`
`By: John D. Vandenberg (Reg. No. 31,312)
`
`john.vandenberg@klarquist.com
`Stephen J. Joncus (Reg. No. 44,809)
`stephen.joncus@klarquist.com
`Klarquist Sparkman, LLP
`One World Trade Center, Suite 1600
`121 S.W. Salmon Street
`Portland, Oregon 97204
`Telephone: (503) 595-5300
`Facsimile: (503) 595-5301
`
`
`
`UNITED STATES PATENT AND TRADEMARK OFFICE
`____________
`
`BEFORE THE PATENT TRIAL AND APPEAL BOARD
`____________
`
`MICROSOFT CORPORATION
`Petitioner
`
`v.
`
`PROXYCONN, INC.
`Patent Owner
`
`____________
`
`Case IPR2012-00026 (TLG)
`Case IPR2013-00109 (TLG)
`Patent 6,757,717 B1
`
`____________
`
`3rd Declaration of Professor Darrell D. E. Long
`Regarding U.S. Patent No. 6,757,717
`
`
`
`
`
`
`
`
`
`
`
`QUALIFICATIONS
`
`I incorporate my qualifications as described in my two prior declarations
`
`I.
`II. COMPENSATION
`
`
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`(Ex. 1007 (1st IPR) and Ex. 1013 (2nd IPR) submitted in this matter.
`
`
`
`Counsel for Microsoft is compensating me at my standard compensation
`
`rate of $500/hour for consulting and $600/hour for testimony in deposition or trial,
`
`plus reimbursement for reasonably incurred expenses. I have no interest in the
`
`outcome of the related litigation or of this proceeding.
`
`III. SUMMARY OF MY STUDY AND CONCLUSIONS
`
`10
`
`
`
`I have reviewed the entire transcript of testimony of Dr. Alon Konchitsky in
`
`11
`
`12
`
`13
`
`14
`
`this matter, dated July 2, 2013. I have considered in particular pages 22–24, 28–
`
`30, 96–97, 109–11, and 141–55 of this transcript. I also have reviewed two
`
`versions of Dr. Konchitsky’s CV, one identified as Ex. 2003 (4 pages) and the
`
`other identified as Ex. 1022 (6 pages).
`
`15
`
`
`
`Based on these materials, I am of the opinion that Dr. Konchitsky is not an
`
`16
`
`17
`
`18
`
`19
`
`20
`
`expert in any field of endeavor material to the ’717 patent or the prior art
`
`references I have discussed in this matter. I am of the opinion that Dr. Konchitsky
`
`does not even qualify as having ordinary skill or knowledge in this field. More
`
`specifically, Dr. Konchitsky’s testimony demonstrates that he does not know what
`
`is common knowledge to experts in this field, about HTTP, caching, file systems,
`
`2
`
`
`
`
`
`
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`etc. He has not written about matters that experts in this field typically would have
`
`written about, has not worked in this field, and does not belong to the
`
`organizations to which experts in this field would typically belong.
`
`IV. FIELD OF THE INVENTION
`
`{For ease of reference, the below description is copied from my second
`
`declaration submitted in this matter.}
`
`The ’717 patent defines its “field of the invention” as accessing data in
`
`communication networks. (’717, 1:10–15). The field also includes the areas of
`
`distributed data storage systems and networking, coding theory including error
`
`detection and correction codes, and cryptographic hash functions commonly called
`
`message digest functions. These were all mature fields for many years prior to
`
`1998–99.
`
`V. LEVEL OF SKILL IN THE ART IN 1998–99
`
`{For ease of reference, the below description is copied from my second
`
`declaration submitted in this matter.}
`
`A person of ordinary skill in this art in 1998–99 would hold a B.S. degree in
`
`computer science and would have as part of his study courses in operating systems,
`
`networking, data compression and computer security. These studies would include
`
`the storage subsystem of computer operating systems which is covered briefly in
`
`most undergraduate operating systems courses, but few require the student to
`
`3
`
`
`
`
`
`
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`examine actual source code. In addition he would have several years of practical
`
`experience working in operating systems, in particular the data storage subsystem.
`
`As a result, actual experience in working with this operating system
`
`subsystem would normally occur after several years of experience working for a
`
`company with a focus on systems software.
`
`Alternatively, a person would develop the level of ordinary skill in the art in
`
`1998–99 by obtaining an M.S. in computer science and by writing his or her thesis
`
`in an area related to data storage and/or computer security.
`
`A person of ordinary skill in the art would understand network protocols.
`
`This was normally part of undergraduate programs in computer science in 1998–
`
`99. A person of ordinary skill in the art would also understand coding theory; in
`
`particular error detection and correction codes, as well as cryptographic hash
`
`functions and message digest functions. Introduction to basic hash functions is a
`
`normal part of most undergraduate curricula, but coding theory is normally part of
`
`specialized courses (although it is commonly part of electrical engineering
`
`programs), and cryptographic hash functions would normally be taught only in
`
`courses in computer security.
`
`I have first-hand experience teaching and working with such persons of
`
`ordinary skill in the art. For example, I have taught students having about that
`
`level of skill in this art since at least as early as 1990.
`
`4
`
`
`
`
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`
`
`VI. DR. KONCHITSKY IS NOT AN EXPERT IN THIS FIELD
`
`He is Not Familiar with Technologies that Any Expert Would Know:
`
`HTTP GET: The HTTP protocols (e.g., HTTP/1.0 and HTTP/1.1) have
`
`been the application-level data communication protocol for the World Wide Web
`
`for more than 20 years. Any expert in this art would understand HTTP well. For
`
`example, many of the references under discussion refer to HTTP: the HTTP/1.1
`
`protocol is cited in DRP (Ex. 1003 (2nd IPR), p. 10), Mattis (Ex. 1004 (2nd IPR),
`
`14:4–5) and is cited prior art to the ’717 patent.
`
`Dr. Konchitsky testified that he was not comfortable testifying about the
`
`basic GET request in the HTTP protocol, and did not know whether a GET request
`
`identified the data it requests by its Uniform Resource Locator (URL).
`
`(Konchitsky TR 22:5–24:6). This lack of knowledge demonstrates a lack of even
`
`ordinary skill in the art. Persons of ordinary skill in the art in 1998–99, and today,
`
`would have no trouble describing an HTTP GET request. The HTTP/1.0 protocol
`
`defines only three methods: GET, HEAD and POST. A HTTP/1.0 GET request
`
`has only three elements: “GET” followed by the URL (e.g., /TheProject.html)
`
`followed by the protocol (e.g., “HTTP/1.0.”) The same is true of the HTTP/1.1
`
`GET method. (See, e.g., DRP, p. 7 (“GET /Example/home.html HTTP/1.1”). Any
`
`expert in this art would know without hesitation that an HTTP GET request
`
`identifies the desired resource by its URL.
`
`5
`
`
`
`
`
`
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`Caching Servers: Caching servers are a fundamental part of distributed data
`
`storage systems, networking and the World Wide Web. Yet, Dr. Konchitsky’s
`
`testimony shows he is utterly unfamiliar with the state of the art in caching servers
`
`in 1998.
`
`He testified that caching servers stored their caches in RAM, that he could
`
`not recall caching servers storing their caches on disk, and that he had no opinion
`
`as to whether any caching servers in 1998 stored their cache in nonvolatile
`
`memory. (Konchitsky TR 109:24–111:12). This testimony is akin to someone
`
`professing to be an expert on baseball history and having no opinion whether the
`
`Yankees ever won a World Series.
`
`Experts and even persons of ordinary skill in the art know that in the late
`
`1990s Web (and Internet) caching servers nearly universally used non-volatile
`
`(typically disk) storage for cache when used to cache Web pages or other files
`
`transmitted over the Internet. For example, Inktomi’s 1999 “Caching Guide” (Ex.
`
`1010 (2nd IPR)) explains the use of disks with web proxy caches. (E.g., id. at pp.
`
`18 (of 45) (“The size of the cache being managed is another consideration for
`
`scalability. The physical disks and their seek-times do not represent a measurable
`
`bottleneck, but the cache server software must be able to manage large amounts of
`
`data effectively.”), 30 (“Determining how large a cache to maintain is a trade-off
`
`between the cache hit rate and the cost of configuring disk storage. Cache capacity
`
`6
`
`
`
`
`
`
`
`can range from single cache servers to clusters of servers to distributed cache
`
`hierarchies.”), 36 (“Inktomi’s Object Store is a custom-designed and optimized
`
`Web-object database. It uses raw disk I/O to provide optimal storage and retrieval
`
`of content objects.”). So do Inktomi’s Mattis patents (Exs. 1015 (1st IPR) and
`
`1004 (2nd IPR).) Someone not knowing this lacks even ordinary skill in the art.
`
`Well-Known Technologies: In his examination, Dr. Konchitsky was asked
`
`a series of questions to test his knowledge of well-known technologies and systems
`
`and research in this field. He failed the test. In my opinion, no expert in this field
`
`could possibly be unfamiliar with more than perhaps one or two of the following:
`
`Harvest, Squid, LBFS (Low-Bandwidth File System), NFS, AFS (Andrew File
`
`System), Sprite, Rsync, xdelta, and Venti. Yet, Dr. Konchitsky was not familiar
`
`with any of them, not even the ones he said that he knew about.
`
`Harvest and Squid: Dr. Konchitsky testified that he was not familiar with
`
`Harvest or Squid. (Konchitsky TR 150:15–20, 153:11–12). Anyone who has
`
`studied distributed storage systems understands web proxy caches and knows of
`
`the seminal Harvest web cache and its derivative open-source Squid web cache.
`
`Each is directed to the same goal at issue here, reducing network bandwidth
`
`consumption. I raised the Harvest cache in my testimony. (Long TR, 51–53, 69).
`
`The following is an excerpt from a 2004 Ph. D. Thesis by my colleague Ethan
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`7
`
`
`
`
`
`
`
`Miller’s Ph. D. student (I sat on his dissertation committee), Ismail Ari, about
`
`Harvest and Squid:
`
`Web proxy caches [66] have successfully been used over the last
`decade to improve response times for web clients. These caches
`have also reduced network bandwidth consumption and server
`CPU utilization, thus contributing to scalable growth of the
`Internet. Web caches allow sharing of web objects among a
`community of web clients, thus they exploit web access patterns
`[36] to save Internet resources. The following is a description of
`three different architectures used for web content caching and
`dissemination.
`
`2.1.3.1 Hierarchical Web Caches
`
`Harvest Squid [66, 21] web caches have a hierarchical structure.
`They are widely–adopted around the world [16] today. They define
`parent–child relationships between levels of a cache hierarchy,
`thus forming a tree topology from the bottom (i.e. edges of the
`Internet) to the top levels. These caches establish and maintain
`connections between themselves and clients during web document
`transfers. The upper–level caches in a cache hierarchy serve misses
`of the lower–level caches.
`
`Santos (Ex. 1004 (1st IPR) cites to Squid. (Santos, p. 13). So does
`
`Inktomi’s Network Caching Guide (Ex. 1010 (2nd IPR), p. 33) (“Inktomi Traffic
`
`Server competes with the public domain Squid cache server.”) (Inktomi is the
`
`assignee on the Mattis patents and its Traffic Server web proxy caching server
`
`used an alias-free object cache (Ex. 1010 (2nd IPR), p. 37), which is what the
`
`Mattis patents describe.)
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`8
`
`
`
`
`
`
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`NFS (Network File System) and LBFS (Low Bandwidth File System): Dr.
`
`Konchitsky testified that he was familiar with NFS and LBFS (Konchitsky TR
`
`150:21–151:12, 151:23–152:5), but his testimony does not demonstrate that he was
`
`familiar with either. Any expert in data storage systems existing in the late 1990s
`
`understands the leading distributed file systems and knows that “NFS” is not a
`
`generic term for just any network file system but rather that Sun Microsystems
`
`developed the NFS protocol and file system. Similarly, LBFS is recognized by
`
`experts in the art not as a generic term for a file system designed for low-
`
`bandwidth networks, as Dr. Konchitsky seemed to have guessed, but rather as the
`
`specific LBFS network file system described in the paper “A Low-bandwidth
`
`Network File System” and developed at MIT. LBFS breaks data into chunks and
`
`sends hashes of those chunks between clients and servers to avoid unnecessarily
`
`consuming network bandwidth by transmitting data already present in the client’s
`
`cache. An expert in this field, asked to describe LBFS, would have referenced that
`
`feature or at least noted that LBFS was developed at MIT. Dr. Konchitsky did
`
`neither and could not say what “LBFS” meant. (Konchitsky TR 150:21–151:12).
`
`AFS (Andrew File System): Dr. Konchitsky testified that he did not know
`
`what the Andrew File System is. (Konchitsky TR 151:13–18). In this field, AFS
`
`is well known and is considered seminal. It is required reading in graduate courses
`
`in distributed systems and data storage, and has been for more than 20 years. It is
`
`9
`
`
`
`
`
`
`
`a distributed network file system developed at Carnegie Mellon. Each AFS client
`
`caches files on its file system.
`
`Sprite: Dr. Konchitsky testified that he did not know what Sprite is.
`
`(Konchitsky TR 151:19–22). Although less well known than AFS and NFS, an
`
`expert in this field would have heard of Sprite—a distributed operating system
`
`developed at UC Berkeley. It contributed important concepts including caching
`
`techniques, especially for locating files via a prefix cache.
`
`Rsync: Dr. Konchitsky purported to describe Rsync (Konchitsky TR 152:6–
`
`25), but his testimony demonstrates that he does not understand Rsync and was
`
`instead confusing it with rsh, a remote login shell that was distributed as part of the
`
`rlogin package in 4.2BSD (Berkeley UNIX). Rysnc is a protocol using chunk-
`
`based encoding to synchronize files over a network while reducing the
`
`consumption of network bandwidth. It was developed by Andrew Tridgell as part
`
`of his 1996 Ph.D. thesis. Contrary to Dr. Konchitsky’s testimony, it was not
`
`mainly for remote logging in over dial-up links with 30 kb/s bandwidth.
`
`Xdelta: This is a compression utility using delta encoding. It is an open
`
`source utility for computing the differences between files, originally based on the
`
`same algorithm as Rsync. Dr. Konchitsky was unable to identify it. (Konchitsky
`
`TR 153:1–4).
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`10
`
`
`
`
`
`
`
`Venti: Venti is a single instance network storage system developed at Bell
`
`Labs that uses hashes of data blocks as the address of the data. Dr. Konchitsky
`
`said he had heard of Venti but then misdescribed it as a “virtual network.”
`
`(Konchitsky TR 153:7–10).
`
`Ross Williams: Ross Williams is an Australian computer scientist with a
`
`seminal patent in data deduplication and compression. Most experts in this field
`
`would know of Ross Williams and the focus of his work. Dr. Konchitsky testified
`
`that he knew of Ross Williams but his testimony shows that he did not, as he mis-
`
`described Mr. Williams work as “different type of file systems.” (Konchitsky
`
`153:23–154:12).
`
`His CV Does Not Claim Expertise in This Field: Dr. Konchitsky’s six-page
`
`CV lists the following areas as his areas of expertise. Reading this list in the
`
`context of his entire CV, none of these areas is the field in which the ’717 patent
`
`resides. He lists “distributed computing” and “client server” but without
`
`specifying which aspects of these broad categories in which he professes expertise.
`
`His listed patents, work experience and writings do not pertain to these areas.
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`11
`
`
`
`
`
`
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`His Work Experience Demonstrates No Expertise or Even Ordinary Skill in
`
`
`
`This Field: As noted, a person of ordinary skill in the art in 1998–99 would have
`
`had several years of practical experience working in operating systems, in
`
`particular the data storage subsystem (or the equivalent post-graduate academic
`
`work). There is nothing in either of Dr. Konchitsky’s two CVs to support the
`
`claim that he has any expertise in these areas.
`
`His Patents Show No Expertise in This Field: The titles of his patents listed
`
`in his CV show no expertise in this field. I searched the U.S. Patent and
`
`Trademark Office database of patents for “Konchitsky” in the “Inventor” field and
`
`any of the following words in the “Description/Specification” field, finding no
`
`hits: server, client, proxy, hash, MD5, fingerprint, digest, signature, cryptographic,
`
`encryption, cache/cached/caching, proxy, “file system,” or HTTP.
`
`He Has Not Published in This Field: A typical expert in this field would
`
`have authored peer-reviewed publications in the field. Peer review by other
`
`12
`
`
`
`
`
`
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`experts is the standard for correctness and quality. Yet, Dr. Konchitsky testified
`
`that he has authored no peer-reviewed publications in any field. (Konchitsky TR
`
`149:15–25).
`
`Most experts in this field would have authored some publications in this
`
`field. Dr. Konchitsky’s Ph.D. thesis (“Multimode Multiband RF Transmitter”)
`
`(Konchitsky TR 150:1–5) is not pertinent to this field. Based on their titles, none
`
`of Dr. Konchitsky’s “publications” or patents listed in the two CVs, pertain to this
`
`field. None concern distributed data storage systems and networking, coding
`
`theory including error detection and correction codes, or cryptographic hash
`
`functions commonly called message digest functions. He testified that he had
`
`published no papers discussing hashing techniques or techniques for generating
`
`message digests or web caching or forward proxy caching or reverse proxy
`
`caching. In short, his writings demonstrate no expertise or experience in the
`
`relevant fields. (Konchitsky TR 154:13–155:2).
`
`He Does Not Belong to the Organizations in This Field: The three leading
`
`professional organizations in this field are USENIX, the Association for
`
`Computing Machinery, and the Computer Society of the IEEE. In my experience,
`
`the majority of experts in this field would belong to one or more of these
`
`organizations. Dr. Konchitsky testified that he belongs to none of them.
`
`(Konchitsky TR 150:6–14).
`
`13
`
`
`
`
`
`
`
`I declare under penalty of perjury under the laws of the United States of
`
`America that the foregoing is true and correct.
`
`Executed on the ___ day of August, 2013, in San Diego, California.
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`Darrell D. E. Long
`
`
`
`
`
`
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`14