Ex Parte Kawahito et al

Board of Patent Appeals and Interferences

Jun 29, 2009

10085455 (B.P.A.I. Jun. 29, 2009)

UNITED STATES PATENT AND TRADEMARK OFFICE
________________

BEFORE THE BOARD OF PATENT APPEALS
AND INTERFERENCES
________________

Ex parte MOTOHIRO KAWAHITO, HIDEAKI KOMATSU,
and JOHN CRAIG WHALEY
________________

Appeal 2008-003996
Application 10/085,455
Technology Center 2100
________________

Decided:1 June 29, 2009
________________

Before ALLEN R. MACDONALD, HOWARD B. BLANKENSHIP, and
DEBRA K. STEPHENS, Administrative Patent Judges.

MACDONALD, Administrative Patent Judge.

DECISION ON APPEAL

1 The two-month time period for filing an appeal or commencing a civil
action, as recited in 37 C.F.R. § 1.304, begins to run from the decided date
shown on this page of the decision. The time period does not run from the
Mail Date (paper delivery) or Notification Data (electronic delivery).
Appeal 2008-003996
Application 10/085,455

2
Appellants appeal under 35 U.S.C. § 134 from the Examiner’s
rejection of claims 1-16, which are all the claims remaining in the
application. We have jurisdiction under 35 U.S.C. § 6(b).
We affirm.

STATEMENT OF THE CASE
According to Appellants, the invention relates to a method for
optimizing a computer program based on the property of data at the time of
execution.2
Exemplary Claims
1. A program optimization method for translating, into machine
code, source code for a program written in a programming language,
and for dynamically optimizing said program comprising the steps of:
performing a dynamic analysis during execution to determine
whether the execution speed of said program can be increased by
fixing, in a specific state, a parameter for a predetermined command
in said program; and
employing results of said analysis for the dynamic generation,
in said program, of a path along which said parameter of said
predetermined command is fixed in said specific state.

5. A program optimization method for translating, into machine
code, the source code for a program written in an object-oriented

2 Spec. 1:5-8.
Appeal 2008-003996
Application 10/085,455

3
programming language, and for optimizing said program comprising
the steps of:
detecting one command dynamically during execution, of the
commands in said program, for which a method call destination can
be identified, and for which the processing speed can be increased by
identifying said method call destination; and
dynamically generating a path wherefor said method call
destination for said detected command is limited in order to increase
the processing speed of said command.

9. The compiler according to claim 8, wherein, in accordance with
the state of said program at execution, said specialization unit
generates, in said program, a branching process for selectively
performing a specialized path and an unspecialized path; and wherein,
while taking into account a delay due to the insertion of said
branching process, said data specialization selector determines a state
in which said parameter of said predetermined command is fixed.

Prior Art
The Examiner relies on the following prior art references to show
unpatentability:
Linden 2002/0066086 A1 May 30, 2002
Shaylor 6,760,907 B2 Jul. 6, 2004

Appeal 2008-003996
Application 10/085,455

4

Examiner's Rejections
1. The Examiner rejected claims 1-4 and 6-16 under 35 U.S.C.
§ 102(e) as being anticipated by Linden.
2. The Examiner rejected claim 5 under 35 U.S.C. § 103(a) as
unpatentable over Linden and Shaylor.

102(e) REJECTION OVER LINDEN

ISSUE
The issue before us is whether the prior art teaches (1) fixing a
parameter for a predetermined command in a specific state and
(2) dynamically generating a path along which that parameter is fixed in that
state.

FINDINGS OF FACT
1. The Examiner found "Linden explicitly discloses the dynamic
decoding (i.e., analysis) stage going through block of source instructions and
analyzes the operation codes such as the addition (i.e., command) of two
numbers (i.e., parameters)" (Ans. 16) (emphasis omitted).
2A. We also find Linden discloses the dynamic compiler 10 comprises a
decoding stage for decoding the source instructions and parameters and for
creating an instruction stream that is optimized based on the source
instructions and parameters (see Linden, ¶ [0034]).
Appeal 2008-003996
Application 10/085,455

5
2B. We also find that Linden discloses "at the decoding stage 12, the
source instructions are fetched at 40" (Linden, ¶ [0037]) (emphasis omitted).
2C. Further, we find Linden discloses "[t]he decoding stage goes through
the block of source instructions and analyzes, examines the operation codes,
and outputs an instruction string which is basically a breakdown of what the
functional operations are doing" (Linden, ¶ [0037]).
2D. In addition, we find that Linden discloses "[t]aking the decoded
instruction, a preliminary optimization of the instruction stream is conducted
at 44 to include . . . optimization tasks" (Linden, ¶ [0038]) (emphasis
omitted).
3. One of the optimization tasks disclosed in Linden, as found by the
Examiner, involves "translating the assignment (i.e., command) of
Register 3 (i.e., one of the parameters) to the outcome of 'Register 1 XOR
Register 1' which is always fixed in a specific state because clearly it does
not matter what the value of variable Register 1 is, the result of XOR-ing a
variable and itself is always the constant 0 (i.e., Boolean false)" (Ans. 16,
emphasis omitted; see also Linden, ¶ [0038]).
4. Further, we find that Linden discloses that "[t]he preliminary
optimization step 44 outputs the optimized instruction stream in preparation
for subsequent flow optimization and encoding." (Linden, ¶ [0038])
(emphasis omitted).

PRINCIPLES OF LAW
In rejecting claims under 35 U.S.C. § 102, “[a] single prior art
reference that discloses, either expressly or inherently, each limitation of a
claim invalidates that claim by anticipation.” Perricone v. Medicis
Appeal 2008-003996
Application 10/085,455

6
Pharmaceutical Corp., 432 F.3d 1368, 1375 (Fed. Cir. 2005) (citing Minn.
Mining & Mfg. Co. v. Johnson & Johnson Orthopaedics, Inc.,
976 F.2d 1559, 1565 (Fed. Cir. 1992)).
Analysis of whether a claim is patentable over the prior art under
35 U.S.C. § 102 begins with a determination of the scope of the claim. We
determine the scope of the claims in patent applications not solely on the
basis of the claim language, but upon giving claims their broadest reasonable
construction in light of the specification as it would be interpreted by one of
ordinary skill in the art. In re Am. Acad. of Sci. Tech Ctr., 367 F.3d 1359,
1364 (Fed. Cir. 2004). The properly interpreted claim must then be
compared with the prior art.
If the Examiner’s burden is met, the burden then shifts to the
Appellants to overcome the prima facie case with argument and/or evidence.
Obviousness is then determined on the basis of the evidence as a whole and
the relative persuasiveness of the arguments. See In re Oetiker,
977 F.2d 1443, 1445 (Fed. Cir. 1992).

ANALYSIS
Appellants’ Arguments and Contentions
Claims 1, 6, 7, 10, 13, and 15
Exemplary claim 1 recites (1) “determine whether the execution speed
of said program can be increased by fixing, in a specific state, a parameter
for a predetermined command in said program,” and (2) “dynamic
generation, in said program, of a path along which said parameter of said
predetermined command is fixed in said specific state” (App. Br. 27, Claims
Appendix).
Appeal 2008-003996
Application 10/085,455

7
Fixing a Parameter for a Pre-determined Command
Appellants contend "Linden creates new instructions for the target
device 'independent of operations specified by the source instructions'
whereas the present invention uses the 'predetermined command in said
program' but dynamically generates a path along which a parameter for the
command is fixed in a specific state" (App. Br. 13). Further, Appellants
contend "Linden does not look at a parameter of a predetermined command;
rather, Linden looks [at the] intent and purpose of instructions"
(App. Br. 13-14).
In addition, Appellants argue (1) "that eliminating operations is not
the same as the claimed step of determining whether execution speed can be
increased by fixing a parameter for a command of the program in a specific
state" (App. Br. 14); (2) "Linden's step for making a result a constant does
not anticipate fixing a variable parameter" (App. Br. 15); and (3) "Linden
does not state that any parameters be fixed" (App. Br. 16).
We find that Linden discloses Register 3 = Register 1 XOR Register 1
(see FF 3). Thus, we find that Linden discloses the assignment command, as
a skilled artisan would have recognized that the "=" after Register 3 is an
instruction to assign what follows the "=" to Register 3 (see FF 3).
Also, we agree with the Examiner that "Register 3" is one of the
parameters. Further, we also agree with the Examiner that Register 3 is
always fixed in a specific state because (1) Register 3 = Register 1 XOR
Register 1, and (2) when you XOR something with itself, you always get a
false or "0" (see FF 3). Thus, we find one of ordinary skill in the art would
have recognized that Linden discloses a parameter (i.e., Register 3) fixed
Appeal 2008-003996
Application 10/085,455

8
(i.e., set at "0") for a predetermined command (i.e., the assignment
command, as represented by the "=").
Performing a Dynamic Analysis and Dynamically Generating a Path Along
Which the Parameter of a Predetermined Command is Fixed
Appellants contend "the Examiner has not cited any teaching in
Linden which anticipates the claim language of 'performing a dynamic
analysis'" (App. Br. 15). Appellants argue "[t]he Examiner has listed terms
including 'dynamically cross-compiling, execution speed, execution time,
overhead . . . dynamic recompiliation . . . decoded instruction, instruction
sequence, result, constant, Register3=0,'" and (2) "[s]imilarly, with respect
to the claim language of 'employing results . . .' the Examiner has listed
terms (e.g., 'instruction sequence, result, constant, Register=3, optimization
step 44, optimized instruction stream' on page 5-6 of the Final Office
Action), but the Examiner does not explain how Linden's use of those terms
anticipates the claim language" (App. Br. 15). Further, Appellants contend
that "the Examiner's duty [is] to apply teachings of the cited art to the claims
in order to establish anticipation, and not to simply list terms from the
reference" (App. Br. 15).
Appellants also contend "Linden does not dynamically create a path
along which a parameter of a predetermined command is fixed in a specific
state" (App. Br. 14). Appellants argue that "Linden teaches translating an
instruction sequence into another sequence where the result is always a
constant (paragraph [[0038]]), but that is not the same as or suggestive of
dynamically creating a path along which a parameter of a predetermined
command is fixed in a specific state" (App. Br. 14). Appellants’ arguments
Appeal 2008-003996
Application 10/085,455

9
with regard to exemplary claim 1, also apply to independent claims 7, 10,
13, and 15 (see App. Br. 12, 13 and 17).
We agree with the Examiner that Linden discloses a decoding stage
that analyzes operation codes (see FF 1). Further, we find a skilled artisan
would have recognized that Linden's dynamic compiler would have
performed the analysis of the operation codes (see FF 2A). Accordingly, we
find Linden discloses performing a dynamic analysis.
In addition, we find that Linden discloses that "[t]he preliminary
optimization step 44 outputs the optimized instruction stream in preparation
for subsequent flow optimization and encoding,” and this instruction stream
corresponds to the path of exemplary claim 1 (see FF 4). Further, we find
that the optimized instruction stream results from the dynamic analysis
(as discussed above) that included, for example, the translation of the
instruction sequence into another sequence where the result is always zero
(see FF 3 ).
Accordingly, we find that a skilled artisan would have recognized that
the path is generated from step 44 that involved setting "Register 3=0", i.e.,
fixing a parameter for a predetermined command in said program (see FF 3-
4). Thus, we find Linden discloses dynamically generating a path along
which a parameter is fixed, and thus, Appellants have not persuaded us of
error in the Examiner’s finding of anticipation regarding claims 1, 7, 10, 13,
and 15. Accordingly, we sustain the Examiner's rejection of claims 1, 7, 10,
13, and 15.
Further, (1) we find claim 6 is commensurate in scope with exemplary
claim 1, as claim 6 recites "detecting dynamically . . . one command . . . for
which a variable can be limited to a predetermined value . . . and generating
Appeal 2008-003996
Application 10/085,455

10
a path along which said constant value for said variable of said detected
command is fixed," and (2) Appellants arguments with regard to claim 6 are
commensurate in scope with Appellants arguments with respect to
exemplary claim 1. Thus, for the reasons discussed above with respect to
exemplary claim 1, Appellants have not persuaded us of error in the
Examiner's finding of anticipation regarding claim 6, and the Examiner's
rejection of claim 6 is sustained.

Claims 2, 3-4, 8, 11-12, 14, and 16
ISSUE
The issue before us is whether the prior art teaches (1) obtaining
statistical data for the appearance frequency of each available state, (2) and a
data specialization selector.

ADDITIONAL FINDINGS OF FACT
5A. The Examiner found support for Linden's teaching or suggestion of
"obtaining statistical data for the appearance frequency of each available
state . . . to dynamically generate the path" at least in part, in paragraph 0012
of Linden (Ans. 19).
5B. Further, the Examiner found
Linden explicitly discloses improving the [overall] execution speed by
cross-compiling (also referred to as recompilation) programs
containing execution loops of instructions that are repeatedly executed
hundreds, thousands of times (i.e., appearance frequency of each
available state) during execution of the program. The same paragraph
also explicitly discloses analyzing and decoding these frequently
executed instructions once to generate the target instruction in order to
avoid repeated interpretation of the instructions in such loops, thereby
Appeal 2008-003996
Application 10/085,455

11
saving (i.e., increasing) execution time/speed of the
translated/compiled code.

(Ans. 19) (emphasis omitted).
5C. The Examiner then found
it is inherent in Linden that the frequency of being executed for each
command is determined (i.e., obtain statistical data for the appearance
frequency of each available state) in order to identify which command
to dynamically recompile (as opposed to repeatedly interpreting the
frequently executed command which would slow down the overall
execution speed of the program) to generate the optimized instruction
stream (i.e., path) for the target machine.

(Ans. 19-20).
6. We find that Linden discloses:
The primary reason that overall execution speed is improved by
cross-compilation is that most programs contain execution loops of
instructions that are repeatedly executed hundreds, thousands, or even
millions of times during a typical execution of the program. The
source instructions are analyzed and decoded only once, i.e., the first
time they are addressed, and the target instruction stream is generated
and stored in memory, usually a RAM memory that may be cache
memory. By avoiding repeated interpretation of the instructions in
such loops, substantial execution time is saved. Consequently,
subsequent emulation of the same source instruction may be
performed quickly because the decoding overhead is nonexistent.

(Linden, ¶ [0012]).

ANALYSIS
Appellants argue "the Examiner has not cited any teaching in Linden
which anticipates the claim language of 'obtaining statistical data for the
appearance frequency of each available state'" and for employing the
obtained statistical data to dynamically generate the path"
Appeal 2008-003996
Application 10/085,455

12
(see App Br. 18-19).
The Appellants rely on the arguments set forth above with regard to
the teachings of Linden as they related to claims 1, 7, 10, 13, and 15.
(App. Br. 18). Appellants also argue "the Examiner lists terms used in the
Linden reference (e.g., 'instruction sequence, result, constant, Register=3,
optimization step 44, optimized instruction stream' on page 6 of the Final
Office Action), but the Examiner does not explain how Linden's use of those
terms anticipates the claim language" (App. Br. 18-19).
Appellants also argue "[s]imilarly, there is no teaching or suggestion
of a data specialization selector as set forth in Claims 8 and 9" (App. Br. 19).
Further, Appellants argue that "the Examiner has a duty to apply teachings
of the cited art to the claims in order to establish anticipation, and not to
simply list terms from the reference" (App Br. 19).
We find that Linden discloses that the reason the overall execution
speed is improved by cross-compilation is that most programs contain
execution loops of instructions that are repeatedly executed hundreds,
thousands, or even millions of times during a typical execution of the
program (see FF 5-6). We find that a skilled artisan would have recognized
that the execution speed would not have improved if the task of cross-
compilation had not been performed to reduce the number of loops of
instructions that were repeated. Further, we find that the task of identifying
the instructions loops that were repeated, in Linden, inherently teaches or
suggests obtaining statistical data for the appearance frequency of each
available state (i.e. instruction loop or set of commands), as recited in
representative claim 2 (see FF 5-6).
Appeal 2008-003996
Application 10/085,455

13
Further, we find that Linden discloses (1) "[t]he source instructions
are analyzed and decoded . . . and stored in memory," and (2) "[b]y avoiding
repeated interpretation of the instructions in such loops, substantial
execution time is saved" (see FF 6). Thus, we also find that a skilled artisan
would have recognized that what is stored in storage is the analyzed and
decoded source instructions, i.e., the new path, as what is stored in storage
avoids repeated interpretation of instructions ( inherently as a result of
obtaining statistical data for the appearance frequency of each instruction
loop or set of commands). Accordingly, we find Linden inherently teaches
or suggests employing the obtained statistical data (i.e., using the results of
the analysis) to dynamically generate the path), as recited in representative
claim 2.
Claim 8 recites that the data specialization selector is "for, when said
program is executed, obtaining statistical data for the appearance frequency
of each state obtained by said impact analysis unit, and for determining the
state in which said parameter of said predetermined command is to be set"
(App. Br. 31). As discussed above, Linden teaches or suggests obtaining
statistical data for the appearance frequency of each state, and setting a
predetermined command. Accordingly, we find that Linden inherently
teaches a data specialization selector, as there must inherently be some
component that performs those tasks.
Appellants argue claims 2, 3-4, 8, 11-12, 14, and 16 as a group
(App. Br. 18, 20). However, Appellants have not persuaded us of error in
the Examiner's finding of anticipation regarding claims 2, 3-4, 8, 11-12, 14,
and 16. Accordingly, we sustain the Examiner's rejection of claims 2, 3-4, 8,
11-12, 14, and 16.
Appeal 2008-003996
Application 10/085,455

14

Claim 9
ISSUE
The issue before us is whether the prior art teaches a specialization
unit that generates a branching process for selectively performing a
specialized path and an unspecialized path.

ADDITIONAL FINDINGS OF FACT
7. The Specification of the present application discloses "[a]ccording to
the invention, the program is so optimized that on a path (a specialized path)
along which the parameter of the predetermined command is fixed and on a
path (an unspecialized path) along which this parameter is not fixed in the
specific stated is selectively executed during the branching process"
(Spec. 4).
8. The Examiner found Linden discloses "creating an optimized
instruction flow stream which omits the instructions corresponding to the
overridden C=A+B operation (i.e., branching process for selectively
performing a specialized path) as opposed to removing the C=A+B
instruction at the decoding stage” (Ans. 20) (emphasis omitted).

ANALYSIS
Claim 9 depends from claim 8 and further recites
wherein, in accordance with the state of said program at execution,
said specialization unit generates, in said program, a branching
process for selectively performing a specialized path and an
unspecialized path; and wherein, while taking into account a delay
due to the insertion of the branching process, said data specialization
Appeal 2008-003996
Application 10/085,455

15
selector determines a state in which the parameter of the
predetermined command is fixed.

(App. Br. 31, Claims Appendix).
Appellants argue
the instructions listed by the Examiner on page 10 of the Final Office
Action, including 'optimized instruction flow stream, optimization
rules, pipeline delay[,]' do not provide teachings which anticipate a
specialization unit which generates a branching process for selectively
performing both a specialized path and an unspecialized path, or a
data specialization selector for determining a state in which the
parameter of a predetermined command is fixed.

(App. Br. 21) (emphasis omitted).
We find the Specification of the present application defines (1) a
specialized path as a path along which the parameter of the predetermined
command is fixed, and (2) an unspecialized path as a path along which the
parameter is not fixed in the specific state (FF 7). We find that the act of
overriding the instruction, C=A+B, corresponds to a branching process, as
the path presents two options: (1) a path that includes the C=A+B
instruction, as the instruction was not omitted, and (2) a path that overrides
the C=A+B instruction (see FF 8).
Accordingly, Linden inherently discloses a specialization unit, as a
skilled artisan would have recognized that there must be a component to
perform the branching process. Thus, Appellants have not persuaded us of
error in the Examiner's finding of anticipation regarding claim 9.
Accordingly, we sustain the Examiner's rejection of claim 9.

Appeal 2008-003996
Application 10/085,455

16
103(a) REJECTION OVER LINDEN AND SHAYLOR
Claim 5
ISSUE
The issue before us is whether the prior art teaches or suggests
detecting a command for which a method call destination can be identified
and for which the processing speed can be increased by identifying the
method call destination.

FINDINGS OF FACT
9. The Examiner found
that Shaylor's target method should be understood as the method
being called (i.e., method call destination) from the calling method.
Furthermore, in order to inline the target method into the calling
method (i.e., copy the code of the target method into the cod[e] of the
calling method), it is inherent that the actual calling
command/instruction (inside the calling method) be detected so as to
determine the exact position for inlining (i.e., inserting the code of the
target method).

(Ans. 22) (emphasis omitted).
10. We find that Shaylor discloses
[a] common program occurrence is that a first method (the caller
method) calls a second method (the target method) in a process
referred to as a 'method call'. This method call process is
advantageous from a programmer's perspective, but consumes many
clock cycles. An important optimization in the compilation of object
oriented program code is called 'inlining.['] In fully compiled
programs, inlining makes these target method calls more efficient by
copying the code of the target method into the calling method.

(Shaylor, col. 2, ll. 48-56).

Appeal 2008-003996
Application 10/085,455

17
PRINCIPLES OF LAW
In rejecting claims under 35 U.S.C. § 103, it is incumbent upon the
Examiner to establish a factual basis to support the legal conclusion of
obviousness. See In re Fine, 837 F.2d 1071, 1073 (Fed. Cir. 1988). In so
doing, the Examiner must make the factual determinations set forth in
Graham v. John Deere Co., 383 U.S. 1, 17 (1966). Discussing the question
of obviousness of claimed subject matter involving a combination of known
elements, KSR Int’l v. Teleflex, Inc., 550 U.S. 398 (2007), explains:
When a work is available in one field of endeavor, design
incentives and other market forces can prompt variations of it,
either in the same field or a different one. If a person of
ordinary skill can implement a predictable variation, § 103
likely bars its patentability. For the same reason, if a technique
has been used to improve one device, and a person of ordinary
skill in the art would recognize that it would improve similar
devices in the same way, using the technique is obvious unless
its actual application is beyond his or her skill. Sakraida [v. AG
Pro, Inc., 425 U.S. 273 (1976)] and Anderson's-Black Rock[,
Inc. v. Pavement Salvage Co., 396 U.S. 57 (1969)] are
illustrative—a court must ask whether the improvement is more
than the predictable use of prior art elements according to their
established functions.

Id. at 417. If the claimed subject matter cannot be fairly characterized as
involving the simple substitution of one known element for another or the
mere application of a known technique to a piece of prior art ready for the
improvement, a holding of obviousness can be based on a showing that
“there was an apparent reason to combine the known elements in the fashion
claimed.” Id. at 418. Such a showing requires:
“some articulated reasoning with some rational underpinning to
support the legal conclusion of obviousness” . . . . [H]owever,
the analysis need not seek out precise teachings directed to the
Appeal 2008-003996
Application 10/085,455

18
specific subject matter of the challenged claim, for a court can
take account of the inferences and creative steps that a person
of ordinary skill in the art would employ.

Id. (quoting In re Kahn, 441 F.3d 977, 988 (Fed. Cir. 2006)).
If the Examiner’s burden is met, the burden then shifts to the
Appellants to overcome the prima facie case with argument and/or evidence.
Obviousness is then determined on the basis of the evidence as a whole and
the relative persuasiveness of the arguments. See In re Oetiker,
977 F.2d at 1445.

ANALYSIS
Appellants rely on the discussion and arguments presented above with
regard to the teachings of Linden (see App. Br. 24). Applicants further argue
that "the Shaylor reference does not supply those teachings which are
missing from Linden" (App. Br. 24).
Appellants argue (1) Shaylor "teaches that the overhead of method
calls can be reduced by 'inlining' which copies the code of a target method
into a calling method,” and (2) "[c]opying the code of the target method into
a calling method is not the same as or suggestive of detecting a command for
which a method call destination can be identified and for which the
processing speed can be increased by identifying the method call
destination" (App. Br. 24).
In addition, Appellants contend Shaylor discloses "inlining causes
code size to increase, which effectively teaches away from the claimed steps
for increasing processing speed for a command" (App. Br. 25). Further,
Appellants argue "the alternative direct method call provided in the Shaylor
Appeal 2008-003996
Application 10/085,455

19
passage is not the same as or suggestive of dynamically generating a path
where a call destination is limited to increase processing speed"
(App. Br. 25).
We agree with the Examiner, that in order to inline the target method
into the calling method, it is inherent that the actual calling
command/instruction be detected, so as to determine the exact position for
inlining (see FF 9). Further, we find that, inherently, inlining increases the
speed, as Linden discloses inlining is (1) for the purpose of optimization,
and (2) makes target method calls more efficient (see FF 10). Thus, we find
Shaylor does not teach away from increasing the processing speed for a
command.
Thus, Appellants have not persuaded us of error in the Examiner's
conclusion of obviousness for claim 5, and the rejection of claim 5 is
sustained.

DECISION
The Examiner's decision rejecting claims 1-16 is affirmed.
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

msc

ANNE V. DOUGHERTY
3173 CEDAR RD.
YORKTOWN HEIGHTS, NY 10598