Page: 1
`Page 3
`
` 1 APPEARANCES (continued):
` 2
` 3 On Behalf of Metasearch Systems, LLC:
` 4 Ryan M. Schultz, Esquire
` 5 ROBINS, KAPLAN, MILLER & CIRESI, LLP
` 6 2800 LaSalle Plaza
` 7 800 LaSalle Avenue
` 8 Minneapolis, Minnesota
` 9 Phone: 612.349.8500
`10 Email: rmschultz@rkmc.com
`11
`
`Kevin Almeroth Ph.D.
`8/26/2014
`Page 1
` 1 UNITED STATES PATENT AND TRADEMARK OFFICE
` 2 ------------
` 3 BEFORE THE PATENT TRIAL AND APPEAL BOARD
` 4 ------------
` 5
` 6 AMERICAN EXPRESS COMPANY, AMERICAN EXPRESS TRAVEL
` 7 RELATED SERVICES COMPANY, INC., EXPEDIA, INC.,
` 8 HOTELS.COM LP, HOTELS.COM GP, LLC, HOTWIRE, INC.,
` 9 ORBITZ WORLDWIDE, INC., PRICELINE.COM, INC.,
`10 TRAVELOCITY.COM LP, and YAHOO! INC.,
`11 Petitioner,
`12 v.
`13 METASEARCH SYSTEMS, LLC
`14 Patent Owner
`15 ------------
`16 Case CBM2014-00001
`17 Patent 8,326,924
`18 ------------
`19
`20 DEPOSITION OF DR. KEVIN ALMEROTH
`21 Taken Tuesday, August 26, 2014
`22 Scheduled for 9:00 a.m.
`23
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`25 REPORTED BY: Dana S. Anderson-Linnell
`
`24
`
`25
`
`Page 2
` 1 DEPOSITION OF DR. KEVIN ALMEROTH taken on Tuesday,
` 2 August 26, 2014, commencing at 8:49 a.m. at Robins,
` 3 Kaplan, Miller and Ciresi, 2800 LaSalle Plaza, 800
` 4 LaSalle Plaza, Minneapolis, Minnesota before
` 5 Dana S. Anderson-Linnell, a Notary Public in and of
` 6 the State of Minnesota.
` 7 **********************
` 8
` 9 APPEARANCES
`10
`11 On Behalf of Petitioners American Express Company,
`12 et al.:
`13 Kristen P. Lantz Reichenbach, Ph.D., Esquire
`14 John D. Vandenberg, Esquire
`15 KARQUIST SPARKMAN, LLP
`16 One World Trade Center
`17 121 S.W. Salmon Street
`18 Portland, Oregon 97204
`19 Phone: 503.595.5300
`20 Email: kristen.reichenbach@klarquist.com
`21 john.vandenberg@klarquist.com
`22
`23 (Appearances continued on the next page.)
`24
`
`Page 4
`
` 1 INDEX
` 2
` 3 WITNESS: Dr. Kevin Almeroth PAGE
` 4 EXAMINATION BY:
` 5 Ms. Reichenbach 6
` 6
` 7 INSTRUCTIONS NOT TO ANSWER: (None.)
` 8
` 9 PRODUCTION REQUESTS: (None.)
`10
`11 INDEX OF EXHIBITS REFERRED TO:
`12
`13 Exhibit 1001 - U.S. Patent Number 8,326,924 6
`14
`15 Exhibit 2001 - U.S. Patent Number 6,789,073 7
`16
`17 Exhibit 2018 - Substitute Claim 13 8
`18
`19 Exhibit 2019 - Substitute Claim 14 8
`20
`21 Exhibit 2023 - U.S. Patent Number 6,185,598 27
`22
`23 Exhibit 2024 - U.S. Patent Number 6,108,703 30
`24
`
`25
`
`800-545-9668
`612-339-0545
`
`25
`
`Paradigm Reporting & Captioning
`www.paradigmreporting.com
`
`#80832
`
`

`

`Kevin Almeroth Ph.D.
`8/26/2014
`Page 5
` 1 INDEX OF EXHIBITS REFERRED TO (continued): PAGE
` 2
` 3 Exhibit 2025 - WWW Caches and Search Engines 34
` 4
` 5 Exhibit 2027 - Declaration of Dr. Kevin
` 6 Almeroth 60
` 7
`
` 8
`
` 9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`25
`
`Page 6
`
`Page: 2
`Page 7
`
` 1 A. Yes.
` 2 Q. How carefully did you read it?
` 3 A. Carefully enough to understand it.
` 4 Q. Did you skim it or how much time
` 5 approximately spending on each page?
` 6 A. Oh, I don't think I could tell you how
` 7 much time on each page. I've looked at it
` 8 multiple times, sometimes searching for
` 9 specific words or phrases.
`10 Q. So would you say you read it -- excuse
`11 me.
`12 A. I mean, to answer your question, I think
`13 I've spent hours reading it.
`14 Q. Okay. Placing before you an exhibit
`15 marked Exhibit 2001.
`16 Do you see the patent number 6,789,073 in
`17 the upper right-hand corner?
`18 A. (Reviews document.) Yes.
`19 Q. Okay. Have you seen this before?
`20 A. Yes.
`21 Q. Do you see that this application for
`22 patent was filed in February 2000? It's on
`23 the left column there?
`24 A. I do.
`25 Q. And I'm going to refer to this as the
`Page 8
`
` 1 DR. KEVIN ALMEROTH,
` 2 called as a witness, being first duly sworn, was
` 3 examined and testified as follows:
` 4
` 5 EXAMINATION
` 6
` 7 BY MS. REICHENBACH:
` 8 Q. Good morning, Dr. Almeroth. My name is
` 9 Kristen Reichenbach, and I'm going to ask you
`10 some questions this morning about your
`11 declaration that was submitted in this
`12 proceeding. First I'm going to start off with
`13 a couple exhibits. I'm placing before you an
`14 exhibit marked Exhibit 1001.
`15 Do you see the Patent Number 8,326,924 in
`16 the upper right-hand corner?
`17 A. (Reviews document.) Yes.
`18 Q. And have you seen this document before?
`19 A. Yes.
`20 Q. I'm going to refer to this as the
`21 '924 Patent, okay?
`22 A. Okay.
`23 Q. Have you read the '924 Patent?
`24 A. Yes, I have.
`25 Q. Have you read all of it?
`
` 1 2000 patent application.
` 2 A. Okay.
` 3 Q. Have you read the 2000 patent
` 4 application?
` 5 A. I have.
` 6 Q. All of it?
` 7 A. Yes.
` 8 Q. And as thoroughly as you state you read
` 9 the '924 Patent?
`10 A. Generally I think I have. I mean,
`11 there's a lot of overlap between the two.
`12 Obviously the '924 has more material. But I
`13 think I've read it as much as the '924.
`14 Q. Okay. I'm handing you two documents.
`15 One is labeled Exhibit Number 2018, which is
`16 the text of claim 13, substitute claim 13 that
`17 has been submitted in this proceeding. And
`18 the second document is Exhibit Number 2019,
`19 which is the text of substitute claim 14
`20 submitted in this proceeding.
`21 Have you seen both of these documents
`22 before?
`23 A. (Reviews documents.) Yes.
`24 Q. And have you read claims 13 and 14?
`25 A. Yes.
`
`800-545-9668
`612-339-0545
`
`Paradigm Reporting & Captioning
`www.paradigmreporting.com
`
`#80832
`
`

`

`Kevin Almeroth Ph.D.
`8/26/2014
`Page 9
` 1 Q. Have you tried to understand them from
` 2 the perspective of someone in 2000 who had
` 3 ordinary skill in the art of metasearching?
` 4 A. Yes.
` 5 Q. Okay. Please look at step A of claim 13.
` 6 Do you see where the claim states that the
` 7 request from the client device is associated
` 8 with at least one travel-related item that may
` 9 be ordered? Do you see that?
`10 A. Yes.
`11 Q. Does this term "travel-related" have any
`12 special meaning that you are aware of?
`13 A. Not that I'm aware of. I think the task
`14 that I was given was largely to focus on the
`15 amendments.
`16 Q. Okay.
`17 A. So --
`18 Q. But you stated --
`19 A. Sorry. I'm not finished.
`20 Q. Okay.
`21 A. So part of what I was trying to do in my
`22 declarations was to use the assumption that
`23 all of the limitations other than what's been
`24 proposed were found in the knowledge broker in
`25 the Mamma.com references, and so I primarily
`Page 10
` 1 focused on the distinctions in the amended
` 2 portions of the claims.
` 3 Q. Okay. But you did state that you read
` 4 the entirety of the claim and considered it
` 5 from the perspective of someone in 2000 who
` 6 had ordinary skill in the art, correct?
` 7 A. That's correct.
` 8 Q. Okay. So the term "travel-related item"
` 9 doesn't to you have a special meaning in the
`10 field of computer science, for example?
`11 A. Again, generally I don't believe so. But
`12 I wasn't really focused on defining the scope
`13 of the claims or attributing any special
`14 meaning or looking at claim construction
`15 issues in particular in the limitations where
`16 I had assumed were already present in prior
`17 art references.
`18 Q. Well, let's just think about the term
`19 from the perspective of your background in
`20 computer science. Do you see the word
`21 "travel-related" as having some special
`22 definition in the field of computer science so
`23 that when that word is used in computer
`24 science, it has a meaning that's different
`25 from an ordinary conversational meaning?
`
`Page: 3
`Page 11
` 1 A. Well, to be clear, if we're -- just so
` 2 that I understand your question, if you're
` 3 removing that term from the claim and then
` 4 asking about it generally, I don't think it
` 5 has any special meaning. But certainly there
` 6 are instances where the context of how that
` 7 term is used might have special meaning.
` 8 Q. Do you think there's a particular context
` 9 in this -- in the claim or any language in the
`10 claim that gives it a special particular
`11 meaning?
`12 A. As I said, I really haven't looked to do
`13 claim construction issues or attribute meaning
`14 or special meaning or really analyze the terms
`15 beyond what I've identified or what I've
`16 considered as part of the amendment.
`17 Q. Well, so just looking at the term today,
`18 I realize that you haven't discussed this
`19 specifically before, but from your perspective
`20 as how the claim is -- or how the term
`21 "travel-related" is used in the claim, do you
`22 see a particular definition of that word in
`23 the claim?
`24 MR. SCHULTZ: Objection, outside
`25 the scope of his declaration.
`
`Page 12
` 1 THE WITNESS: As I sit here now, I
` 2 don't. But it's a question that I would spend
` 3 more time on analyzing before I would give you
` 4 a definitive answer. I haven't really looked
` 5 at the claim to see if it defines
` 6 travel-related items within the claim, so
` 7 it's -- it would be hard for me to give you my
` 8 expert opinion on that topic.
` 9 BY MS. REICHENBACH:
`10 Q. If you want, you can take time to look
`11 through the claim again right now if that
`12 would help.
`13 A. I'm not sure that it would. I think if I
`14 were asked my expert opinion as to whether or
`15 not the claim provided a definition, it's
`16 something I would want to think carefully
`17 about. I don't think I really, you know, can
`18 sit here now -- I certainly can read the
`19 claims. I don't see anything in here that
`20 gives it a specific definition, but that
`21 really is just giving a very superficial look.
`22 Q. Okay. So based on your experience just
`23 in the field of computer science and
`24 considering the phrase "travel-related" and
`25 "travel-related item," you would say that you
`
`800-545-9668
`612-339-0545
`
`Paradigm Reporting & Captioning
`www.paradigmreporting.com
`
`#80832
`
`

`

`Kevin Almeroth Ph.D.
`8/26/2014
`Page 13
` 1 do not know of any technology-based test for
` 2 determining whether or not an item is
` 3 travel-related or not?
` 4 MR. SCHULTZ: Objection.
` 5 THE WITNESS: I'm not sure what you
` 6 mean by "technology-based test." I think
` 7 outside of the scope of claim 13 there
` 8 certainly might be a test. It really depends
` 9 on the system or how it's used or what I'm
`10 looking at or how it's defined or what the
`11 characteristics are. So I don't really think
`12 I have enough information to answer that
`13 question.
`14 BY MS. REICHENBACH:
`15 Q. Okay. Let's look again at the -- you
`16 have claim 13, which is Exhibit 2018, in front
`17 of you, correct?
`18 A. Yes.
`19 Q. Okay. Let's look at step C of the claim.
`20 A. Okay.
`21 Q. Do you see the requirement where it says:
`22 The database comprises at least one previously
`23 stored search result comprising price
`24 information received in response to at least
`25 one previous search query sent by the
`
`Page 14
`
` 1 metasearch engine?
` 2 A. I see those words.
` 3 Q. Do you see the phrase "previous search
` 4 query sent by the metasearch engine" that's in
` 5 the last two lines of step C?
` 6 A. Yes.
` 7 Q. So I want you to consider this phrase
` 8 from the perspective of the skilled artisan in
` 9 2000 in the area of metasearching. Might this
`10 phrase "previous search query sent by the
`11 metasearch engine" mean that the metasearch
`12 engine sent the query previous to step C?
`13 A. I haven't really looked to answer that
`14 question before. The best I can say in
`15 answering your question is it might. It
`16 certainly seems like looking at the claim that
`17 it would be possible to have sent that search
`18 query previously.
`19 Q. So there's nothing in your view in the
`20 claim that rules out that reading?
`21 A. I haven't really looked at it to answer
`22 the question of ordering of parts of the
`23 limitation with respect to the limitation as a
`24 whole. I would have to think about it more
`25 carefully. I just haven't looked at that
`
`Page: 4
`Page 15
`
` 1 particular aspect of the claim.
` 2 Q. So do you agree, though, that the claim
` 3 would make sense with that understanding, that
` 4 previous search queries sent by the metasearch
` 5 engine means previous to step C?
` 6 A. As I said, as I look at it right now, I
` 7 don't see anything that jumps out of the page
` 8 to me as to why that wouldn't be a reasonable
` 9 reading. But as I said, I haven't really gone
`10 through to try and establish the ordering of
`11 steps. I understand in a method claim there's
`12 no presumption of ordering unless it's
`13 dictated by the claim, but I haven't had to go
`14 into that level of detail in forming my
`15 opinions as to whether or not that previous
`16 search query could happen or must happen or
`17 might happen before step C.
`18 Q. So as you look at the claim now, do you
`19 see any order dictated by the language of that
`20 phrase, "previous search query sent by the
`21 metasearch engine"?
`22 A. Nothing definitive with respect to --
`23 maybe the better way to say it is I just
`24 haven't looked. It seems a reasonable
`25 interpretation, but I really haven't looked to
`Page 16
` 1 answer that question and sort of trace out
` 2 where I think the boundaries are of the timing
` 3 relationships even within limitation C.
` 4 Q. Okay. Well, let's consider might that
` 5 phrase "previous search query" mean that the
` 6 metasearch engine sent the search query
` 7 previous to step B?
` 8 A. I have not even really tried to answer
` 9 that question.
`10 Q. Does that seem like the claim would make
`11 sense with that reading to you?
`12 A. I haven't looked to answer that question
`13 or sit down with the claim and try and assess
`14 the requirements of ordering. I just haven't
`15 tried to answer that question as part of
`16 offering my opinions in the declaration.
`17 Q. Okay. But you don't see anything today
`18 in your review of the claim that would rule
`19 out that reading of the claim?
`20 MR. SCHULTZ: Objection, outside
`21 the scope of his declaration.
`22 THE WITNESS: As I sit here right
`23 now, I don't see anything that requires or
`24 precludes the possibility that that previous
`25 search query couldn't have happened before
`
`800-545-9668
`612-339-0545
`
`Paradigm Reporting & Captioning
`www.paradigmreporting.com
`
`#80832
`
`

`

`Kevin Almeroth Ph.D.
`8/26/2014
`Page 17
` 1 step B. But again, I haven't really looked at
` 2 it in detail and it wasn't really a necessary
` 3 question I needed to answer in order to
` 4 perform the analysis in my declaration.
` 5 BY MS. REICHENBACH:
` 6 Q. Okay. So I will have you consider just
` 7 one more meaning here as that might the phrase
` 8 "previous search query" mean prior to step A?
` 9 MR. SCHULTZ: Objection.
`10 THE WITNESS: I think like the
`11 previous questions, I haven't really looked at
`12 the ordering requirements of claim 13 to see
`13 whether or not the previous search query term
`14 in limitation C has any temporal relation to
`15 happening before or after step B. I just --
`16 sorry, step A. I just haven't looked.
`17 BY MS. REICHENBACH:
`18 Q. But is your answer again the same, that
`19 you don't see anything today in the claims
`20 that would rule out that meaning?
`21 MR. SCHULTZ: Objection.
`22 THE WITNESS: Well, that's not
`23 quite what I said earlier with respect to the
`24 claims. Taking a very cursory examination of
`25 the claim as I sit here right now, I don't see
`Page 18
`
` 1 anything. But I don't think that's
` 2 representative of my expert opinion, because I
` 3 really haven't had a chance to look at it in
` 4 detail.
` 5 BY MS. REICHENBACH:
` 6 Q. You stated previously, though, that your
` 7 focus was on the added portions of the claim,
` 8 which step C is part of the added portion of
` 9 the claim, is it not?
`10 A. Absolutely. But as part of my analysis I
`11 didn't need to consider the ordering of the
`12 steps, because from the perspective of the
`13 prior art and the assumptions I needed to make
`14 it wasn't really a question that created a
`15 distinctiveness over the prior art. There
`16 wasn't prior art that did everything here but
`17 in a different order and so I had to analyze
`18 what I thought the order was. That level of
`19 detail really never needed to be addressed in
`20 the analysis that I did.
`21 Q. Do you recall reading anything in the
`22 '924 Patent that addressed this issue and
`23 would rule out any of these three alternative
`24 meanings?
`25 A. I don't recall anything from reading the
`
`Page: 5
`Page 19
` 1 '924 specification that would rule out any of
` 2 those meanings. But again, as the line of
` 3 questioning has hopefully communicated, it
` 4 wasn't a level of detail I needed to get into
` 5 as part of my analysis.
` 6 Q. Does your answer change for the 2000
` 7 patent application?
` 8 A. It does not.
` 9 Q. Okay. Let's talk about caching. In
`10 2000, caching was a well-known technique,
`11 wasn't it?
`12 A. It's a fairly broad question. There's
`13 different kinds of caching. Certainly at
`14 least some of the types of caching were fairly
`15 well known in -- more specifically, I think
`16 known to persons of skill in the art at the
`17 time.
`18 Q. Okay. Did a person of ordinary skill in
`19 the art in metasearching in 2000 have a
`20 well-accepted understanding of the word
`21 "caching"?
`22 A. From that particular perspective, I think
`23 such a person would need some context on what
`24 caching was. There's memory caching. There's
`25 caching of the kind I've talked about in my
`Page 20
` 1 declaration, web object caching, file caching.
` 2 It exists in lots of different contexts. Even
` 3 a person of skill in the art, and you've
` 4 mentioned a person of skill in the art of
` 5 metasearching which I mean I have my
` 6 definition I'm not sure I would call that a
` 7 person of skill in the art in metasearching a
` 8 person of skill in the art as it relates to
` 9 the patents in suit I think would be aware of
`10 many kinds of caching.
`11 Q. Well let's use that term as you
`12 understand it in your declaration. So what
`13 definition of caching would the person as you
`14 understand it of ordinary skill in the art
`15 have for the term caching?
`16 A. I think caching generally and I don't
`17 know that I can give you sort of a textbook
`18 definition I can probably give you some
`19 examples of caching but generally caching was
`20 of the type where objections in particular
`21 static objects were stored in places other
`22 than what I will call the origin server such
`23 that that content could be more easily
`24 retrieved from somebody who was trying to get
`25 it.
`
`800-545-9668
`612-339-0545
`
`Paradigm Reporting & Captioning
`www.paradigmreporting.com
`
`#80832
`
`

`

`Kevin Almeroth Ph.D.
`8/26/2014
`Page 21
` 1 Q. So would you agree that part of caching
` 2 was to fore go searching that origin server as
` 3 you stated if the data was already available
` 4 and valid in a cache?
` 5 A. No.
` 6 Q. And what do you disagree with?
` 7 A. Well, I don't think that's a blanket
` 8 statement that can be applied to the concept
` 9 of caching. For example, in many of the
`10 caching systems at the time, one of the
`11 objectives was to still be able to confirm
`12 with the content provider that the document
`13 had been requested so that the content
`14 provider could do a count of how much interest
`15 there was in a particular object. There's
`16 also the concept that you mentioned in your
`17 question this idea of whether it's a valid
`18 cache object. And oftentimes it's difficult,
`19 if not impossible, to determine the validity
`20 or freshness of a cache object without doing
`21 some sort of confirmation. So there were
`22 certainly systems where you would make a
`23 request for an object and say: I have a cache
`24 of this object as of this date. Is there an
`25 updated version? And a web server, either the
`Page 22
` 1 origin server or proxy would respond and say:
` 2 No, there's no updated version. You have the
` 3 latest version. And so the client would use
` 4 that most recent version that it already had.
` 5 Q. Okay. So do you agree that if the
` 6 version that is in the cache is available and
` 7 current or valid, then the version in the
` 8 cache would be used instead of searching again
` 9 or retrieving again that resource from the
`10 origin server?
`11 A. Possibly. I think there were some
`12 systems that could be configured to do things
`13 like that, but I wouldn't say that that's the
`14 way systems only operated.
`15 Q. So you stated just now that there are
`16 some systems that might not have operated that
`17 way. Can you name any systems that might not
`18 have operated that way?
`19 A. I believe the Farber Patent -- or
`20 actually instead of the Farber Patent, let me
`21 use the Akamai Patent that's mentioned in my
`22 declaration. I think it's -- I forget the
`23 number or the inventors of Leighton. The way
`24 that that system would work is you would make
`25 a request for the page. There's at least one
`
`Page: 6
`Page 23
` 1 instance where the request for a web page
` 2 would go to the origin server regardless of
` 3 what was cached and then objects within that
` 4 page could be fetched from a caching server.
` 5 I believe that there were timeout values
` 6 associated with cache objects. And so even
` 7 from an Akamai server and even though the
` 8 Akamai server had the cache object, it might
` 9 be the case that the object was re-fetched
`10 from the origin server before being provided
`11 to the user. Another system I believe is
`12 Squid. And I think Squid was highly
`13 configurable. And I believe it would have
`14 options where you could identify certain
`15 content as being precluded from caching even
`16 though it might already exist in the cache.
`17 Squid existed also at multiple levels, meaning
`18 you could have a squid cache on a user's
`19 computer, at an organization boundary or
`20 somewhere within an ISP. And I think another
`21 system that was also highly configurable at
`22 the time was called Harvest. And then just
`23 another reference, there's the Farber Patent.
`24 And I don't recall if Farber provided an
`25 option of retrieving a -- an item that
`Page 24
` 1 potentially was already cached using some
` 2 criteria other than the validity of the
` 3 object.
` 4 Q. Isn't it true that whether or not the
` 5 object was re-fetched depended on what content
` 6 was found in the cache, whether it be valid or
` 7 up to date?
` 8 A. In which system?
` 9 Q. In the systems that you described you
`10 mentioned that the object might still be
`11 re-fetched. My question is whether or not the
`12 decision to re-fetch was always dependent on
`13 what the -- what item was found in the cache
`14 database, isn't that correct?
`15 A. No, I don't think it always depended on
`16 that. As I said, I think these systems were
`17 highly configurable and there might have been
`18 instances where that was the case and
`19 instances where it wasn't the case. It really
`20 just depends on the system. There was also
`21 guidelines from the Internet Engineering Task
`22 Force that tried to address this kind of
`23 problem. One of the problems that was
`24 recognized was it would be difficult for a
`25 content provider, somebody who would run an
`
`800-545-9668
`612-339-0545
`
`Paradigm Reporting & Captioning
`www.paradigmreporting.com
`
`#80832
`
`

`

`Kevin Almeroth Ph.D.
`8/26/2014
`Page 25
` 1 origin server to tell how popular its objects
` 2 were if it was using a third-party content
` 3 delivery network or caching system. And so
` 4 there were mechanisms within the RFC and
` 5 systems that had been produced and
` 6 commercialized that contemplated, you know,
` 7 not allowing objects to be cached even though
` 8 a user could pretty much do whatever they
` 9 wanted in terms of storing the objects on
`10 their computer. So depending on how the web
`11 page was set up, objects might be re-fetched
`12 even if they existed in a local cache.
`13 Q. Do you know of any caching system that
`14 always retrieved the object from the origin
`15 server even if it was found in the cache?
`16 A. No. That's almost the definition of not
`17 doing caching. I don't really know of a
`18 system that would be designed or operate in a
`19 way that would never use the cache. Maybe
`20 again one of these highly configurable systems
`21 like squid or Harvest you could essentially
`22 turn the cache off even though there are
`23 objects that were still resident in the cache.
`24 It might have been possible. So after I
`25 thought about it while I was answering the
`Page 26
` 1 question, my answer sort of changed a little
` 2 bit. So to be clear, I don't think so. But
` 3 certainly one of these highly-configurable
` 4 systems it might have been the case that you
` 5 could turn off caching and it would meet the
` 6 requirements of your question.
` 7 Q. So it sounds like to me, and let me know
` 8 if I'm incorrect, that caching will always
` 9 consider what is in the cache -- let me step
`10 back a second.
`11 That whether or not the object is in the
`12 cache will always be considered in a caching
`13 system and the determination of whether or not
`14 to retrieve the object will be based on an
`15 evaluation of the object in the cache?
`16 A. So that question had two parts. Let me
`17 sort of break it down. The first part seemed
`18 to say that it was always the case that a
`19 cache system would consider what was in the
`20 cache. I'm close to being willing to agree
`21 with you on that, but I'm suspect of the
`22 question -- or the characterization that it's
`23 always the case. There's probably some
`24 caching system out there where it cared less
`25 about what objects were in the cache and more
`
`Page: 7
`Page 27
` 1 about other tasks. It's just hard for me to
` 2 know everything so I can answer that question
` 3 definitively.
` 4 For the second part, I don't think I can
` 5 agree with that, because it might be the case
` 6 that a caching system uses criteria other than
` 7 the object to determine whether to retrieve
` 8 the object. And there could be call sorts of
` 9 characteristics, who the provider is, the date
`10 an object was last retrieved, those kinds of
`11 things. The criteria can vary fairly
`12 dramatically across different kinds of caching
`13 systems.
`14 Q. Okay. I'm going to hand you an exhibit
`15 that's marked as 2023. This is U.S. Patent
`16 Number 6,185,598.
`17 Is this familiar to you?
`18 A. (Reviews document.) It is.
`19 Q. And why is it familiar to you?
`20 A. For a couple of reasons. It was one of
`21 the patents that I identified in my
`22 declaration as being relevant in the context
`23 of discussing caching within web-based
`24 systems. I've also seen it before in at least
`25 a couple of other cases in which I've
`
`Page 28
` 1 testified. It was used as either prior art.
` 2 Or in another case it was used as part of an
` 3 accusation of infringement.
` 4 Q. Can you turn to page 7, which is
` 5 figure 5. Do you see where it says -- there's
` 6 a decision box labeled C-3 that says:
` 7 Resource is cached locally? And if the answer
` 8 to the question is no, it goes to a box that
` 9 says: Request and obtain resource from
`10 reflector/origin. If the answer to the
`11 question is yes, it moves on to C-4, which
`12 says: Send resource to client.
`13 Do you see that?
`14 A. I do.
`15 Q. Okay. Please turn to page 13 in the
`16 patent. Please look at column 10, line 39.
`17 Do you see where it says "the repeater,"
`18 column 10, line 39?
`19 A. Yes.
`20 Q. Do you see where it says: C-3, the
`21 repeater then determines whether the requested
`22 resource is cached locally. If the requested
`23 resource is in the repeater's cache, it is
`24 retrieved. On the other hand, if a valid copy
`25 of the requested resource is not in the
`
`800-545-9668
`612-339-0545
`
`Paradigm Reporting & Captioning
`www.paradigmreporting.com
`
`#80832
`
`

`

`Kevin Almeroth Ph.D.
`8/26/2014
`Page 29
` 1 repeater's cache, the repeater modifies the
` 2 incoming URL creating a request that it issues
` 3 directly to the originating reflector which
` 4 processes it as in B-1-B-6.
` 5 Do these excerpts conform with how a
` 6 skilled artisan in 2000, as you've used that
` 7 term, would have understood caching?
` 8 A. I think that's part of it. I think as
` 9 we've discussed in the questioning so far,
`10 that's certainly one way of doing it but not
`11 the only way.
`12 Q. So your understanding is this is
`13 consistent with how a person of skill would
`14 you understand caching, although there may be
`15 additional features, is that correct?
`16 A. I think this is one example of caching.
`17 I think that there might be other embodiments
`18 even within the '598 Patent that describe
`19 options for slightly different ways of
`20 processing cache requests. I think a person
`21 of skill in the art would understand that
`22 there were other ways of designing systems.
`23 But I think at a high level the idea where
`24 some device is receiving or intercepting
`25 requests for objects and there's a
`
`Page 30
` 1 determination as to whether or not that object
` 2 is stored locally, and if it is, it's returned
` 3 to the client, I think that's generally
` 4 consistent with the most understood high-level
` 5 idea of what caching is.
` 6 Q. Handing you a document marked as
` 7 Exhibit 2024. This document is U.S. Patent
` 8 Number 6,108,703.
` 9 Do you recognize this patent?
`10 A. (Reviews document.) I do.
`11 Q. And was this patent also cited to in your
`12 declaration?
`13 A. It was.
`14 Q. Please turn to page 10 and look at
`15 column 12, line 26. Do you see where it says
`16 step five? It says: Step five, the browser
`17 then makes a request for an object named
`18 a123.ghosting.akamai.com/.../www.provider.com/
`19 tech/images/space.story.gif from the close-by
`20 ghost. Note that the name of the original
`21 server, www.provider.com, preferably is
`22 included as part of the URL. The software
`23 running on the ghost parses the page name into
`24 the original host name and the real page name.
`25 If a copy of the file is already stored on the
`
`Page: 8
`Page 31
` 1 ghost, then the data is returned immediately.
` 2 If, however, no copy of the data on the ghost
` 3 exists, a copy is retrieved from the original
` 4 server or another ghost server.
` 5 Does the ghost server described here
` 6 conform with how the skilled artisan, as you
` 7 use that term, would ha