Oracle America, Inc. v. Google Inc.

Filing 208

MOTION for Leave to File Supplement Invalidity Contentions filed by Google Inc.. (Attachments: #1 Affidavit Declaration of Mark Francis in support of Motion for Leave to Supplement Invalidity Contentions, #2 Proposed Order Proposed Order Granting Google Inc's Motion for Leave to Supplement Invalidity Contentions)(Sabnis, Cheryl) (Filed on 7/8/2011)

1
2
3
4
5
6
7
8
9
10
11
12

KEKER & VAN NEST LLP
ROBERT A. VAN NEST - #84065
[e-mail]
CHRISTA M. ANDERSON - #184325
[e-mail]
DANIEL PURCELL - #191424
[e-mail]
710 Sansome Street
San Francisco, CA 94111-1704
Telephone: [phone]
Facsimile: [phone]

KING & SPALDING LLP
SCOTT T. WEINGAERTNER (Pro Hac Vic
[e-mail]
ROBERT F. PERRY
[e-mail]
BRUCE W. BABER (Pro Hac Vice)
[e-mail]
1185 Avenue of the Americas
New York, NY 10036-4003
Telephone: [phone]
Facsimile: [phone]

KING & SPALDING LLP
DONALD F. ZIMMER, JR. (SBN 112279)
[e-mail]
CHERYL A. SABNIS (SBN 224323)
[e-mail]
101 Second Street, Suite 2300
San Francisco, CA 94105
Telephone: [phone]
Facsimile: [phone]

GREENBERG TRAURIG, LLP
IAN C. BALLON (SBN 141819)
[e-mail]
HEATHER MEEKER (SBN 172148)
[e-mail]
1900 University Avenue, Fifth Floor
East Palo Alto, CA 94303
Telephone: [phone]
Facsimile: [phone]

Attorneys for Defendant
GOOGLE INC.

13
14
15

UNITED STATES DISTRICT COURT

16

NORTHERN DISTRICT OF CALIFORNIA

17

SAN FRANCISCO DIVISION

18
19

Case No. 3:10-cv-03561 WHA

ORACLE AMERICA, INC.,

20

Plaintiff,

21

v.

22

GOOGLE INC.,

23

DECLARATION OF MARK H.
FRANCIS IN SUPPORT OF
DEFENDANT GOOGLE INC.’S
MOTION FOR LEAVE TO
SUPPLEMENT INVALIDITY
CONTENTIONS

Defendant.

24
25
26
27
28

DECLARATION OF MARK H. FRANCIS IN SUPPORT OF
GOOGLE’S MOTION FOR LEAVE TO SUPPLEMENT INVALIDITY CONTENTIONS, CIV. NO. CV 10-03561-WHA

1

I, Mark H. Francis, declare as follows:

2

I am an associate in the law firm of King & Spalding LLP, counsel to Google Inc.

3

in the present case. I submit this declaration in support of Defendant Google Inc.’s

4

Motion for Leave to Supplement Invalidity Contentions. I make this declaration based on

5

my own personal knowledge. If called as a witness, I could and would testify

6

competently to the matters set forth herein.

7
8
9

1.

Both before and after serving its Invalidity Contentions, Google sought out

new bases for invalidity.
2.

A number of attorneys, general researchers, experts, and commercial search

10

firms participated in Google’s search to identify many thousands of potentially relevant

11

prior art publications.

12

3.

All such prior art publications were reviewed by an individual with a

13

background in software programming and determined to be relevant to one or more of the

14

patents-in-suit.

15

4.

Google’s prior art search encompassed a wide variety of sources and

16

techniques to identify prior art dating back to the 1960’s, including searching on

17

commercial and publicly-available electronic databases such as Google Scholar, the

18

Patent Office’s patent database, the Institute for Electrical and Electronics Engineers

19

(IEEE) database, the Association for Computing Machinery (ACM) database, and

20

electronic catalogs provided by a number of public and private universities with extensive

21

print collections of programming and systems materials.

22
23
24

5.

Much of this searching was based on a manual identification and review of

these prior art materials.
6.

One search technique employed to locate prior art was the use of

25

bibliographies to perform forward and reverse citation searches on many of the relevant

26

publications.

27
28

7.

For many of the older prior art publications, citation searching required

manual adjustment of search criteria due to inconsistencies in citation formats.
1
DECLARATION OF MARK H. FRANCIS IN SUPPORT OF
GOOGLE’S MOTION FOR LEAVE TO SUPPLEMENT INVALIDITY CONTENTIONS, CIV. NO. CV 10-03561-WHA

1

8.

Review of electronic materials and conversations with experts in the field

2

identified a number of potential prior art systems and materials that were not easily

3

accessible through traditional search engines.

4

9.

Google’s searchers manually reviewed extensive collections of prior art

5

materials relating to Multics, IBM System 360, and other systems dating back to the

6

1960’s.

7
8
9
10
11
12
13

10.

Many of these older prior art documents were not indexed or easily

searchable.
11.

Some of the identified prior art documentation was only available from

individuals’ personal collections.
12.

Once located, many such prior art documents were difficult to understand

without first learning about the particular computer platforms they discussed.
13.

Each of these older prior art systems had its own distinctive terminology

14

and many had system architectures foreign to all but experienced computer programmers

15

and researchers.

16

14.

Google’s attorneys and experts examined a number of physical prior art

17

documents in university libraries across the country looking for relevant material in

18

textbooks, conference proceedings, dissertations, and product documentation.

19

15.

Google’s attorneys also searched through various websites (eBay.com,

20

abebooks.com and others) and purchased prior art textbooks and product documentation

21

that contained prior art or identified other relevant publications or products.

22

16.

Google has also conducted searches of Oracle’s massive document

23

production to identify relevant prior art material, but the voluminous nature of that

24

production, combined with limited available metadata and search context, lack of

25

searchable text, and disorganized dump of the production, has made searching difficult

26

and time consuming.

27
28

17.

Google has expended over fifteen hundred hours person hours in the search

process, involving over fifteen individuals, including technical consultants and experts.
2
DECLARATION OF MARK H. FRANCIS IN SUPPORT OF
GOOGLE’S MOTION FOR LEAVE TO SUPPLEMENT INVALIDITY CONTENTIONS, CIV. NO. CV 10-03561-WHA

1
2
3

18.

The Tafvelin reference was discovered in April 2011, the Daley reference

was discovered in April 2011, and the Vyssotsky reference was discovered in June 2011.
19.

Attached to this declaration as Exhibit A is a true and correct copy of

4

Oracle’s Supplemental Response to Google Interrogatory No. 13 (dated April 25, 2011),

5

stating that the inventors’ documents had been destroyed (“Oracle no longer has custodial

6

data for the following people because they left employment at Sun Microsystems some

7

time ago and their data was removed from Sun’s systems pursuant to Sun’s policies: Lars

8

Bak … Nedim Fresko, Robert Griesemer, … Li Gong, … Richard Tuck, …Frank

9

Yellin.”) (highlights added).

10

20.

Attached to this declaration as Exhibit B is a true and correct copy of an

11

excerpt from the transcript of the April 14, 2011 deposition of Lisa J. Ripley

12

acknowledging that the inventors’ documents had been destroyed (highlights added).

13

21.

Attached to this declaration as Exhibit C is a true and correct copy of a

14

letter from Oracle to Google on June 22, 2011, stating that Oracle “recently identified”

15

documents associated with two of the named inventors.

16

22.

Attached to this declaration as Exhibit D is a true and correct copy of a

17

letter from Google to Oracle on May 3, 2011 asking for consent to file an unopposed

18

motion for leave.

19

23.

Attached to this declaration as Exhibit E is a true and correct copy of a

20

letter from Oracle to Google on May 3, 2011 declining to consent to a motion for leave

21

and demanding “an identification of the new prior art references and an explanation of

22

the grounds for good cause to amend.”

23
24
25

24.

Attached to this declaration as Exhibit F is a true and correct copy of a

letter from Google to Oracle on May 6, 2011 outlining its good cause to supplement.
25.

Attached to this declaration as Exhibit G is a true and correct copy of an e-

26

mail from Google to Oracle on May 16, 2011 enclosing a draft of Google’s supplemental

27

invalidity contentions and supporting exhibits.

28

26.

Attached to this declaration as Exhibit H is a true and correct copy of a
3

DECLARATION OF MARK H. FRANCIS IN SUPPORT OF
GOOGLE’S MOTION FOR LEAVE TO SUPPLEMENT INVALIDITY CONTENTIONS, CIV. NO. CV 10-03561-WHA

1

letter from Google to Oracle on May 25 2011, memorializing Mr. Peters’, counsel for

2

Oracle, concession that Oracle was not prejudiced by Google’s supplemental invalidity

3

contentions.

4

27.

Attached to this declaration as Exhibit I is a true and correct copy of a letter

5

from Oracle to Google on May 31, 2011, refusing to consent to an unopposed motion for

6

leave, but not disputing Mr. Peters’ concession at the meet-and-confer.

7

28.

Attached to this declaration as Exhibit J is a true and correct copy of an

8

excerpt from B. Ramakrishna Rau, LEVELS OF REPRESENTATION OF PROGRAMS AND THE

9

ARCHITECTURE OF UNIVERSAL HOST MACHINES, Coordinated Science Laboratory,

10

University of Illinois (1978) (“Rau”) (highlights added), as produced to Oracle in this

11

case.

12

29.

Attached to this declaration as Exhibit K is a true and correct copy of an

13

excerpt from Oracle’s Second Supplemental Infringement Contentions, alleging that

14

JavaOS 1.0 practices the invention claimed in the ‘702 patent (highlights added).

15

30.

Attached to this declaration as Exhibit L is a true and correct copy of a May

16

29, 1996 press release from JavaSoft, a subsidiary of Oracle (then named Sun

17

Microsystems Inc.), announcing the release of JavaOS more than a year before the

18

October 31, 1997 filing date of the application which issued as the ‘702 patent, available

19

at http://web.archive.org/web/19961220110704/http://www.sun.com/smi/Press/sunflash/

20

9605/sunflash.960529.11819.html.

21

31.

Attached to this declaration as Exhibit M is a true and correct copy of a

22

letter from Oracle to Google on July 6, 2011 claiming that its identification of a prior art

23

version of JavaOS as practicing the patent was “in error” and attempting to amend its

24

infringement contentions without leave of the Court. All references to material subject to

25

the “HIGHLY CONFIDENTIAL – SOURCE CODE” designation have been redacted or

26

omitted.

27
28

32.

Attached to this declaration as Exhibit N is a true and correct copy of an

excerpt from Google’s January 2011 Invalidity Contentions (highlight added).
4
DECLARATION OF MARK H. FRANCIS IN SUPPORT OF
GOOGLE’S MOTION FOR LEAVE TO SUPPLEMENT INVALIDITY CONTENTIONS, CIV. NO. CV 10-03561-WHA

1

33.

Attached to this declaration as Exhibit O is a true and correct copy of an

2

excerpt from Li Gong et al., GOING BEYOND THE SANDBOX: AN OVERVIEW OF THE NEW

3

SECURITY ARCHITECTURE IN THE JAVA DEVELOPMENT KIT 1.2, USENIX Symposium on

4

Internet Technologies and Systems (December 8-11, 1997) (“Gong”) (highlights added),

5

as produced to Oracle in this case.

6

I declare under penalty of perjury that the foregoing facts are true and correct.

7

Executed on July 8, 2011 in New York, New York.
/s/ Mark H. Francis /s/
Mark H. Francis

8
9
10

I hereby attest that Mark H. Francis concurs in the e-filing of this document.

11

/s/ Cheryl A. Sabnis /s/

12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
5

DECLARATION OF MARK H. FRANCIS IN SUPPORT OF
GOOGLE’S MOTION FOR LEAVE TO SUPPLEMENT INVALIDITY CONTENTIONS, CIV. NO. CV 10-03561-WHA

GOOGLE’S MOTION FOR LEAVE TO SUPPLEMENT INVALIDITY CONTENTIONS
CIV. NO. CV 10-03561-WHA

Exhibit A

1

PROPOUNDING PARTY:

Defendant Google Inc.

2

RESPONDING PARTY:

Plaintiff Oracle America, Inc.

3

SET NO.:

Three (Interrogatory 13)

4

Pursuant to Rules 26 and 33 of Federal Rules of Civil Procedure, Plaintiff Oracle

5

America, Inc. (“Oracle”) hereby submits the following supplemental response and objections to

6

Defendant Google Inc.’s (“Google”) Third Set of Interrogatories.

7

INTERROGATORY NO. 13:

8
9

Describe with particularity any Documents that Oracle has a reasonable belief were at one
time in the possession, custody, or control of Sun or Oracle and that would be responsive to any

10

of Google’s Requests for Production of Documents but that are no longer in the possession,

11

custody, or control of Oracle and explain, with specificity, why each such Document is no longer

12

in Oracle’s possession, custody, or control.

13

FIRST SUPPLEMENTAL RESPONSE TO INTERROGATORY NO. 13:

14

Oracle objects to this Interrogatory to the extent it seeks information (e.g., descriptions of

15

documents) protected by the attorney-client privilege, the work product doctrine, the common

16

interest privilege, and/or any other applicable privilege, immunity, or protection.

17

Subject to the foregoing objection, Oracle responds: Oracle is not aware of any specific

18

responsive documents that were at one time, but are no longer, in Oracle’s possession, custody or

19

control. As described below, data and documents belonging to certain identified Oracle and Sun

20

Microsystems custodians are no longer in Oracle’s possession:

21

•

22

Data from the computer of Oracle employee Vineet Gupta was lost when Mr.
Gupta’s computer was stolen in 2008.

23

•

24

employment at Sun Microsystems some time ago and their data was removed from

25

Sun’s systems pursuant to Sun’s policies: Lars Bak, Manoharan Balasubramaniam,

26

David Brownell, Danese Cooper, Alan Brenner, Graham Hamilton, Vincent Hardy,

27

Kathleen Knopoff, Anil Vijendran, Nedim Fresko, Robert Griesemer, William (Bill)

Oracle no longer has custodial data for the following people because they left

28
PLAINTIFF’S SUPPLEMENTAL RESPONSE TO DEFENDANT’S INTERROGATORY NO. 13
CASE NO. CV 10-03561 WHA
pa- 1459323

1

1

Joy, Li Gong, Peter Lord, Shannon Lynch, David P. Stoutamire, Omar Tazi, Laurie

2

Tolson, Richard Tuck, Kenneth Urquhart, Frank Yellin.

3
4
5
6
7
8
9

Dated: April 25, 2011

MICHAEL A. JACOBS
MARC DAVID PETERS
DANIEL P. MUINO
MORRISON & FOERSTER LLP
By: /s/ Marc David Peters
Attorneys for Plaintiff
ORACLE AMERICA, INC.

10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
PLAINTIFF’S SUPPLEMENTAL RESPONSE TO DEFENDANT’S INTERROGATORY NO. 13
CASE NO. CV 10-03561 WHA
pa- 1459323

2

GOOGLE’S MOTION FOR LEAVE TO SUPPLEMENT INVALIDITY CONTENTIONS
CIV. NO. CV 10-03561-WHA

Exhibit B

HIGHLY CONFIDENTIAL
ATTORNEYS' EYES ONLY
Page 74

Page 76

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. Well, in this case, it would be -- I would be
researching the home directory server, the file server,
or the mail server.
Q. And did you determine that the servers had
been end-of-lifed?
A. I don't recall specifically in regards to
her. There's over, I think, 250 custodians on this case,
I could not go line-by-line and tell you -- I mean,
obviously we're tracking it internally, but I can't tell
you right here sitting here more specifically as to her
data.
Q. Okay. Well, these are the inventors and
they're also people that Oracle has made an affirmative
statement in response to Request For Productions that
Oracle does not have custodial data for these
individuals. And so is there anyone other than you who
could explain in more detail the steps that were taken to
support these statements?
A. No, because I did the work, and what I can
tell you, if you're speaking to the inventors, I was
given the names, I went out and researched what servers
their data had resided on, and for the majority of them,
they'd been gone too long, and the servers were gone.
But I verified the servers didn't exist any longer, and
if a server happened to still exist, which I think that
Page 75

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'm not sure there would be records. A lot
of this is based on my memory and just knowledge of the
infrastructure because based on my past history of
employment at Sun, I was involved in system
administration of systems that supported our
infrastructure. But as I've testified, I've made the
effort to go out and try to essentially connect to those
systems to verify they're not in existence.
Q. Is there any formal documentation of the end
of life of servers?
A. Perhaps in our past history, there were
project plans, but obviously with the acquisition, we've
gone -- undergone a lot of changeover, and I could not
speak to specifics as to what's still retained.
Q. And would you have the same answer -- well,
I'll just ask it: What records exist that would reflect
the destruction of specific archive tapes?
A. Our -- I don't know of specific documentation
prior to 2007.
Q. Do you know -- well, were you asked to look
into custodial data for Oracle employees that were never
at Sun?
A. No, I was not personally.
Q. And do you know who would have been asked
similar questions as you were to the -- to the inventors
Page 77

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

may have been one, maybe two people that a server still
existed, I went head and searched those servers to make
sure the data wasn't still there.
Q. And sitting here today, you can't tell me for
any of the inventors, which are Li Gong, Frank Yellin,
Lars Bak, Robert Griesemer, Nedim Fresko, which specific
servers you're talking about?
A. Not off the top of my head.
Q. Are you aware of an interrogatory that was
served relating to document preservation?
A. I've -- there's been multiple interrogatories
that have been discussed, but I can't recall. You know,
I can't speak to the specific interrogatory.
Q. Did you review any interrogatories in
preparation for this litigation, I mean this deposition?
A. Interrogatories? No. As I've already
testified, I'm involved in ongoing meetings discussing
requests for different materials, and so interrogatories
are discussed.
Q. So what records exist that would reflect the
end-of-lifing of the servers you referred to earlier as
well as the -- well, let's start there. What records
exist that would reflect the end-of-lifing of the servers
you referred to in responses to your inventors' custodial
data?

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

for people who were at Oracle with respect to whether
their data was still available?
A. I'm sorry, can you repeat your question? I
didn't catch your train of thought on that.
Q. If there was a similar situation as to
identifying whether custodial data existed, but it was an
Oracle employee and not a prior Sun employee, do you know
who that inquiry would have been directed at?
A. So normally the legal assistant or paralegal
for the case contacts Oracle IT.
Q. And in preparation for this deposition, did
you discuss any situations where Oracle employee
custodial data might not still be available with anyone
who would have been responsible for that inquiry?
A. Not spoke to anyone specifically, but as part
of my integration into Oracle, I'm aware of their -- to
some degree of their processes, and as such, know that
they are similar to us in that former employee data is
available for a limited amount of time unless it's on
legal hold, and then it's removed as part of normal
business operations.
(Exhibits Google 9 and 10 marked.)
MR. SNYDER: Introduced as Google Exhibit 9,
a document entitled "Policies, Procedures and
Guidelines," and it's stamped OAGOOGLE0000062856 to 857.

20 (Pages 74 to 77)
[phone]

VERITEXT REPORTING COMPANY
www.veritext.com

[phone]

6ab8bada-feb0-4f31-a41b-929debde1f2a

HIGHLY CONFIDENTIAL
ATTORNEYS' EYES ONLY
Page 86

Page 88

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. Would that document be provided to all
employees?
A. Yes.
Q. And do you have an understanding as to what's
meant in the first bolded sentence on the cover page or
the first bolded clause, "unless the records have a
current business purpose"?
A. Basically if you need it still for your
ongoing daily business tasks.
Q. So the understanding of that whole sentence
is that it's suggesting that the records shouldn't be
retained unless you're currently using them for a
business task or they're subject to a legal hold?
A. Correct.
Q. And we discussed litigation holds earlier.
Are you personally aware of when the first litigation
hold was put in place for this litigation?
A. No. As I've previously testified, I don't
know any specific dates, just that holds were issued
prior to the claim being filed.
Q. Do you know how Oracle treats Sun's documents
with respect to retention? And by that, the question is
does Oracle continue to use Sun's policy for those
documents or did they recategorize them under Oracle's
policy?
Page 87

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

preserved or retained.
Q. And who would have been responsible for doing
that?
A. The individual employees or the line of
business.
Q. And if someone identified something under
this that needed to be retained, it wouldn't necessarily
be considered a permanent record, would it?
A. If it falls into one of these retention
schedules, yes, then it's considered a permanent record
is supposed to be preserved off-site at Iron Mountain.
Q. So anything that was -- fell into this
category should have been printed and retained in hard
copy?
A. If it was deemed to be a permanent record,
yes.
Q. Well, that's what I'm trying to get at. Let
me go back to what it says.
A. Okay.
Q. Is it automatically deemed a permanent record
if it falls in this category?
A. (No audible response.)
Q. I'll strike that previous question.
I want to re -- if we go back to Exhibit 9,
there's two things. One, it says: "If an email message
Page 89

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. They're being recategorized under Oracle's
policy, with the exception obviously if things are on
legal hold.
Q. And is that process still ongoing?
A. It is.
Q. So if you -- let's go back to Exhibit 10. At
page 63194 and under Retention Code 07102, do you see
that it says, "Records related to domestic patents and
products and processes held by Sun, including those
products developed by Sun personnel on company time,
includes patent registration and related correspondence"?
A. I do.
Q. And that the retention is seven years?
A. I see that.
Q. And it says, "Retention begins when the
patent has expired"?
A. I see that.
Q. Do you have any understanding of whether
related correspondence would include email correspondence
from inventors?
A. At Sun, email itself was not considered a
record. It would be the contents of an email that could
potentially be a record, and that's why they say to refer
to the retention schedule to determine if some content of
an email would be considered a record and needed to be

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

is considered a permanent record, it should be printed
and stored in hard copy format," and it refers to a
general retention schedule to determine if it is
designated as a permanent record. But it also refers in
the very last sentence to the general retention schedule
to determine if messages need to be retained.
Is there a difference between documents that
just need to be retained versus ones that are permanent
records?
A. Yes, because you could have something that's
on legal hold that's not considered a permanent record,
that has to be retained, obviously, to comply with the
legal hold.
Q. So looking at the -- if I look at the general
retention schedule, I don't see any real distinct -- I
don't see anywhere where it says certain documents are -should be permanent records and some should simply be
retained.
I guess, is -- is what you said earlier, that
anything that -- an individual would identify as applying
to, for instance, retention code 07102, would be a
permanent record that they should then print out and
store in hard copy format?
A. Well, I think one of the difficulties is this
is an older document and this is a newer document

23 (Pages 86 to 89)
[phone]

VERITEXT REPORTING COMPANY
www.veritext.com

[phone]

6ab8bada-feb0-4f31-a41b-929debde1f2a

GOOGLE’S MOTION FOR LEAVE TO SUPPLEMENT INVALIDITY CONTENTIONS
CIV. NO. CV 10-03561-WHA

Exhibit C

755 PAGE MILL ROAD
PALO ALTO
CALIFORNIA 94304-1018
TELEPHONE: [phone]
FACSIMILE: 650 494 0792
WWW MOFO COM

June 22, 2011

MORRISON & FOERSTER LLP
NEW YORK, SAN FRANCISCO,
LOS ANGELES, PALO ALTO,
SACRAMENTO, SAN DIEGO,
DENVER, NORTHERN VIRGINIA,
WASHINGTON, D.C.
TOKYO, LONDON, BRUSSELS,
BEIJING, SHANGHAI, HONG KONG

Writer’s Direct Contact
650.813.5878
[e-mail]

Via E-Mail
[e-mail]
Steven T. Snyder
King & Spalding
100 N. Tryon Street, Suite 3900
Charlotte, NC 28202
Re:

Oracle America, Inc. v. Google Inc. – Oracle Production

Dear Steve:
We recently identified certain source code and documents associated with Nedim Fresko and
Richard Tuck. The source code, designated HIGHLY CONFIDENTIAL – SOURCE CODE,
is currently available for inspection in the source code repository at our office. The
documents are being processed and should be ready for production next week. We will
supplement germane discovery responses to reflect this production of Fresko and Tuck
materials.
Sincerely,
/s/ Jessica J. Tipton
Jessica J. Tipton

pa-1469935

GOOGLE’S MOTION FOR LEAVE TO SUPPLEMENT INVALIDITY CONTENTIONS
CIV. NO. CV 10-03561-WHA

Exhibit D

1185 Avenue of the Americas
New York, New York 10036-4003
www.kslaw.com
Mark H. Francis
Direct Dial: [phone]
Direct Fax: [phone]
[e-mail]

May 3 2011
VIA E-MAIL
Marc D. Peters, Esq.
Michael A. Jacobs, Esq.
Morrison & Foerster LLP
755 Page Mill Road
Palo Alto, CA 94304-1018
[e-mail]
Re:

Oracle America, Inc. v. Google Inc., No. 3:10-CV-03561-WHA (N.D. Cal.)

Dear Marc,
Google would like to move the Court for leave to supplement its Patent L.R. 3-3
Invalidity Contentions. Google has good cause to supplement in view of issues raised by Oracle
during the course of discovery, Oracle’s supplemental infringement contentions, newly
discovered prior art references and new invalidity theories developed by Google after serving its
initial Invalidity Contentions. Moreover, Google believes that supplementing its invalidity
contentions is appropriate in view of the Court’s instructions to the parties regarding their
respective disclosures. See, e.g., April 6, 2011 Transcript of Proceedings, Dkt. 110 at 7:19-20
(“you should err on the side of more disclosure and more answer, and not hide the ball”).
Google may have good cause to supplement its contentions again after the Court’s claim
construction is finalized, but we would like to provide this initial supplementation to Oracle as
soon as possible.
Please let us know if Oracle will agree to not oppose Google’s motion for leave to
supplement its invalidity contentions.

Sincerely,
/s/ Mark H. Francis
Mark H. Francis

GOOGLE’S MOTION FOR LEAVE TO SUPPLEMENT INVALIDITY CONTENTIONS
CIV. NO. CV 10-03561-WHA

Exhibit E

755 PAGE MILL ROAD
PALO ALTO
CALIFORNIA 94304-1018
TELEPHONE: [phone]
FACSIMILE: 650.494.0792
WWW.MOFO.COM

May 3, 2011

MORRISON & FOERSTER LLP
NEW YORK, SAN FRANCISCO,
LOS ANGELES, PALO ALTO,
SACRAMENTO, SAN DIEGO,
DENVER, NORTHERN VIRGINIA,
WASHINGTON, D.C.
TOKYO, LONDON, BRUSSELS,
BEIJING, SHANGHAI, HONG KONG

Writer’s Direct Contact
650.813.5932
[e-mail]

Via E-Mail [e-mail]
Mark H. Francis
King & Spalding LLP
1185 Avenue of the Americas
New York, NY 10036
Re:

Oracle America, Inc. v. Google Inc.
Case No. 10-03561-WHA

Dear Mark:
When we discussed a possible amendment to Google’s Invalidity Contentions four weeks
ago at the April 6 in-person meet-and-confer in the Court’s jury room, Oracle asked for an
identification of the new prior art references and an explanation of the grounds for good
cause to amend, such as an explanation of why the new art was not identified earlier. Your
letter of today leaves us in the dark. Oracle still needs that information to evaluate Google’s
request.
Sincerely yours,

Marc David Peters

pa-1462416

GOOGLE’S MOTION FOR LEAVE TO SUPPLEMENT INVALIDITY CONTENTIONS
CIV. NO. CV 10-03561-WHA

Exhibit F

1185 Avenue of the Americas
New York, New York 10036-4003
www.kslaw.com
Mark H. Francis
Direct Dial: [phone]
Direct Fax: [phone]
[e-mail]

May 6, 2011
VIA E-MAIL
Marc D. Peters, Esq.
Michael A. Jacobs, Esq.
Morrison & Foerster LLP
755 Page Mill Road
Palo Alto, CA 94304-1018
[e-mail]
Re:

Oracle America, Inc. v. Google Inc., No. 3:10-CV-03561-WHA (N.D. Cal.)

Dear Marc,
As a follow up to my May 2, 2011 letter to you, Google would like to immediately move
the Court for leave to supplement its Patent L.R. 3-3 Invalidity Contentions and has good cause
for doing so for at least the following reasons:
1. Oracle raised purported deficiencies with Google’s Invalidity Contentions during the
course of discovery. For example, on February 3, 2011, you sent Scott Weingaertner a
letter requesting additional disclosures pertaining to Google’s 35 U.S.C. §§ 101 and 112
arguments. Google believes that supplementation is warranted – and in fact directly
requested by Oracle – and that this additional disclosure would enable the parties to
address these invalidity issues more constructively as the case proceeds.
2. On February 25, 2011, Oracle responded to Google’s Interrogatory No. 11 with a 55-page
argument that the asserted patents are not invalid under 35 U.S.C. §§ 102 or 103 despite
the prior art referenced in Google’s Invalidity Contentions. In view of these allegations,
Google believes that supplementation is warranted and that additional disclosure would
enable the parties to address these invalidity issues more constructively as the case
proceeds.
3. Oracle served Google with its First Supplemental Infringement Contentions on February
18, 2011 and its Second Supplemental Infringement Contentions on April 1, 2011. While
Google’s investigation of the factual information and legal theories introduced in these

Marc D. Peters, Esq.
May 6, 2011
supplementations is ongoing, the supplemental Invalidity Contentions are intended to
address some of the issues raised in Oracle’s supplemental infringement contentions.
4. Google continued in good faith to investigate potential prior art and invalidity positions
after serving its Invalidity Contentions. In some instances, it discovered new prior art
when preparing the ex parte and inter parte re-examination requests which were recently
submitted to the patent office (copies of which were also provided to Oracle). In other
instances, prior art initially referenced in connection with one of the asserted patents in
the Invalidity Contentions was more recently determined to also be applicable to a
different one of the asserted patents.
5. Supplemental invalidity positions have been developed by Google since serving its initial
Invalidity Contentions, even with respect to prior art initially served upon Oracle, largely
based on Oracle’s interrogatory responses, Oracle’s representations to the Court in its
claim construction briefings, statements by Oracle’s counsel at the April 6 and April 20
hearings regarding the patented technology, and the Court’s tentative claim constructions.
The items below represent a summary of the additions and revisions prepared by Google,
along with a brief description of some reasons for the supplementations:


The supplemental Invalidity Contentions will provide additional disclosure
regarding Google’s invalidity positions under 35 U.S.C. § 101, addressing issues
specifically raised by Oracle in its February letter;



The supplemental Invalidity Contentions will provide additional disclosure
regarding Google’s invalidity positions under 35 U.S.C. § 112, addressing issues
specifically raised by Oracle in its February letter;



With respect to the ‘104 patent, the following charts were revised or added:
o Exhibit A-1 (D. Gries, Compiler Construction for Digital Computers, John
Wiley & Sons, Inc., 1971) was revised in view of Oracle interrogatory
responses (e.g., Google Interrogatory No. 11) and in view of the Court’s
tentative claim construction for intermediate form (object) code;
o Exhibit A-6 (Applicants Admitted Prior Art as admitted in U.S. Patent No.
RE 38,104) was added in view of Court’s tentative claim constructions and
statements that Oracle’s counsel made on the record regarding the terms
resolving and storing;
o Exhibit A-7 (S. Tafvelin, “Dynamic Microprogramming and External
Subroutine Calls in a Multics-Type Environment,” BIT 15 (1975) with R.C.
Daley & J.B. Dennis, “Virtual Memory, Processes, and Sharing in
MULTICS,” Communications of the ACM, Vol. 11, No. 5, 1968) is
supplemental art discovered after serving Google’s Invalidity Contentions
and particularly added in view of Oracle’s interrogatory responses (e.g.,
Google Interrogatory No. 11) with respect to what constitutes “executable”
code;
2

Marc D. Peters, Esq.
May 6, 2011
o Exhibit A-8 (B. Ramakrishna Rau, “Levels of Representation of Programs
and the Architecture of Universal Host Machines,” IEEE (1978)) is a
supplemental invalidity chart based on previously produced art and
particularly added in view of Oracle’s interrogatory responses (e.g., Google
Interrogatory No. 11) with respect to what constitutes “executable” code;
o Exhibit A-9 (Applicants Admitted Prior Art as admitted in U.S. Patent No.
RE 38,104, in view of B. Ramakrishna Rau, “Levels of Representation of
Programs and the Architecture for Universal Host Machines,” IEEE (1978))
is a supplemental invalidity chartbased on previously produced art and
particularly added in view of Oracle’s interrogatory responses (e.g., Google
Interrogatory No. 11) with respect to what constitutes “executable” code,
and also added in view of Court’s tentative claim constructions and
statements that Oracle’s counsel made on the record regarding resolving and
storing;
o Exhibit A-10 (S. Tafvelin, “Dynamic Microprogramming and External
Subroutine Calls in a Multics-Type Environment,” BIT 15 (1975), in view of
R.C. Daley & J.B. Dennis, “Virtual Memory, Processes, and Sharing in
MULTICS,” Communications of the ACM, Vol. 11, No. 5, 1968) is
supplemental art discovered after serving Google’s Invalidity Contentions
and particularly added in view of Oracle’s interrogatory responses (e.g.,
Google Interrogatory No. 11) with respect to what constitutes “executable”
code;
o Exhibit A-11 (Applicants Admitted Prior Art as admitted in U.S. Patent No.
RE 38,104, in view of Richard G. Bratt, U.S. Patent No. 4,525,780 (issued
Jun. 25, 1985)) is supplemental art discovered after serving Google’s
Invalidity Contentions and particularly added in view of Court’s tentative
claim constructions and statements that Oracle’s counsel made on the record
regarding the terms resolving and storing;
o Exhibit A-12 (Gries, Davidson, or Tafvelin, view of Rau or Daley) is
supplemental art discovered after serving Google’s Invalidity Contentions
and particularly added in view of Court’s tentative claim constructions and
statements that Oracle’s counsel made on the record regarding the terms
resolving and storing, it was also added in view of Oracle’s interrogatory
responses (e.g., Google Interrogatory No. 11) with respect to what
constitutes “executable” code; and
o Exhibit A-13 (U.S. Pat. No. 5,367,685, issued on 11/22/1994 to Gosling) is
a supplemental invalidity chart based partly on Court’s tentative claim
constructions and statements that Oracle’s counsel made on the record
regarding the terms resolving and storing.


With respect to the ‘720 patent, the following charts were developed as part of
Google’s preparation for filing an inter partes patent reexamination request and is
primarily based on newly discovered art and/or provides supplemental invalidity
charts based on previously produced art:

3

Marc D. Peters, Esq.
May 6, 2011
o Exhibit C-5 (U.S. Patent No. 6,823,509, issued on 11/23/2004 to Alan
Michael Webb & U.S. Patent Application Publication No. 2003/0088604,
published on 5/8/2003 naming Norbert Kuck et al. as inventors & M. J.
Bach, The Design of the Unix Operating System, Bell Telephone Labs., Inc.
(1986));
o Exhibit C-6 (U.S. Patent No. 6,854,114, issued on 2/8/2005 to Harlan
Sexton et al. & U.S. Patent No. 6,075,938, issued on 6/10/1998 to E.
Bugnion et al.);
o Exhibit C-7 (U.S. Patent No. 6,854,114, issued on 2/8/2005 to Harlan
Sexton et al. & U.S. Patent No. 6,330,709, issued on 12/11/2001 to M.
Johnson et al.);
o Exhibit C-8 (Sriram Srinivasan, Advanced Perl Programming, O’Reilly &
Associates, Inc. (1997) & M. J. Bach, The Design of the Unix Operating
System, Bell Telephone Labs., Inc. (1986));
o Exhibit C-9 (J. Dike, “A user-mode port of the Linux kernel”, Proceeding
ALS’00 Proceedings of the 4th annual Linux Showcase & Conference Volume 4, USENIX Association Berkeley, CA, USA (2000) & U.
Steinberg, “Fiasco μ-Kernel User-Mode Port” Dresden University of
Technology Institute of System Architecture (2002));
o Exhibit C-10 (U.S. Pat. No. 6,405,367, issued on 06/11/2002 to Bryant &
M. J. Bach, The Design of the Unix Operating System, Bell Telephone
Labs., Inc. (1986)); and
o Exhibit C-11 (U.S. Pat. No. 6,405,367, issued on 06/11/2002 to Bryant &
U.S. Patent Application Publication No. 2004/0010787, published on
1/15/2004 naming E. Traut et. al. as inventors).


With respect to the ‘520 patent, the following charts were revised or added:
o Exhibit D-3 (B.T. Lewis et al., “Clarity MCode: A Retargetable
Intermediate Representation for Compilation,” ACM, IR ‘95, 1/95, San
Francisco, California, USA, 1995) was supplemented to include
inadvertently omitted claims 14 and 17 (according to Oracle’s infringement
contentions, these claims are essentially duplicates of claims 10 and 2,
respectively, which were addressed in the original exhibit);
o Exhibit D-5 (B.T. Lewis et al., “Clarity MCode: A Retargetable
Intermediate Representation for Compilation,” ACM, IR ‘95, 1/95, San
Francisco, California, USA, 1995 & Gosling et al., The Java™ Language
Specification 1.0, Sun Microsystems, Inc. (1996) & The Java™ Virtual
Machine Specification, Sun Microsystems Computer Corp., Release 1.0
Beta DRAFT, (Aug. 21, 1995)) is a supplemental invalidity chart based on
previously produced art and particularly added in view of Oracle’s
interrogatory responses (e.g., Google Interrogatory No. 11); and
o Exhibit D-6 (B.T. Lewis et al., “Clarity MCode: A Retargetable
Intermediate Representation for Compilation,” ACM, IR ‘95, 1/95, San
Francisco, California, USA, 1995 & Dave Dyer, Java Decompilers
4

Marc D. Peters, Esq.
May 6, 2011
Compared, JavaWorld.com (July 1, 1997) & Proebsting et al., Toba: Java
for Applications A Way Ahead of Time (WAT) Compiler, Proceedings of
the Third USENIX Conference on Object-Oriented Technologies and
Systems (1997)) is supplemental prior art discovered after serving Google’s
Invalidity Contentions in the course of preparing a re-examination request.


With respect to the ‘205 patent, the following charts were added:
o Exhibit E-9 (Deutsch, Wakeling, Lewis, Yellin, Nilsen, or Hookway, in view
of Tarau or Magnusson) is a supplemental invalidity chart based on
previously produced art, largely clarifying some of the § 103 invalidity
combinations on which Google may rely, although all of the included
references and disclosures were already disclosed in § 102 charts provided
in Google’s initial Invalidity Contentions.

Supplemental invalidity positions may yet be developed as Google’s investigation
continues with respect to invalidity and additional prior art may yet be discovered. In addition,
Google expects that its invalidity positions may be further revised or supplemented in view of
the Court’s pending claim constructions and discovery from Oracle. Google will endeavor to
timely provide any further disclosures to Oracle as they develop.
As noted in my May 3rd letter to you, the Court has strongly encouraged the parties to
supplement their disclosures in view of any new information, or to address any alleged
insufficiencies identified by the receiving party. This is exactly what Google is aiming to
accomplish with these Supplemental Invalidity Contentions.
Please let us know if Oracle will agree to not oppose Google’s motion for leave to
supplement its Invalidity Contentions. Google would appreciate a timely response so we know if
this issue needs to be raised with the Court in an opposed or unopposed manner.
Sincerely,
/s/ Mark H. Francis
Mark H. Francis

5

GOOGLE’S MOTION FOR LEAVE TO SUPPLEMENT INVALIDITY CONTENTIONS
CIV. NO. CV 10-03561-WHA

Exhibit G

Mark Francis
From:
Sent:
To:
Cc:

Subject:
Attachments:

Francis, Mark
Monday, May 16, 2011 7:58 PM
Marc D. Peters
[e-mail]; Oracle MoFo Service List; Google-Oracle-OutsideCounsel;
[e-mail]; [e-mail];
[e-mail]
RE: Oracle America, Inc. v. Google Inc. - Correspondence to M. Peters
11 05 16 - GOOGLE First Supp Invalidity Contentions.pdf; Amd&SuppExhibits.zip

Marc,
Enclosed are Google's First Supplemental Invalidity Contentions, along with a zip file containing the amended and
supplemental exhibits.
Best Regards,
Mark
__________________________
Mark H. Francis
King & Spalding LLP
1185 Avenue of the Americas
New York, NY 10036
[phone]
[phone] (fax)
[e-mail]

From: Peters, Marc D. [mailto:[e-mail]]
Sent: Monday, May 16, 2011 1:56 AM
To: Francis, Mark
Cc: [e-mail]; Oracle MoFo Service List; Google-Oracle-OutsideCounsel
Subject: RE: Oracle America, Inc. v. Google Inc. - Correspondence to M. Peters
Dear Mark,
Michael discussed this issue with Scott Friday morning in connection with the case management
plan. Separate and apart from that, it would be helpful if you could send us any proposed charts
that you have already prepared and a summary of the contentions for the other grounds (e.g.,
101, 112). This would help us and our client evaluate Google's request.
Best regards,
Marc

1

GOOGLE’S MOTION FOR LEAVE TO SUPPLEMENT INVALIDITY CONTENTIONS
CIV. NO. CV 10-03561-WHA

Exhibit H

1185 Avenue of the Americas
New York, New York 10036-4003
www.kslaw.com
Mark H. Francis
Direct Dial: [phone]
Direct Fax: [phone]
[e-mail]

May 25, 2011
VIA E-MAIL
Marc D. Peters, Esq.
Morrison & Foerster LLP
755 Page Mill Road
Palo Alto, CA 94304-1018
[e-mail]
Re:

Oracle America, Inc. v. Google Inc., No. 3:10-CV-03561-WHA (N.D. Cal.)

Dear Marc,
During the meet-and-confer call on May 23, 2011, we again asked for Oracle’s consent to
file an unopposed motion for leave to supplement Google’s Invalidity Contentions. You agreed
that there was no prejudice to Oracle as a result of these supplemental contentions, but indicated
you would not provide us with Oracle’s position until after the Court ruled on certain case
management issues. The Court issued an Order on those case management issues that same day,
yet we have not heard from you regarding Google’s pending request.
Google has been requesting Oracle’s non-opposition to a motion for leave to supplement
since May 3, 2011 and Oracle has still not provided an answer. We have promptly responded to
your many requests, including a detailed description of Google’s good cause for supplementing
its contentions (May 6, 2011), a complete copy of the Supplemental Invalidity Contentions and
supporting exhibits (May 16, 2011), and even red-line versions illustrating all revisions from the
original Invalidity Contentions (May 17, 2011). Your continued delay in responding to this
request is disappointing because we expected Oracle to be cooperative on this straightforward
issue. Please let us know if Oracle will consent to an unopposed motion by no later than the next
meet-and-confer call, after which Google will file its motion – with or without consent.
Sincerely,

Mark H. Francis

GOOGLE’S MOTION FOR LEAVE TO SUPPLEMENT INVALIDITY CONTENTIONS
CIV. NO. CV 10-03561-WHA

Exhibit I

755 PAGE MILL ROAD
PALO ALTO
CALIFORNIA 94304-1018
TELEPHONE: [phone]
FACSIMILE: 650.494.0792
WWW.MOFO.COM

May 31, 2011

MORRISON & FOERSTER LLP
NEW YORK, SAN FRANCISCO,
LOS ANGELES, PALO ALTO,
SACRAMENTO, SAN DIEGO,
DENVER, NORTHERN VIRGINIA,
WASHINGTON, D.C.
TOKYO, LONDON, BRUSSELS,
BEIJING, SHANGHAI, HONG KONG

Writer’s Direct Contact
650.813.5932
[e-mail]

Via E-Mail [e-mail]
Mark H. Francis
King & Spalding LLP
1185 Avenue of the Americas
New York, NY 10036
Re:

Oracle America, Inc. v. Google Inc.
Case No. 10-03561-WHA

Dear Mark:
We have had the chance to review your letter of May 6 regarding Google’s desire to
supplement its invalidity contentions. As I told you last week, we had not focused much on
the issue of invalidity supplementation before then because we expected that the issue would
have been rendered moot through the process of submitting plans to the Court regarding
narrowing the issues for trial. As part of a compromise case plan, Oracle had offered to
accept any supplementation that Google wanted to make, but Google declined that
compromise, and here we are.
The trouble has with Google’s proposed supplementation is that we do not see good cause
for it, particularly since it so dramatically expands the case with new invalidity theories at a
time when we are supposed to be refining and focusing the case.
Google seeks to add nineteen new prior art charts, almost tripling the number applying to the
‘104 patent alone and increasing the total number from thirty-five to fifty-four. Google seeks
to add a host of new defenses, including section 101 and 112 defenses. Some of these
theories are not well-developed. For example, Google’s new best mode theories are made on
“information and belief.” Contentions are not supposed to be placeholders.
The supplemental invalidity theories that Google now desires to assert are ones that could
have been made before, which means there is no good cause to amend. All of Google’s new
non-prior-art-based theories are based on information (the patents and file histories) that
were in Google’s possession at the outset of the case. Almost all of the “new” art was in
Google’s possession when it prepared its January contentions, as demonstrated by the long
list of uncharted references in that document. Oracle’s interrogatory response pointing out

pa-1467540

Mark H. Francis
May 31, 2011
Page Two
the defects in Google’s invalidity theories does not provide “good cause” to amend either—
that just leads to the “shifting sands” approach that the patent local rules are supposed to
eliminate. We asked for additional information on certain defenses in February but, despite
Google’s apparent willingness to supply it then, when we asked for a date certain for that
supplementation, Google declined. The touchstone of good cause to amend is diligence, and
Google just hasn’t shown diligence in developing its new theories.
We don’t see the new invalidity theories as playing any important role in the case. For
example, Google now wants to contend that the ‘104 claims are anticipated by their own
specification (chart A-13). That’s certainly a novel argument, but one not likely to succeed
(not to mention being directly contradicted by Google’s desire to contend that that same
specification fails to provide an enabling written description of those same claims). Nor is
the new “failure to pay proper application fees” theory Google’s strongest one.
I write not to make sport of Google’s creativity, but to point out that Google’s proposed
supplementation is a “kitchen sink” document, containing every possible invalidity theory
that Google has thought of, regardless of its relative merit. Under the circumstances, given
the clear direction we both have received from the Court about narrowing the case to a
manageable size, Google’s proposal is a large step in the wrong direction. While Oracle can
agree to the corrections in the D-3 chart, as I’ve mentioned before, Oracle cannot agree at
this point to expand the case with a slew of new invalidity theories.
Sincerely yours,

Marc David Peters

pa-1467540

GOOGLE’S MOTION FOR LEAVE TO SUPPLEMENT INVALIDITY CONTENTIONS
CIV. NO. CV 10-03561-WHA

Exhibit J

LEVELS OF REPRESENTATION OF PROGRAMS AND THE ARCHITECTURE OF UNIVERSAL HOST MACHINES

B. Ramakrishna Rau
Coordinated Science Laboratory
University of Illinois, Urbana, IL 61801

The issue of high level language support is
treated in a systematic top-down manner. Program
representations are categorized into three classes
with respect to a host processor: high level
representations, directly interpretable representations and directly executable representations.
The space of intermediate languages for high level
language support is explored and it is shown that
whereas the ideal intermediate language from the
point of view of execution time is a directly
executable one, the best candidate from the viewpoint of memory requirements is a heavily encoded
directly interpretable representation. The concept of dynamic translation is advanced as a means
for achieving both goals simultaneously; the program is present in the memory in a compact static
representation, but its working set is maintained
in a dynamic representation which minimizes
execution time. The architecture and organization
of a universal host machine, incorporating this
strategy, is outlined and the potential performance gains due to dynamic translation are studied.

is drawn upon which lies the conventional machine
language. On one side of this line is the domain
of high level languages, compilers, interpreters
and main memory. On the other side lie the microprograms, nanoprograms, emulators and a host of
other micro- and nano- entities. This viewphint
arises, in part, from the use of microprogrammable
machines predominantly for the purpose of emulating the instruction sets of otter machines. This
classical concept of microprogramming tends to
obfuscate the issue which may be phrased as follows: given a certain (open ended) set of high
level languages, what is the nature of the host
hardware that is best suited to supporting them
and what is the process by which programs, written
in these high level languages, are supported? A
fresh perspective can be valuable; the host
machine should be viewed as a special purpose
machine designed to provide high level language
support. The architecture and organization should
evolve as the outcome of a top-down design process
rather than as a carry-over from the classical
view of microprogramming. This is the objective
of this paper and so as to avoid any preconceived
notions, the terminology of microprogramming is
avoided as far as is possible.

Microprogramming was originally conceived by
Wilkes as a systematic means of implementing the
control structure of a computerl. The microprogram, embedded in a read-only memory, interprets the instruction set visible to the
programmer. In view of the permanence of the
microprogram and its transparency to the user, the
interpreted instruction set was, reasonable
enough, thought of as representing the architecture of the machine. Accordingly, the emphasis
was on the interpreted instruction set.

The architecture and instruction set of a
host is determined by the class of languages that
are to be supported by it. If this class is
restricted and consists of similar languages, the
application of the host is fairly specific and
the instruction set will contain powerful instructions which closely resemble the semantics of the
high-level languages that are supported by the
host. Several examples of high-level machine
designs fall into this categoryS-10. On the other
hand, if the class of languages is large and
vague, commonality of semantics will exist only
at a very low level and the instruction set of the
host machine will be primitive. This provides
generality and flexibility. A host of this type
is termed a universal host machine (UHM). A
number of examples of UHM's are available2,11-14

Summary

With the advent of writeable control store,
the situation has changed and, yet, the perspective has remained much the same. Writeable
control store is viewed as a means of providing
a "soft architecture," i.e., one that can be
changed dynamically to match the needs of the
moment which might, for instance, entail the support of a high level language. Experience with
the Burroughs Bl7002,3 and the work of Hoevel4
has demonstrated the effectiveness of such a
strategy. However, the emphasis still is on the
interpreted instruction set. An artificial line

Given a host architecture and a high level
language, one could either interpret the latter
directly, compile it into the machine language or
compile it into an intermediate language which is
then interpreted. Hoevel derives conditions
under which the last alternative is superior to
the other two 15
These conditions are generally
satisfied for the types of universal host architectures that exist or are under consideration.

67
CIU411-8/73/0000-0067$00. 75

Oracle America v. Google, 3:10-cv-03561-WHA

©

1978 IEEE

GOOGLE-00338720

existence of cache memories28-30 and virtual
memories,31,32 The fraction of the address space
that is currently being referenced heavily is
termed the working set.27 The function of the
dynamic translator is to maintain in the dynamic
translation buffer ~) a representation of the
instruction working set that is more tightly bound
than the static representation. If the size of
the DTB is reasonably large and if the contents
of the DTB are selected carefully, it is possible
to ensure that a large fraction of all instructions
executed will be present in the DTB. This fraction
is termed the hit ratio. When the hit ratio is
close to unity, most instructions when executed
will be found in the more tightly bound representation. The time penalty associated with binding
will be experienced only rarely and will not be a
major factor in determining the execution time.
If, at the same time, the size of the DTB is small
in comparison to the size of the loosely bound
representation, the memory requirements will not
have been increased substantially and the conflicting requirements of a compact representation
and low execution time will be met simultaneously.

interpreter must complete whatever binding remains.
However, this binding persists only over the
period of execution of an instruction and must be
repeated each time that instruction is executed.
From the point of view of persistence of binding,
the compiler and interpreter are at opposite
extremes. We introduce the notion of a dynamic
translator, the persistence of whose binding lies
in between that of the compiler and the interpreter. Once the dynamic translator binds an
instruction (totally or partially), it remains
bound over a period of time that spans a certain
number of successive executions of that instruction. Such a strategy assumes, of course, that
the program is not self-modifying -- an assumption
that is valid when programs are written in highlevel languages.
One could conceive of a hierarchy of representations each with a different level of binding
and degree of persistence: the source program
which exists until destroyed, the DIR which lasts
until the source is modified, the link-edited
version which exists for one execution of the
program, possibly a number of lower levels, each
increasingly bound and persisting for decreasing
fractions of the program execution period and,
finally, a completely bound representation of an
instruction which only lasts for the duration of
that instruction's execution.

The concept of a DTB is related to that of the
dynamic address translation mechanism provided
with virtual memories. When addressing a virtual
memory, the virtual address must be bound to a
physical address. This involves indirection
through one or more segment and page tables on
each memory reference. This overhead is reduced by
retaining in an associative array the mapping
between the virtual and physical addresses for the
pages which have been referenced most recently.
The DTB may be viewed as a caGhe on a virtual
memory in which the program is stored in the more
tightly bound representation.

The significance of the dynamic translator is
that it raises the possibility of simultaneously
achieving high speed interpretation and a compact
static intermediate representation. Since the
binding performed by a dynamic translator persists
over a number of executions of an instruction,
the time spent in binding is spread out over those
instructions, thereby reducing the average time
spent in binding per instruction executed. It is
possible then to use a highly encoded DIR without
increasing the interpretation time by very much
if the binding is made to persist over a sufficient number of successive executions of the same
instruction. This persistence of binding is
effected by saving the bound representation of the
instruction which will be less compact than the
encoded DIR version. Attempting to retain this
bound version for extended periods of time for a
number of different instructions will entail the
use of large amounts of memory. In fact, if
the bound version were never discarded, one would
soon obtain and have to provide storage for a
translated version of the entire program, thereby
defeating the purpose of using an encoded DIR.

When the dissimilarities between the representations corresponding to minimum execution
time and minimum storage requirements, respectively,
are great, it is possible that a number of levels
of dynamic translation will be required. However,
in the rest of this paper, we shall concern ourselves with only one level of dynamic translation.
Typically, three different representations are of
interest: the HLR in which the program is written,
the ~ (intermediate) representation into
which it is compiled and the dynamic representation
which is obtained by dynamically translating the
static representation of the working set, Of
these, only the latter two will be in the directly
addressable memory during execution.
The use of dynamic translation permits the
decoupling of the design decisions involved in
selecting the intermediate representation. The
static representation may be selected solely to
minimize the size of the program. Ideally, it
should be a high level, highly encoded DIR. The
dynamic representation, on the other hand, should
be selected to speed up execution and should,
ideally, be a high level P-DER.

The effectiveness of the dynamic translator
hinges on the ability to save the bound representation for just a short period of time which,
nevertheless, spans a large number of executions
of the instruction. The existence of loops and
recursive calls implies this ability, In fact,
the more general '~rinciple of locality" states
that over any interval of time, the vast majority
of memory references are concentrated on a small
subset of the address space. This princi~le has
been empirically validated over and again 6-28
and is the fundamental justification for the

71

Oracle America v. Google, 3:10-cv-03561-WHA

GOOGLE-00338724

GOOGLE’S MOTION FOR LEAVE TO SUPPLEMENT INVALIDITY CONTENTIONS
CIV. NO. CV 10-03561-WHA

Exhibit K

1

•

2
3

MIDP 1.0 and subsequent versions.
2.

The ’205 Patent

The following instrumentalities of Oracle practice the asserted claims of the ’205 patent:

4

•

JDK 1.2 and subsequent versions;

5

•

JRE 1.2 and subsequent versions;

6

•

HotSpot 1.0 and subsequent versions;

7

•

Java SE for Embedded 1.4.2 and subsequent versions;

8

•

CDC RI 1.0.1 and CDC-HI 1.0 and subsequent versions of each;

9

•

CDC AMS 1.0, 1.0_1, 1.0_2, Personal Basis and Personal Profile versions;

10

•

CLDC RI 1.1.1;

11

•

CLDC-HI 1.0 and subsequent versions;

12

•

Foundation Profile 1.0.2 and subsequent versions;

13

•

J2EE 1.2 (later called Java EE) and subsequent versions;

14

•

Java ME SDK 3.0 EA and subsequent versions;

15

•

Java Real-Time System 1.0 and all subsequent versions;

16

•

Personal Profile HI and RI 1.0 and subsequent versions; and

17

•

Personal Basis Profile HI and RI 1.0 and subsequent versions.

18
19

3.

The ’702 Patent

The following instrumentalities of Oracle practice the asserted claims of the ’702 patent:

20

•

PersonalJava (“PJava”) 1.0 and subsequent versions;

21

•

EmbeddedJava (“EJava”) 1.0 and subsequent versions;

22

•

JavaOS 1.0 (and all variants, including Java PC) and subsequent versions;

23

•

CDC RI 1.0 and CDC-HI 1.0, and all subsequent versions of each;

24

•

CDC AMS 1.0, 1.0_1, 1.0_2, Personal Basis and Personal Profile versions;

25

•

CLDC RI 1.1.1 and CLDC-HI 1.0.1, and all subsequent versions of each;

26

•

Personal Profile HI and RI 1.0 and subsequent versions;

27

•

Personal Basis Profile HI and RI 1.0 and subsequent versions;

28

•

Foundation Profile 1.0 and subsequent versions; and

ORACLE’S SECOND SUPPLEMENTAL INFRINGEMENT CONTENTIONS
CASE NO. 3:10-CV-03561-WHA
pa-1456177

10

GOOGLE’S MOTION FOR LEAVE TO SUPPLEMENT INVALIDITY CONTENTIONS
CIV. NO. CV 10-03561-WHA

Exhibit L

Companies endorse JavaOS in desktop, consumer and embedded environments
Industry leaders to provide development tools, applications

San Francisco, CA - May 29, 1996 - JavaSoft, an operating company of Sun Microsystems, Inc. (NASDAQ:SUNW) today
announced "JavaOS"(TM), a highly compact operating system designed to run Java applications directly on microprocessors
in anything from net computers to pagers.
In related news, several industry leaders today announced that they intend to license JavaOS. Additionally, several leading
software companies announced their intention to provide development tools for JavaOS.
A dynamically extensible operating environment, JavaOS brings the design advantages of the Java(TM) programming
language to an operating system. As perhaps the smallest and fastest OS that runs Java, JavaOS enables Java on a broad
range of devices. JavaOS will run equally well on a network computer, a PDA, a printer, a game machine, cellular telephone,
or countless other devices that require a very compact OS and the ability to run Java.
"JavaOS is elegantly simple and extremely powerful at the same time. It was designed with a single purpose -- to be just
enough OS of just the right kind to run the Java Virtual Machine(TM), which brings Java to a huge new range of electronic
appliances," said Jim Mitchell, CTO, JavaSoft. "No other software platform has the reach that JavaOS provides for Java."

Industry endorses JavaOS
To date, Acer Inc., Acer Peripherals Inc., Alcatel Business Systems, Axil Computers, ETEN Information Systems Co.,
Hua-Hsing Information Corp., Hyundai Electronics, Taiwan's Institute for Information Industry (III), Taiwan's Industrial
Technology Research Institute (ITRI), Lite-ON Technology Corp., LG Electronics, Mitac Inc., Mitsubishi Electric Corp.,
Nestor Technology, Nokia, Omron Corporation, Oracle, Proton, Sun Microsystems Computer Company, Sun Moon Star,
SunRiver Data Systems, Tatung Company, THOMSON-Sun Interactive Alliance, THOMSON multimedia S.A., Toshiba
Corporation, UMAX, Visionetics Internationalities Technology and Xerox have stated their intent to license JavaOS.
Borland International, Corel Corporation, Dun & Bradstreet Software, Hugh Symon Group, Justsystems Corp., Metrowerks,
SunSoft, Inc. and Symantec Corp., endorse JavaOS and intend to build tools or applications for the platform.
ARM Ltd., Cirrus Logic, Fujitsu Microelectronics Inc., LSI Logic, National Semiconductor, and Sun Microelectronics are
among the companies that intend to implement JavaOS on their microprocessors. To date, JavaSoft expects JavaOS to be run
on a broad variety of microprocessors, including ARM, CompactRISC, Intel X86, NS486, PowerPC, microJAVA,
microSPARC, picoJAVA, SPARClite and others.
"We're having in depth discussions with companies who want to use JavaOS to create products that really push the
envelope," said Mitchell. "Over the next twelve months we anticipate a broad range of product announcements based on
JavaOS, including Internet and Intranet devices that will set a new standard for ease of installation and use in a networked
environment."

Development Tools for JavaOS
JavaSoft is collaborating with leading tools developers to define open APIs, such as the JavaOS Debugging API, that they'll
use to ensure their tools work seamlessly with JavaOS. Borland, Metrowerks, SunSoft and Symantec intend to adapt and

enhance their Java development environments for JavaOS, bringing real choice of development tools to the embedded
systems arena.
"We're very pleased to have such a prominent group of tools developers on board to support JavaOS," said Mitchell. "We're
leveraging the expertise and volume market presence each of these companies has already developed in creating strong tools
for the basic Java platform, and expect them to be able to deliver effective tools for JavaOS very rapidly."

JavaOS runs on industry microprocessors
In addition, JavaOS has been built to be fully ROMable for embedded applications, and can run with as little as 512K ROM
and 256K RAM. For network computers, an entire system with JavaOS, the HotJava(TM) Browser and space for
downloading Web content and applets requires only 3MB ROM and 4MB RAM. JavaOS can be this small because it is
almost completely written in Java.
JavaSoft, headquartered in Cupertino, CA, is an operating company of Sun Microsystems Inc. The company's mission is to
develop, market and support the Java technology and products based on it. Java supports networked applications and
enables developers to write applications once that will run on any machine. JavaSoft develops applications, tools and
systems platforms to further enhance Java as the programming standard for complex networks such as the Internet and
corporate intranets.
JavaOne, the first JavaSoft-sponsored developers conference for Java, will take place May 29-31, 1996 at San Francisco's
Moscone Center. JavaOne's keynote addresses will be webcast at http://java.sun.com/javaone, and all conference session
materials will be available at the same url.
With annual revenues of more than $6 billion, Sun Microsystems, Inc. provides products and services that enable customers
to build and maintain open network computing environments. Widely recognized as a proponent of open standards, the
company is involved in the design, manufacture and sale of products, technologies and services for commercial and technical
computing. Sun's SPARC(TM) workstations, multiprocessing servers, SPARC microprocessors, SolarisTM operating software
and ISO-certified service organization each rank No. 1 in the UNIX® industry. Sun's Java(TM) platform-independent
programming environment, provides a comprehensive solution to the challenge of programming for complex networks,
including the Internet. Sun Microsystems was founded in 1982, and is headquartered in Mountain View, California.

Sun, the Sun logo, Sun Microsystems, The Network is the Computer, Solaris, Java, HotJava, JavaSoft are trademarks or registered trademarks of Sun Microsystems, Inc. in the
United States and in other countries. All SPARC trademarks are used under license and are trademarks or registered trademarks of SPARC International, INC. in the United States
and other countries. Products bearing SPARC trademarks are based upon an architecture developed by Sun Microsystems, Inc. UNIX is a registered trademark in the United States
and in other countries exclusively licensed through X/Open Company, Ltd. All other product or service names mentioned herein are trademarks of their respective owners.

PR Contacts for Press and Analysts:
JavaSoft
Lisa Poulson
[phone]
http://java.sun.com

Questions or comments regarding this service? [e-mail]
Copyright 1996 Sun Microsystems, Inc., 2550 Garcia Ave., Mtn. View, CA 94043-110 USA. All rights reserved.

GOOGLE’S MOTION FOR LEAVE TO SUPPLEMENT INVALIDITY CONTENTIONS
CIV. NO. CV 10-03561-WHA

Exhibit M

755 PAGE MILL ROAD
PALO ALTO
CALIFORNIA 94304-1018

MORRISON & FOERSTER LLP

TELEPHONE: [phone]
FACSIMILE: 650.494.0792

NORTHERN VIRGINIA, DENVER,
SACRAMENTO

WWW.MOFO.COM

July 6, 2011

NEW YORK, SAN FRANCISCO,
LOS ANGELES, PALO ALTO,
SAN DIEGO, WASHINGTON, D.C.

TOKYO, LONDON, BRUSSELS,
BEIJING, SHANGHAI, HONG KONG

Writer’s Direct Contact
650.813.5932
[e-mail]

HIGHLY CONFIDENTIAL – SOURCE CODE
Via E-Mail [e-mail]
Scott T. Weingaertner
King & Spalding LLP
1185 Avenue of the Americas
New York, NY 10036-4003
Re:

Oracle America, Inc. v. Google Inc., Case No. 10-03561-WHA

Dear Scott:
In our preparations for trial, we have learned that Oracle’s identification of JavaOS 1.0 as an
embodiment of the ’702 patent in Oracle’s infringement contentions pursuant to Patent L.R.
3-1(g) was in error. The correct identification is JavaOS 1.1 and subsequent versions.
Patent L.R. 3-1(g) provides: “If a party claiming patent infringement wishes to preserve the
right to rely, for any purpose, on the assertion that its own apparatus, product, device,
process, method, act, or other instrumentality practices the claimed invention, the party shall
identify, separately for each asserted claim, each such apparatus, product, device, process,
method, act, or other instrumentality that incorporates or reflects that particular claim.”
Because JavaOS 1.0 was not an embodiment of the ’702 patent, Oracle will not assert that it
was in this litigation and need not preserve the right to do so.

pa-1473237

Scott T. Weingaertner
July 6, 2011
Page Two
The error has not affected the case in any significant way. JavaOS was not mentioned in
Google’s January 18, 2011 invalidity contentions, nor was it mentioned or charted in
Google’s proposed supplemental invalidity contentions.
I am available if you wish to discuss further.
Sincerely,

Marc David Peters

pa-1473237

GOOGLE’S MOTION FOR LEAVE TO SUPPLEMENT INVALIDITY CONTENTIONS
CIV. NO. CV 10-03561-WHA

Exhibit N

1

the requisite disclosure to prepare its own disclosures. If Oracle seeks and is granted permission

2

to amend its Infringement Contentions, then Google specifically reserves the right to modify,

3 amend, or supplement these Invalidity Contentions for that reason. Google also specifically
4 reserves its right to modify, amend, or supplement these Invalidity Contentions for other reasons
5 as provided by any applicable Rule or Order, including the Patent Local Rules.
6

Google believes that the references discussed in Section I.A each independently

7 anticipate the Asserted Claim under 35 U.S.C. § 102 and/or alone render obvious the Asserted
8 Claim under 35 U.S.C. § 103. To the extent that any one of the previously discussed anticipatory
9 references is found not to anticipate one or more Asserted Claims, various references described
10 herein and/or the knowledge and skill of a person of ordinary skill in the art at the time of the
11 invention, in combination with any one of the anticipatory references, render the claim invalid as
12 obvious. Google’s contentions that the references in this section, in various combinations, render
13 the Asserted Claims obvious under 35 U.S.C. § 103 are in no way an admission or suggestion
14 that each reference does not independently anticipate a particular claim under 35 U.S.C. § 102.
15
16
17

Table 2. List of 35 U.S.C. § 103 Prior Art Combinations
Chart

Prior Art Reference

A-5

J.W. Davidson, “Cint: A RISC Interpreter for the
C Programming Language,” SIGPLAN ‘87 Papers
of the Symposium on Interpreters and Interpretive
Techniques, 1987,
in view of
AT&T, System V Application Binary Interface
Motorola 68000 Processor Family Supplement,
Prentice Hall Int’l, 1990.

18
19
20
21
22

‘104 Patent, claims 11-41.

One of ordinary skill in the art at the time would
have had motivation to combine these references
at least because they both relate to the System V
operating system.

23
24
25

Claims Rendered Obvious

27

Palay, U.S. Patent No.
with
James Gosling, U.S. Patent No. 5,367,685 (issued
Nov. 22, 1994).

28

One of ordinary skill in the art at the time would

26

B-1

‘702 Patent claim 16.

9
DEFENDANT GOOGLE INC.’S INVALIDITY CONTENTIONS
CIVIL ACTION NO. CV 10-03561-WHA

GOOGLE’S MOTION FOR LEAVE TO SUPPLEMENT INVALIDITY CONTENTIONS
CIV. NO. CV 10-03561-WHA

Exhibit O

ttJ~
PD···········J
.
p.

XP-0021 00907

@

7-..... 0

'o..:?. ~J.'.~··· =

Going Beyond the Sandbox: An Overview of the New Security
Architecture in the Java™ Development K it 1.2
Li Gong, :'vfarianne :'vlueller, Hemma Prafullchandra, and Roland Schemers
JavaSoft, Sun Microsystems, Inc.
('gong,mrm,hemma,schemers }@eng.sun.com

Abstract

1.1

This pap<r descrabes the new security architectur< that h.s ban implemented. as part of 1011"1.2,
· lht forthcoming }at·an Development Kit. In 90ing
btyond tht sandbar security modtl in tht ~riginal
r~least of Java, JD/\'1 . 2 provtdt5 fint ·graintd acctss con trol via an easily configurablt security pol.
icy. t,(.,reover, JDKJ .!! introduces the concept of
protection domain and a few related security pramtlives tha t help to makt the underlying protection
mechanism mort robust.

The Original Security :V!odel

The original security model provided by Ja,·a is
known as the sandbox model, wh ich exists in order
to provide a very restricted environment ·in whir.h
to run untrusted code (called applet) obtained from
t he open network. The esse.nce of the sandbox
model, as illustrated by Figure 1, is that loc3.1 code is
trusted to have full access to vital system resources
(such as the file system) while downloaded remote
code is not trusted and can access only the limited
resources prov ided inside the sandbox.

Introduction

JDK 1.0.2 Security Mod~!

Since the inception of Ja-·a (8, I q, there has been
st rong and growing interest around the s-.r.u rity of
Ja,·a as well ~ new secu rity issu-.s rai.-d by the
d eployment of Java. from a tec hnolog y provider's
point o)f view, Java securitr includes two asper.ls [61:
• Provide Java (primarily through JDK) as a se·
cu re, ready-built platform on· which to run Java
enabled applications in a secu re fashion.
• Provide .-security tools and services il]lple·
men ted in Java that enable a wider rfoge of
security-sensiti,·e applications, for example, in
t he enterprise world.
This paper focuses on issues related to the first
aspect, where the customers for such technologies
include vendors that bundle or embed Java in their
products (such as browsers and operating systems).
It is worth emphasizing that this work by itself
does not claim to break significant new gtound in
terms of the theory of computu security. Instead,
it offers a real world example where well-known se·
curity principles [5, 12, 13, 161 are put into engineering practice to construct a practical and widely
dep loyed secure system.

JSENIX Association

~

local

code
JVM

valuable tesources

figure I: J OK I.O .x Security '.lode I
This sandbox model is deployed through the J ava
Development Toolkit in versions l.O.x, and is generally adopted by applications built with JDK, in·
eluding Java-enabled web browsers.
Overall security is enforced through a number of
mechanisms. First of all, the language is designed
tq be type-safe, and easy to use. The hope is that
the burden on the progtammer is such that it is less
likely to make subtle mistakes, compared with using
other programming languages such as C or C++.
Language features such as automatic memory tl)an·

USENIX Symposium on Internet Technologies and Systems ·:December 8·11, 1997

103

... ..-...... .............

SNSOOCtO: <.XP 2100907A,.

··- ·-

.:

~

~.-·

.

Oracle America v. Google, 3 :10-cv-03561-WHA

GOOGLE-00313922