Northeastern University et al v. Google, Inc.,

Filing 78

REPLY to 71 Claim Construction Brief filed by Google, Inc.,. (Attachments: # 1 Exhibit A, # 2 Exhibit B, # 3 Exhibit C, # 4 Exhibit D, # 5 Exhibit E, # 6 Exhibit F, # 7 Exhibit G, # 8 Exhibit H, # 9 Exhibit I, # 10 Exhibit J, # 11 Exhibit K, # 12 Exhibit L, # 13 Exhibit M, # 14 Exhibit N, # 15 Exhibit O, # 16 Exhibit P)(Albert, Francis)

Northeastern University et al v. Google, Inc.,

Doc. 78 Att. 3

EXHIBIT C

Dockets.Justia.com

1 2 3 4 VS. 5

IN THE UNITED STATES DISTRICT COURT EASTERN DISTRICT OF TEXAS MARSHALL DIVISION NORTHEASTERN UNIVERSITY and JARG CORP. * * * * * GOOGLE INC. * Civil Action No. 2:07-CV-486(TJW)

3DJH

6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 Job No.: 213552
9HULWH[W &RUSRUDWH 6HUYLFHV

******************************************* ORAL AND VIDEOTAPED DEPOSITION OF KENNETH BACLAWSKI AUGUST 27, 2009 ******************************************* ORAL AND VIDEOTAPED DEPOSITION OF KENNETH BACLAWSKI, produced as a witness at the instance of the Defendant and duly sworn, was taken in the above-styled and numbered cause on the 27th day of August, 2009, from 9:04 a.m. to 5:08 p.m., before MARSHA EVANS, Certified Shorthand Reporter in and for the State of Texas, reported by machine shorthand, at the offices of Vinson & Elkins, 2801 Via Fortuna, Suite 100, Austin, Texas, pursuant to the Federal Rules of Civil Procedure and the provisions stated on the record or attached hereto.

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25

content labels are represented using a data structure called a KEYNET." label? A. What's the reference to content

3DJH

What is that referring to? The term content is used to refer to the

collection of objects or documents that are being indexed. The term label is a term for the annotation

that is actually indexed by the -- by this KEYNET -KEYNET system. Q. So a content label is a -- is a representation

of some information in the corpus; is that right? A. No, not necessarily. It could have additional It

annotations that are not actually in the content. could have information that is explicitly in the content.

It's -- it's really an independent annotation

in general, although as I say it could be very closely linked to the content. Q. And what does it mean that both the queries

and the content labels are represented using the data structure called a KEYNET? A. It means that the same data structure is used

for both of these. Q. And is that important for some reason that

they use the same data structure? A. Yes. It's necessary for them to have the same

data structure so that one can match the queries with
9HULWH[W &RUSRUDWH 6HUYLFHV

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25

the objects or the -- or the documents that are in the collection. Q. You referred earlier to a KEYNET being a

3DJH

graphical structure, and this paper refers to it as a directed graph. Is that the same -- same idea? Objection, form. Graphical structure The edges are

MR. RYAN:

THE WITNESS:

consists of vertices and edges.

typically directed, meaning it goes from one vertex to another vertex, hence the name directed graph. Although sometimes undirected edges are used. Q. (By Mr. Pollack) If you'll turn to page 2738,

Exhibit 17, in this range there's a paper entitled An Abstract Model For Semantically Rich Information Retrieval. It looks like it's dated March 31st, '94,

coauthored by yourself and someone named Dan Simovici. See that? A. Q. A. Yes. Who is Dan Simovici? Dan Simovici is a professor at the University He was the coauthor of this

of Massachusetts Boston. paper. Q. A.

What was his role in the KEYNET project? He was not very much involved in it. The only

connection that he had was in this particular paper we
9HULWH[W &RUSRUDWH 6HUYLFHV

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25

identified uniquely by an object identifier (OID) and is described by a content label."

3DJH

Is that reference to

content label, is that the same content label that we referred to earlier in our discussions associated with the KEYNET papers? A. The context here is a little different. It

just says that it's written in a form of artificial language specified by the ontology. speaking it's a similar concept. Q. Okay. The second paragraph states that, "An But roughly

ontology can also specify weight information such as the strength of a relationship or the degree of prototypicality of an attribute value." Can you

explain that a little bit more to me what you understand to mean that an ontology can specify weight information? A. Weight information has to do with the kind of

fuzzy aspect of information retrieval, so that one can distinguish between stronger and weaker matches to concepts and relationships. Q. Okay. So the weight information could be, for

example, some added information contained within a content label, for example? A. label.
9HULWH[W &RUSRUDWH 6HUYLFHV

Yes.

Weight information can be in a content

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25

Q. queries.

The next paragraph starts talking about The second sentence reads, "A fragment of a

3DJH

content label or a query is a part of the content label or query consisting of a limited number of attributes and attribute values joined by relationships." What

does it mean -- well, what are attributes and attribute values, first of all? A. The attributes and attribute values there are

the vertices of this content label, and then you have the attribute values joined by relationships. Q. So am I correct in understanding that, for

example, if a query complies with this structure that we're talking about earlier about being a graphical representation of vertices and edges, the fragment of that query might be a subset of that -- a piece of that graph? A. Q. Would that be accurate? Yes. It's part of it.

And this breaking up of a query or a content

label into -- into fragments, that conceptually is related to what we talked about earlier when we were looking at the fragmentation algorithm in the KEYNET papers. here? MR. RYAN: Objection, form. The KEYNET paper is an Is that the concept that's being referred to

THE WITNESS:

9HULWH[W &RUSRUDWH 6HUYLFHV

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25

A.

I just recently reviewed a couple of papers

3DJH

for a journal on -- I can't remember the title. Knowledge and communities and cultures, something like that. Q. I don't remember the precise name. I'm going to just refer you back to -- we used What's

the term earlier, but the term fuzzy query.

generally your understanding of what a fuzzy query is? MR. RYAN: Objection, form. Fuzzy -- a fuzzy query is

THE WITNESS:

one for which the query itself and the results have weights or degrees of importance, so that a result is more or less -- you know, varying degrees of strength with respect to the -- the query. MR. POLLACK: Why don't we take five. Off the record at

THE VIDEOGRAPHER: 4:59 p.m.

(Recess from 5:00 to 5:07) THE VIDEOGRAPHER: record at 5:06 p.m. Q. (By Mr. Pollack) Professor, I just want to We're back on the

clear something up. question earlier.

I might have asked an ambiguous We talked about after resigning as a

board member that you haven't had any further -- done any further work for Jarg. I just want to clarify,

have you had any additional consulting work that you've
9HULWH[W &RUSRUDWH 6HUYLFHV