Ex Parte Anderson

Board of Patent Appeals and Interferences

May 6, 2009

10729574 (B.P.A.I. May. 6, 2009)

UNITED STATES PATENT AND TRADEMARK OFFICE
____________

BEFORE THE BOARD OF PATENT APPEALS
AND INTERFERENCES
____________

Ex parte THOMAS G. ANDERSON
____________

Appeal 2008-4881
Application 10/729,574
Technology Center 2100
____________

Decided:1 May 6, 2009
____________

Before JAY P. LUCAS, JOHN A. JEFFERY, and
CAROLYN D. THOMAS, Administrative Patent Judges.

JEFFERY, 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 Date (electronic delivery).
Appeal 2008-4881
Application 10/729,574

Appellant appeals under 35 U.S.C. § 134 from the Examiner’s
rejection of claims 3-32 and 34-38. We have jurisdiction under 35 U.S.C.
§ 6(b). We affirm.

STATEMENT OF THE CASE
Appellant invented a method for efficient interaction of a user with a
human-computer interface employing a haptic input device (e.g. a game
controller with tactile feedback). Specifically, two paths are established: (1)
an “object fundamental path,” and (2) a “device fundamental path.” The
interface can detect motion of an input device along a device fundamental
path, and use this detected motion to cause motion of an object in a
computer application along an object fundamental path. The interface can
also supply force feedback to the user resisting off-path motion of the device
to guide the user to move the device along the device fundamental path.2
Claim 3 is illustrative:
3. In a human-computer interface, a method of allowing a user of a
haptic input device to affect the motion of an object in a computer
application, comprising:

a) Establishing an object fundamental path representing a path of
motion of the object in the computer application;

b) Establishing a device fundamental path in correspondence with
the object fundamental path;

c) Detecting motion of the haptic input device;

2 See generally Spec. ¶¶ 0006-08; Abstract.
2
Appeal 2008-4881
Application 10/729,574

d) Moving the object in the computer application along the object
fundamental path responsive to a component of haptic input device motion
along the device fundamental path; and

e) Applying a force to the haptic input device responsive to a
component of haptic input device motion not along the device fundamental
path; and

f) Applying a force to the haptic input device responsive to
interaction of the object with the application.

The Examiner relies on the following as evidence of unpatentability:
Frid-Nielsen US 5,655,093 Aug. 5, 1997
Bertram US 6,191,785 B1 Feb. 20, 2001
Rosenberg US 6,219,032 B1 Apr. 17, 2001
Meredith US 6,220,963 B1 Apr. 24, 2001
Baynton US 6,277,030 B1 Aug. 21, 2001
Rosenberg
(“Rosenberg II”)
US 6,288,705 B1 Sep. 11, 2001
Shih US 6,552,722 B1 Apr. 22, 2003
Gould US 6,583,782 B1 Jun. 24, 2003
Stewart US 6,801,187 B2 Oct. 5, 2004
(filed Jun. 22, 2001)

1. The Examiner rejected claims 3, 4, 7, 8, 11-13, 19-24, and 26-30
under 35 U.S.C. § 102(b) as anticipated by Rosenberg (Ans. 4-10).
2. The Examiner rejected claims 3, 5, 6, 35, and 36 under 35 U.S.C.
§ 102(e) as anticipated by Stewart (Ans. 10-11).
3. The Examiner rejected claim 9 under 35 U.S.C. § 103(a) as
unpatentable over Rosenberg and Frid-Nielsen (Ans. 11-13).
3
Appeal 2008-4881
Application 10/729,574

4. The Examiner rejected claim 10 under 35 U.S.C. § 103(a) as
unpatentable over Rosenberg and Bertram (Ans. 13-14).
5. The Examiner rejected claims 14 and 25 under 35 U.S.C. § 103(a) as
unpatentable over Rosenberg and Rosenberg II (Ans. 15-16).
6. The Examiner rejected claim 15 under 35 U.S.C. § 103(a) as
unpatentable over Rosenberg and Gould (Ans. 16-18).
7. The Examiner rejected claims 16-18 under 35 U.S.C. § 103(a) as
unpatentable over Rosenberg, Gould, and Shih (Ans. 18-20).
8. The Examiner rejected claim 34 under 35 U.S.C. § 103(a) as
unpatentable over Rosenberg, Gould, and Rosenberg II (Ans. 20-21).
9. The Examiner rejected claim 32 under 35 U.S.C. § 103(a) as
unpatentable over Meredith and Rosenberg (Ans. 21-22).
10. The Examiner rejected claims 31 and 38 under 35 U.S.C. § 103(a) as
unpatentable over Baynton, Rosenberg, and Meredith (Ans. 22-24).
11. The Examiner rejected claim 37 under 35 U.S.C. § 103(a) as
unpatentable over Stewart and Rosenberg II (Ans. 24-25).
Rather than repeat the arguments of Appellant or the Examiner, we
refer to the Brief and the Answer3 for their respective details. In this
decision, we have considered only those arguments actually made by
Appellant. Arguments which Appellant could have made but did not make
in the Brief have not been considered and are deemed to be waived. See 37
C.F.R. § 41.37(c)(1)(vii).

3 Throughout this opinion, we refer to the Appeal Brief filed Jan. 25, 2006
and the Examiner’s Answer mailed May 9, 2007.
4
Appeal 2008-4881
Application 10/729,574

THE ANTICIPATION REJECTION OVER ROSENBERG
Claims 3, 7, 8, and 10-13
In rejecting representative claim 3,4 the Examiner relies principally on
Rosenberg’s application of force feedback to an interface device in the form
of a “groove” force. According to the Examiner, this groove corresponds to
an “object fundamental path” since it represents the path of motion of
interface objects in computer applications (e.g., cursors and scroll bar
“thumbs”). The Examiner adds that Rosenberg likewise teaches a “device
fundamental path” in view of the path of the interface device and its
correspondence to the groove. Furthermore, the Examiner notes that
Rosenberg’s system can also apply “collision forces” to the interface device
responsive to various interactions with the object and the associated
application (Ans. 4-6.)
Appellant argues that while Rosenberg’s groove force feedback keeps
a cursor within a region of the graphical user interface (GUI) (i.e., the scroll
bar), bumps or other forces are applied when the user crosses window
boundaries within the GUI. As such, Appellant argues, Rosenberg does not
apply forces to an input device (1) based on interactions with an object with
an application, and (2) while the cursor is in a groove. Based on this
deficiency, Appellant concludes that Rosenberg does not apply forces to the

4 Appellant argues claims 3, 7, 8, 10, 11, and 13 together as a group. See Br.
13-14. Although Appellant did not include claim 12 in this group, it was not
separately argued. See Br. 13-19. Accordingly, we include claim 12 in this
group and select claim 3 as representative. See 37 C.F.R. § 41.37(c)(1)(vii).
5
Appeal 2008-4881
Application 10/729,574

device while the user moves an input device along a “device fundamental
path.” (Br. 13-14.) Appellant adds that Rosenberg controls cursors—not
objects—in an application. (Br. 14.)
The issues before us, then, are as follows:

ISSUES
Has Appellant shown that the Examiner erred in rejecting claim 3
under § 102 by finding that:
(1) Rosenberg applies a force to a haptic input device responsive to (a)
a component of haptic input device motion not along the “device
fundamental path,” and (b) interaction of an object of a computer application
with the application?
(2) Rosenberg’s cursor and scroll bar “thumb” reasonably correspond
to an “object” as claimed?

FINDINGS OF FACT
The record supports the following findings of fact (FF) by a
preponderance of the evidence:
1. Rosenberg discloses a system that provides force feedback to a
user 22 operating a human/computer interface device 14 in conjunction with
a GUI displayed by a host computer system 12. A user object 34 (e.g., a
joystick or mouse) controls a graphical object (e.g., a cursor) within the
GUI. The host computer sends a signal to the interface device to apply a
force sensation to the physical object, where the force sensation is associated
with graphical objects and operating system functions of the GUI. The force
sensation is also determined by the location of the cursor in the GUI with
6
Appeal 2008-4881
Application 10/729,574

respect to targets associated with graphical objects (i.e., icons, windows,
pull-down menus, scroll bars, and buttons). (Rosenberg, Abstract; col. 2, l.
53 − col. 3, l. 36; col. 6, ll. 37-65; col. 13, l. 65 − col. 14, l. 9; Fig. 1.)
2. Figure 14 lists position control host commands and associated
parameters. These commands comprise various position control forces 334
including a groove force. (Rosenberg, col. 37, l. 28 − col. 38, l. 29; Fig. 14.)
3. The groove force as a function of displacement is graphed in
Figure 15 and provides a linear detent sensation along a given degree of
freedom shown by ramps 344. As such, the user object feels like it is
captured in a “groove” via a restoring force centered about a center groove
position “C” to keep the stick in the groove. A deadband DB, however,
allows the user object to move freely near the center groove position. The
magnitude (stiffness) parameter specifies the amount of force or resistance
applied. (Rosenberg, col. 38, l. 38 − col. 39, l. 9; Fig. 15.)
4. The groove force can be applied to a GUI 500. As shown in Figure
20c, an external force of target 559 can be provided as external grooves 561.
When cursor 506 is moved into a groove, the grooves apply resistive forces
to resist movement out of the groove, yet freely allow movement within the
groove. (Rosenberg, col. 57, l. 30 − col. 58, l. 8; Fig. 20c.)
5. In Figure 21, GUI 500 includes a scroll bar 581 with a guide 582 in
which a “thumb” 580 moves via cursor 506. An attractive external force is
associated with the guide so that the cursor is attracted to a field origin point
“N” within thumb 580. Alternatively, the dead region of guide 582 has zero
width so that the cursor is always attracted to a point halfway across the
width of the guide (i.e., the middle line “L”). (Rosenberg, col. 59,
l. 49 − col. 60 l. 23; Fig. 21.)
7
Appeal 2008-4881
Application 10/729,574

6. “Collision” forces can be applied to user object 34 if the cursor
moves over or into certain edges, objects, or regions in GUI 500 that are
designated as “solid” objects. (Rosenberg, col. 60, ll. 46-62; Fig. 21.)
7. Sensations of physical bumps or textures can be applied to user
object 34 as the user moves the cursor across the screen of a GUI to indicate
that the cursor has been positioned within a given region or crossed a
particular boundary. (Rosenberg, col. 44, l. 47 − col. 45, l. 8; Fig. 18.)
8. Cursor 506 can be assigned a mass of its own so that the user
object will feel collision forces in accordance with the mass of cursor 506,
the velocity of the cursor moving across the screen, and an assigned
compliance of the cursor and the object moved into. (Rosenberg, col. 60, ll.
54-58.)
9. Thumb 580 can be assigned inertia forces such that user can feel
the inertia “mass” of the thumb when moving it along guide 582.
(Rosenberg, col. 60, ll. 24-26.)
10. Upon moving the cursor to the desired target, the user maintains
the cursor at the target while providing a “command gesture” associated with
a physical action such as (1) pressing a button; (2) squeezing a trigger; (3)
depressing a pedal; (4) or making some other gesture to command the
execution of a particular operating system function associated with the
graphical object/target. For example, the “click” (press) of a button located
on a mouse or joystick while the cursor is on an icon allows an application
program to execute. Also, clicking this button while the cursor is on part of
a window allows the user to drag the window across the screen by moving
8
Appeal 2008-4881
Application 10/729,574

the user object. The command gesture can also modify forces (e.g., remove
the forces applied in a certain region or target in GUI 500). (Rosenberg, col.
45, l. 61 − col. 46, l. 12.)
11. Users can designate particular magnitudes of forces associated
with targets via a menu command or other standard method. (Rosenberg,
col. 52, ll. 54-59.) Additionally, for safety reasons, users can also override
and deactivate actuators 30 via a safety switch 41. (Rosenberg, col. 13, ll.
37-64.)
12. Appellant’s Specification notes that “[t]he user is provided a
switch which, when actuated, causes the interface to provide an interaction
according to a device fundamental path. The switch can comprise . . . a
defined motion of the input device (e.g., a tight circle motion) . . . .” (Spec.
9:13-17.)

PRINCIPLES OF LAW
Anticipation is established only when a single prior art reference
discloses, expressly or under the principles of inherency, each and every
element of a claimed invention as well as disclosing structure which is
capable of performing the recited functional limitations. RCA Corp. v. Appl.
Dig. Data Sys., Inc., 730 F.2d 1440, 1444 (Fed. Cir. 1984); W.L. Gore &
Assoc., Inc. v. Garlock, Inc., 721 F.2d 1540, 1554 (Fed. Cir. 1983).
“Inherency . . . may not be established by probabilities or possibilities.
The mere fact that a certain thing may result from a given set of circum-
stances is not sufficient.” In re Robertson, 169 F.3d 743, 745 (Fed. Cir.
1999) (citations omitted).

9
Appeal 2008-4881
Application 10/729,574

ANALYSIS
Claims 3, 7, 8, and 10-13
Based on the record before us, we find no error in the Examiner’s
anticipation rejection of representative claim 3 which calls for, in pertinent
part, applying a force to a haptic input device responsive to (a) a component
of haptic input device motion not along the device fundamental path, and (b)
interaction of an object of a computer application with the application.
At the outset, we agree with the Examiner (Ans. 28) that the cursor in
Rosenberg reasonably constitutes an “object” in a computer application as
claimed. Not only does the Specification fail to specifically define the term
“object” as the Examiner indicates, Rosenberg actually uses this very term in
the Abstract in connection with the cursor. See FF 1 (noting that a cursor is
an exemplary “graphical object”). We reach a similar conclusion with
respect to Rosenberg’s scroll bar “thumb” (FF 5) as we see no reason why it,
too, could not function at least as a graphical “object.”
Turning to the rejection, the Examiner equates the groove of
Rosenberg’s “groove force” with the recited “object fundamental path.” The
Examiner also finds that Rosenberg also establishes a “device fundamental
path” associated with the interface device that corresponds to this “object
fundamental path.” (Ans. 5-6.) These findings are undisputed.
As the Examiner indicates (Ans. 27), the claims do not recite that the
force applied to the input device occurs while the user moves the input
device along the device fundamental path. Appellant’s arguments in this
10
Appeal 2008-4881
Application 10/729,574

regard (Br. 13-14) are simply not commensurate with the scope of the claim
which, as the Examiner indicates, merely recites applying force to the input
device responsive to the two conditions (a) and (b) above.
As such, limitation (e) of claim 3 is fully met by Rosenberg’s
applying forces to the input device responsive to a component of device
motion that is not on the device fundamental path (e.g., when the cursor is
not in the groove). See, e.g., FF 6 (applying collision forces if cursor moves
over or into “solid” objects); see also FF 7 (applying bumps or textures to
the user object as the user moves the cursor across the screen).
Moreover, the disputed limitations are fully met by the resisting forces
associated with Rosenberg’s groove force (FF 3-4) that can be applied to a
GUI (FF 4-5). Not only do these implementations apply forces to the input
device responsive to components of forces not along the device fundamental
path (i.e., in a direction away from the groove), but they also apply forces
responsive to object interactions with the application (e.g., moving the
cursor within the groove). See id.
For the foregoing reasons, Appellant has not persuaded us of error in
the Examiner’s rejection of representative claim 3. Therefore, we will
sustain the Examiner’s rejection of that claim, and claims 7, 8, 10, 11, and
13 which fall with claim 3.

Claim 4
We will also sustain the Examiner’s rejection of claim 4 which calls
for, in pertinent part, that the applied forces provide a perception of
momentum and inertia of the input device corresponding to that of the
11
Appeal 2008-4881
Application 10/729,574

object. We agree with the Examiner (Ans. 30-31) that nothing in the claim
precludes the Rosenberg’s collision forces applied to the user object in
accordance with the cursor’s mass and velocity. See FF 8 and 6. Nor does
the claim preclude the user’s feeling inertia mass forces as they move the
thumb along the guide in the scroll bar. See FF 9 and 4.
Although Appellant argues that objects with simulated mass or inertia
move within groove-constrained slider bars (Br. 14-15), Rosenberg’s cursor
does, in fact, move within such a scroll bar. See FF 5. In any event, the
claim is fully met by other motion that is outside scroll bars. See, e.g., FF 8
and 6.
For the foregoing reasons, Appellant has not persuaded us of error in
the Examiner’s rejection of claim 4. Therefore, we will sustain the
Examiner’s rejection of that claim.

Claims 19 and 20
We will also sustain the Examiner’s rejection of representative claim
19 which calls for, in pertinent part, a characteristic of the object being
responsive to input device motion off the device fundamental path. We find
no error in the Examiner’s reliance on Rosenberg’s command gesture
functionality (FF 10) as meeting this limitation. Notably, these command
gestures are associated with motion of the input device (e.g., pressing a
button, squeezing a trigger, depressing a pedal) (Id.) that is off the device
fundamental path.
While these motions execute particular operating system functions as
Appellant indicates (Br. 15), they nonetheless impart distinctive
characteristics to the object (e.g., cursor) associated with these functions—
12
Appeal 2008-4881
Application 10/729,574

even under Appellant’s definitions of the term. For example, when the user
presses a mouse button while the cursor is on an icon, the corresponding
application executes (FF 10) and, as a result, the selection characteristics of
the cursor change commensurate with the newly-launched application. That
is, after launching the new application, the “distinguishing traits, qualities, or
properties” regarding the cursor’s selection capabilities change accordingly.
Likewise, clicking this button while the cursor is on part of a window
allows the user to drag the window across the screen by moving the user
object. (FF 10.) In this instance, the cursor’s target movement
characteristics change after clicking by providing the ability to drag the
underlying window. Additionally, these command gestures can even
remove the very forces applied to the object (Id.): a feature that also would
change its characteristics.
For the foregoing reasons, Appellant has not persuaded us of error in
the Examiner’s rejection of representative claim 19. Therefore, we will
sustain the Examiner’s rejection of that claim, and claim 20 not separately
argued.

Claim 21
We will also sustain the Examiner’s rejection of claim 21 which calls
for, in pertinent part, the magnitude of the force being partially dependent on
the position of the object along the object fundamental path. Although
Appellant argues that Rosenberg does not vary forces as the cursor moves
along the slider bar (Br. 15-16), we agree with the Examiner (Ans. 34) that
Rosenberg centering force applied to the cursor in the scroll bar (FF 5) meets
13
Appeal 2008-4881
Application 10/729,574

this limitation. By attracting the cursor to the guide’s midpoint, the
magnitude of the force would be at least partially dependent on the object’s
position along the object fundamental path.
For the foregoing reasons, Appellant has not persuaded us of error in
the Examiner’s rejection of claim 21. Therefore, we will sustain the
Examiner’s rejection of that claim.

Claim 22
We will also sustain the Examiner’s rejection of claim 22 which calls
for, in pertinent part, the magnitude of the force being partially dependent on
the interaction of the object with the application. We agree with the
Examiner (Ans. 37) that Rosenberg’s varying the force according to the
cursor’s distance from the center of the groove (FF 3 and 5) reasonably
meets this limitation since nothing in the claim precludes the user’s
movement of the cursor itself as “interaction” with the application.
We further note that nothing in the claim precludes the other forces
applied based on various cursor movements (i.e., interactions) (see, e.g., FF
6 and 7), or even command gesture interactions which also can vary the
force magnitude. See FF 10. And as the Examiner indicates (Ans. 38), users
can specify particular magnitudes of forces associated with targets (FF 11)
such that force magnitudes would be partially dependent on object
interaction with those particular targets.
For the foregoing reasons, Appellant has not persuaded us of error in
the Examiner’s rejection of claim 22. Therefore, we will sustain the
Examiner’s rejection of that claim.

14
Appeal 2008-4881
Application 10/729,574

Claim 23
We will also sustain the Examiner’s rejection of claim 23 which calls
for, in pertinent part, the force magnitude being partially dependent on a
user-assistance parameter of the interface. We agree with the Examiner
(Ans. 39-40) that nothing in the claim precludes the recited “user-assistance
parameter” as reading on Rosenberg’s magnitude (stiffness) parameter that
specifies the amount of force or resistance applied in connection with the
groove force. See FF 3. We see no reason why the force magnitude would
not at least partially depend on this parameter—a dependence all but
indicated by its very title. We further see no reason why this parameter (or
any of the other parameters associated with the position control commands
and forces (FF 2)) would not constitute a “user-assistance” parameter. The
Examiner’s point in this regard (Ans. 41-42) is well taken.
For the foregoing reasons, Appellant has not persuaded us of error in
the Examiner’s rejection of claim 23. Therefore, we will sustain the
Examiner’s rejection of that claim.

Claim 24
We will also sustain the Examiner’s rejection of claim 24 which calls
for, in pertinent part, establishing the user-assistance parameter by a measure
of the user’s proficiency in manipulating the input device. Rosenberg’s
system is certainly designed to assist users of varying proficiency levels
effectively use GUIs via force feedback as the Examiner indicates (Ans. 45).
And while users could increase the force feedback magnitude parameter if
they needed more assistance as the Examiner surmises (see FF 3 and 11),
15
Appeal 2008-4881
Application 10/729,574

there is simply nothing in Rosenberg that such an adjustment was actually
performed based on such user proficiency considerations, let alone a
measure of the user’s proficiency as claimed.
Although the Examiner makes a point regarding what skilled artisans
would understand regarding the relationship between the magnitude
parameter and a user’s skill level (Ans. 45), this line of reasoning is, at best,
based on an inference that does not necessarily follow from the teachings
within the four corners of the Rosenberg reference. It is well settled that
“[i]nherency . . . may not be established by probabilities or possibilities. The
mere fact that a certain thing may result from a given set of circumstances is
not sufficient.” Robertson, 169 F.3d at 745 (citations omitted). As such,
even if it were probable that less skilled users (i.e., users needing more
assistance) would have increased the magnitude parameter as the Examiner
seems to suggest, that alone would not be enough for anticipation. Rather,
such a relationship between the magnitude parameter and a measure of user
proficiency must be necessarily present in the reference. It is not.
Nevertheless, the scope of claim 24 does not preclude “establishing” a
user-assistance parameter by selectively activating or deactivating
Rosenberg’s force feedback feature via a safety switch. See FF 11. By
overriding the force feedback feature, a “user-assistance parameter” would,
in effect, be established by a measure of the user’s perceived proficiency in
manipulating the input device, at least during adverse or unsafe conditions.
See id.
Moreover, the force characteristics of the groove force itself (FF 3)
would, in effect, establish a “user-assistance parameter” by a measure of the
user’s proficiency in manipulating the input device in view of the deadband
16
Appeal 2008-4881
Application 10/729,574

DB that allows free movement near the center groove position, as well as the
gradual increasing of force away from the center. See id. These zones of
force were not arbitrarily created, but are rather based on a user’s
proficiency in manipulation of an object within a groove—a feature that
ultimately assists the user to that end. See Rosenberg, col. 1, l. 65 – col. 2, l.
59.
For the foregoing reasons, Appellant has not persuaded us of error in
the Examiner’s rejection of claim 24. Therefore, we will sustain the
Examiner’s rejection of that claim.

Claims 26, 27, and 30
We will also sustain the Examiner’s rejection of representative claim
26 which calls for, in pertinent part, establishing a device fundamental path
by (1) determining when the user supplies a motion-initiation signal, and
then (2) establishing the device fundamental path according to a defined
device path and a cursor position when the motion-initiation signal was
supplied.
We see no error in the Examiner’s position (Ans. 46-47) that the
motion-initiation signal is supplied when the user positions the cursor within
a region associated with a groove force (e.g., a scroll bar). See FF 3-5. In
this case, the groove force (i.e., the device fundamental path) would not only
be established after moving the cursor within the target, but would also be
established according to the cursor’s position.

For the foregoing reasons, Appellant has not persuaded us of error in
the Examiner’s rejection of representative claim 26. Therefore, we will
17
Appeal 2008-4881
Application 10/729,574

sustain the Examiner’s rejection of that claim, and claims 27 and 30 not
separately argued.

Claims 28 and 29
We will also sustain the Examiner’s rejection of representative claim
28 which calls for the motion-initiation signal to comprise a switch actuated
by the user. We see no error in the Examiner’s position (Ans. 49-50) that
positioning the cursor within a region associated with a groove force (e.g., a
scroll bar) constitutes actuating a “switch.” This conclusion is consistent
with Appellant’s Specification which expressly states that a switch can
include defined motion of an input device. (FF 12.) As we indicated above
in connection with claim 26, such positioning of the cursor would cause a
motion-initiation signal to be supplied and therefore, in effect, function as a
“switch” by virtue of this motion.
For the foregoing reasons, Appellant has not persuaded us of error in
the Examiner’s rejection of representative claim 28. Therefore, we will
sustain the Examiner’s rejection of that claim, and claim 29 not separately
argued.

THE ANTICIPATION REJECTION OVER STEWART
Regarding claim 3, the Examiner finds that the displayed geometric
model in Stewart corresponds to the “object fundamental path” since it
designates a path that the input device must follow to “feel” the geometric
18
Appeal 2008-4881
Application 10/729,574

object that the model represents. Additionally, the model’s “virtual surface”
is said to correspond to the “device fundamental path,”5 and that force is
applied to the input device to constrain the user’s hand onto the virtual
surface. The Examiner also notes that force feedback is also applied when
the user modifies the geometric model. (Ans. 10-11.)
Appellant argues that claim 3 is limited to an interface that allows
distinct object path and object interaction, but the Examiner’s relies on a
different mode of operation (i.e., the modification mode) for this feature.
(Br. 19-20.)
The Examiner responds that claim 3 does not require applying the
forces in steps (e) and (f) simultaneously. As such, the Examiner reasons,
the claim does not preclude the user switching from the browsing mode to
the modification mode, and the respective forces applied in each mode.
(Ans. 51-53.)
The issue before us, then, is as follows:

ISSUE
Has Appellant shown that the Examiner erred in finding that Stewart
applies a force to a haptic input device responsive to (1) a component of the
input device motion not along the device fundamental path, and (2)
interaction of the object with the application in rejecting representative claim
3 under § 102?

5 But see Br. 10 (characterizing the Examiner’s reliance on Stewart’s virtual
surface as corresponding to the “object path” in claim 3).
19
Appeal 2008-4881
Application 10/729,574

FINDINGS OF FACT
The record supports the following additional findings of fact (FF) by a
preponderance of the evidence:
13. Stewart discloses a system 10 for interactively evaluating and
manipulating a geometric model 22 (e.g., a vehicle design) via a haptic
interface 12 operatively coupled to a computer system 16. The haptic
interface includes a haptic end effector device 18 (e.g., a stylus, pen, or
similar gripping device) that transmits information between a user 14 and a
geometric model as the user browses or edits the model’s surface using the
end effector device. (Stewart, Abstract; col. 4, ll. 22-45; Figs. 1-3.)
14. The haptic interface 12 controls position, orientation, and force
feedback between the user 14, computer system 16, and the geometric
model. To this end, an active force is applied to the end effector device 18
to constrain the user’s hand onto a virtual surface 20 representing a surface
of the geometric model 22. By forcing the end effector device 18 to stick to
the virtual surface 20 representing the geometric model, the user receives
tactile information to enable the user to explore, feel, and manipulate the
geometric model and assess design quality. (Stewart, col. 4, ll. 25-28; col. 5,
l. 62 − col. 7, l. 13; Figs 1-4A (Step 115 (browse mode)).)
15. Similarly, forcing the end effector device 18 to stick to the virtual
surface 20 enables the user to edit the surface via haptic sculpting. To this
end, the user can enter the edit mode via an activation signal. (Stewart, col.
4, ll. 54-65; col. 7, ll. 14-38; Fig 4A (Step 120).)
16. Figure 4C details a methodology for editing a fixed point in
parametric space on virtual surface 20. In Step 235, force feedback is
computed representing the stick-to-surface/stick-to-pen force and surface
20
Appeal 2008-4881
Application 10/729,574

property feedback force. In Step 240, these forces are added together and
applied to end effector device 18. (Stewart, col. 10, ll. 9-55; Fig. 4C.)

ANALYSIS
Claim 3
We will sustain the Examiner’s anticipation rejection of claim 3 based
on Stewart. As the Examiner indicates (Ans. 51-53), the scope of claim 3
does not preclude applying forces in the browse and edit modes,
respectively. As such, we agree with the Examiner that Stewart’s system
applies a force to an input device responsive to a component of the input
device motion not along the device fundamental path while browsing. See
FF 13-14. We also find no error in the Examiner’s position that Stewart also
applies a force to the input device responsive interaction of the object with
the application in the edit mode—a mode that would involve interaction of
the object with the application. See FF 15-16. Appellant’s argument that
Stewart’s object path also serves as the interaction with the application (Br.
20) is simply not commensurate with the scope of the claim.
For the foregoing reasons, Appellant has not persuaded us of error in
the Examiner’s anticipation rejection of claim 3 over Stewart. Therefore, we
will sustain the Examiner’s rejection of that claim.

Claims 5 and 6
We will also sustain the Examiner’s rejection of claims 5 and 6 calling
for the shape of the object and device fundamental paths to be dependent on
the application state, respectively. We agree with the Examiner (Ans.
21
Appeal 2008-4881
Application 10/729,574

53-55) that nothing in the claim precludes Stewart’s editing mode that would
yield a particular “application state” from a number of such states. As such,
the shape of the object fundamental path (and its corresponding device
fundamental path) in a subsequent browsing session would therefore be
dependent on this particular application state. See FF 13-16.
For the foregoing reasons, Appellant has not persuaded us of error in
the Examiner’s anticipation rejection of claims 5 and 6 over Stewart.
Therefore, we will sustain the Examiner’s rejection of those claims.

Claims 35 and 36
We will also sustain the Examiner’s rejection of claim 35. Although
this claim is broader than claim 3 in that only one applied force (not two) is
recited, claim 35 nonetheless calls for a computer presentation of the
interaction of objects simulating physical objects in the preamble.
We agree with the Examiner (Ans. 55-56) that nothing in the claim
precludes Stewart’s displayed representation of the input device as
corresponding to a simulated physical object in addition to the virtual
surface. As the Examiner indicates (Ans. 56), this simulated physical object
moves along a defined object path (the virtual surface). We see no error in
this position.
For the foregoing reasons, Appellant has not persuaded us of error in
the Examiner’s anticipation rejection of claim 35 over Stewart. Therefore,
we will sustain the Examiner’s rejection of that claim, and claim 36 not
separately argued.

22
Appeal 2008-4881
Application 10/729,574

THE OBVIOUSNESS REJECTIONS OF CLAIMS 9, 10, 14-18, 25, AND 34
Claim 9
In rejecting claim 9, the Examiner relies on Frid-Nielsen for teaching
modifying a cursor’s appearance when it is placed on a scroll bar to indicate
valid inputs, and concludes that the claim would have obvious over the
collective teachings of Rosenberg and Frid-Nielsen. (Ans. 11-13.)
Appellant argues that not only does Frid-Nielsen fail to mention force
feedback, neither cited reference changes an object’s visual representation
based on its on-path or off-path condition as claimed. Appellant adds that
there is no suggestion to combine the references as the Examiner proposes
since, among other things, Frid-Nielsen does not suggest any force feedback,
and Rosenberg fails to suggest combining its force feedback teachings with
different cursor representation techniques. (Br. 22-23.)

Claim 10
In rejecting claim 10, the Examiner relies on Bertram for teaching
plural scroll bars, each having a thumb, and concludes that it would have
been obvious to modify Rosenberg’s interface to have plural scroll bars (i.e.,
an additional object fundamental path) with a groove associated therewith.
(Ans. 13-14.)
Appellant argues that the proposed combination does not teach or
suggest two objects, but rather an interface with two scrollbars into which a
user could move a single cursor. (Br. 23.) The Examiner, however,
considers a scroll bar “thumb” to be an object, and, as such, the interface
would have two object paths (two scroll bars) and two objects—one within
each scroll bar. (Ans. 60-61.)
23
Appeal 2008-4881
Application 10/729,574

Claims 14 and 25
In rejecting representative claim 14, the Examiner relies on Rosenberg
II for teaching applying a force to an input device to urge it towards a
particular region of the device’s range of motion (i.e., the “starting region”)
and combines this teaching with Rosenberg in concluding the claim would
have been obvious. (Ans. 15-16.)
Appellant argues that Rosenberg II does not teach a “starting region”
based on an object path, but rather centers the mouse to accommodate
subsequent motion in any direction. (Br. 24.) The Examiner, however,
notes that Rosenberg II’s automatic centering process effectively urges the
mouse to a starting region (i.e., the center). The Examiner adds that
Rosenberg II also urges the input device away from the edge of its range of
motion (i.e., within an “isometric” region) and towards the “isotonic” region.
(Ans. 61-65.)

Claim 15
In rejecting claim 15, the Examiner relies on Gould for teaching that
the path that the cursor moves on a screen (i.e., the “object fundamental
path”) can have a different shape than the path in which the interface device
moves (the “device fundamental path”).
Appellant argues that, unlike Rosenberg, Gould teaches an alternative
to force feedback—not force feedback. As such, Appellant contends, not
only does Gould fail to cure the deficiencies of Rosenberg, but combining
the references as proposed would actually render Gould unsatisfactory for its
intended purpose as an alternative to force feedback. (Br. 25.)
24
Appeal 2008-4881
Application 10/729,574

The Examiner, however, responds that not only is it known to
combine virtual and actual force feedback, doing so would neither render
Rosenberg nor Gould unsatisfactory for their intended purposes. Rather, the
combination is based on adding virtual force feedback to Rosenberg’s use of
actual force feedback and, as such, the cited prior art does not teach away
from its combination. (Ans. 65-67.)

Claims 16-18
Regarding representative claim 16, the Examiner cites Shih for
teaching a two- or three-dimensional “object fundamental path” (i.e., a
geometric constraint) that represents the path of a virtual tool object that is
moved by an interface device in a “device fundamental path” in concluding
the claim would have been obvious. (Ans. 18-20.)
Although Appellant acknowledges that Shih’s virtual sculpting
application would allow moving a tool in three dimensions, Appellant
nonetheless argues that, by its very nature, Shih’s virtual sculpting
application requires the device and object paths to be same shape to effect
sculpting. As such, Appellant contends, the proposed combination to yield
differently-shaped paths would destroy Shih’s utility for its intended
purpose. (Br. 25-26.)
The Examiner, however, contends that since the virtual object to be
sculpted and the representation of the input device are displayed to the user,
the user can simply observe the tool’s relation to the virtual object rather
than relying solely on force feedback. In any event, the Examiner notes that
25
Appeal 2008-4881
Application 10/729,574

force feedback may nonetheless be applied to the input device that would
result in a differently shaped path than the displayed object fundamental
path. (Ans. 67-68.)

Claim 34
Regarding claim 34, Appellant argues that the groove interaction in
Rosenberg requires that the paths have the same shape and a one-to-one
correspondence. Appellants add Gould discloses differently shaped paths
are an alternative to force feedback, and the non-one-to-one correspondence
in Rosenberg II pertains to motion that not based on any object or device
paths and that combining the references as proposed would change the basic
principle of operation of the cited references. (Br. 26-27.)
The Examiner, however, notes that Rosenberg need not have a one-to-
one correspondence between the motion of the cursor and the input device,
and reiterates that merely adding virtual force feedback of Gould to
Rosenberg would not defeat the intended purpose of the prior art. The
Examiner adds that Rosenberg II’s non-one-to-one correspondence may
accommodate any input device motion, not just motion based on object or
device paths. (Ans. 68-70.)

THE OBVIOUSNESS REJECTION OVER MEREDITH AND ROSENBERG
Regarding claim 32, the Examiner finds that Meredith’s computerized
pool cue and control system discloses all of the claimed subject matter
except for applying for the input device responsive to a component of input
26
Appeal 2008-4881
Application 10/729,574

device motion not along the device fundamental path, and relies on
Rosenberg for this teaching in concluding the claim would have been
obvious. (Ans. 21-22.)
Appellant argues that Meredith does not teach force feedback, let
alone that the interface affects pool cue motion. Rather, Appellant contends,
users in Meredith freely manipulate a real pool cue that would be replaced
with simulated groove forces if modified as proposed. According to
Appellant, such a modification would change Meredith’s principle of
operation and render it unsuitable for its intended purpose. (Br. 27.)
The Examiner, however, disagrees and contends that adding force
feedback to Meredith would still allow the user to freely move the pool cue.
(Ans. 71.)

THE OBVIOUSNESS REJECTION OVER BAYNTON, ROSENBERG, AND MEREDITH
Claim 31
Regarding claim 31, the Examiner finds that Baynton discloses a
computerized golf swing apparatus with force feedback, but does not
disclose establishing an object fundamental path where (1) the object
comprises a golf club, and (2) the object moves along the object fundamental
path responsive to a component of input device motion along the device
fundamental path. The Examiner, however, relies on Rosenberg and
Meredith for these teachings in concluding the claim would have been
obvious. (Ans. 22-24.)
Appellant argues that Baynton is a golf swing training apparatus that
not only lacks force feedback based on object interaction in a computer
application, but also lacks any suggestion of golf simulations or computer
27
Appeal 2008-4881
Application 10/729,574

games. According to Appellant, the proposed modification to Baynton
would require redesigning Baynton to sense golf club motion rather than
control its motion—a completely different purpose. (Br. 27-28.)
The Examiner responds that Rosenberg and Meredith collectively
teach (1) applying a pool cue input to a computer; (2) applying force
feedback; and (3) providing interaction of a simulated pool cue with other
objects. According to the Examiner, Baynton shows that golf—like pool—
is a popular activity and that skilled artisans could have likewise applied the
teachings of Rosenberg and Meredith to a golf club such as that disclosed by
Baynton. (Ans. 71-72.)

Claim 38
Regarding claim 38, Appellant argues that Baynton cannot be
modified to have differently-shaped object and device paths as the Examiner
proposes if it is to function for its intended purpose, namely to teach correct
golf swings to athletes. (Br. 28.) The Examiner, however, notes that
Rosenberg teaches such a feature and that such a modification is therefore
suggested by the cited prior art. (Ans. 73.)

THE OBVIOUSNESS REJECTION OVER STEWART AND ROSENBERG II
Regarding claim 37, the Examiner finds that Stewart discloses all of
the claimed subject matter except for moving the object and applying force
responsive to a signal indicating that the user desires such interaction. The
Examiner, however, relies on Rosenberg II’s application of force feedback
responsive to detecting a user-activated safety switch in concluding the
feature would have been obvious. (Ans. 24-25, 74.)
28
Appeal 2008-4881
Application 10/729,574

Appellant argues that Rosenberg II’s safety switch disables all haptic
interaction—a result that differs from the claimed signal that disables forces
specific to the path-based motion. (Br. 29.)
The issues before us, then, are as follows:

ISSUES
(1) Has Appellant shown that the Examiner erred in finding that
Rosenberg and Frid-Nielsen collectively teach or suggest a perceptively
different visual representation of the haptic input device depending on
whether the input device is on the device fundamental path in rejecting claim
9 under § 103?
(2) Has Appellant shown that the Examiner erred in finding that
Rosenberg and Bertram collectively teach or suggest establishing (1) a
second object fundamental path representing a path of motion of a second
object in the computer application, and (2) a second device fundamental path
corresponding to the second object fundamental path in rejecting claim 10
under § 103?
(3) Has Appellant shown that the Examiner erred in finding that
Rosenberg and Rosenberg II collectively teach or suggest applying a force to
the input device to urge it to a starting region of its range of motion in
rejecting claim 14 under § 103?
(4) Has Appellant shown that the Examiner erred in finding that
Rosenberg and Gould collectively teach or suggest an object fundamental
path and device fundamental path with different shapes in rejecting claim 14
under § 103?
29
Appeal 2008-4881
Application 10/729,574

(5) Has Appellant shown that the prior art cited to reject claim 14
teaches away from its combination?
(6) Has Appellant shown that combining the references to arrive at the
invention of claims 15, 32, 34, and 38 would render the prior art unsuitable
for its intended purpose?
(7) Has Appellant shown that the Examiner erred in finding that
Baynton, Rosenberg, and Meredith collectively teach a golf simulation
where the object fundamental path comprises a path suited for perception of
the swing of a golf club in rejecting claim 31 under § 103?
(8) Has Appellant shown that the Examiner erred in finding that
Stewart and Rosenberg II collectively teach moving the object in accordance
with a signal indicating that the user desires path interaction in rejecting
claim 37 under § 103?
(9) Is the Examiner’s reason to combine the teachings of these
references supported by articulated reasoning with some rational
underpinning to justify the Examiner’s obviousness conclusion?

FINDINGS OF FACT
The record supports the following additional findings of fact (FF) by a
preponderance of the evidence:
17. Frid-Nielsen discloses an “intelligent screen cursor” 275 that
indicates the valid inputs of a pointing device at all times. Unlike the cursor
225 shown in Figure 5A, the intelligent cursor 275 shown in Figure 5B
graphically displays the valid inputs available to the user at various points
(1)-(4) along its path of movement with respect to scroll bar 217. (Frid-
Nielsen, col. 8, ll. 30-48; Figs. 5A and 5B).
30
Appeal 2008-4881
Application 10/729,574

18. Bertram in Figure 3 shows a GUI window 60 with vertical and
horizontal scroll bars 64 and 70. Each scroll bar has an associated slider 62
and 68, respectively. (Bertram, col. 7, l. 1 − col. 8, l. 16; Fig. 3.)
19. Rosenberg II discloses a system with enhanced cursor control and
force feedback in a mouse device 10 using an indexing feature. In one
implementation, an indexing force (i.e., a resistive spring or restoring force)
opposes the mouse’s motion from an “isotonic” region 443 to an “isometric”
region 442.6 The magnitude and direction of the isometric spring force is
based on the mouse position within the isometric region 442. (Rosenberg II,
col. 29, ll. 24-47; col. 31, l. 44 − col. 32, l. 32; Figs. 5 and 9.)
20. Rosenberg II discloses an “auto centering” feature that uses the
actuators 64 of the force feedback mouse to automatically reduce the offset
between the local and display frames. To this end, the mouse is
automatically moved to a location in the local frame corresponding to the
center of the display frame. Alternatively, the user can auto center the
mouse via a special button or switch. (Rosenberg II, col. 37, ll. 44-61.)
21. Gould discloses a virtual force feedback system that alters the
motion of a cursor as it travels. As shown in Figure 10A, the motion of
pointing device 12 is linear, but the motion of cursor 180 does not match this
linear path in the vicinity of attractive forces 172 associated with oval 174.
(Gould, col. 6, ll. 44-62; col. 16, ll. 48-60; Fig. 10A.)

6 An “isometric region” borders an “isometric limit” which is a physical
limit to the mouse workspace (e.g., a relatively small edge region). See
Rosenberg II, col. 30, l. 62 − col. 31, l. 3; see also id. at col. 31, ll. 22-30.
An “isotonic” region, however, borders an isometric region and allows
normal isotonic mouse positioning. See id. at col. 31, ll. 24-26.
31
Appeal 2008-4881
Application 10/729,574

22. Shih discloses a haptic virtual environment including a virtual
object 26 (e.g., a three-dimensional block) and a virtual tool 28 that is
manipulated via a haptic interface device 10. Unless the haptic rendering
process 16 uses the virtual tool 28 to remove material from the virtual object
26, then the rendering process does not allow the virtual tool to penetrate the
virtual object. (Shih, col. 8, l. 21 − col. 9, l. 23; Figs. 2A and 2B.)
23. Meredith discloses a computerized pool cue 12 and controller 12
comprising a control ring 52 and case 32. During operation, the user fits the
cue through the controller’s control ring 52 and maneuvers the case to a
desired position. Movement of the case and the cue is observed on a display
device 122 (e.g., a computer screen) with respect to game balls on the
screen. By actually seeing the cue move on the screen, the user can, among
other things, align the cue’s tip 18 with respect to the ball to be hit.
(Meredith, col. 3, ll. 7-16; col. 5, ll. 15-30; Figs. 1, 2, 4, and 12.)
24. Baynton discloses a golf swing training and correction system
comprising a computer-controlled mechanical manipulation mechanism for
the controlling the position and orientation of a golf club 1000 within a
predetermined swing path geometry. The computer control system monitors
the coordinates associated with the golf club position throughout the entire
swing path and applies an appropriate corrective force to the golf club.
(Baynton, Abstract; col. 5, ll. 63-67; col. 6, l. 54 − col. 7, l. 7; col. 12, ll. 23-
33; Figs. 1 and 10.)
25. Baynton’s control system includes a PC with a GUI designed to,
among other things, display information about both the calculated swing and
the actual swing. (Baynton, col. 11, ll. 7-12.)
32
Appeal 2008-4881
Application 10/729,574

26. The signals provided to each position control actuator in Baynton
could be used to provide visual feedback such as a display on a CRT.
(Baynton, col. 12, ll. 34-36.)
27. In Rosenberg, the position of cursor 506 is directly related to the
position of user object 34. Thus, when cursor 506 is moved left on screen
20, user object is moving in a corresponding direction. But the distance that
user object 34 moves may not be the same distance that cursor moves on the
screen, and is typically related by a predetermined function. (Rosenberg,
col. 48, ll. 56-64.)
28. In Rosenberg II, safety or “deadman” switch 150 is included in
the interface device to (1) enable the user to override and deactivate
actuators 64, or (2) require a user to activate actuators 64. Safety switch 150
is coupled to actuators 64 such that the user must continually activate or
close the safety switch during manipulation of the mouse to activate the
actuators 64. If the safety switch is deactivated, power is cut to the
actuators. (Rosenberg II, col. 17, ll. 47-67; Fig. 4.)

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) (noting that 35 U.S.C.
§ 103 leads to three basic factual inquiries: (1) the scope and content of the
33
Appeal 2008-4881
Application 10/729,574

prior art; (2) the differences between the prior art and the claims at issue;
and (3) the level of ordinary skill in the art). Furthermore, the Examiner’s
obviousness rejection must be based on
“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
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.

KSR Int’l v. Teleflex, Inc., 550 U.S. 398, 418 (2007) (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
Appellant 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).
If the Examiner’s proposed modification renders the prior art
unsatisfactory for its intended purpose, the Examiner has failed to make a
prima facie case of obviousness. See In re Gordon, 733 F.2d 900, 902 (Fed.
Cir. 1984).

ANALYSIS
Claim 9
We will sustain the Examiner’s rejection of claim 9 which calls for a
perceptively different visual representation of the haptic input device
depending on whether the input device is on the device fundamental path.
We see no error in the Examiner’s position (Ans. 56-60) that skilled artisans
34
Appeal 2008-4881
Application 10/729,574

would have combined Frid-Nielsen’s teaching of changing the visual
representation of a cursor in accordance with its position with respect to a
scroll bar (see FF 17) with Rosenberg—a reference which teaches providing
force feedback associated with a cursor’s position with respect to a scroll bar
as we noted previously. See FF 4-5. Providing visual feedback to users to
inform them of available input device functions as taught by Frid-Nielsen in
conjunction with the force feedback system of Rosenberg is tantamount to
the predictable use of prior art elements according to their established
functions—an obvious improvement. See KSR, 550 U.S. at 417. The
Examiner’s reason to combine the teachings of these references is therefore
supported by articulated reasoning with some rational underpinning to
justify the Examiner’s obviousness conclusion.
For the foregoing reasons, Appellant has not persuaded us of error in
the Examiner’s obviousness rejection of claim 9. Therefore, we will sustain
the Examiner’s rejection of that claim.

Claim 10
We will sustain the Examiner’s obviousness rejection of claim 10
calling for, in pertinent part, establishing (1) a second object fundamental
path representing a path of motion of a second object in the computer
application, and (2) a second device fundamental path corresponding to the
second object fundamental path. We see no error in the Examiner’s position
(Ans. 60-61) that combining Bertram’s teaching of plural scroll bars with
associated sliders (FF 18) with Rosenberg’s interface would provide two
object fundamental paths (and associated device fundamental paths). We
see no reason why each scroll bar’s “thumb” could not constitute a separate
35
Appeal 2008-4881
Application 10/729,574

object as the Examiner indicates (Ans. 60-61). Providing such functionality
in Rosenberg is tantamount to the predictable use of prior art elements
according to their established functions—an obvious improvement. See
KSR, 550 U.S. at 417. The Examiner’s reason to combine the teachings of
these references is therefore supported by articulated reasoning with some
rational underpinning to justify the Examiner’s obviousness conclusion.
For the foregoing reasons, Appellant has not persuaded us of error in
the Examiner’s obviousness rejection of claim 10. Therefore, we will
sustain the Examiner’s rejection of that claim.

Claims 14 and 25
We will sustain the Examiner’s rejection of claim 14 which calls for,
in pertinent part, applying a force to the input device to urge it to a starting
region of its range of motion. We agree with the Examiner (Ans. 61-64) that
Rosenberg II’s automatically centering the mouse (FF 20) at least suggests
urging the mouse to a starting region of its range of motion, namely the
center. As the Examiner indicates (Ans. 63), Appellant’s argument that
Rosenberg II does not determine a starting region based on object path (Br.
24) is not commensurate with the scope of the claim. Additionally, the
Examiner’s point (Ans. 63-64) that motion of the input device along the
device fundamental path starting from this “starting region” should not
generally require motion outside its range of motion is well taken. We reach
a similar conclusion with respect to Rosenberg II’s urging the mouse from
an isometric region to an isotonic region. See FF 19.

36
Appeal 2008-4881
Application 10/729,574

For the foregoing reasons, Appellant has not persuaded us of error in
the Examiner’s obviousness rejection of claim 14. Therefore, we will
sustain the Examiner’s rejection of that claim, and claim 25 not separately
argued.

Claim 15
We will also sustain the Examiner’s rejection of claim 15 which calls
for the object fundamental path and device fundamental path to have
different shapes. At the outset, we note that nothing in the claim precludes a
joystick implementation of Rosenberg (see FF 1). In this case, the device
fundamental path in Rosenberg would be, at least in part, curved or arc-
shaped in multiple dimensions, and would have a different shape than the
object fundamental path represented on the display. Since Rosenberg itself
meets claim 15, the Examiner’s reliance on Gould is merely cumulative.
Nevertheless, even assuming, without deciding, that Gould’s virtual
force feedback (FF 21) is an alternative to actual force feedback (such as that
disclosed by Rosenberg) as Appellant seems to suggest (Br. 25), we see no
reason why such a virtual force feedback could not function as an adjunct to
Rosenberg’s actual force feedback system as the Examiner indicates (Ans.
66). Such an enhancement would be tantamount to the predictable use of
prior art elements according to their established functions—an obvious
improvement. See KSR, 550 U.S. at 417.
We recognize, however, that if the Examiner’s proposed modification
renders the prior art unsatisfactory for its intended purpose, the Examiner
has failed to make a prima facie case of obviousness. See Gordon, 733 F.2d
at 902. But we see no reason why merely adding a virtual force feedback
37
Appeal 2008-4881
Application 10/729,574

feature as suggested by Gould to Rosenberg’s actual force feedback as the
Examiner proposes would not render either reference unsuitable for its
intended purpose. The Examiner’s reason to combine the teachings of these
references is therefore supported by articulated reasoning with some rational
underpinning to justify the Examiner’s obviousness conclusion.
For the foregoing reasons, Appellant has not persuaded us of error in
the Examiner’s obviousness rejection of claim 15. Therefore, we will
sustain the Examiner’s rejection of that claim.

Claims 16-18
We will also sustain the Examiner’s rejection of representative claim
16 which calls for the device fundamental path to define a curve in three
dimensions. As we noted above, we find that Rosenberg’s joystick
implementation (FF 1) also fully meets this limitation.
Nevertheless, Appellant acknowledges (Br. 26) that Shih allows
moving the virtual tool in three dimensions. And as the Examiner indicates
(Ans. 67-68), movement of the displayed virtual tool may not exactly match
that of the input device (see FF 22) and therefore force feedback could be
used in such a condition. Furthermore, sculpting is not the only technique
envisioned by Shih’s haptic rendering process. See id. As such, we fail to
see how applying these techniques to the other cited prior art would render
the prior art unsuitable for its intended purpose.
For the foregoing reasons, Appellant has not persuaded us of error in
the Examiner’s obviousness rejection of claim 16. Therefore, we will
sustain the Examiner’s rejection of that claim, and claims 17 and 18 not
separately argued.
38
Appeal 2008-4881
Application 10/729,574

Claim 34
We will also sustain the Examiner’s rejection of claim 34 which calls
for the correspondence between the device fundamental path and the object
fundamental path to not be one-to-one. At the outset, we note that nothing
in the claim precludes a joystick implementation of Rosenberg (see FF 1) as
we noted previously. As such, the device fundamental path established by
the joystick would, at least in part have a different shape than the object
fundamental path represented on the display and therefore not have a one-to-
one correspondence. Since Rosenberg itself meets claim 34, the Examiner’s
reliance on Gould is merely cumulative.
Nevertheless, as we indicated previously, we see no error in
combining Gould with Rosenberg to provide virtual force feedback as an
adjunct to actual force feedback. And we further see no error in the
Examiner’s position (Ans. 69-70) that skilled artisans could combine
Gould’s teaching of differently-shaped paths with Rosenberg. As the
Examiner indicates, while Rosenberg’s groove is intended to aid the user in
moving a cursor along an interface object (e.g., a scrollbar) (FF 3-5), we are
not persuaded that providing differently-shaped paths in Gould would
somehow destroy Rosenberg’s functionality. Rather, such an enhancement
is tantamount to the predictable use of prior art elements according to their
established functions—an obvious improvement. See KSR, 550 U.S. at 417.
For the foregoing reasons, Appellant has not persuaded us of error in
the Examiner’s obviousness rejection of claim 34. Therefore, we will
sustain the Examiner’s rejection of that claim.

39
Appeal 2008-4881
Application 10/729,574

Claim 32
We will also sustain the Examiner’s rejection of claim 32 which calls
for, in pertinent part, a pool simulation application where the object
fundamental path comprises a path suited for perception of a the motion of a
pool cue. We disagree with Appellant’s contention (Br. 27) that adding
force feedback to Meredith’s pool simulation would render it unsuitable for
its intended purpose. To the contrary, we see no reason why adding force
feedback such as that suggested by Rosenberg would not still allow the user
in Meredith to use the pool cue in that system as the Examiner indicates
(Ans. 71). See FF 23. While there may be some resistance forces applied
via this modification, the user would still be able to use the cue, its
associated controller, and the simulation. Indeed, applying feedback forces
in Meredith’s system may actually cause the user to play better as the
applied forces could function as a training aid. See id.
For the foregoing reasons, Appellant has not persuaded us of error in
the Examiner’s obviousness rejection of claim 32. Therefore, we will
sustain the Examiner’s rejection of that claim.

Claim 31
We will also sustain the Examiner’s obviousness rejection of claim 31
based on the collective teachings of Baynton, Rosenberg, and Meredith. It is
true that Baynton’s mechanical golf training system differs from that of
Meredith’s display-based simulation. Compare FF 23 with FF 24.
Nevertheless, we see no reason why the pool simulation system with force
feedback of the Meredith/Rosenberg system could not be adapted for golf
simulations in light of Baynton as the Examiner suggests (Ans. 72).
40
Appeal 2008-4881
Application 10/729,574

Not only do both games involve striking a ball with a handheld
implement to direct the ball in a desired direction and velocity, Baynton all
but suggests that visual feedback is used in addition to force feedback as part
of the golf training system. For example, Baynton notes that a GUI on the
control system’s PC can display information about the actual swing (FF 25),
and control actuator signals can be used to provide visual feedback on a
display (FF 26). Since (1) Baynton envisions visually displaying aspects of
the golf swing in addition to applying force feedback, and (2) in view of the
similar fundamental aspects of pool and golf (e.g., hitting a ball with a
handheld implement to direct it in a desired direction and velocity), we see
no reason why the Meredith/Rosenberg system could not be adapted for golf
simulations in lieu of, or in addition to, pool simulations.
For the foregoing reasons, Appellant has not persuaded us of error in
the Examiner’s obviousness rejection of claim 31. Therefore, we will
sustain the Examiner’s rejection of that claim.

Claim 38
Regarding claim 38, we are not persuaded of error in the Examiner’s
position (Ans. 73) that Baynton, Rosenberg, and Meredith collectively teach
a golf simulation apparatus where the object and device fundamental paths
have different shapes. As we indicated previously with respect to claim 31,
we see no reason why the Meredith/Rosenberg system could not be adapted
for golf simulations in lieu of, or in addition to, pool simulations. Such a
modification would hardly render this system unsuitable for its intended
purpose. And as the Examiner indicates (Ans. 73), Rosenberg at least
suggests providing differently-shaped object and device paths based, in part,
41
Appeal 2008-4881
Application 10/729,574

on the disparity between the relative movement of the cursor and user object.
See FF 27. As such, we find that the cited prior art collectively teaches and
suggests all limitations of claim 38.
For the foregoing reasons, Appellant has not persuaded us of error in
the Examiner’s obviousness rejection of claim 38. Therefore, we will
sustain the Examiner’s rejection of that claim.

Claim 37
We will also sustain the Examiner’s obviousness rejection of claim 37
which calls for, in pertinent part, moving the object in accordance with a
signal indicating that the user desires path interaction. We see no error in
the Examiner’s reliance (Ans. 74) on the user’s activation of the safety
switch in Rosenberg II (FF 28) for this feature. We agree with the Examiner
that since the user must affirmatively actuate the safety switch to activate the
actuators to effect force feedback, then a signal would be accepted indicating
that path interaction was desired. Even if we assume, without deciding, that
this safety switch disables all haptic interaction as Appellant argues (Br. 29),
the scope of the claim simply does not preclude this feature.
For the foregoing reasons, Appellant has not persuaded us of error in
the Examiner’s obviousness rejection of claim 37. Therefore, we will
sustain the Examiner’s rejection of that claim.

42
Appeal 2008-4881
Application 10/729,574

CONCLUSIONS
Appellant has not shown that the Examiner erred in rejecting claims 3,
4, 7, 8, 11-13, 19-24, and 26-30 under § 102 over Rosenberg. Nor has
Appellant shown that the Examiner erred in rejecting claims 3, 5, 6, 35, and
36 over Stewart under § 102.
Also, Appellant has not shown that the Examiner erred in rejecting
claims 9, 10, 14-18, 25, 31, 32, 34, 37, and 38 under § 103.

ORDER
The Examiner’s decision rejecting claims 3-32 and 34-38 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

pgc

V. Gerald Grafe, esq.
P.O. Box 2689
Corrales, NM 87048
43