United States District Court, S.D. New York
August 9, 2005.
CONVOLVE, INC. and MASSACHUSETTS INSTITUTE OF TECHNOLOGY, Plaintiffs,
COMPAQ COMPUTER CORP. and SEAGATE TECHNOLOGY LLC, Defendants.
The opinion of the court was delivered by: GEORGE DANIELS, District Judge
MEMORANDUM DECISION AND ORDER
Plaintiffs Colvolve, Inc. ("Convolve") and Massachusetts
Institute of Technology ("M.I.T.") (collectively, "plaintiffs")
bring suit, alleging that defendants Compaq Computer Corp.
("Compaq") and Seagate Technology LLC ("Seagate") (collectively,
"defendants") have infringed several of their patents. Plaintiffs
seek injunctive relief and monetary damages. A Markman hearing
was held to address the construction of the patent claims.
Plaintiff Convolve owns United States Patent Nos. 5,639,267
("the '267 patent") and 6,314,473 ("the '473 patent"). Plaintiff
M.I.T. owns United States Patent No. 4,916,635 ("the '635
patent"). Plaintiffs allege patent infringement, trade secret
misappropriation, and breach of contract. Specifically,
plaintiffs assert infringement of claims 1-4, 11, 21, and 24 of
the '635 patent, claims 19, 29, 39, and 55 of the '267
patent,*fn1 and claims 1, 3, 4, and 7-15 of the '473 patent. The '635 patent, entitled "Shaping Command Inputs to Minimize
Unwanted Dynamics," discloses a method to reduce vibration and
noise in physical systems. One example given is of a load-bearing
crane which, as the load is moved, develops vibrations within the
system. As the operator moves the crane, the load begins to
swing. This swing and resulting vibration can increase stress on
the crane, forcing the operator to wait for the load to come to
rest before performing another operation, such as lowering the
load to the ground. The '635 patent discloses a method for
reducing unwanted dynamics, such as the swinging of the crane
load, by translating the sequence of operations performed by the
operator into a sequence of operations that achieves the same end
position while minimizing the unwanted dynamics introduced into
the system. The method of altering the inputs to a system is
generally referred to within the art as "input shaping." ('635
patent Col. 1 line 29).
The '473 patent, entitled "System for Removing Selected
Unwanted Frequencies in Accordance with Altered Settings in a
User Interface of a Data Storage Device" discloses the
application of input shaping to computer disk drives. Computer
disk drives consist of platters coated with a magnetic substance,
and a sensor attached to an arm which scans over the platters as
they rotate. The spinning of the platters and movement of the arm
introduce vibrations into the disk drive. The arm must be
precisely positioned over the platter to accurately read the data
contained therein, and vibration increases the time for the arm
to reach the correct location. Performance of hard drives can be
increased by spinning the platters or moving the arm more
rapidly, at the cost of increased vibration. As disk drive
capacity increases, the arm must become increasingly precise. All
of these factors indicate the desirability of eliminating
vibration in disk drives. The '473 patent translates the signals
sent to the disk drive into a sequence which reduces unwanted
vibration caused by the movement of the arm. The patent
identifies certain frequencies which, when reduced, cause the disk drive to operate more
quietly, and other frequencies that allow the arm to seek the
correct position more quickly. ('473 patent Col. 7 lines 16-20).
Patent infringement analysis involves two steps. In the first
step the court determines the proper construction of the patent
claims by establishing the scope and boundaries of the subject
matter that is patented, as a matter of law. In the second step,
the trier of fact compares the properly constructed patent with
the alleged infringing device to determine if infringement has
occurred. Claim construction is a question of law for the
court.*fn2 See Markman v. Westview Instruments, Inc.,
52 F.3d 967 (Fed. Cir. 1995), aff'd, 517 U.S. 320 (1996). The
claims of a patent define "the metes and bounds of the right
which the patent confers on the patentee to exclude others from
making, using, or selling the protected invention." Corning
Glass Works v. Sumitomo Elec. U.S.A., Inc., 868 F.2d 1251, 1257
(Fed. Cir. 1989). Proper interpretation looks first to the
intrinsic evidence: the language of the claims themselves, the
patent specification, and the prosecution history if it is in
evidence. See Gart v. Logitech, Inc., 254 F.3d 1334, 1340
(Fed. Cir. 2001). The claims are at the center of the analysis.
See Interactive Gift Express, Inc. v. Compuserve, Inc.,
256 F.3d 1323, 1331 (Fed. Cir. 2001). There is a heavy presumption
that claim terms have the plain and ordinary meaning given to
them by one of ordinary skill in the art. See Bell Atl.
Network Servs., Inc., v. Covad Communications Group, Inc.,
262 F.3d 1258, 1268 (Fed. Cir. 2001). Where the claim language is
unambiguous, the intrinsic record should only be consulted to
determine whether a deviation from the plain meaning is
required.*fn3 See Interactive Gift Express, Inc., 256 F.3d at 1331. Dictionaries may be used
at any point in litigation to aid the court in better
understanding both the technology and terminology used by those
skilled in the art, but the dictionary definition cannot indicate
a meaning different from the intrinsic record. See Texas
Digital Sys., Inc. v. Telegenix, Inc., 308 F.3d 1193, 1202 (Fed.
Furthermore, claims must be read in light of the specification,
of which they are a part. See Markman, 52 F.3d at 979. Claim
terms must be constructed so as to be consistent with their use
in the specification.*fn4 The presumption that claim terms
have their ordinary meaning is rebutted when the inventor has
demonstrated a clear intent to act as his own lexicographer.
See C.R. Bard, Inc. v. United States Surgical Corp,
388 F.3d 858 (Fed. Cir. 2004) (providing an explicit definition or words
of manifest exclusion demonstrate disavowal of claim scope). The
specification also contains a preferred embodiment of the
invention, which discloses the best mode of practicing the
invention contemplated by the inventor. The preferred embodiment
does not impart a limitation on the claims. The prosecution
history, a contemporaneous record of proceedings before the
Patent and Trademark Office, must also be considered if it is in
evidence. Amendments made by the patentee to overcome or
distinguish a reference serve to limit claim terms, as do
statements made to the examiner during prosecution. However, if
the applicant wishes to disclaim claim scope during prosecution,
such disclaimer must be clear and unambiguous. See Invitrogen
Corp. v. Biocrest Mfg., L.P., 327 F.3d 1364, 1367 (Fed. Cir.
2003). After the intrinsic evidence has been considered, the court may
look to extrinsic evidence to guide the interpretation. Extrinsic
evidence includes expert and inventor testimony, treatises, and
all other evidence external to the text of the patent and the
prosecution history. See Markman, 52 F.3d at 979-90.
THE '635 PATENT
The '635 patent is directed to a method for removing unwanted
dynamics from a physical system. The patent discloses a method,
called "shaping," which modifies inputs to a system so as to
minimize those unwanted dynamics. In the preferred embodiment, a
mathematical model of the system is generated, from which the
response of the system to an input can be determined. ('635
patent Col. 4 lines 63-69). Based on this model, a sequence of
inputs*fn5 is generated such that the unwanted dynamics of
each individual input cancels out the unwanted dynamics of the
others when an impulse is applied to the system. In different
refinements of the system, the number of inputs in the generated
sequence is increased. ('635 patent Cols. 5-7). As more inputs
are added to the sequence, the ability of the model to reduce
unwanted dynamics to an acceptable level is improved. ('635
patent Col. 6 lines 12-49). The ability of the model to maintain
unwanted dynamics below an acceptable level is termed
"robustness." ('635 patent Col. 4 lines 14-18).
The parties dispute the meaning of several terms which appear
in multiple claims. Each of the terms has the same meaning unless
otherwise noted. Claim 1 is cited by both parties as a
representative claim. (Pls.' Claim Construction Br. For U.S.
Patent No.'s '635, '267, and '473 at 37; Defs.' Responsive Claim
Construction Brief for U.S. Patent Nos. '635 and '267 at 9). A. CLAIM 1:
The first element of claim 1 provides: "A method for generating
an input to a physical system to minimize unwanted dynamics in
the physical system response . . ." ('635 patent Col. 10 lines
41-42) (emphasis added).
Plaintiffs allege that the term "input" encompasses any
"information that is delivered to the system from outside the
system", relying on the plain dictionary definition of the term.
(Pls.' Claim Construction Br. At 39). The dictionary definition
of "input" is "data or similar information fed into a computer or
accounting machine." Merriam-Webster's Third New International
Dictionary Unabridged, Merriam-Webster, Inc., 1993.*fn6
Defendants respond, relying principally on the specification,
that an "input" is "the second signal that is applied to a
control system to cause a desired action in the control system
such that the altering of the first signal to create the second
signal is unaffected by the actual action of the control system."
(Joint Claim Construction and Prehearing Statement at Exhibit B,
'635 Patent 1). The principal distinction in defendants'
definition is that the signal sent to the physical system is "an
altered signal generated outside the control loop." (Defs.'
Proposed Claim Construction Order for U.S. Patent Nos. '635; '267
and '473 at 10). Defendants seek to exclude situations where the
input shaping is performed within the control loop. The control
loop is a configuration where a system receives an input, acts in
response to it, then measures whether the system reached the
desired state. If the system did not respond as anticipated, then
the system is adjusted repeatedly until the correct state is
reached. The plain meaning of "input" is consistent with its use
throughout the patent. Particularly relevant are two uses in
claim 11. The preamble provides: "Method for shaping an arbitrary
command input to a dynamic physical system . . ." ('635 patent
Col. 11 lines 17-19). The third element provides: "controlling
the physical system based on the shaped command input . . ."
('635 patent Col. 11 lines 24-25). Any interpretation of the term
"input" must be consistent across these uses. There is a
"presumption that the same terms appearing in different portions
of the claims should be given the same meaning unless it is clear
. . . that the terms have different meanings at different
portions of the claims." Fin Control Systems Pty, Ltd., v. OAM,
Inc., 265 F.3d 1311, 1318 (Fed. Cir. 2001). The use in the
preamble of claim 11 clearly indicates that the "input" therein
is the input to the physical system that is going to be shaped,
as the second element provides that the input is shaped by
"convolving the impulse sequence with the arbitrary command
input." ('635 patent Col. 11 lines 22-23). The use in the third
element is modified by "shaped command," indicating that the
input has been shaped as disclosed in the first two elements. The
plain meaning of "input" offered by plaintiffs is consistent with
all these uses. "Input" covers, generally, a piece of information
which is delivered to a system. In the preamble of claim 11, this
input is the unshaped input to a physical system which the
invention addresses. In element three, it is the shaped element
produced by the invention.
Defendants offer two main arguments for their proposed
construction. The first is that the patent only discloses a
system where the shaped command input is generated outside the
control loop. (Defs.' Responsive Claim Construction at 11). An
inquiry into whether or not a patent adequately discloses an
invention is a question of validity. See, e.g., Moba v.
Diamond Automation, 325 F.3d 1306 (Fed. Cir. 2003). It is
improper to determine validity before properly constructing the
claims. See Markman, 52 F.3d 967. It is apparent that
plaintiffs have not disclaimed the situation where input shaping is performed within
the control loop. Whether or not they have adequately disclosed a
method which can be used within the control loop is question of
validity. Plaintiffs have asserted a construction that is
supported by the plain language of the specification, and
accordingly bear the risk that the patent will subsequently be
Defendants' second argument is that Plaintiffs' construction
"would necessarily cover an unaltered input that would result in
the very unwanted vibrations the alleged inventions are intended
to eliminate." (Defs.' Responsible Claim Construction at 10).
This assertion lacks merit. Adopting plaintiffs' construction
merely indicates that an "input" follows the plain and ordinary
meaning of the term. Concerns that the patent will be broadened
beyond what is claimed, "a method for generating an input . . .
to minimize dynamics," are unfounded as infringement requires
that each limitation of the claim be met, either literally or
under the `doctrine of equivalents.' See Frank's Casing Crew &
Rental Tools, Inc., v. Weatherford Int'l, Inc., 389 F.3d 1370,
1377 (noting "literal infringement requires that each and every
limitation set forth in a claim appear in an accused product");
Riles v. Shell Exploration and Production Co., 298 F.3d 1302,
1309 (stating "[i]nfringement under the doctrine of equivalents
requires that the accused product contain each limitation of the
claim or its equivalent.") An accused infringer must meet the
other limitations of the claim in order to infringe, including
that the method "minimize unwanted dynamics." If an input does
not "minimize unwanted dynamics," then it is not covered by claim
1 because it has failed to meet all the elements of that claim.
Accordingly, plaintiffs' proposed construction for the term
"input" is adopted. The term "input," as used in independent
claims 1, 11, 21, and 24, and dependent claim 2, is interpreted
to mean: "information that is delivered to the system from
outside the system." 2. Expressions
The first element of claim 1 provides: "establishing
expressions quantifying the unwanted dynamics of the physical
system" ('635 patent Col. 10 lines 44-45) (emphasis added).
Plaintiffs argue that the term "expression" indicates any means
of symbolizing, and so in this context, an "expression" would be
anything that quantifies the unwanted dynamics of the system.
(Pls.' Claim Construction Br. For U.S. Patent No.'s '635, '267,
and '473 at 40). Plaintiffs base this meaning on the dictionary
definition of an expression. The dictionary definition most
relevant to the context of the patent is: "A sign or a character
or a finite sequence of signs or characters (as logical or
mathematical symbols) representing a quantity or operation."
Merriam-Webster's Third New International Dictionary
Unabridged, Merriam-Webster, Inc., 1993. Defendants insist that
the meaning be limited to mathematical equations which are in the
time domain. (Defs.' Proposed Claim Construction Order for U.S.
Patent Nos. '635, '267, and '473 at 1). A mathematical equation
in the time domain is one which expresses some value, here the
unwanted dynamic, as a function of time.
The correct interpretation must be consistent with the use of
the term in the specification. See Texas Digital,
308 F.3d at 1202. The terms "expression" and "expressing" or "expressed"
appear several times in the specification: "establishing
expressions quantifying the unwanted dynamics" ('635 patent Col.
3 lines 62-63); "a system's unwanted dynamic response will be
expressed as a function of its transient input" ('635 patent Col.
4 lines 54-57); "[t]he same result can be obtained mathematically
by adding two impulse responses and expressing the result for all
times greater than the duration of the input" ('635 patent Col. 5
lines 17-21); "[e]limination of vibration after the input has
ended requires that the expression for Aamp equal zero at the
time at which the input ends . . ." ('635 patent Col. 5 lines
44-46); "[i]n order to quantify the residual vibration level for a system, a vibration-error expression must
be defined . . ." ('635 patent Col. 6 lines 1-3); "[i]t can be
shown that the same expressions that guarantee zero derivatives
with respect to frequency also guarantee zero derivatives with
respect to damping ratio" ('635 patent Col. 6 lines 41-44);
"[t]hese sequences can be obtained by expressing the same (or
some other equivalent) constraint equations . . ." ('635 patent
Col. 7 lines 10-12).
Imposing defendants' "time domain" limitation would be
inappropriate. While many of the "expressions" discussed in the
specification are solved for time, see, i.e., '635, Col. 5,
lines 44-46; '635, Col. 5, lines 44-45, all of these examples
occur in the preferred embodiment of the specification. Since
limitations cannot be imposed based solely on the preferred
embodiment, defendants' argument requires that the patentee
evinced a clear intent to act as his own lexicographer.
Consistent and explicit use of a claim term in the specification
to mean something other than the ordinary meaning indicates this
intent. See Intellicall, Inc. v. Phonometrics, Inc.,
952 F.2d 1384, 1388 (Fed. Cir. 1992) (noting "[s]o long as the meaning of
an expression is made reasonably clear and its use is consistent
within a patent disclosure, an inventor is permitted to define
the terms of his claims.") The patentee did not consistently and
uniformly use "expression" to unambiguously refer to time-domain
equations, See, i.e., '635, Col. 6, lines 46-48 (referring to
a graph in the frequency domain as "the vibration-error
expression for the same three sequences of impulses . . .") The
use in the specification is not sufficiently definite to
demonstrate the necessary "unequivocal intent," therefore the
plain and ordinary meaning is presumed.
Within the context of the specification, and to be consistent
with the plain meaning, "expression" cannot include all possible
methods that quantifies a dynamic. There must be a solution that
minimizes that which is quantified expression. ('635, Col. 10,
lines 50-52). The only type of "expression" offered by plaintiffs which quantifies
and can be solved is a mathematical characterization. (Pls.'
Reply to Defs.' Responsive Claim Construction Br. for U.S. Patent
Nos. '635 and '267 at 3). Plaintiff has argued that "expression"
should not be limited to "equations where there must be an equal
sign" (Transcript of Proceedings Held on 3/31/2004 at 262). Given
the character of the invention, which is designed to meet or
exceed a threshold level of unwanted dynamics, there is no reason
to impose a requirement for "an equal sign." Accordingly, the
term "expression," as used in independent claim 1 is interpreted
to mean: "a mathematical characterization that
3. First Constraint
The second element of claim 1 provides: "establishing first
constraints bounding the available input to the physical system"
('635 patent Col. 10 lines 46-47) (emphasis added).
Plaintiffs argue that the "first constraint" imposes a limit on
the "input" generated by the method such that it does not exceed
the capabilities of the physical system. The method transforms
each input into a sequence of impulses in a way that vibration is
reduced. There is no certainty that the physical system is
capable of responding to the new sequence of impulses. Defendants
argue that the constraint is "a limit on the size of the input."
(Defs.' Responsive Claim Construction Brief for '635 and '267
Patents at 15). The term "first constraints" is defined by the
language of the specification as the physical limits of what
input the system can receive. ('635 patent Col. 7 lines 17-39).
Defendants' reading is only justified if the only limitations on
the inputs of the system are on the size. For some physical
systems this may be the case. The language of the patent itself,
however, extends to any limitations on the possible inputs to the physical system, and therefore "first constraint" is not solely a
limit on the size of the available input.
Accordingly, the term "first constraint" as used in independent
claim 1 is interpreted to mean: "a limit on the input based on
the physical system."
4. Second Constraint
The third element of claim 1 provides: "establishing second
constraints on variation in system response with variations in
the physical system characteristics." ('635 patent Col. 10 lines
48-49) (emphasis added).
Plaintiffs argue that in this context "second constraints"
refer to robustness. It is undisputed that robustness refers to
the ability of the model of the physical system to successfully
reduce unwanted dynamics in the physical system when the model is
not exact. (Pls.' Claim Construction Br. For U.S. Patent No.'s
'635, '267, and '473 at 9; Defs.' Responsive Claim Construction
Br. for '635 and '267 Patents at 16). The model of the physical
system which quantifies the response of the system to an input
will only eliminate all dynamics when the model is exact. An
inexact model can be due to operation of the physical system in
an environment which can influence its behavior, such as
variation in temperature, humidity, or precision of manufactured
components of the system. In the crane example, the addition of
wind to the system could cause the load to swing in ways that the
model cannot anticipate. Plaintiff argues that the term should be
given the meaning "constraints on the range of response over a
range of differences in the physical system." (Pls.' Claim
Construction Br. at 41). Defendants respond that the only valid
interpretation of the term is that derivatives of specific time
domain equations are set to zero. (Defs.' Responsive Br. for '635
and '267 at 16). This definition is based on an example in the
preferred embodiment, where the model for the physical system is
expressed as a system of equations. ('635 patent Cols. 5-6). The "time domain
equations" referred to by defendants are a set of equations added
to the system which constrain the solutions to the system.
Since there is no plain and ordinary meaning for the term
"second constraints," its meaning must be found in the
specification. The term is defined at the outset to be "variation
in system response with variations in system characteristics."
('635, Col. 4, lines 4-5). The preferred embodiment describes the
"second constraint" as an additional constraint where the
derivatives of the physical response to an impulse are taken with
respect to frequency. ('635, Col. 6, lines 12-49). Second
constraints are limitations imposed on the "expressions"
described in part A.2, supra. Expressions are not restricted to
equations in the time domain. The expression modeling the
physical system could exist in, i.e., the frequency domain. It
is nonsensical to impose a constraint on a variable which is not
in the equation. Moreover, restricting the term "second
constraints" to setting derivatives of an equation to zero
ignores the breadth claimed in the patent. The specification
contemplates imposing constraints on the expressions modeling the
physical system. The full specification, not just the preferred
embodiment, breathes meaning into the claims. See Amhil
Enters. v. Wawa, Inc., 81 F.3d 1554, 1559 (Fed. Cir. 1996)
(refusing to limit the claim to the preferred embodiment since
"[a] preferred embodiment, however, is just that, and the scope
of a patentee's claims is not necessarily or automatically
limited to the preferred embodiment. The entire specification,
including all of the claims, the prosecution history, and the
prior art may all affect the interpretation ultimately placed on
claim language.") It would therefore be inappropriate to restrict
the meaning of "second constraints" to defendants'
interpretation, which emphasizes the preferred embodiment over
the specification as a whole, and would be incompatible with the
meaning of the term "expression." Accordingly, the court finds the term "second constraint" as
used in independent claim 1 is interpreted to mean: "variation in
system response with variations in system characteristics."
The term "minimize" appears in both the preamble and the fourth
element of claim 1: "finding a solution which is used to generate
the input which minimizes the value of the expressions while
satisfying the first and second constraints . . ." ('635 patent
Col. 10 lines 50-51) (emphasis added).
Plaintiffs argue that the term "minimize" should be understood
to mean "reduce to a minimum." Defendants argue that "minimize"
should be "reduce to zero." Neither party has indicated that this
term is used in any way other than it's ordinary meaning. The
dictionary definition of "minimize" is "to reduce to the smallest
possible number, degree, or extent." Merriam-Webster's Third New
International Dictionary Unabridged, Merriam-Webster, Inc.,
1993. Plaintiffs assert that within the context of the claim,
"minimize" does not require reduction to zero. In support of
their argument, plaintiffs recite the claim language that
expressions are "minimized" subject to the constraints in the
other elements of the claim. The constraints, as constructed
supra, impose strict requirements on the input generated.
Physical systems have a range of inputs which they are designed
to receive. The input must be within this range ('635 patent Col.
7, lines 36-39) ("If at any time during the convolution, the
actuator limit is exceeded, the sequence is unacceptable."), and
within the range of tolerances the system permits ('635 patent
Cols. 5-7). Within these constraints, and subject to real world
conditions, the claimed method cannot necessarily eliminate all
vibrations. (Decl. of Dr. Bell ¶ 20-21). See also '635 patent
Col. 6, lines 8-11 ("Acceptable response is defined as less than
5% of total move size residual vibration.") Defendants correctly indicate that the abstract of the patent
states an objective "to eliminate unwanted dynamics", and a
disclosed method in the specification which describes amplitudes
of undesired vibrations as being reduced to zero. ('635 patent
abstract; Col. 5, lines 44-61). Reducing unwanted dynamics to
zero is only one step of the method taught by the invention. The
abstract provides "A sequence of impulses is determined which
eliminates unwanted dynamics of a dynamic system." ('635 patent
abstract). This "sequence of impulses" is not the "input which
minimizes the value of the expressions." In the preferred
embodiment, the method taught determines how to translate an
arbitrary input into another input. If the model is perfect with
respect to the real-world operating environment, the patent
contemplates eliminating all residual vibration. The imposition
of the first and second constraints, above, are specifically
imposed because of the difficulty of formulating a model which
can capture every variation of the system's environment. See
'635 patent Col. 5 lines 67-68 ("The two-impulse input, however,
cancels vibration only if the system natural frequency and
damping ratio are known exactly.") Interpreting "unwanted
dynamics" to incorporate a threshold for acceptable residual
vibration is consistent with the specification and the awareness
of this difficulty. The disclosed method establishes a model
which, in an idealized and abstracted world, eliminates a
dynamic, such as vibration, from the system. ('635 patent Col.
5-6, lines 67-3) (noting that "vibrations [are canceled] only if
the system natural frequency and damping ratio are known
exactly.") The preferred embodiment specifically accommodates for
uncertainty in the natural frequency of the system, and to
variations in the damping ratio. This is exactly what is
contemplated by the "first constraints" and "second constraints"
in the '635 patent claims. Defendants concede that they have no support for the assertion
that reducing resulting vibrations to zero is possible.*fn8
Reading a limitation into the claims which is so substantial as
to exclude the preferred embodiment, when an alternative
interpretation is clearly consistent with the language of the
specification, is unwarranted. See Vitronics Corp. v.
Conceptronic, Inc., 90 F.3d 1576, 1583 (Fed. Cir. 1994). The
preferred embodiment contemplates a model where unwanted dynamics
are reduced as much as possible, but with a tolerance for error.
This tolerance would not be necessary if the invention reduced
unwanted dynamics to zero in every instance. Presumably this is
why plaintiffs opted for a broader term, encompassing solutions
where the input is reduced as much as possible subject to
constraints, but not necessarily to zero.
Accordingly, the term "minimize" as used in independent claims
1 and 11 is interpreted to mean: "reduce to a minimum."
6. Find A Solution
The third element of claim 1 provides: "finding a solution
which is used to generate the input which minimizes the value of
the expressions . . ." (emphasis added).
Plaintiffs assert that the term "finding a solution" means
"determining an answer." Defendants initially contended that
"finding a solution" means "the sequence of impulse signals or
finite duration pulses that sets the expressions of the amplitude
of the resonant vibration equal to zero" (Defs' Responsive Brief
for '635 and '267 Patents at 18), but conceded plaintiffs' definition at the Markman hearing (Transcript of Proceedings
Held on 3/30/2004 at 176). Accordingly, the term "finding a
solution," as used in independent claims 1 and dependent claims
2-4, is interpreted to mean: "determining an answer."
THE '473 PATENT
The '473 patent is generally directed to a system for reducing
unwanted dynamics in disk drives. Dynamics can arise in a disk
drive due to the movement of the drive arm, an internal
mechanism, over the platters within the disk containing the data.
Reducing these dynamics can provide faster seek time, by reducing
the time it takes for the drive arm to settle over a specific
location of the platter, or lower noise level since noise in hard
drives is caused by vibration. The '473 patent discloses an
application of the methods for shaping inputs to a disk drive,
and providing a user interface to select between a faster seek
time or lower noise level. As for the other two patents, '267 and
'635, each term has the same meaning in all the allegedly
infringed claims unless otherwise noted.
A. CLAIM 10:
1. User Interface
The first element of claim 10 provides:
providing a user interface for controlling one of a
seek time of the data storage device and an acoustic
noise level of the data storage device; (emphasis
There are two disputes over the term "user interface." The
first dispute is whether a "user interface" is limited to being
either a graphic user interface (GUI) or electromechanical
switches, or whether a "user interface" more broadly encompasses
any means to alter the drive performance parameters. The second
dispute is whether the "user interface" can only be used by the
human end-user of the system. Plaintiffs assert that a "user
interface" broadly encompasses any means to alter the parameters. They further contend that a
"user interface" can more generally be used by any person or
other system which is capable of setting the parameters.
Within the relevant art,*fn9 there are distinctions
between a "user interface" and a "graphic user interface." The
dictionary meaning of "User interface" is "the means of
communication between a human user and a computer system,
referring in particular to the use of input/output devices with
supporting software. Examples include the graphical user
interface (GUI) and command-line interface (CLI)." A Dictionary
of Computing, Oxford University Press, 1996. Defendants assert
that the specification indicates that "user interface" should be
construed as "a graphic user interface (`GUI') of a type common
to windowing operating systems such as Microsoft Windows or
electro-mechanical switches." ('473 patent Col. 6:44-9:44). This
fails to distinguish between a "user interface" and a "graphic
user interface". Defendants claim the specification defines the
term "user interface" to be synonymous with "graphical user
[A]s an alternative to the GUI, or to supplement the
GUI, the disk drive may include one or more
electro-mechanical manual switches for controlling
all or part of the drive's operation in the manner
set forth above.
('473 patent Col. 9:38-42).
This language, however, clearly indicates the breadth of the
term "user interface" to include both a GUI or electromechanical
switches. This language indicates that a GUI is one type of "user
interface," and an electro-mechanical switch is another, albeit
simpler, type of "user interface." The "alternative to the GUI"
and "supplement the GUI" language indicates that the
electro-mechanical switch is not intended to limit the definition
of "user interface," but rather demonstrates the range of user
interfaces encompassed by the invention. See Johnson Worldwide Assocs. v.
Zebco Corp., 175 F.3d 985
, 991 (Fed. Cir., 1999) ("Varied use of
a disputed term in the written description demonstrates the
breadth of the term rather than providing a limited definition.")
Defendants' construction would require that "user interface"
ensnare both a complex and simple user interface, but not the
spectrum of user interfaces in between. The inventor did not
demonstrate a manifest intent to disclaim every type of user
interface not disclosed in the specification.
Plaintiffs' construction of "user" as "a person or device that
uses the user interface" is too broad. It fails to give meaning
to the adjective "user," which distinguishes the interface from
other types of interfaces used in the computing field, for
example, an advanced programming interface (API) which allows one
software program to "interface" with another. The plain meaning
of "user interface" requires a user, and the term "interface"
typically indicates a means by which anything can interact with a
device. Plaintiffs maintain that the term "user" cannot be
limited to "a person" because this would exclude the situation
where a person indirectly uses the interface. They argue, in
essence, that any device can be broken down into a series of
simpler devices. A potential infringer could, therefore,
circumvent the patent by placing an intermediary device between
the user and the invention. For instance, when using a GUI, the
user typically positions the cursor on the screen using a mouse.
A potential infringer could claim that the user is interfacing
with the mouse, and the mouse is interfacing with the computer,
thereby avoiding infringement. To avoid such pedantic arguments
by accused infringers, the term is constructed specifically to
include circumstances where a person has, at some point, dictated
what the parameters should be. A patent cannot be constructed to
deprive a claim term of all meaning, and plaintiffs' construction of
"user interface" would render "user" meaningless.
Defendants also contend that the user of the "user interface"
must be the human end-user of the drive. They claim that the
description of the GUI in the specification compels an
interpretation imposing the end-user restriction. However, the
user can be any person who wants to make use of the device; in
this case, setting the drive parameters. The plain meaning of
"user" does not require an individual who is the end-user of a
product. Even the word "end-user" militates against the proposed
construction: "end" modifies "user," so in it's regular usage
"user" must mean something other than end-user. To impose a
definition more limited than the normal meaning, there must be a
clear intent on the part of the author to use the more limited
definition. Johnson Worldwide Assoc., 175 F.3d at 990 (stating
that anything other than the "ordinary and accustomed meaning"
inheres only "if the patentee has chosen to be his or her own
lexicographer by clearly setting forth an explicit definition for
a claim term" . . . [or] . . . "the term or terms chosen by the
patentee so deprive the claim of clarity that there is no means
by which the scope of the claim may be ascertained from the
language used.") (internal citations omitted). Defendants do not
assert that the term end-user appears in the patent, nor does any
language limiting the potential user of the product appear.
Claim terms are to be given their normal broad meaning absent
an explicit disavowal of scope. See Novartis Pharm. Corp. v.
Eon Labs Mfg., 363 F.3d 1306, 1314 (Fed. Cir. 2004). The
"Windows-like operating system," one of the types of user
interfaces defendants claim fits under their definition, is still
a user interface when it is used by a technician configuring the
computer before sale. There is no contemplation within the patent that only the end-user of the drive be able to
use the interface, and defendants' examples also militate toward
no limitation based on the type of user.
Defendants contend that plaintiffs' proposed construction could
be read to include disk drives that have been irrevocably
configured at the factory. This meaning is beyond the scope of a
"user interface." A user interface enables the alteration of
parameters of the system. Drives manufactured in such a way as to
only be capable of a one-time setting, which is then fixed in the
hardware, do not possess a capability to alter the parameters.
They only have a capability to set the parameters once. This is
beyond the ordinary meaning of "user interface" in the art.
Accordingly, the term "user interface," as used in independent
claims 1, 10, 11, and dependent claims 4, 12, and 13, is
interpreted to mean:
"hardware, firmware, or a combination thereof that
allows a person, directly or indirectly, to alter
2. Data Storage Device
The first element of claim 10 also provides:
. . . for controlling one of a seek time of the data
storage device and an acoustic noise level of the
data storage device; (emphasis added)
The term "data storage device" is used several times in claim
10. The interpretation of this term is linked with the
interpretation of "outputting," below. Plaintiffs assert that the
data storage device comprises the electromechanical components of
the disk drive that perform commands specified by a controller.
Defendants assert that "data storage device" should be understood
to mean a hard disk drive, including controller electronics. Defendants' construction would exclude
from the scope of the claim those devices where shaped commands
were generated within the drive itself.
The meaning adduced by both parties to "data storage device" is
that it is a disk drive. Plaintiff states that those skilled in
the art of control systems typically use a term to refer to that
which is controlled, and that therefore what a layperson
considers a disk drive would be characterized by a control
engineer as a disk drive and a controller. (Pls.' Claim
Construction Br. For U.S. Patent No.'s '635, '267, and '473 at
26). The specification supports such a construction. The patent
in Col 10:1-21 describes a preferred embodiment of the control of
a disk drive. Included in this description is the possibility
that the processor controlling the disk drive is located in the
PC, or, "alternatively, a separate controller dedicated to the
disk drive which receives commands from [that] processor." '473
patent Col 10:6-8. Defendants claim that the specification
discloses only two possible structures which can control the disk
drive: the first is a processor within the PC, and the second is
a controller which receives "commands" from that processor.
(Defs.' Responsive. Claim Construction Br. for '473 Patent at 5).
Since the third element of claim 10 uses the term "command" to
refer to shaped inputs, defendant infers that the commands must
be shaped by the processor located within the PC. Defendant
combines a context-specific use within one claim with a more
general use in the specification. The term "command" as it is
used in the specification refers to signals communicated from a
keyboard or mouse (Col. 6:19), to a signal which will be shaped
(Col. 8:12), and to signals in a disk drive generally (Col.
2:22-28). Claim 10 refers to a specific kind of "command," one
that has been shaped by a processor. The command referred to in
Col. 10:6-8 could therefore be a signal which has been shaped by
a disk drive engine running on a processor in the PC, or an unshaped
signal from the PC which will be shaped by a processor dedicated
to the disk drive and not located in the PC. Defendants argue
that since "disk drives," as lay people use the term, commonly
include on-board electronics dedicated to controlling the drive,
a construction that permits on-board electronics to send
"commands" is not supported. Such a configuration is within the
scope of the claim. Accordingly, the term "data storage device,"
as used in claim 10, is interpreted to mean:
"a computer disk drive, including all electronic and
mechanical components, which receives shaped commands
from a processor which may be integrated into the
3. Outputting Commands to the Data Storage Device
The third element of claim 10 provides: "outputting commands
to the data storage device causing the data storage device to
alter seek trajectory . . ." (emphasis added).
The parties dispute whether the underlined portion of the claim
imposes a limitation that the commands must be generated in the
computer. Plaintiffs asserts that the commands can originate from
any source. Defendants seek to exclude from the scope of the
claims situations where electronics located within the disk drive
itself perform a shaping method on commands sent to the data
storage device. The specification discloses in the preferred
embodiment a software application, the disk drive engine, which
"generates and outputs commands to the disk drive." ('473 patent,
Col. 6 lines 43-45). This application is executed either on the
main processor of the computer of which the disk drive is a
component, or "a separate controller dedicated to the disk drive
which receives commands" from the main processor of the computer.
('473 patent, Col. 10 lines 6-8). Defendants' argument depends in
part on their construction of data storage device to include any
electronics built into a commercially packaged disk drive. The construction of "data storage device" adopted is consistent with
the claim language requiring that commands be "output" to the
data storage device. This reduces the question to the source of
Defendants are correct that plaintiffs do not provide any
guidance regarding the source of the "commands" being "output."
In order for the claim to be self-consistent, the "commands" must
be the "input signals to the data storage device." The dictionary
definition of "output" is "the result of data-processing activity
when it is presented external to the system . . . it may be ready
for input to another system or process . . ." A Dictionary of
Computing, Oxford University Press, 1996. Within the context of
the invention, the disclosed method for shaping inputs is
performed. The result of this method, in the form of "commands,"
are sent (as output) from the processor performing the shaping to
the data storage device (as input). The word "output" when
applied to a signal does not inherently require that the signal
be sent from a different device than that receiving the signal. A
feedback system "outputs" a signal to itself, which it then uses
as an input.
Accordingly, the term "output," as used in independent claims
1, 7-11, and 14-15, is interpreted to mean: "sending a command to
a device, which may originate from within the device."
4. By Shaping Input Signals and Selected Unwanted Frequencies
The third element of claim 10 also provides:
". . . causing the data storage device to alter seek
trajectory shape by shaping input signals to the
data storage device to reduce selected unwanted
frequencies from a plurality of frequencies . . ."
(emphasis added) The two underlined terms are in dispute. Plaintiff argues that
the term "shaping" includes frequency-based filtering, and that
"selected unwanted frequencies" can be some or all of the
unwanted frequencies. Defendant insists that filtering, which is
found in prior art references, cannot be encompassed by the
invention. They also argue that the term "selected" precludes
selection of all unwanted frequencies.
Defendants' assertion that including filtering in the term
"shaping" would result in invalidity is premature. The
specification unambiguously indicates that shaping includes
conventional filtering techniques. ('473 patent, Col. 22:60
67). The patentee's offered construction does not conflict with
the explicit language of the claim, and follows from the plain
meaning of the term. Rhine v. Casio, 183 F.3d 1342, 1346 (Fed.
Cir. 1999) (the court should "sustain the patent and the
construction claimed by the patentee himself, if this can be done
consistently with the language which he has employed.") This
construction may later be found invalid, but it is inappropriate
to argue invalidity during claim construction. Rhine v. Casio,
183 F.3d 1342, 1346 (Fed. Cir. 1999) (defendant "cannot avoid a
full-blown validity analysis by raising the specter of invalidity
during the claim construction phase.")
Accordingly, the term "shaping" as used in claim 10 is
interpreted to mean: "applying a transformation to a signal."
The third element of claim 10 also provides: "shaping input
signals to the data storage device to reduce selected unwanted
frequencies" ('473 patent Col. 44 lines 43-45) (emphasis added).
Plaintiffs assert that "selected unwanted frequencies" should
be interpreted to mean some unwanted frequencies, and could
include, but need not include, all unwanted frequencies. (Pls.'
Claim Construction Br. For U.S. Patent No.'s '635, '267, and '473
at 29). Defendants respond that selected cannot encompass all unwanted frequencies. (Defs'
Responsive Brief for '473 Patent at 24).
Neither party has presented any special meaning of the term
"selected" in the art. The dictionary definition most relevant to
the context of the patent is "choose from a number or group."
Merriam-Webster's Third New International Dictionary
Unabridged, Merriam-Webster, Inc., 1993. Plaintiffs note
particularly that when the method reduces certain specific
unwanted frequencies, such as would be the case if less than all
unwanted frequencies are selected, other unwanted frequencies may
also be reduced. (Transcript of Markman Hearing on 3/31/2004 at
278). As a natural consequence of the robustness built into the
system, plaintiffs maintain, when any specific frequency is
reduced, there will be collateral reductions in other
frequencies. (Transcript of Markman Hearing at 279). These
arguments do not directly address the question of "selection" of
frequencies. A method which reduces frequencies other than those
selected is not excluded from the patent since there is no
limiting language requiring that only specific frequencies are
Defendants agree on the ordinary meaning of the term "selected"
(Transcript of Markman Hearing at 254), but assert that the
prosecution history of the '473 patent discloses a surrender of
any method where all unwanted frequencies are reduced. In an
interview on May 9, 2001, the PTO asserted that U.S. Patent
6,011,373 (hereinafter "McConnell patent") teaches selection by
the user between fast response time and noise reduction mode.
Plaintiffs amended the language "reduce unwanted frequencies" to
"reduce selected unwanted frequencies from a plurality of
frequencies" in the '473 patent. The PTO rejected the "reduce
unwanted frequencies" language since "McConnell et al teach that
the user can select between the fast response time and noise
reduction modes. However, McConnell et al can't reduce selected unwanted
frequencies . . ." (Akerley Dec. Exh. V, PLFS-C 008898). The
McConnell invention "allow[s] the user to select a command input
from a given group of command inputs . . . by ranking the
importance of three characteristics . . . response time,
robustness, and noise generation potential." (McConnell patent
Col. 11 lines 21-25). Whereas the '473 patent identifies specific
unwanted frequencies to reduce, and generates a command input
which reduces them, the McConnell patent takes a set of
pre-determined command inputs, categorizes these commands by the
three characteristics above, and applies the input best suited to
the user's preference.
Amendments made during prosecution are a guide to understanding
the language of the patent, although they do not "enlarge,
diminish, or vary the limitations in the claim." Markman,
52 F.3d at 980. Plaintiffs did not give up any claim to a reduction
of all unwanted frequencies. The amendment made to overcome
rejection by the PTO clearly indicates that the '473 patent is
differentiated from the McConnell patent by the fact that the
McConnell patent does not perform selective reduction of unwanted
frequencies. The McConnell patent applies methods known to reduce
all frequencies, and picks one based on the user specified
criteria. The term "selected" was imposed to distinguish the '473
patent from the McConnell patent, and it is a significant
distinction that the '473 patent targets specific frequencies to
reduce as opposed to indiscriminately reducing all frequencies.
The fact that under this interpretation the '473 patent could
result in a reduction of more than the selected frequencies is
unimportant. The inventor has disclosed a method which provides
that certain targeted frequencies will definitely be reduced, which the PTO considered sufficiently novel in light of
the prior art to grant a patent.
Accordingly, the term "selected unwanted frequencies" as used
in claim 10 is interpreted to mean: "at least the chosen unwanted
Based on the foregoing, the term "input" means "information
that is delivered to the system from outside the system"
The term "expression" means "a mathematical characterization
The term "first constraint" means "a limit on the input based
on the physical system."
The term "second constraint" means a "variation in system
response with variations in system characteristics."
The term "minimize" is interpreted to mean "reduce to a
The term "find a solution" is interpreted to mean "determining
The term "user interface" is constructed to mean "hardware,
firmware, or a combination thereof that allows a person, directly
or indirectly, to alter parameters."
The term "data storage device" is constructed to mean "a
computer disk drive, including all electronic and mechanical
components, which receives shaped commands from a processor which
may be integrated into the drive."
The term "outputting commands" is constructed to mean "sending
a command to a device, which may originate from within the
The term "shaping input signals" is constructed to mean
"applying a transformation to a signal."
The term "selected unwanted frequencies" is interpreted to mean
"at least the chosen unwanted frequencies."