Ex Parte 7584301 et al

Patent Trial and Appeal Board

Dec 23, 2013

90011765 (P.T.A.B. Dec. 23, 2013)

UNITED STATES PATENT AND TRADEMARK OFFICE
UNITED STATES DEPARTMENT OF COMMERCE
United States Patent and Trademark Office
Address: COMMISSIONER FOR PATENTS
P.O. Box 1450
Alexandria, Virginia 22313-1450
www.uspto.gov
APPLICATION NO. FILING DATE FIRST NAMED INVENTOR ATTORNEY DOCKET NO. CONFIRMATION NO.
90/011,765 06/27/2011 7584301 11517.0018 2935
87916 7590 12/23/2013
2nd Reexam Group - Novak Druce + Quigg LLP
1000 Louisiana Street
Fifty-Third Floor
Houston, TX 77002
EXAMINER
DESAI, RACHNA SINGH
ART UNIT PAPER NUMBER
3992
MAIL DATE DELIVERY MODE
12/23/2013 PAPER
Please find below and/or attached an Office communication concerning this application or proceeding.
The time period for reply, if any, is set in the attached communication.
PTOL-90A (Rev. 04/07)

UNITED STATES PATENT AND TRADEMARK OFFICE
_____________

BEFORE THE PATENT TRIAL AND APPEAL BOARD
______________

Ex parte BROCADE COMMUNICATIONS SYSTEMS, INC.
______________

Appeal 2013-008244
Reexamination Control No. 90/011,765
Patent 7,584,301 B1
1

Technology Center 3900
______________

Before JOHN C. MARTIN, HOWARD B. BLANKENSHIP,
and MAHSHID D. SAADAT, Administrative Patent Judges.

MARTIN, Administrative Patent Judge.

DECISION ON APPEAL
This is an appeal under 35 U.S.C. § 134(b) from the Examiner’s
rejection of original claims 1, 2, 9, 15, 24, and 28 of U.S. Patent
7,584,301 B1 (hereinafter “’301 patent”), which are all of the claims subject
to reexamination in this ex parte reexamination proceeding. Final Office

1
Issued September 1, 2009, based on Application 10/839,919, filed May 6,
2004, and modified by Certificates of Correction dated September 14, 2010,
and May 15, 2012.
Appeal 2013-008244
Reexamination Control 90/011,765
Patent 7,584,301 B2

2
Action mailed May 16, 2012 (hereinafter “Final Action”) at 1, Part II, ll. 1a
& 4. We have jurisdiction under 35 U.S.C. §§ 6(b) and 306.
We AFFIRM.

I. STATEMENT OF THE CASE
A. This Ex Parte Reexamination Proceeding
This ex parte reexamination proceeding was initiated by a
“REQUEST FOR EX PARTE REEXAMINATION” (hereinafter “Request”),
filed on June 27, 2011, by Third-Party Requester A10 Networks, Inc.
(hereinafter “A10 Networks”).
The sole reference relied on in the rejection on appeal is “3-DNS
Reference Guide, Version 4.2” (hereinafter “Reference Guide”).
2

B. Related Litigation and Reexamination Proceedings
The ’301 patent is the subject of litigation styled: Brocade
Communications Systems, Inc. v. A10 Networks, Inc., Case No. 10-CV-
03428-LHK (N.D. Cal.). PATENT OWNER’S APPEAL BRIEF
PURSUANT TO 37 C.F.R. § 41.37 (hereinafter “Brief”) at 4.
Appellant has identified twelve other ex parte reexamination
proceedings and thirteen inter partes reexamination proceedings that “may

2
Identified as follows in the Information Disclosure Statement (IDS) that
accompanied the Request: “3-DNS Ref. Guide, version 4.2, July 1, 2002
(261 pages).”
Appeal 2013-008244
Reexamination Control 90/011,765
Patent 7,584,301 B2

3
be related to, directly affect, or be directly affected by, or have a bearing on
the Board’s decision in the pending appeal.” Id. at 2-3. The only one of
these related proceedings that involves the ’301 patent is Inter Partes
Proceeding 95/001,812 (hereinafter “’812 inter partes proceeding”), which
was initiated by a REQUEST FOR INTER PARTES REEXAMINATION
filed by A10 Networks on November 8, 2011. A Right of Appeal Notice
(RAN), mailed in the ’812 inter partes proceeding on December 19, 2012,
includes (at 5) a rejection of original claims 3-8, 10-14, 16-23, 25-27, and
29-31 for anticipation by Reference Guide. A petition
3
by Patent Owner to
terminate the ’812 inter partes proceeding as a result of a final decision in
the above-identified litigation was dismissed in a USPTO decision mailed
August 30, 2013.
4

B. The Invention Described in the ’301 Patent
The invention described in the ’301 patent relates generally to load
balancing among servers. ’301 patent 1:6-7.
Figure 1 of the ’301 patent is reproduced below.

3
PATENT OWNER’S PETITION UNDER 37 C.F.R. § 1.182 TO
TERMINATE THE REEXAMINATION PROCEEDINGS.
4
DECISION DISMISSING PETITION TO TERMINATE INTER PARTES
REEXAMINATION PROCEEDING.
Appeal 2013-008244
Reexamination Control 90/011,765
Patent 7,584,301 B2

4

Figure 1 illustrates an example of a GSLB (global server load balancing)
system that can employ an embodiment of the invention. Id. at 4:12-13. A
GSLB switch 12 connected to Internet 14 acts as a proxy to an authoritative
Domain Name System (DNS) server 16 for a domain (e.g.,
“foundrynet.com”). Id. at 4:13-17. GSLB switch 12 communicates via the
Appeal 2013-008244
Reexamination Control 90/011,765
Patent 7,584,301 B2

5
Internet 14 and routers 19 and 21 with site switches 18A and 18B at site 20,
site switches 22A and 22B at site 24, and any other similarly configured site
switches. Id. at 4:24-29. Site switches 18A, 18B, 22A and 22B connect
routers 19 and 21 to servers 26A to 26N. Id. at 4:27-29.
A client program 28 communicates via Internet 14 with a local DNS
server 30. Id. at 4:48-50. When a browser on the client program 28 requests
a web page, a query is sent to local DNS server 30 to resolve the symbolic
host name (e.g., www.foundrynet.com) to an IP address of a host server. Id.
at 4:50-55. A list of IP addresses corresponding to the resolved host name is
either retrieved from local DNS server 30’s cache or obtained from the
GSLB switch 12. Id. at 4:55-60.
A list of IP addresses is obtained GSLB switch 12, which acts as a
proxy to DNS server 16 for the domain www.foundrynet.com, as follows.
After forwarding the client query to DNS server 16, GSLB switch 12
evaluates the list of IP addresses identified by DNS server 16 in the manner
described under the heading “BACKGROUND INFORMATION”:
When the DNS server sends the list of IP addresses in response
to a client query, the GSLB switch evaluates this list by
applying a GSLB policy in conjunction with using the
information provided by the site switches. The GSLB policy
contains, among other possible specifications, the metrics that
are to be applied to the list of IP addresses and the order in
which they are to be applied to select the best IP address. After
the GSLB switch evaluates the addresses returned by the DNS
server using the metrics in the policy, the GSLB switch sends
Appeal 2013-008244
Reexamination Control 90/011,765
Patent 7,584,301 B2

6
the ordered address list, having the optimum address for access
listed at the top, to the client.
Id. at 1:45-62.
The improved GSLB system described in the ’301 patent addresses
the following problem encountered when using the prior art GSLB system:
Currently, the user (such as a system administrator) can
only define such a GSLB policy (or other load balancing
policies) globally. This global GSLB policy is applicable to all
the domains for which the GSLB switch is providing GSLB.
For instance, consider the example where the GSLB switch is
providing GSLB for www.foo.com and www.test.com. (Also
note in this example that for the domain www.foo.com, “www”
is referred to as the “host” and “foo.com” is referred to as the
“zone”). The user may define a GSLB policy with the
following example metrics and metric-order:
Health check
Geographic
Least-response
This global policy would apply to both www.foo.com and
www.test.com.
If the user wants a round-trip time (RTT) metric to be
used for selection of the best IP address for the domain
www.foo.com but not for the domain www.test.com, then there
is currently no provision for this capability.
Id. at 2:8-27.
The invention described in the ’301 patent solves this problem by
permitting “[u]sers [to] define a host-level policy (alternatively or
additionally to the global GSLB policy) and apply the host-level policy to
Appeal 2013-008244
Reexamination Control 90/011,765
Patent 7,584,301 B2

7
hosts in GSLB domains.” Id. at 3:18-21. Thus, “[i]n effect the user can
enable different policies for different hosts.” Id. at 3:21-22.
Figure 2 of the ’301 patent is reproduced below.

Figure 2 is a block diagram illustrating a GSLB switch in accordance with
the invention. Id. at 2:50-51. GSLB switch 12 includes a GSLB switch
controller 201 that is communicatively coupled to various modules,
including first and second storage units 212 and 210. Id. at 6:29-32; 7:3-16.
First storage unit 212 contains one or more configured GSLB host-level
policies. Id. at 7:3-6. These host-level policies can in turn specify, for each
host, the metrics to be used for selection of IP addresses, the order in which
the metrics are to be applied, the parameters for the metrics (such as
tolerance, limits, and the like), or other associated settings. Id. at 7:6-10.
Appeal 2013-008244
Reexamination Control 90/011,765
Patent 7,584,301 B2

8
Second storage unit 210 contains one or more global GSLB policies, some
of which may be default policies in some embodiments. Id. at 7:10-13.
Figure 3 of the ’301 patent is reproduced below.

Figure 3 is a flowchart 300 of an embodiment of a technique to configure
(i.e., create) or modify a host-level policy on GSLB switch 12. Id. at 7:25-
27. Beginning at block 302, the user “specifies” a host-level policy, such as
P1 (id. at 7:38-39), which we understand to mean the user enters the policy
Appeal 2013-008244
Reexamination Control 90/011,765
Patent 7,584,301 B2

9
identifier P1. If policy P1 does not already exist (as determined in blocks
304, 306), the user can create a new policy P1 (block 308). Id. at 7:39-45.
When a new policy P1 is created, the user specifies its metrics and metric
order (block 310), after which policy name P1, its metrics, and metric order
are stored in the host-policy database (block 312). Id. at 7:39-49; 9:34-36.
If, on the other hand, policy P1 already exists (as determined at blocks 304,
306), it can be modified by the user (block 314), after which the policy
stored in the database is accordingly updated (block 316). Id. at 9:37-44.
Some of Appellant’s arguments for patentability are based on Figure
4, reproduced below, which shows how a load balancing policy is associated
with a domain.
Appeal 2013-008244
Reexamination Control 90/011,765
Patent 7,584,301 B2

10

Figure 4 is a flowchart 400 depicting “application of a host-level policy to a
host according to one embodiment.” Id. at 9:45-46. At block 402, the user
“specifies” the host (e.g., H1) to which the user wishes to apply a host-level
GSLB policy. Id. at 9:46-50. At block 404, GSLB controller 201
Appeal 2013-008244
Reexamination Control 90/011,765
Patent 7,584,301 B2

11
determines “whether the host H1 has been defined on the GSLB switch 12”
(id. at 9:50-51), which we understand to mean the controller determines
whether host name H1 is stored on the GSLB switch. If host H1 has not
been defined on GSLB switch 12, a message or other notification is printed
(or otherwise presented) to the user at block 406 requesting the user to
define host H1 on GSLB switch 12. Id. at 9:52-54.
After it has been determined that host H1 has been defined, “the user
can specify at a block 408 the host-level policy (e.g., the policy P1) that the
user wishes to associate with this host H1” (id. at 9:55-59), which we
understand to mean the user enters policy identifier P1. The GSLB switch
then checks (block 410) the host-policy database to determine “if the policy
P1 has been defined by the user.” Id. at 9:59-61. If policy P1 is defined on
the GSLB switch (block 412), any previous policy that is associated with the
host H1 is disassociated from that host (block 416), after which policy P1 is
“associated with” host name H1 (block 418). Id. at 9:66-10:2.
Figure 5 is reproduced below.
Appeal 2013-008244
Reexamination Control 90/011,765
Patent 7,584,301 B2

12

Figure 5 is a flowchart 500 depicting use of a host-level policy for selection
of IP addresses during the GSLB process. Id. at 10:3-4. At block 502, a
Appeal 2013-008244
Reexamination Control 90/011,765
Patent 7,584,301 B2

13
client program 28 queries for a domain for which GSLB switch 12 is
providing load balancing. Id. at 10:5-6. The authoritative DNS server 16
sends back a reply containing the IP addresses for the requested domain
(block 504), and GSLB switch 12 extracts the host/domain for the above
query (e.g., the host H1) (block 506). Id. at 10:6-12. If GSLB switch 12
determines (blocks 508, 510) that a host-level GSLB policy (e.g., P1) is
associated with host H1, GSLB switch 12 retrieves the information for
policy P1 from the GSLB host-policy database (block 512). Id. at 10:12-20.
GSLB switch 12 then uses this host-level policy P1 to select the best IP
address for client program 28 (block 514). Id. at 10:20-22.
The ’301 patent describes an example in which GSLB switch 12
provides load balancing for the following three domains:
(a) www.gslb1.com;
(b) ftp.gslb1.com; and
(c) ftp.foo.com.
Id. at 10:29-37. The user defines (and stores in the GSLB host-policy
database) a host-level policy named “test” as follows:
Metric order: health check, least response
Use DNS best-only (i.e., return only the best IP address to
the client, and discard the other IP addresses in the
response).

Id. at 10:38-43. The user also configures a global GSLB policy as follows:
Metric order: health check, least response[.]
Appeal 2013-008244
Reexamination Control 90/011,765
Patent 7,584,301 B2

14
The user then applies the host-level policy “test” to the host “www” for the
zone gslb1.com and to the host “ftp” for the zone foo.com. Id. at 10:47-49.
Since the user does not associate the domain ftp.gslb1.com with any host-
level policy, this domain will be associated with the global GSLB policy by
default. Id. at 10:49-52.

C. The Claims on Appeal
The independent claims on appeal are claims 1, 15, and 24. These
claims read as follows:
1. A method, comprising:
storing, in a load balance switch, a first load balancing policy
associated with a first domain and a second load
balancing policy associated with a second domain
different from the first domain, wherein said first load
balancing policy specifies an order in which a first
plurality of metrics of said first load balancing policy are
to be applied; and
load balancing traffic, by said load balance switch, to network
addresses associated with the first and second domains,
using corresponding said first and second load
balancing policies that are respectively associated to
the first and second domains,
wherein said metrics of said first load balancing policy each
include at least one parameter, wherein said at least one
parameter includes at least one of a tolerance, limit,
threshold, DNS parameter, preference, and alternative
tie-breaker metric use.
Appeal 2013-008244
Reexamination Control 90/011,765
Patent 7,584,301 B2

15
15. An article of manufacture, comprising:
a storage medium having instructions stored thereon that are
executable by a processor of a load balance switch to
enable load balancing, by:
associating, by said load balance switch, a first load
balancing policy to a first domain associated with a first
site switch, wherein said first load balancing policy
specifies an order in which a first plurality of metrics of
said first load balancing policy are to be applied;
associating, by said load balance switch, a second load
balancing policy to a second domain associated with a
second site switch and different from said first domain;
and
using said first and second load balancing policies by the load
balance switch to respectively rank addresses associated
with said first and second domains,
wherein said metrics of said first load balancing policy each
include at least one parameter, wherein said at least one
parameter includes at least one of a tolerance, limit,
threshold, DNS parameter, preference, and alternative
tie-breaker metric use.
24. An apparatus, comprising:
a load balance switch that includes:
a first storage unit to store a plurality of load balancing
policies, which are to be applied by said load
balance switch to rank addresses of different
domains,
Appeal 2013-008244
Reexamination Control 90/011,765
Patent 7,584,301 B2

16
wherein each of said load balancing policies specify an order of
application of a plurality of metrics of each respective
load balancing policy; and
a controller coupled to the first storage unit to use a
respective one of said stored load balancing
policies to rank addresses of each one of said
domains,
wherein said metrics, of each respective load balancing
policy, each include at least one parameter,
wherein said at least one parameter includes at least
one of a tolerance, limit, threshold, DNS parameter,
preference, and alternative tie-breaker metric
use.
’301 patent cols. 11, 13, 14.
5

C. The Rejection
Claims 1, 2, 9, 15, 24, and 28 stand rejected under 35 U.S.C. § 102(b)
for anticipation by Reference Guide. Final Action 3.
Appellant treats dependent claims 2, 9, and 28 as standing or falling
with the independent claims (i.e., claims 1, 15, and 24). Br. 19, 21.

D. The Issues

5
The copies of claims 1, 15, and 24 in the Claim Appendix (Br. 25-26) omit
the paragraph format employed by these claims in the ’301 patent. Also, the
copy of dependent claim 9 in the Claim Appendix does not include the
change made by the May 15, 2012, Certificate of Correction.
Appeal 2013-008244
Reexamination Control 90/011,765
Patent 7,584,301 B2

17
1. Whether Reference Guide is available as a reference under
35 U.S.C. §102(a) against the claims, i.e., has been shown to have a
publication date prior to the May 6, 2004, filing date of the ’301 patent.
2. Whether the rejection is based on an unreasonably broad claim
interpretation.

II. DISCUSSION
A. The Date of the Reference
Whether a given reference is a “printed publication” depends on
whether it was “publicly accessible” during the prior period. Bruckelmyer v.
Ground Heaters, Inc., 445 F.3d 1374, 1378 (Fed. Cir. 2006).
The standard of proof required to show unpatentability over prior art
during a reexamination proceeding is proof by a preponderance of the
evidence:
In a reexamination proceeding . . . there is no presumption of
validity and the “focus” of the reexamination “returns
essentially to that present in an initial examination,” In re Etter,
756 F.2d [852,] 857 [(Fed. Cir. 1985) (en banc)], at which a
preponderance of the evidence must show nonpatentability
before the PTO may reject the claims of a patent application. In
re Caveney, 761 F.2d 671, 674 . . . (Fed. Cir. 1985).
Ethicon, Inc. v. Quigg, 849 F.2d 1422, 1427 (Fed. Cir. 1988). A
preponderance of the evidence is evidence sufficient to persuade the finder
of fact that the existence of an asserted fact is more probable than its
nonexistence. Bosies v. Benedict, 27 F.3d 539, 542 (Fed. Cir. 1994).
Appeal 2013-008244
Reexamination Control 90/011,765
Patent 7,584,301 B2

18
As noted above, the IDS filed with the Request in this ex parte
proceeding identifies the reference as follows: “3-DNS Ref. Guide, version
4.2, July 1, 2002 (261 pages).” The following copyright dates appear at
page ii of Reference Guide under the heading “Legal Notices” and
subheading “Copyright”:
(a) “© 2002 Dell Computer Corporation” and
(b) “Copyright 1998-2002, F5 Networks, Inc.”
These dates are collectively referred to by the Examiner as “the Copyright
Date” of 2002. Answer 15.
The Request, in order to establish a July 1, 2002, date for Reference
Guide, relies on a “snapshot” or screen shot showing the results of an
Internet search:
The 3-DNS Ref. Guide was published at least as early as July 1,
2002, as evidenced, for example, by the date listed as associated
with the 3-DNS Ref. Guide in the snapshot of an internet search
results page that is provided after the last page of the 3-DNS
Ref. Guide attached to this Request. This publication date is
consistent with the 2002 copyright date listed on page ii of the
3-DNS Ref. Guide.
Request 5.
6
This snapshot is reproduced below.

6
Requester subsequently provided additional evidence that is relied on by
the Examiner and discussed infra.
Appeal 2013-008244
Reexamination Control 90/011,765
Patent 7,584,301 B2

19

The entry that is enclosed in the box is enlarged below:

We assume that the copy of Reference Guide that accompanied Request was
obtained from the website identified in this entry or from another website.
As correctly pointed out by Appellant, “websites can be used as prior
art only if a posting date can be found.” Reply Br. 14-15. Specifically,
MPEP § 2128 (8th ed., rev. Aug. 2012) provides in relevant part:

Date of Availability
Appeal 2013-008244
Reexamination Control 90/011,765
Patent 7,584,301 B2

20
Prior art disclosures on the Internet or on an on-line database
are considered to be publicly available as of the date the item
was publicly posted. *>Absent evidence of the date that the
disclosure was publicly posted, if< the publication >itself< does
not include a publication date (or retrieval date), it cannot be
relied upon as prior art under 35 U.S.C. 102(a) or (b)*>.
The non-final Office action (hereinafter “Non-Final Action”), mailed
October 13, 2011, rejected claims 1, 2, 9, 15, 24, and 28 for anticipation by
Reference Guide under 35 U.S.C. § 102(b). Non-Final Action 3, para. 4.
Appellant responded to the rejection by arguing, inter alia, that Reference
Guide has not been shown to have been publicly accessible prior to the
critical date (i.e., the May 6, 2004, filing date of the ’301 patent).
RESPONSE TO NON-FINAL OFFICE ACTION OF OCTOBER 4, 2011,
mailed January 12, 2012, at 3-5. Specifically, Appellant argued that the “3-
DNS Reference Guide” is not listed on the U.S. Copyright Office’s website
(id. at 3) and that “when the Patent Owner’s representative attempted to
reconstruct the same Google® search illustrated in the screen shot provided
by the Requester, he could not reproduce the same search results.” Id. at 4-
5.
The Final Action, as evidence of public accessibility prior to the
critical date, relies (at 4) on the above-noted 2002 copyright date that
appears in Reference Guide and also (at 4-6) on the following additional
Appeal 2013-008244
Reexamination Control 90/011,765
Patent 7,584,301 B2

21
documents, which the Examiner identifies as having been submitted as
Exhibits B-E in an IDS filed on April 25, 2012:
7

1. Release Note: 3-DNS Controller, version 4.5, March 5, 2007,
pages 2-1, 2-28, 3-1 to 3-12, and 5-1 to 5-24 (F5 Networks, Seattle,
Washington) (hereinafter “Version 4.5 Release Note”) (Ex. B),
http://support.f5.com/kb/ en-us/archived_products/3-
dns/releasenotes/product /relnotes4_5.html;
2. F5 Network’s Newest Wide Area Traffic Management Solution
Enables Enterprises to Meet Global e-Business Objectives, Business Wire
(Feb. 19, 2002) (hereinafter “Business Wire”) (Ex. C);
3. Release Note: 3-DNS Controller, Release Note, February 13, 2002
(hereinafter “Version 4.2 Release Note”) (Ex. D), http://support.f5.com/kb/
en-us/archived_products/ 3-dns/releasenotes/product/relnotes4_2.html; and
4. Cheng Huang, et al., A DNS Reflection Method for Global Traffic
Management (USENIXATC’10 Proceedings of the 2010 USENIX
conference on USENIX annual technical conference, 2010) (hereinafter
“Huang”) (Ex. E).

7
Our review of the eDAN record for the ’812 inter partes proceeding shows
that these documents were submitted on March 29, 2012, with THIRD
PARTY REQUESTER’S COMMENTS UNDER 35 U.S.C. § 314(B)(2) and
37 C.F.R. § 1.947,” which at page 4 identifies the sources of these
(Continued on next page.)
Appeal 2013-008244
Reexamination Control 90/011,765
Patent 7,584,301 B2

22
Appellant argues that that the statement of a 2002 copyright date in
Reference Guide is insufficient, in and of itself, to establish public
accessibility prior to the critical date by a preponderance of the evidence,
citing In re Lister, 583 F.3d 1307 (Fed. Cir. 2009). Reply Br. 13 & n.12.
8

However, even assuming for the sake of argument that the presence of this
copyright date in Reference Guide is insufficient to establish public
accessibility prior to the critical date by a preponderance of the evidence, the
remaining evidence (with the exception of Huang) supports such a finding
for the following reasons.
9

We begin our analysis by noting that Appellant does not deny that the
evidence shows that the 3-DNS version 4.2 software was publicly accessible

documents.
8
Appellant describes the holding in Lister as follows:
The Federal Circuit ruled that a patent applicant’s own
manuscript submitted for copyright registration was not a
“printed publication” bar because there was “insufficient
evidence” that this manuscript was “publicly accessible”
(including in the United States Copyright Office’s automated
catalog which would only allow searching for the name of the
author and the first word of the title) more than one year prior
to the filing date of the patent application.).
Reply Br. 13 n.12.
9
Huang is insufficient because Huang’s identification (at page 6) of
reference [1] as “DNS reference guide. WhitePaper, F3 Networks, Inc.
2002” may be based solely on the copyright date given in Reference Guide.
Appeal 2013-008244
Reexamination Control 90/011,765
Patent 7,584,301 B2

23
in 2002.
10
Instead, Appellant argues that it cannot be assumed that
Reference Guide was available at that time. See Br. 22 (“With regard to the
press release announcing version 4.2 of its 3-DNS system, the release of
software does not establish public availability of the document cited in this
reexamination proceeding.”).
The evidence persuades us that a document named “DNS-3 Reference
Guide” was publically accessible in 2002. Specifically, the Version 4.2
Release Note (Ex. D), which is dated February 13, 2002, refers to the
“DNS-3 Reference Guide” in the following two passages at pages 5 and 7,
respectively:
bigpipe utility
The bigpipe utility is a command line utility that you can use to
configure several platform-level settings for the 3-DNS. To
view the available bigpipe commands, type bigpipe at the
command line. For information on syntax and usage, refer to
the 3-DNS Reference Guide, Appendix C, bigpipe Command
Reference.

10
The Version 4.2 Release Note (Ex. D) shows (at 1) an “Updated Date:
02/13/2002” and explains (id.) that it “documents version 4.2 of the 3-DNS
system.” Also, Business Wire (Ex. C), dated February 19, 2002, states (at 2)
that “3-DNS version 4.2 is available now.” Finally, the Version 4.5 Release
Note (Ex. B) gives (at 1) a release date of “10/30/2002” for software version
4.5 and explains “how to install the 3-DNS Controller, version 4.5 onto
existing systems running version 4.2 and later.” Id.
Appeal 2013-008244
Reexamination Control 90/011,765
Patent 7,584,301 B2

24
The bleed_requests variable (CR17316)
The bleed_requests variable has been renamed the
drain_requests variable in the 3-DNS Reference Guide,
Appendix A, 3-DNS Configuration File.
We find that Version 4.2 Release Note is sufficient to establish that a
document named “3-DNS Reference Guide” was publicly available as of the
release date of Release Note, Ver. 4.2.
Appellant further argues that there may have been multiple versions of
“3-DNS Reference Guide” and that the version on which the rejection is
based has not been shown to have been publicly accessible prior to the
critical date:
[A]rguendo, if it is accepted that a product manual describing 3-
DNS version 4.2 existed at the time the software was released,
the Examiner still has not shown that the version being cited in
this reexamination was publicly available. It is possible that
there are multiple revisions or drafts. Neither the Examiner nor
the Appellant knows if the cited reference was a publicly
available version. The Examiner appears to allege that since a
reference guide is usually provided with a software release, that
it is acceptable to find that the version of 3-DNS cited in this
proceeding was publicly available. Such an allegation is
tenuous at best for establishing by a preponderance of the
evidence, for this proceeding, that the version of 3-DNS cited
was publicly available.
Reply Br. 4, para. 3. While we agree that that there may have been multiple
versions of “3-DNS Reference Guide,” we find that the version relied on in
the rejection was publicly accessible prior to the May 6, 2004, critical date
for the following reasons. The subject matter relied on in the rejection
Appeal 2013-008244
Reexamination Control 90/011,765
Patent 7,584,301 B2

25
appears in Chapter 8 (“Load Balancing”) and Chapter 13 (“Topology”),
discussed in detail infra. Answer 6-9. Specifically, the rejection is based on
the descriptions in these chapters of using preferred, alternate, and default
load balancing modes in order to select an optimal server IP address in a
pool of addresses in a “wide IP” consisting of one or more pools of
addresses. Id. This subject matter appears to be an essential part of the 3-
DNS system described in the “3-DNS Reference Guide” and thus
presumably would have been included in all versions, including the version
relied on in the rejection. We direct Appellant’s attention to In re Epstein,
32 F.3d 1559, 1567 (Fed. Cir. 1994):
Appellant . . . points out that the software products that were
“first installed” or “released” may have been different from the
products as described in the later-published abstracts. Again, in
the absence of evidence to support this speculation, we do not
find appellant’s argument persuasive. To the contrary, we are
persuaded by the relative importance of the various prior art
features upon which the PTO relies when those features are
considered in the context of the prior art products in which they
appear. After reviewing the abstracts and the examiner’s
positions in this case, it appears to us that the features relied
upon are not of the type that would be altered over the life of
the product, either to upgrade the system or to eliminate
existing bugs in the system. To the contrary, the relatively
broad features relied upon by the examiner relate to the central
purposes of the systems in which the features appear.
(Emphasis added.)
Appeal 2013-008244
Reexamination Control 90/011,765
Patent 7,584,301 B2

26
We therefore are not persuaded of error in the Examiner’s finding that
the subject matter relied on in Reference Guide was publicly accessible prior
to the critical date of May 6, 2004.
11

B. Claim Interpretation Principles
Appellant contends that the Examiner has not properly construed the
claims in a manner consistent with the ’301 patent specification.
“During reexamination, as with original examination, the PTO must
give claims their broadest reasonable construction consistent with the
specification.” In re ICON Health & Fitness, Inc., 496 F.3d 1374, 1379
(Fed. Cir. 2007) (citing In re Am. Acad. of Sci. Tech Ctr., 367 F.3d 1359,
1364 (Fed. Cir. 2004)). “[T]he ordinary and customary meaning of a claim
term is the meaning that the term would have to a person of ordinary skill in
the art in question at the time of the invention, i.e., as of the effective filing
date of the patent application.” Phillips v. AWH Corp., 415 F.3d 1303, 1313
(Fed. Cir. 2005) (en banc). Furthermore, it is improper, when in giving
claims their broadest reasonable interpretation, to read limitations from the
specification into the claims. See Phillips, 415 F.3d at 1323 (“[A]lthough

11
The Brief additionally argues that “[w]ith regard to the Skene patent,
again, the fact that the Skene patent claims priority back to 1999 does not
establish the public availability of the version of 3-DNS cited in this
proceeding.” Br. 23. However, no Skene patent is mentioned by the
Examiner or by the Request in this ex parte reexamination proceeding, and
this argument is not repeated in the Reply Brief.
Appeal 2013-008244
Reexamination Control 90/011,765
Patent 7,584,301 B2

27
the specification often describes very specific embodiments of the invention,
we have repeatedly warned against confining the claims to those
embodiments. . . . In particular, we have expressly rejected the contention
that if a patent describes only a single embodiment, the claims of the patent
must be construed as being limited to that embodiment.”) (citations omitted).
Instead, as explained in Decisioning.com Inc. v. Federated Dep’t Stores
Inc., 527 F.3d 1300, 1307-08 (Fed. Cir. 2008), the focus should be on the
claim terms:
“[T]here is sometimes a fine line between reading a claim in
light of the specification, and reading a limitation into the claim
from the specification.” Comark Commc’ns, Inc. v. Harris
Corp., 156 F.3d 1182, 1186 (Fed. Cir. 1998); accord Phillips v.
AWH Corp., 415 F.3d 1303, 1323–24 (Fed. Cir. 2005) (en
banc) (“In the end, there will still remain some cases in which it
will be hard to determine whether a person of skill in the art
would understand the embodiments to define the outer limits of
the claim term or merely to be exemplary in nature.”). This
court has recognized that “attempting to resolve that problem in
the context of the particular patent is likely to capture the scope
of the actual invention more accurately than either strictly
limiting the scope of the claims to the embodiments disclosed in
the specification or divorcing the claim language from the
specification,” Phillips, 415 F.3d at 1323–24, and, thus, that
there can be “no magic formula or catechism for conducting
claim construction,” id. at 1324. We must read the specification
in light of its purposes in order to determine “whether the
patentee is setting out specific examples of the invention to
accomplish those goals, or whether the patentee instead intends
for the claims and the embodiments in the specification to be
strictly coextensive.” Id. at 1323. “The manner in which the
Appeal 2013-008244
Reexamination Control 90/011,765
Patent 7,584,301 B2

28
patentee uses a term within the specification and claims usually
will make the distinction apparent.” Id. Ultimately, our “focus
remains on understanding how a person of ordinary skill in the
art would understand the claim terms.” Id.
(Emphasis added).
Appellant argues:
The written description can assist a person having ordinary skill
in the art in understanding what portions of the prior art are not
covered by the claims. PSC Computer Products, Inc. v.
Foxconn Intern., 355 F.3d 1353, 1359 (Fed. Cir. 2004) (“In
essence, the written description [found within the specification]
can, at times, assist the public in understanding the notice given
in the claims by explaining which portions of the relevant art
the patent does not cover.”) (Emphasis original).
Br. 14-15 (brackets in original). However, PSC further states: “Consistent
with its scope definition and notice functions, the claim requirement
presupposes that a patent applicant defines his invention in the claims, not in
the specification. After all, the claims, not the specification, provide the
measure of the patentee’s right to exclude.” 355 F.3d at 1359 (emphasis
added).

C. The Merits of the Anticipation Rejection
Reference Guide describes the features of the “3-DNS System®”
(hereinafter “3-DNS system”). Reference Guide 1-1 (i.e., Chapter 1, page
1).
Appeal 2013-008244
Reexamination Control 90/011,765
Patent 7,584,301 B2

29
As already noted, the rejection of claim 1 is based on Reference Guide’s
Chapters 8 (“Load Balancing”) and 13 (“Topology”). Answer 6-9.
When a name resolution request is received from a local DNS server,
the 3-DNS system uses a load balancing mode to select the best available
virtual server from a “wide IP pool” (i.e., a pool of a wide IP). Id. at 8-1.
“A wide IP is a mapping of a fully-qualified domain name (FQDN) to a set
of virtual servers that host the domain’s content, such as a web site, an e-
commerce site, or a CDN.”
12
Id. at 8-10. A wide IP contains one or more
pool definitions. Id. A pool is a group of virtual servers that the 3-DNS
load balances. Id.
The 3-DNS system uses load balancing in two different ways:
♦ Load balancing among multiple pools
The 3-DNS supports multiple pools. Configurations that
contain two or more pools use a load balancing mode first to
select a pool. Once the 3-DNS selects a pool, the system
then uses a load balancing mode to choose a virtual server
within the selected pool. . . .

♦ Load balancing within a pool
Within each pool, you specify three different load balancing
modes that the system uses in sequential order: preferred
method, alternate method, and fallback method. The
preferred method is the first load balancing mode that the 3-
DNS uses for load balancing. If the preferred method fails,
the system then uses the alternate method for load balancing.
If this load balancing mode fails, the system uses the

12
A “CDN” is a content delivery network. Reference Guide, Glossary 2.
Appeal 2013-008244
Reexamination Control 90/011,765
Patent 7,584,301 B2

30
rollback load balancing mode. If the fallback method fails,
the 3-DNS returns the client to standard DNS for resolution.
Id. The rejection (discussed below) is alternatively based on single-pool
wide IP and a two-pool wide IP.
Table 8.1 (id. at 8-2) is reproduced below.
13

13
We downloaded the Reference Guide figures reproduced herein, which
have better clarity than the figures found in the record copy of Reference
Guide, from http://support.f5.com/content/kb/en-us/archived_products/3-
dns/
manuals/product/3dns4_2ref/_jcr_content/pdfAttach/download/file.res/3-
DNS_Reference_Guide%2c_version_4.2.pdf.
Appeal 2013-008244
Reexamination Control 90/011,765
Patent 7,584,301 B2

31

Table 8.1 shows a complete list of the supported load balancing modes and
indicates where each mode can be used in the 3-DNS system configuration.
Id. at 8-2. The first column (“Load Balancing mode”) in Table 8.1 lists
sixteen balancing modes available to the 3-DNS system. These load
balancing modes are described at pages 8-3 to 8-5. Selection of the “None”
mode results in skipping the associated load balancing method. Id. at 8-3.
Also, some of balancing modes have default settings. For example, in
describing the “Ratio” mode, Reference Guide explains that “[t]he default
Appeal 2013-008244
Reexamination Control 90/011,765
Patent 7,584,301 B2

32
ratio weight for a server or a pool is set to 1.” Id. at 8-4.
The second column (“Use for pool load balancing”) in Table 8.1
shows that of any of the following five balancing modes can be used to
select a pool when the wide IP consists of a plurality of pools: Global
Availability, Random, Ratio, Round Robin, and Topology. The remaining
columns identify the balancing modes that can be used to select the
appropriate server from the servers from the selected pool. Specifically, the
third and fifth columns (“Use for preferred method” and “Use for default
method”) show that any one of the sixteen load balancing modes can be
selected as the preferred method or the default method. The fourth column
(“Use for alternate method”) identifies ten balancing modes available for
selection as the alternative method.
Figure 8.7 (at 8-12) is reproduced below.

Appeal 2013-008244
Reexamination Control 90/011,765
Patent 7,584,301 B2

33

Figure 8.7 shows the correct syntax for a “wideip” statement. Id. As
shown in this figure, the domain name that corresponds to the wide IP will
be inserted in the sixth line, which reads: name <“domain_name”>. The
remaining lines in this example permit identification of the selection criteria,
Appeal 2013-008244
Reexamination Control 90/011,765
Patent 7,584,301 B2

34
including identification of the balancing modes selected for the preferred,
alternate, and backup balancing methods.
We understand the rejection to be based on reading the recited “first
load balancing policy associated with a first domain” (claim 1) on a first
wideip statement associated with a first domain and reading the recited
“second load balancing policy associated with a second domain different
from the first domain” (claim 1) on a second wideip statement associated
with a second, different domain. Appellant contends that neither the claimed
“first load balancing policy associated with a first domain” nor the claimed
“second load balancing policy associated with a second domain” can be read
on a wideip statement.
The rejection is alternatively based on (i) the single-pool example
defined by the wideip statement shown in Figure 13.3 (at page 13-5)
(Advisory Action 7-8
14
); and (ii) the two-pool example defined by the
wideip statement shown in Figure 8.8 (at page 8-14). Advisory Action 8.
For the following reasons, we agree that claims 1, 15, and 24, including the
above-quoted language of claim 1, read on each of these examples.

1. The single-pool example
Figure 13.3 is reproduced below.

14
Advisory Action mailed July 20, 2012.
Appeal 2013-008244
Reexamination Control 90/011,765
Patent 7,584,301 B2

35

Figure 13.3 shows “a sample wide IP definition where topology is the
[preferred] load balancing mode within a pool.” Id. at 13-5. As shown in
this figure, the wide IP identifies “topology” as the preferred load balancing
method and “rtt” as the alternate load balancing method. Furthermore, the
Examiner correctly describes this figure as describing “a wide IP statement
with a single pool containing three addresses.” Answer 7.
The Examiner reads the recited “first load balancing policy associated
with a first domain . . . , wherein said first load balancing policy specifies an
order in which a first plurality of metrics of said first load balancing policy
are to be applied” (claim 1) on the Figure 13-3 embodiment as follows:
“Because only one pool is defined for the domain, the ‘preferred,’
‘alternate’, and ‘fallback’ methods are applied (in order) to all three
addresses associated with the domain. For example, the topology metric will
be applied before the ‘rtt’ metric, which will be applied before the default
‘fallback’ metric.” Id. The Examiner’s reference to a fallback metric is
misplaced, as no fallback method or metric is specified in the wideip
Appeal 2013-008244
Reexamination Control 90/011,765
Patent 7,584,301 B2

36
statement. Nevertheless, we find that the preferred “topology” metric and
alternate “rtt” metric in this wideip statement are sufficient to satisfy the
above-quoted language of claim 1.
Appellant’s arguments to the contrary are not persuasive.
15
One
argument is:
In the ‘301 patent, the claimed load balancing policy is
associated with or applied across all IP addresses hosted by the
domain -- not just a portion or a pool of the IP addresses. In
contrast, 3-DNS, at best, discloses associating or applying a
load balancing policy across only a pool or a portion of
addresses to rank addresses in a domain.
Reply Br. 6 (emphasis added). According to Appellant: “A pool of
addresses is only a subset of addresses within a domain. A pool of
addresses, by definition, never includes all addresses of a domain. FIG. 13.3
. . . only illustrates load balancing on a pool level (i.e., ‘poolA’), and not a
domain level.”). Id. at 9. This argument is unpersuasive for two reasons.
First, assuming for the sake of argument that claim 1 would have been
understood to require application of the first load balancing policy to all of
the network addresses associated with a domain, we agree with the Examiner
that the claim thus interpreted reads on the single-pool example of Figure
13.3. See Answer 7 (“[A] wide IP statement may contain only one pool,

15
The incorrect argument at page 16 of the Brief that “FIG. 13.3 of 3-DNS
clearly illustrates that only one metric (i.e. rtt) of an alternate policy is
applied to the pool” is not repeated in the Reply Brief.
Appeal 2013-008244
Reexamination Control 90/011,765
Patent 7,584,301 B2

37
consisting of all the servers associated with that domain.”). Appellant’s
above-quoted assertion that “[a] pool of addresses [in Reference Guide], by
definition, never includes all addresses of a domain” (Reply Br. 9) is
incorrect. Reference Guide explains that a wide IP represents all of the
server addresses for a domain, which can be located in a single pool. See
Reference Guide 8-10 (“A wide IP is a mapping of a fully-qualified domain
name (FQDN) to a set of virtual servers that host the domain’s content . . .
.”); id. at 8-11 (“Note that you must configure at least one pool in the wide
IP, but you may configure any number of pools.”). Appellant has not
identified any description in Reference Guide of using a wide IP to map only
a portion of a set of virtual servers that host a domain’s content.
Second, we agree with the Examiner that the claim language does not
require associating the recited first load balancing policy with all of the
network addresses associated with the first domain. See Answer 6 (“The
plain language of the independent claims requires only applying a load
balancing policy to rank addresses associated with a domain. Under the
broadest reasonable interpretation, this includes ranking less than all of the
addresses associated with that domain.”). Appellant, as support for the “all
of the network addresses” claim interpretation, cites the examples given in
the specification of the ’301 patent. See Reply Br. 6 (“In every instance
outlined below, the specification of the ’301 patent always applies or
associates a load balancing policy to a domain (i.e., host) and not just a pool
or subset of addresses in a domain.”). However, it is improper to read
Appeal 2013-008244
Reexamination Control 90/011,765
Patent 7,584,301 B2

38
limitations from the examples given in the specification into the claims.
Phillips, 415 F.3d at 1323. Insofar as the “first policy load balancing
policy” and the “first domain” are concerned, the claim 1 recitations of “a
first load balancing policy associated with a first domain” and “first and
second load balancing policies that are respectively associated to the first
and second domains” are broad enough to read on a first load balancing
policy that applies to only a portion of the server IP addresses associated
with the first domain (e.g., the server IP addresses in one of a plurality of
pools that correspond to a domain). The same is true of the claim 15 phrases
“associating, by said load balance switch, a first load balancing policy to a
first domain” and “using said first and second load balancing policies by the
load balance switch to respectively rank addresses associated with said first
and second domains.” The same conclusion also applies to the language of
claim 24.
In another argument, Appellant contends that the rejection is improper
because it is improperly based on reading both the recited “first domain” and
“first load balancing policy” on a single entity in the form of a wideip
statement:
The claims, even under the broadest reasonable interpretation,
must be interpreted so that each claim term equates to a single
feature within the prior art reference for an anticipation
rejection to stand. . . .
[T]hroughout this proceeding, the Examiner has
construed the Wide IP statement as the claimed “load balancing
policy.” Because the claims separately recite a “domain” and
Appeal 2013-008244
Reexamination Control 90/011,765
Patent 7,584,301 B2

39
“load balancing policy,” the “domain” cannot also be a part of
the entity that is being construed as the “load balancing policy.”
Reply Br. 11-12 (footnote omitted). While we agree with Appellant that the
claims terms “load balancing policy” and “domain” refer to different
entities, we do not agree that the rejection fails to take this into account. The
Examiner reads the recited “load balancing policy” on the portion of the
Figure 13.3 wideip statement that (i) provides the name of the single pool
(i.e., “poolA”) and (ii) identifies the preferred and alternate balancing modes
(i.e., “topology” and “rtt,” respectively) to be applied to the IP addresses in
the pool. Answer 7. Although the name of the associated domain appears in
the wideip statement, this domain is an entity separate from the load
balancing policy set forth in the wideip statement. See Answer 12 (“The
Wide IP statement separately associates a ‘domain’ and ‘load balancing
policy.’ As shown in Figure 8.7, the Wide IP statement may contain the
claimed ‘load balancing policy’ that is ‘associated with,’ and a separate
element from a domain.”) (emphasis added).
Appellant additionally argues that “the specification of the [’]301
patent requires separately defining the domain and load balancing policy in
different data structures that are established before they are associated.”
Reply Br. 11. We understand Appellant to be arguing (1) that the domain
and load balancing policy must be defined in different data structures and
(2) that the domain and load balancing policy must be defined before they
are associated with each other. As support, Appellant cites the flowchart
Appeal 2013-008244
Reexamination Control 90/011,765
Patent 7,584,301 B2

40
depicted in Figure 4 of the ’301 patent. Id. at 11 & n.8. As explained above,
this flowchart shows in relevant part:
(1) determining whether a host name (H1) specified by the user has
been defined on the GSLB switch (block 404), and
(2) if the answer is yes, determining whether the host-level policy (P1)
specified by the user has been defined on the GSLB switch (blocks 410,
412), and,
(3) if the answer is yes, associating the policy P1 with host H1 in the
GSLB switch (blocks 416, 418).
To the extent Appellant is arguing that the domain and load balancing
policies must be defined in different data structures, we do not agree. In
fact, the ’301 patent explains that the host policy information can be stored
in a single data structure:
[F]irst storage unit 212 . . . contains one or more configured
GSLB host-level policies. These host-level policies can in turn
specify, for each host, the metrics to be used for selection of IP
addresses, the order in which the metrics are to be applied, the
parameters for the metrics (such as tolerance, limits, and the
like), or other associated settings.
In an embodiment, the first storage unit 212 and the
second storage unit 210 [storing global-level policies] can be in
the form of databases having tables. It is appreciated, however,
that either or both of these storage units can be embodied by
any suitable data structure (including file systems, directory
structures, variables, static or dynamic code, or any other
suitable technique or structure that can accept and store policy
settings and data).
Appeal 2013-008244
Reexamination Control 90/011,765
Patent 7,584,301 B2

41
Id. at 7:4-21 (emphasis added). Furthermore, assuming for the sake of
argument that the Figure 4 flowchart would have been understood to
represent the use of separate data structures for defining the domain and a
load balancing policy, it is improper to read limitations from the examples
given in the Specification into the claims. Phillips, 415 F.3d at 1323.
We also do not agree that the domain and load balancing policy must
be defined before they are associated with each other, as allegedly shown in
Figure 4 of the ’301 patent. Appellant has not explained why the language
of claims 1 and 24 supports such an interpretation. Regarding claim 15,
which recites “associating, by said load balance switch, a first load balancing
policy to a first domain associated with a first site switch,” Appellant argues:
“the step of ‘associating, by said load balance switch’ is performed by the
load balancing switch, which requires that the load balancing policy and
domain are separately defined before they are associated.” Reply Br. 12
(emphasis altered).
16
We do not agree. In our view, the claim language is
broad enough to permit the first domain and first load balancing policy to be
defined at the same time that they are associated with each other.

16
This Reply Brief argument presumably replaced the following argument
that appears at page 21 of the Brief: “In claim 15, the step of ‘associating,
by said load balance switch’ is performed by the load balancing switch,
which requires that the load balancing policy and the domain not be
associated prior to the step of associating.”
Appeal 2013-008244
Reexamination Control 90/011,765
Patent 7,584,301 B2

42
For the foregoing reasons, we are not persuaded of error in the
Examiner’s rejection to the extent it is based on the single-pool example
shown in Figure 13.3.

2. Two-pool example
For reasons like those given above, we also are not persuaded of error
in the rejection to the extent it is alternatively based on the two-pool
example shown in Figure 8.8 (Reference Guide 8-14), which is reproduced
below.

Figure 8.8 shows a sample wideip statement for a wide IP named
mx.wip.domain.com (and having an alias of mail.wip.domain.com). Id. at
8-13. The wide IP contains two pools, named “pool_1” and “pool_2.” Id.
Appeal 2013-008244
Reexamination Control 90/011,765
Patent 7,584,301 B2

43
The pool load balancing metric for selecting a pool is “ratio,” which is
defined so that pool_1 receives three times as many requests as pool_2. Id.
Furthermore, pool_1 has preferred and alternative modes (i.e., metrics) of
“rtt” and “random,” whereas pool_2 has only a preferred mode of “ratio,”
which is set so that the first listed virtual server receives twice as many
connections as the second listed virtual server. Id.
When claim 1 is compared to Figure 8.8, the recited “first domain”
reads on the name “mx.wip.domain.com” and/or the alias
“mail.wip.domain.com.” The recited “first load balancing policy associated
with a first domain” is broad enough to read on the balancing mode
associated with pool_1, on the balancing mode associated with pool_2, or on
the balancing modes associated with both pools.

III. CONCLUSION
For the above reasons, the rejection of independent claims 1, 15, and
24 for anticipation by Reference Guide is sustained. For the same reasons,
the rejection on this ground is sustained with respect to dependent claims 2,
9, and 28, which are not separately argued.

IV. DECISION
The Examiner’s decision that claims 1, 2, 9, 15, 24, and 28 are
unpatentable over the prior art is affirmed.
AFFIRMED
Appeal 2013-008244
Reexamination Control 90/011,765
Patent 7,584,301 B2

44

ack

For Patent Owner:

2ND REEXAM GROUP
NOVAK DRUCE + QUIGG LLP
1000 Louisiana Street
53rd Floor
Houston, Texas 77002

For Third-Party Requester:

Timothy J. May
Finnegan Henderson, Farabow, Garrett & Dunner, L.L.P.
901 New York Avenue, N.W.
Washington, D.C. 20001-4413