`advertisers engaged in the practice of what was commonly referred to as "cache-
`
`busting"—disabling or circumventing cache operation so that resources could not
`
`be stored in cache. If a resource was not found in cache, then a signal requesting
`
`that resource would necessarily travel all the way to a primary server without being
`
`blocked by local or intermediate cache. (GOOG 1007, pp. 2-3). "HTTP/1.1 already
`
`allows origin servers to prevent caching of responses, and we have evidence that at
`
`least some of the time, this is being done for the sole purpose of collecting counts
`
`of the number of accesses of specific pages." (Id.)
`
`38. Many cache-busting methods existed, and were openly discussed
`
`among industry professionals:
`
`Many content providers have resorted to "beating the
`cache" when attempting to obtain full access statistics
`and
`tracking
`information. HTTP/1.0 specifies a
`'Pragma: no-cache' header, which is an instruction to the
`cache server not to cache that object. However, due to the
`misuse of the header by content providers, many proxy
`
`
`
`- 17 -
`
`
`
`
`
`
`
`CBM Review of USPN 6,286,045
`Declaration of Henry Houh (Exhibit 1003)
`
`cache administrators have resorted to ignoring the header
`and caching the resource anyway. Cache busting
`methods include appending a random segment to the
`URL which confuses the cache into thinking that all
`the resources are different, and generating all pages
`dynamically through a cgi-script, the results of which
`do not get cached. (GOOG 1008, pp. 7-8, emphasis
`added.)
`
`39. Reporting from a W3C meeting in which this topic was discussed by
`
`members of the field, the Skardal report states: "There are phenomenons in the
`
`Web working against caching. One is the 'hit count problem' where content
`
`providers modify the URL's to be CGI/non-cacheable URL's in order to improve
`
`'hit count', which is needed for advertisement counting." (GOOG 1009, p. 2.)
`
`40. Several of these cache-busting techniques are specifically discussed in
`
`the '045 patent as being well-known for such functionality "using standard HTML
`
`and HTTP protocols." (GOOG 1001, 18:18-20.)
`
`41. For example, the
`
`'045 patent describes incorporating variable
`
`components into links on a webpage or hypertext document as an option for
`
`implementing the unblockable aspect of the "invention." (GOOG 1001, 18:18-23.)
`
`However, this was known in the art, as is admitted by the '045 patent. The '045
`
`patent acknowledges that the use of variable components in links was part of the
`
`
`
`- 18 -
`
`
`
`
`
`
`"standard HTML and HTTP protocols," and were known not to be cached. (Id.)
`
`CBM Review of USPN 6,286,045
`Declaration of Henry Houh (Exhibit 1003)
`
`The '045 patent lists examples of several different variable components that could
`
`be used. Among the listed components is the use of cgi-bin, which is discussed
`
`herein as already being known in the art. (GOOG 1001, 18:27-29.)
`
`42.
`
`In another example described by the '045 patent, it was well-known
`
`that "caching proxy servers exist which will specifically avoid caching content
`
`related to any URL address containing the strings "cgi-bin" and "?" which are
`
`strings conventionally used in the construction of URL addresses for which
`
`responses are dynmically [sic] generated and, therefore, are unsuitable for
`
`caching." (GOOG 1001, 18:48-54; see also GOOG 1014, pp. 15, 17 (regarding the
`
`use of "?" with HTTP headers and URLs).)
`
`43.
`
`In yet another example described by the '045 patent, "many types of
`
`tags [] can be used for this purpose," including the Expiry tag, the Last-Modified
`
`Tag, Cache-Control Tags, or the pragma:no-cache tag. (Id., 17:65-18:17.) Such
`
`cache control headers provide a way to prevent a cache from blocking a request
`
`signal, and were known in the art. This was a topic discussed among industry
`
`professionals at the Sixth International WWW Conference: "[a] common resource
`
`intensive solution to the problem of reliably determining pages [sic] views and
`
`users is to use the cache specific headers in HTTP to effectively defeat all attempts
`
`at caching pages. Despite the reasons for employing such an approach, this
`
`
`
`- 19 -
`
`
`
`
`
`
`technique is commonly referred to as cache-busting. This is accomplished in
`
`CBM Review of USPN 6,286,045
`Declaration of Henry Houh (Exhibit 1003)
`
`several ways including sending 'Cache-control: proxy-revalidate', or 'Expires:
`
`<past date>' headers." (GOOG 1006, p. 5.)
`
`44.
`
`It is of note that such techniques were not always successful in
`
`ensuring that the request would not be blocked by cache. Particularly with regard
`
`to the use of "cgi-bin" and "?" as described by the patent, the '045 patent
`
`acknowledges that "some caching proxy servers will be led to conclude that the
`
`URL address should not be cached." (GOOG 1001, 18:59-61.) The use of the word
`
`"some" would lead a POSA to believe that while some proxy servers would not
`
`cache the data, there are in fact also other proxy servers that would nonetheless
`
`cache the data associated with the URL address despite the use of the strings "cgi-
`
`bin" or "?".
`
`45. As the practice evolved, the industry recognized that disabling caches
`
`increased server loads, so the natural evolution of Web implementation led to
`
`efforts to overcome the counting-related drawbacks of cache without increasing
`
`network traffic or server load. (GOOG 1006, p. 2.)
`
`46. For example, the PageMeter tool was said to balance accurate page
`
`counts with cache functionality: "Until now, web metering products could not
`
`accurately count page requests because of caching by both proxy servers and
`
`browsers. With Express PageMeter, however, the web server is notified each time
`
`
`
`- 20 -
`
`
`
`
`
`
`the page is requested, even when it is served from cache." (GOOG 1010, p. 1,
`
`CBM Review of USPN 6,286,045
`Declaration of Henry Houh (Exhibit 1003)
`
`emphasis added.)
`
`47. At the time the '045 patent was filed, it was already known that one
`
`way to balance the tension between counting and minimizing traffic on the
`
`network was to respond to a cache-busting signal with a redirect signal such that
`
`the actual resource could then be retrieved from cache if it was available. A
`
`redirect signal was well-known functionality within HTTP. (GOOG 1013, pp. 27-
`
`28.)
`
`48. An example of a redirect signal used in conjunction with an
`
`unblockable signal is U.S. Patent No. 5,751,956 to Kirsch, filed February 21,
`
`1996—well in advance of the '045 patent's filing date. (GOOG 1011.)
`
`49. Kirsch "is generally related to the control of network information
`
`server systems supporting World Wide Web based data pages and, in particular, to
`
`a server system and process for efficiently redirecting external server hyperlink
`
`references for purposes of controlling, moderating, and accounting for such
`
`references." (GOOG 1011, 1:26-30.) In Kirsch, a URL for a particular resource,
`
`such as an advertisement, is embedded in a web page. (Id., 6:47-59.) When the
`
`URL is executed, the URL is transmitted by the client to the "direct server," where
`
`it can be counted. (Id., 5:25-40; 6:47-59; 7:19-34, 7:45-50.) Kirsch indicates that
`
`this URL may be dynamically generated. (Id., 8:20-36.) As stated by the '045
`
`
`
`- 21 -
`
`
`
`
`
`
`patent, dynamic URLs were known to prevent a terminal or proxy server from
`
`CBM Review of USPN 6,286,045
`Declaration of Henry Houh (Exhibit 1003)
`
`blocking the transmission to the originating server. (GOOG 1001, 18:36-42.) In
`
`Kirsch, the "direct server" extracts the corresponding redirect signal from the
`
`dynamic URL, and then returns the redirect to the client. (GOOG 1011, 6:47-59.)
`
`The client then executes the redirect signal per normal HTTP processes. (Id.,
`
`10:65-11:4.)
`
`50. As such, the use of redirect signals, as proposed by the '045 patent, or
`
`even the use of redirect signals with proxy servers, was not a new technology and
`
`was described in the HTTP 1.1 Internet Draft. (GOOG 1014, pp. 37-39.)
`
`51. The HTTP 1.1 Internet Draft itself combines unblockable requests
`
`with redirect signals, though not explicitly for the purpose of counting. For
`
`example, the HTTP 1.1 describes the GET and HEAD methods. (GOOG 1014, p.
`
`43.) With a HEAD request, a server would return header information to determine
`
`whether the cached object had changed since its previous caching. (Id.) The HTTP
`
`1.1 Internet Draft states, “[i]f the new field values indicate that the cached entity
`
`differs from the current resource entity (as would be indicated by a change in
`
`Content-Length, Content-MD5, or Content-Version), then the cache MUST mark
`
`the cache entry stale.” (Id.) This required a signal to be sent to the originating
`
`server, but which would minimize the amount of bandwidth being used to
`
`determine the status of the cached object. This signal to the server would not be
`
`
`
`- 22 -
`
`
`
`
`
`
`blocked by cache, and a POSA would have known that such a signal could be used
`
`CBM Review of USPN 6,286,045
`Declaration of Henry Houh (Exhibit 1003)
`
`for accurate request counting. If the object had changed, a GET request could be
`
`issued for the object, otherwise the object may be retrieved from a cache or proxy
`
`server. (Id.) In some cases, the use of a redirect message would not even be
`
`necessary if the object is already stored in a known cache location, and the use of a
`
`redirect message would actually unnecessarily increase network traffic.
`
`52. As such, the industry had already recognized the problem identified
`
`by the '045 patent, and the claimed "solution" was nothing more than the natural
`
`evolution of the industry as technology developed, which was reached before the
`
`'045 patent was ever filed.
`
`VIII. Claim Construction
`A.
`"Banner"
`53. With regard to the term "banner" the '045 patent states:
`
`invention, the term
`the present
`For purposes of
`"banner"