Ex Parte Knitter et al

Patent Trial and Appeal Board

Mar 27, 2013

11408375 (P.T.A.B. Mar. 27, 2013)

UNITED STATES PATENT AND TRADEMARK OFFICE
____________

BEFORE THE PATENT TRIAL AND APPEAL BOARD
____________

Ex parte JAY D. KNITTER and WADE J. SATTERFIELD
____________

Appeal 2010-007159
Application 11/408,375
Technology Center 2400
____________

Before JOHN A. JEFFERY, BRUCE R. WINSOR, and
BARBARA A. BENOIT, Administrative Patent Judges.

JEFFERY, Administrative Patent Judge.

DECISION ON APPEAL
Appellants appeal under 35 U.S.C. § 134(a) from the Examiner’s
rejection of claims 1-25. We have jurisdiction under 35 U.S.C. § 6(b). We
affirm-in-part.

STATEMENT OF THE CASE
Appellants’ invention uses session state identifiers to reconcile a web
server’s session state with that of a client’s web browser. See generally
Abstract; Spec. ¶¶ 24-34; Fig. 2. Claim 1 is illustrative:
1. A web-server implemented method of reconciling a web-server
session state with a client web-browser state, said method comprising:
Appeal 2010-007159
Application 11/408,375

2
responding to an nth request from a client web browser by reading an
mth session-state identifier included in said request, where n-1 > m >1, where
integer values of m from 1 to n correspond to a chronological succession of
web-server session states of a session; and

assuming an mth session state associated with said mth session-state
identifier.

THE REJECTIONS1
1. The Examiner rejected claims 1, 9, and 15 under 35 U.S.C. § 112,
¶ 2 as indefinite. Ans. 3-4.2
2. The Examiner rejected claims 21-25 under 102(e) as anticipated by
Feenan (US 2006/0155857 A1; published July 13, 2006; filed Jan. 6, 2005).
Ans. 4-6.
3. The Examiner rejected claims 1-3, 6, 7, 9-11, 14-17, and 20 under
35 U.S.C. § 103(a) as unpatentable over Feenan and Habusha (US 6,477,590
B1; issued Nov. 5, 2002). Ans. 6-10.
4. The Examiner rejected claims 4, 5, 12, 13, 18, and 19 under
35 U.S.C. § 103(a) as unpatentable over Feenan, Habusha, and Sattler (US
2006/0294240 A1; published Dec. 28, 2006; filed June 24, 2005). Ans. 10-
12.
5. The Examiner rejected claim 8 under 35 U.S.C. § 103(a) as
unpatentable over Feenan, Habusha, Sattler, and Lui (US 2007/0083813 A1;
published Apr. 12, 2007; filed Oct. 11, 2005). Ans. 12-13.

1 Since the Examiner withdrew rejections of claims 22 and 24 under § 112
¶¶ 1 and 2 (Ans. 2, 13), those rejections are not before us.
2 Throughout this opinion, we refer to (1) the Appeal Brief filed October 7,
2009 (“App. Br.”); (2) the Examiner’s Answer mailed December 24, 2009
(“Ans.”); and (3) the Reply Brief filed February 15, 2010 (“Reply Br.”).
Appeal 2010-007159
Application 11/408,375

3
THE INDEFINITENESS REJECTION
The Examiner finds that independent claims 1, 9, and 15 are indefinite
because the range of the SSID variable “m” is not clearly defined. Ans. 3-4,
14. According to the Examiner, the variable “m” (1) must be less than “n-1”
according to the first recited definition; (2) may be equal to request variable
“n” according to the second recited definition; and (3) must be less than “n”
according to the Abstract. Ans. 3-4.
Appellants acknowledge that while one recited limitation permits
“m=n,” the other does not. Reply Br. 3. Appellants, however, emphasize
that since the ranges in these respective limitations overlap such that one
does not invalidate the other, the claim does not permit “m=n” when
considering these two ranges together, but permits “m=n-1.” Id. Since both
recited constraints are said to be met under this condition, Appellants
contend that the claims are not vague or ambiguous. Reply Br. 2-3.

ISSUE
Under § 112, ¶ 2, has the Examiner erred in rejecting claim 1 by
finding that the recited relationships between “n” and “m,” namely “n-1 > m
≥ 1” for integer values of “m” from “1 to n” renders the claim indefinite?

ANALYSIS
We begin by noting that since the Examiner withdrew one of two
grounds in connection with the indefiniteness rejection, namely the ground
based on the terms “session state,” “session,” and “SSID” (Ans. 14; Reply
Br. 2), we review only the remaining ground based on the recited
mathematical expression relating the variables “m” and “n.”
Appeal 2010-007159
Application 11/408,375

4
We further note, as does the Examiner (Ans. 3-4), that the
mathematical expression relating the variables “m” and “n” recited in claim
1 differs from that indicated in the Abstract. Compare claim 1 (“n-1 > m ≥
1”) with Abstract (“n > m ≥ 1”). This inconsistency likewise exists between
originally-filed claim 1 and the originally-filed Abstract. Compare claim 1
filed Apr. 22, 2006 (“n-1 > m ≥ 1”) with Abstract (“n > m ≥ 1”) filed Apr.
22, 2006. But apart from the originally-claimed expression and Appellants’
referring to various method steps in the particular example in the flow chart
of Figure 2, Appellants do not squarely identify—nor can we find—where
the particular mathematical expression recited in claim 1 is located in the
Specification. See App. Br. 4, 17 (referring to method segments M23 and
M24 in connection with the recited responding step containing the
mathematical expression at issue).
Turning to the rejection, we first note that Appellants do not dispute
that there is at least one scenario (“m=n”) that is not permitted by the
expressions of claim 1, despite the claim’s specifying integer values of “m”
from 1 to “n”: the upper limit of which yields this impermissible scenario.
Reply Br. 3. Nevertheless, while we can see some scenarios that would be
permitted by the recited expressions, we fail to see how the scenario “m=n-
1” is always permitted under the recited constraints which undermines
Appellants’ argument based on this scenario. See Reply Br. 2-3 (arguing
that “m=n-1” is within the scope of the claims since it meets both recited
constraints).
For example, if n=2 and m=1, substituting these values in the recited
expression yields (2-1 > 1 ≥ 1) or (1 > 1 ≥ 1) which is mathematically
impossible, since the value “1” is not greater than itself as emphasized. A
Appeal 2010-007159
Application 11/408,375

5
similar inconsistency results if n=3 and m=2, for this yields (3-1 > 2 ≥ 1) or
(2 > 2 ≥ 1) which is, again, mathematically impossible.
The recited expressions, however, do yield mathematically consistent
results for other values of “n” and “m.” For example, if n=3 and m=1, then
this would yield (3-1 > 1 ≥ 1) or (2 > 1 ≥ 1) which is consistent.
At best, then, the claim encompasses some scenarios that are
mathematically impossible under the recited constraints, yet others that are
not. This inconsistency not only renders the claim somewhat vague with
respect to its scope, but also results in an ambiguity that renders the claim
amenable to at least two plausible constructions that likewise renders the
claim indefinite. See Ex parte Miyazaki, 89 USPQ2d 1207, 1211 (BPAI
2008) (precedential).
We are therefore not persuaded of error in the Examiner’s
indefiniteness rejection of representative claim 1, and claims 9 and 15 not
separately argued with particularity.

THE ANTICIPATION REJECTION
The Examiner finds that Feenan discloses every recited element of
independent claim 1 receiving an ith HTTP request of a session having a jth
SSID embedded therein where i and j are counting number indices and j < i,
a relationship which is said to correspond to activating previous session
states. Ans. 4-5. According to the Examiner, since the jth session state
information is read and identified from a session state cookie, a session state
link allows activating previous session states which may be different website
pages, thus meeting the recited condition that j = (i-1) and j<(i-1). Ans. 5.
Appeal 2010-007159
Application 11/408,375

6
Appellants argue that Feenan discloses session identifiers—not
session state identifiers. App. Br. 20-22; Reply Br. 3. Appellants add that
Feenan does not disclose the recited numerical relations between the SSIDs
and requests. App. Br. 22-23.

ISSUE
Under § 102, has the Examiner erred in rejecting claim 21 by finding
that Feenan discloses a (1) SSID, and (2) determining from a jth SSID that
an ith HTTP request was made from a jth page provided in a web session
when j = (i-1) and j<(i-1)?

ANALYSIS
We will not sustain the Examiner’s rejection of representative claim
21. The Examiner refers to Feenan’s session cookie 320 that is generated by
cache array 350 following a client’s logging on to an analysis application
(www.mybank.com/analysis) 382 on a site 300 (www.mybank.com). Ans.
14-15; Feenan ¶ 0046; Fig. 3. This cookie not only identifies the site and
analysis application, but also the cluster (cache array 350). Feenan ¶ 0046.
Notably, this session cookie is also referred to as a session state cookie and
indicates affinity with a particular origin server, namely origin server 362.
Id. Similarly, a counterpart session state cookie 330 indicates affinity with
origin server 364 should the client log into another application and establish
another session via that server. Id.
In another example, Feenan refers to a cookie (“MYSESSION”)
associated with a particular session that ensures that subsequent requests are
directed to the same origin server. Feenan ¶ 0035. Feenan further notes that
Appeal 2010-007159
Application 11/408,375

7
once a client connects with an origin server, a session is bound to
connections for all subsequent requests. Feenan ¶ 0036. For example, if a
session begins with one of 10 origin servers (e.g., “OS5”), the session is
bound to that server. Id. Moving to another server (e.g., “OS6”), however,
results in not finding the previous state which requires re-establishing the
session. Id.
Feenan refers to a “session” as anytime a user logs onto a website, and
“session state” as maintaining a virtual active link on a website. Feenan
¶ 0047. When considering Feenan’s “session state” terminology in context,
this virtual active link refers to a particular origin server associated with a
particular session. We therefore agree with Appellants (App. Br. 22) that
each session in Feenan has one identified session state, namely that
associated with the particular origin server to which that session is bound.
Accord Feenan claim 1 (reciting (1) binding a user session and an origin
server in a session cookie, and (2) routing subsequent requests for the user
session to the same origin server based on information in the cookie).
At best, then, Feenan’s session cookie identifies a session state with
respect to binding a particular session to a particular origin server. But even
considering this information as an SSID, we cannot say—nor has the
Examiner shown—that Feenan necessarily determines from a jth SSID that
an ith HTTP request was made from a jth page provided in a web session
when j = (i-1) and j<(i-1)3 as claimed. The Examiner’s position that Feenan
meets this limitation since “activating previous session states which may be

3 We note in passing the incongruity of “when j = (i-1) and j<(i-1)” recited
in claim 21 (emphasis added), which is not relevant to the dispute before us.

Appeal 2010-007159
Application 11/408,375

8
different pages of a web site” (Ans. 5; emphasis added) is untenable. Even
if we were to accept the Examiner’s position, the Examiner’s permissive
language undermines the Examiner’s anticipation rejection since activating
previous session states does not necessarily yield different pages. In any
event, as noted above, each session in Feenan has one identified session
state, namely that associated with a particular origin server to which the
session is bound. We therefore cannot say that the recited numerical
relations between SSIDs and requests are necessarily present in Feenan.
We are therefore persuaded that the Examiner erred in rejecting
(1) independent claim 21, (2) independent claim 23 which recites
commensurate limitations, and (3) dependent claims 22, 24, and 25 for
similar reasons.

THE OBVIOUSNESS REJECTION OVER FEENAN AND HABUSHA
The Examiner finds that Feenan discloses every recited element of
representative claim 1 except for the recited integer values to correspond to a
chronological succession of web-server session states of a session, but cites
Habusha as teaching this feature in concluding that the claim would have
been obvious. Ans. 6-8, 15.
Appellants argue that (1) Feenan lacks an SSID, and (2) Habusha
teaches away from including an identifier in a request and therefore reading
an identifier from a request as claimed. App. Br. 24-26; Reply Br. 4.

ISSUES
(1) Under § 103, has the Examiner erred in rejecting claim 1 by
finding that Feenan and Habusha collectively would have taught or
Appeal 2010-007159
Application 11/408,375

9
suggested reading an “mth” SSID included in an “nth” request from a client
web browser, where n-1 > m ≥ 1?
(2) Is the Examiner’s combining the teachings of these references
supported by articulated reasoning with some rational underpinning to
justify the Examiner’s obviousness conclusion?

ANALYSIS
We sustain the Examiner’s obviousness rejection of representative
claim 1. As noted above, each session in Feenan has one identified session
state, namely that associated with the particular origin server to which that
session is bound. Thus, Feenan’s cookies that identify this state reasonably
correspond to an SSID given the scope and breadth of this limitation.
Appellants’ arguments regarding Feenan’s lacking an SSID (App. Br. 24;
Reply Br. 4) are therefore not commensurate with the scope of the claim.
Nor do Appellants’ arguments persuasively rebut the Examiner’s
reliance on Feenan for teaching responding to subsequent requests by
reading this cookie-based identifier. Ans. 7. Appellants’ contention
regarding Feenan’s lacking the recited numerical relations between request
numbers and SSIDs, namely j = (i-1) and j<(i-1) (App. Br. 24), is unavailing,
for this expression is not recited in claim 1.
We also see no error in the Examiner’s reliance on Habusha merely to
show that serializing numbers corresponding to particular states of a packet-
based communication is known, and that applying this technique to Feenan’s
reading SSIDs from a request would have been obvious. See Ans. 6-8, 15.
In Habusha, nodes 82 and 84 have sequence numbers associated with
sending and receiving packets for identification and tracking purposes.
Appeal 2010-007159
Application 11/408,375

10
Habusha, col. 6, ll. 5-23. When the first node sends a packet to the receiving
node, the sending node increments its packet sent number by one. Habusha,
col. 6, ll. 11-14. Upon receipt, the receiving node increments its received
number by one. Habusha, col. 6, ll. 15-17. Although sequence numbers are
not sent with packets as Appellants indicate (App. Br. 25), that does not
obviate Habusha’s fundamental teaching of using integer values (sequence
numbers) that correspond to a chronological succession of states of a
messaging session.
Nor does Habusha teach away from modifying Feenan as Appellants
contend. App. Br. 24-26; Reply Br. 4. First, the Examiner relies on
Feenan—not Habusha—for reading an SSID in a request. Ans. 7. In any
event, we see no reason why Habusha’s fundamental technique could not be
applied to Feenan’s SSID as the Examiner proposes. To be sure, Feenan’s
cookie-based SSID identifies one session state, namely that associated with
the particular origin server to which that session is bound as noted above.
But that does not mean that this SSID cannot also identify other web-server
session states, particularly in view of the Examiner’s reliance on Feenan’s
Paragraph 0011 which notes that as each query of an application executes, it
can generate results that can store and share results with subsequent requests
coming from the same end user (i.e., a session state).
Thus, while a session may only have one state with respect to a
particular origin server as noted above, skilled artisans would recognize that
this particular state may also have other associated states corresponding to
consecutive queries from the same end-user for sharing results. These
query-based states associated with a particular origin server are at least
suggested by Feenan. See Feenan ¶¶ 0011, 0035 (noting that if a user that
Appeal 2010-007159
Application 11/408,375

11
has added to a web-based shopping cart decides to leave the cart and start
another application, the user can return to the shopping cart without having
to start a new session). Accounting for those states via integers that
correspond to the states’ chronological succession is at least suggested by
Feenan when considered in light of Habusha. Accordingly, we find the
Examiner’s combining the teachings of these references supported by
articulated reasoning with some rational underpinning to justify the
Examiner’s obviousness conclusion.
We are therefore not persuaded that the Examiner erred in rejecting
representative claim 1, and claims 2, 3, 6, 7, 9-11, 14-17, and 20 not
separately argued with particularity.

THE OBVIOUSNESS REJECTION OVER FEENAN AND SATTLER
We will not, however, sustain the Examiner’s rejection of dependent
claim 4 reciting, in pertinent part, a user navigating among pages cached by
a web browser. In the rejection, the Examiner finds that Sattler discloses
this feature. Ans. 11 (citing Sattler ¶ 0001). Although this passage teaches
various browser navigational commands including activating a back button,
it does not indicate from what source the previous page is obtained, let alone
a browser cache, as Appellants indicate. App. Br. 27.
Apparently recognizing this shortcoming, the Examiner in the
Response to Arguments takes a different position, namely that “one of
ordinary skill in the art would appreciate that pages are normally cached by a
browser to improve web service performance.” Ans. 15. Since no authority
is cited for this proposition, the Examiner’s new position is tantamount to
officially noticing this fact as “capable of such instant and unquestionable
Appeal 2010-007159
Application 11/408,375

12
demonstration as to defy dispute.” See In re Ahlert, 424 F.2d 1088, 1091
(CCPA 1970) (citations omitted). Appellants challenge this reliance on
Official Notice as inappropriate since “caching is not a necessary or inherent
feature of a browser.” Reply Br. 5 (emphasis added).
Despite the fact that what is “necessary or inherent” is more germane
to anticipation, not obviousness, Appellants have nonetheless seasonably
challenged the Examiner’s belated taking of Official Notice in the Answer at
the earliest practicable opportunity, which in this case, is the Reply Brief.
In view of this challenge, it is therefore incumbent upon the Examiner to
provide documentary evidence to support the Examiner’s challenged
finding—a crucial deficiency on this record. See MPEP
§ 2144.03(C) (noting if the Examiner’s assertion of Official Notice is
adequately traversed, the Examiner must provide documentary evidence in
the next Office Action to maintain the rejection). Notwithstanding the
Examiner’s shift in position, where, as here, there is no concrete evidence on
this record to support that challenged position, we are constrained by this
record to reverse the rejection. 4 Accord In re Zurko, 258 F.3d 1379, 1385-
86 (Fed. Cir. 2001) (noting that “the deficiencies of the cited references
cannot be remedied by the Board’s general conclusions about what is ‘basic
knowledge’ or ‘common sense’ to one of ordinary skill in the art” absent
some concrete evidence in the record to support these findings).

4 We note, however, that should prosecution reopen after this opinion, the
Examiner should consider Appellants’ statement in Paragraph 04 of the
Specification that “[a]s a user moves from page to page, the browser can
cache the prior pages on the user’s computer” in connection with the
requisite evidentiary support needed to support the obviousness rejection of
claim 4. See Spec. ¶ 04 (describing the invention’s background).
Appeal 2010-007159
Application 11/408,375

13
We are therefore persuaded that the Examiner erred in rejecting
(1) claim 4; (2) claims 12 and 18 which recite commensurate limitations;
and (3) claims 5, 13, and 19, which each depend from one of claims 4, 12,
and 18 respectively, for similar reasons.

THE OBVIOUSNESS REJECTION OVER FEENAN, HABUSHA, SATTLER, AND LUI
We will, however, sustain the Examiner’s rejection of claim 8 over
Feenan, Habusha, Sattler, and Lui. Ans. 12-13. Claim 8 depends from
claim 1 and recites, in pertinent part, detecting the use of two or more
windows in a session. We see no error in the Examiner’s reliance on Lui for
teaching this feature in concluding that claim 8 would have been obvious.
Ans. 13 (citing Lui ¶ 0307).
Although the Examiner relies on Sattler in connection with this
rejection—a reference which we found to be problematic regarding
dependent claim 4 as noted above—Appellants do not dispute the
Examiner’s reliance on Sattler in connection with claim 1 from which claim
8 depends.5 Nevertheless, to the extent that the Examiner’s reliance on
Sattler is cumulative to the cited other references in connection with the
Examiner’s earlier findings regarding claim 1, we deem the Examiner’s error
in this regard harmless.
Turning to the rejection, Appellants’ argument regarding reading an
SSID from a request (App. Br. 28; Reply Br. 5-6) is unpersuasive for the
reasons previously discussed. Appellants also admit that Lui uses two or

5 See Ans. 12 (rejecting claim 8 over “Feenan in view of Habusha and
Sattler, as applied to claim 1 above, and further in view of Lui . . . .”)
(emphasis added; bolding omitted).
Appeal 2010-007159
Application 11/408,375

14
more windows in a session, but does not detect two or more web session
windows. App. Br. 28; Reply Br. 6.
On this record, we find the weight of the evidence favors the
Examiner’s position. Even assuming, without deciding, that Lui does not
pertain to a web session as Appellants contend, Appellants provide no
persuasive argument or evidence proving that applying Lui’s known
technique of detecting two or more windows in a session could not be
applied to a web session as in Feenan. It is well settled that if a technique
has been used to improve one device (detecting two or more windows in a
session), and an ordinarily skilled artisan would recognize that it would
improve similar devices in the same way (applying that technique to web
sessions), using the technique is obvious unless its actual application is
beyond his or her skill. KSR Int’l Co. v. Teleflex Inc., 550 U.S. 398, 417
(2007). Appellants’ contention that claim 8 relates to detecting two web
sessions (Reply Br. 6) is unavailing, for the claim does not recite two web
sessions, but rather (1) detecting the use of two or more windows in a
session, and (2) tracking the session in each window. We are therefore not
persuaded that the Examiner erred in rejecting claim 8.

CONCLUSION
The Examiner did not err in rejecting (1) claims 1, 9, and 15 under
§ 112, and (2) claims 1-3, 6-11, 14-17, and 20 under § 103. The Examiner,
however, erred in rejecting (1) claims 21-25 under § 102, and (2) claims 4,
5, 12, 13, 18, and 19 under § 103.

Appeal 2010-007159
Application 11/408,375

15
ORDER
The Examiner’s decision rejecting claims 1-25 is affirmed-in-part.
No time period for taking any subsequent action in connection with
this appeal may be extended under 37 C.F.R. § 1.136(a)(1)(iv).

AFFIRMED-IN-PART

rwk