IN THE UNITED STATES DISTRICT COURT FOR THE NORTHERN DISTRICT OF CALIFORNIA
May 11, 2005
SAMSUNG ELECTRONICS CO, LTD, PLAINTIFF,
QUANTA COMPUTER, INC ET AL, DEFENDANTS.
The opinion of the court was delivered by: Vaughn R Walker United States District Chief Judge
Following the dismissal of various claims, this suit now focuses on United Stats Patent No 5,333,273, which discloses a "protected hot key function for microprocessor-based computer system." The invention is in the field of computer circuitry and, more particularly, keyboard-input processing for ISA-compatible personal computers. The court held a claim construction hearing on May 6, 2005, pursuant to Markman v Westview Instruments, Inc, 517 US 370 (1996). Based on the hearing, the parties' memoranda and the applicable Federal Circuit law, the court construes the claims of the patent as follows.
The patent in suit has been the subject of considerable litigation, both in this court and in the Southern District of Texas before Magistrate Judge Johnson and Judge Rainey. Between the patent itself, the parties' memoranda, Judge Johnson's report and recommendation (the "Texas R&R," reproduced at Joint Statement (Doc #196) Ex 1) and Judge Rainey's claim construction order (the "Texas Order," also reproduced at Joint Statement (Doc #196) Ex 1), the invention has been exhaustively described. Accordingly, the barest outline will suffice here.
As the patent explains, activity on ISA-compatible computers typically proceeds via an "interrupt," which causes the computer's CPU to stop one activity and start another. Various interrupts (conventionally labeled in ISA-compatible computers as IRQ0, IRQ1, IRQ2 and so on) are associated with various functions. Conventionally, the "IRQ1" interrupt is associated with keyboard input; when the CPU receives an IRQ1 interrupt, it stops its activity to receive a "scan code" from the keyboard. This scan code signifies that a certain key has been depressed or released, and the CPU will then execute software instructions in response to the particular scan code. Such software instructions are specific to the software that a user has loaded onto the computer. For example, pressing the "A" key on the keyboard while running a word processing program will, typically, cause the letter "a" to appear in the active document. More exotically, particular keys or key combinations that call up a program or perform a certain function are often known as "hot keys." If two loaded programs are both designed to respond to the same hot key, the resulting conflict can cause the computer not to function in the way the user expects.
The patent is directed to a "protected hot key function" whose behavior is not as readily altered or customized by user-loaded software as conventional hot keys. The invention describes an "additional function key," which, when depressed on its own or in combination with a conventional key or keys, causes a second interrupt -- i e, an interrupt other than the IRQ1 interrupt conventionally used in ISA-compatible computers. Apparently, at the time of the invention, little if any software existed that would respond to this second interrupt, thus ensuring that computer manufacturers could reliably assign predictable behaviors to hot keys using the additional function key (or to the additional function key on its own) without interference from user-loaded software.
The construction of patent claims is a question of law to be determined by the court. Markman v Westview Instruments, Inc, 517 US 370 (1996). The goal of claim construction is "to interpret what the patentee meant by a particular term or phrase in a claim." Renishaw PLC v Marposs SpA, 158 F3d 1243, 1249 (Fed Cir 1998). In determining what a patentee meant by a term or phrase, the court looks first to the claim itself.
The claims of the patent provide the concise formal definition of the invention. They are the numbered paragraphs which "particularly [point] out and distinctly [claim] the subject matter which the applicant regards as his invention." 35 USC § 112. It is to these wordings that one must look to determine whether there has been infringement. Courts can neither broaden nor narrow the claims to give the patentee something different than what he has set forth. No matter how great the temptations of fairness or policy making, courts do not rework claims. They only interpret them.
EI Du Pont de Nemours & Co v Phillips Petroleum Co, 849 F2d 1430, 1433 (Fed Cir 1988).
"The claims define the scope of the right to exclude; the claim construction inquiry, therefore, begins and ends in all cases with the actual words of the claim." Renishaw, 158 F3d at 1248. "The words used in the claim are viewed through the viewing glass of a person skilled in the art." Brookhill-Wilk 1, LLC v Intuitive Surgical, Inc, 326 F3d 1215, 1220 (Fed Cir 2003) (citing Tegal Corp v Tokyo Electron Am, Inc, 257 F3d 1331, 1342 (Fed Cir 2001)). "Absent a special and particular definition created by the patent applicant, terms in a claim are to be given their ordinary and accustomed meaning." York Prods, Inc v Central Tractor Farm & Family Ctr, 99 F3d 1568, 1572 (Fed Cir 1996). The court may, if necessary, consult a variety of sources to determine the ordinary and customary meaning of a claim term, including the claim terms themselves, dictionaries, the written description, the drawings and the prosecution history, if in evidence. Brookhill-Wilk 1, 326 F3d at 1220. "Such intrinsic evidence is the most significant source of legally operative meaning of disputed claim language." Vitronics Corp v Conceptronic, Inc, 90 F3d 1576, 1582 (Fed Cir 1996). With respect to dictionary definitions, "[i]f more than one dictionary definition is consistent with the use of the words in the intrinsic record, the claim terms may be construed to encompass all such consistent meanings." Texas Digital Systems, Inc v Telegenix, Inc, 308 F3d 1193, 1203 (Fed Cir 2002).
The court begins its construction of claim terms by consulting intrinsic evidence of the meaning of disputed claim terms, which includes the claims, the specification and the prosecution history (if in evidence). Lacks Industries, Inc v McKechnie Vehicle Components USA, Inc, 322 F3d 1335, 1341 (Fed Cir 2003) (citation omitted). "If upon examination of this intrinsic evidence the meaning of the claim language is sufficiently clear, resort to 'extrinsic' evidence, such as treatises and technical references, as well as expert testimony when appropriate, should not be necessary." Digital Biometrics, Inc, v Identix, Inc, 149 F3d 1335, 1344 (Fed Cir 1998). "[I]f after consideration of the intrinsic evidence, there remains doubt as to the exact meaning of the claim terms, consideration of extrinsic evidence may be necessary to determine the proper construction." Id.
"[A] court may constrict the ordinary meaning of a claim term in * * * one of four ways[:]" (1) "if the patentee acted as his own lexicographer and clearly set forth a definition of the disputed claim in either the specification or prosecution history;" (2) if the intrinsic evidence shows that the patentee distinguished the term from prior art on the basis of a particular embodiment, expressly disclaimed subject matter, or described a particular embodiment as important to the invention; (3) "if the term chosen by the patentee so deprives the claim of clarity as to require resort to other intrinsic evidence for a definite meaning; and (4) "if the patentee phrased the claim in step-or means-plus-function format," then "a claim term will cover nothing more than the corresponding structure or step disclosed in the specification, as well as equivalents thereto * * *." CCS Fitness, Inc v Brunswick Corp, 288 F3d 1359, 1366-67 (Fed Cir 2002) (internal citations and quotation marks omitted).
Limitations from the specification, such as from the preferred embodiment, cannot be read into the claims absent an express intention to do so. Teleflex, Inc v Ficosa North Am Corp, 299 F3d 1313, 1326 (Fed Cir 2002) ("The claims must be read in view of the specification, but limitations from the specification are not to be read into the claims."). But "a construction that excludes a preferred embodiment 'is rarely, if ever, correct.'"
C R Bard, Inc v United States Surgical Corp, 388 F3d 858, 865 (Fed Cir 2004) (citing Vitronics, 90 F3d at 1583).
With these legal principles in mind, the court now turns to the construction of the disputed claim language of the patent.
Independent claims 1 and 5 and several dependent claims are at issue. For many terms, the parties have stipulated to constructions (either of their own accord or by adopting the Texas construction). The court accepts those stipulations for purposes of this action. The claims -- with disputed language underscored the first time it appears -- are rescribed below. (The dependent claims about which there are no disputes are not rescribed.)
1. A system for providing a built-in function in an ISA-compatible computer in response to activation of a selected combination of user activated keys, comprising: a keyboard having a set of conventional alphanumeric and function keys and further having at least one additional function key; a keyboard controller connected to said keyboard to monitor said conventional keys and said additional function key to detect when at least one of said keys is activated, said keyboard controller having first and second interrupt signal lines connected to said ISA-compatible computer, said keyboard controller responsive to an activation of at least one of said conventional keys to activate a first interrupt signal to said ISA-compatible computer on said first interrupt signal line, said keyboard controller responsive to an activation of said additional function key in combination with at least one of said conventional alphanumeric keys to generate a second interrupt signal to said ISA-compatible computer on said second interrupt signal line; a first conventional interrupt handling routine within said ISA-compatible computer responsive to said first interrupt signal from said keyboard controller to input data scan codes from said keyboard; and a second non-conventional interrupt handling routine within said ISA-compatible computer responsive to said second interrupt signal from said keyboard controller to input an identification of said activated alphanumeric key and to perform a predetermined function selected by said identified alphanumeric key.
4. The system for providing a built-in function as defined in claim 1, further comprising a central processing unit that indexes a first memory location pointer in response to said first interrupt signal, said central processing unit further indexing a second memory location pointer in response to said second interrupt signal
5. A system for servicing keyboard interrupts in an ISA-compatible computer, comprising: a keyboard having a plurality of keys including conventional alphanumeric keys, conventional symbol keys, conventional function keys and conventional cursor control keys, said keyboard further including at least one non-conventional function key, said keyboard generating a scan code in response to an activation of at least one of said keys, said scan code varying depending upon which of said keys is activated; and a keyboard controller coupled to said keyboard, said keyboard controller further coupled to said ISA-compatible computer by first and second interrupt signal lines, said keyboard controller generating a first interrupt signal on said first interrupt signal line upon receipt of a scan code corresponding to one of said conventional keys, said ISA-compatible computer programmed to execute a program to input said scan code in response to said first interrupt signal, said keyboard controller generating a second interrupt signal on said second interrupt signal line upon receipt of a scan code corresponding to said non-conventional function key, said ISA-compatible computer programmed to execute at least one special routine upon receipt of said second interrupt signal.
Patent at 13:36-14:48.
Thus, the terms to be construed are:
* "ISA-compatible computer" (claims 1 and 5)
* "additional function key" (claim 1) and "non-conventional function key" (claim 5), which the parties agree should be given the same construction
* "keyboard controller" (claims 1 and 5)
* "connected to" (claim 1)
* "second interrupt signal" (claims 1 and 5)
* "data scan code" (claim 1) and "scan code" (claim 5), which the parties agree should be given the same construction
* "indexes a first memory location pointer" (claim 4)
* "indexing a second memory location pointer" (claim 4)
* "coupled to" (claim 5)
* whether "upon receipt of a scan code corresponding to said non-conventional function key" (claim 5) means "upon receipt of a scan codes that is generated by the activation of a single non-conventional function key alone"
The court construes these terms in a sequence that reflects their logical interdependence.
"ISA-compatible computer" (claims 1 and 5)
The patent explains that "ISA" refers to "Industry Standard Architecture." Patent at 1:13-14. ISA "was developed by IBM Corporation for use in its AT-type computers that utilize an Intel 80x86 microprocessor." Id at 1:17-20. "The architecture of ISA-compatible computer systems is well known and [is not] described in detail" in the patent. Id at 3:48-50. The Texas R&R concluded that "an 'ISA-compatible computer' is a computer that can handle ISA standard defined interrupts." Texas R&R at 14. This was adopted by the Texas Order.
Defendants agree with the Texas construction. Defs Br (Doc #207) at 24. Plaintiff asks the court to "clarify" the Texas construction with the caveat that "[n]ot all ISA standard defined interrupts need to be used in an ISA-compatible computer." Pl Br (Doc #199) at 23. Plaintiff is anxious that under the Texas construction, defendants might "argu[e] to the jury that an ISA-compatible computer must actually use all of the ISA standard defined interrupts." Id.
The parties agree that the patent states that IRQ15 (which is used in the preferred embodiments of the patent as the "second interrupt signal") is not defined on a conventional ISA system. Id; Defs Br (Doc #207) at 24. Defendants quarrel with plaintiff's proposal on the ground that it "muddies the waters by leaving open the possibility that some of [the IRQs that are indispensable to a functioning ISA-compatible computer] might not be used." Id. That is, defendants contend that, with respect to defining interrupts, ISA compatibility has some room for flexibility, but a certain minimal set of interrupts must be used.
The Texas construction -- "a computer than can handle ISA standard defined interrupts" -- is sufficient, at least for the time being. Nothing in this language excludes the possibility that some interrupts might not be used in a given computer. Plaintiff is free to argue that point to the jury. It may be that the parties' true dispute is not over an ISA-compatible computer's ability to handle interrupts, but rather over what interrupts must be defined for a computer to be dubbed "ISA-compatible." If so, the parties may apply to the court for a further construction of "ISA-compatible computer," presenting relevant briefing and authorities when they do.
Accordingly, the court construes "ISA-compatible computer" as "a computer that can handle ISA standard defined interrupts."
"additional function key" (claim 1) and "non-conventional function key" (claim 5)
The parties list these terms as disputed, Joint Statement (Doc #196), but they do not appear to brief their dispute. Plaintiff seems to contend that "additional function key" is the "Fn" key, while defendants contend that it should mean "an extra function key." While the "Fn" key is disclosed as part of a preferred embodiment, see Patent fig 5, there is nothing to limit the claim to this preferred embodiment.
The court essentially agrees with defendants that the terms "additional function key" and "non-conventional function key" can be construed according to their plain meaning. Indeed, the court concludes on the record before it that no construction is needed at all. Accordingly, the court declines to construe these terms.
"data scan code" (claim 1) and "scan code" (claim 5)
The parties (and the Texas Order) agree that a "scan code" (or "data scan code") is "a code number that the keyboard generates or outputs whenever a key is depressed or released." Plaintiff and defendants part company on whether "a row and a column signal" -- the electrical signals generated from the grid of wires that underlays the keys of the keyboard -- constitutes a scan code. Plaintiff argues that a row and column signal can constitute a scan code; defendants argue that they cannot.
Throughout the patent, the discussion of "scan codes" concentrates on how they are passed among different components of the system. See, e g, Patent at 4:25-27 ("The communication of keyboard scan code information from the keyboard controller 128 to the microprocessor 110 is illustrated pictorially in FIG 2."). There is (almost) no discussion of how the scan codes are created, and hence what forms they may take; the specification simply says that "the keyboard generates a scan code." Id at 4:18. But the specification makes clear why it is unnecessary to specify how scan codes are created:
The keyboard 150 operates in a known conventional manner to repeatedly scan a plurality of contact switches associated with the keys on the keyboard to determine if one or more contact switches is open or closed. When an open contact switch is closed by depressing a key or a closed contact switch is opened by releasing a key, the keyboard 150 generates a scan code which is communicated to the keyboard controller 128 via the signal lines 152.
Id at 4:12-21. In other words, the specification leaves "scan code" undefined (and hence ambiguous) and instructs that "scan code" is to be construed as it would be understood by one skilled in the art.
Elsewhere in the specification, the operation of the keyboard is explained in more detail:
When the keyboard microprocessor 250 detects an active row signal, it uses the row signal and the currently active column signal to uniquely identify which of the keys on the keyboard is depressed. The key is identified with a scan code in a conventional manner.
Id at 7:51-55. Plaintiff posits that "us[ing] the row signal and * * * column signal to uniquely identify" the depressed key means that a row and column signal is a scan code, because scan codes identify keys. But in the excerpt block-quoted above, "identify" in the first sentence is best interpreted as "determine," while "identified" in the second sentence means "identified to the keyboard controller." The two sentences would be redundant (or inconsistent) if "identify" and "identified" had the same meaning. Admittedly, this part of the specification is poorly drafted. The more revealing aspect of this excerpt is the suggestion that scan codes are "conventional, i e, well known to those skilled in the art. Moreover, by using different words, the specification itself suggests a distinction between "signals" and "scan codes."
In light of the specification's intentional lack of a definition of "scan code," it is appropriate for the court to look to dictionary definitions. Conventional dictionary definitions make a distinction between "signals," which are physical phenomena, and "codes," which are descriptions or identifiers. In the relevant senses, a "code" is a "symbol used to identify something that lacks a specific name" and a "signal" is a "detectable physical quantity or impulse (as a voltage, current, magnetic field strength) by which messages or information can be transmitted." Webster's Third New International Dictionary (Merriam-Webster, Unabridged ed 1981) 437, 2115.
Two technical references cited by defendants are particularly illuminating, and are consistent with the conventional dictionary definitions. First, the Microsoft Press Computer Dictionary defines "scan code" as a "code number transmitted to an IBM or compatible computer whenever a key is pressed or released." Microsoft Press Computer Dictionary 306-07 (Microsoft, 1991), reprinted in Dang Decl (Doc #209) Ex 6. It is difficult to see how a "signal" can be a "number." Second, How Computers Work explains that a "key's circuit carries a signal to the [keyboard] microprocessor" which in turn "generates a number, called a scan code". Ron White, How Computers Work 112 (ZD Press, 1993) (emphasis added), reprinted in Dang Decl (Doc #209) Ex 5. This reference can only be understood to exclude the possibility that a row and column signal is a scan code.
Even resort to extrinsic evidence to construe "scan code" confirms these dictionary meanings. Defendants' expert makes the point very clearly:
The microprocessor within the keyboard determines if a key is depressed by scanning the keyboard's matrix, which can be thought of as a grid of circuits (row wires and column wires) underneath the keys. Each key has an electrical switch associated with it, with the electrical switch located at an intersection of a row wire and a column wire. The microprocessor scans each column of the keyboard matrix, reading all the rows of a selected column to determine whether any of the switches associated with any of the respective keys is pressed or released. The row and column signals on the row wires and column wires of the keyboard matrix only represent the present electrical state of a given key as described above. * * * Row and column signals are signals communicated between the keyboard microprocessor and keyboard matrix to enable the microprocessor to monitor actuation and release of keys of the keyboard. However, the row and column signals are never provided from the keyboard microprocessor to the host computer. In fact, the keyboard microprocessor has to first convert or translate the pairing of a row signal and a column signal into a scan code. It is this scan code that is provided to the host computer.
Buscaino Decl (Doc #208) ¶15 (emphasis in original).
This lends considerable credence to the specification's apparent distinction between "signals" and "codes": "Signals" are raw electrical states (indicating, for example, an open or closed circuit), while "codes" are translated or abstracted information describing an electrical state.
Plaintiff's own expert describes the operation of a keyboard in much the same way, also respecting the difference between signals (the detection of which he refers to as "looking for closed switches") and codes (which he refers to as "values"):
The process that the keyboard microprocessor is performing is to scan each column successively looking for closed switches and reporting their value back to the CPU when one is found. Because the keyboard microprocessor performs a scan of each column to find which key or keys have been pressed, the values that correspond to the pressed keys are called "scan codes."
Wedig Decl (Doc #201) ¶19. In fact, plaintiff's expert even describes scan codes as being "generated." Id ("Once one or more pressed keys have been detected and one or more scan codes generated * * *."). If the row and column signals are used to detect a key press, it makes sense to speak of subsequently "generating" a scan code only if the row and column signals are not themselves scan codes.
Plaintiff strenuously suggests that two pieces of extrinsic evidence -- manuals for the Phoenix 80286 Advanced Rom Bios, see Wedig Supp Decl (Doc #215) Ex A -- show that row and column signals can be scan codes. One diagram depicts "keyboard make/break scan codes" flowing to an "8042 keyboard controller" which in turn issues a "make/break system scan code." Id at 8-4. Another diagram explains that "each time a key is pressed, the keyboard hardware itself generates a keyboard make or break scan code. In and of themselves, keyboard scan codes are not understandable to the system." Id at 10-5. While this may be evidence that scan codes can take different forms, it is not evidence that row and column signals are themselves scan codes. This is not the only example of dispute among the parties about whether scan codes are standardized and, if so, what standard applies. But that is all quite beside the point for purposes of construing "scan code," because there is no basis in the claims for limiting the term to one particular industry standard or another. What is clear is that row and column signals are not scan codes.
Accordingly, the court adopts defendants' proposal (with a grammatical correction) and construes "scan code" and "data scan code" as "a code number that the keyboard generates whenever a key is depressed or released, said code number created by converting a pairing of a row signal and a column signal in the keyboard matrix."
"keyboard controller" (claims 1 and 5)
The parties largely agree on the construction of this term: The "keyboard controller" is "a component that activates interrupt signals in response to receipt of data scan codes from the keyboard and, upon request, transmits the data scan codes to the computer." The Texas Order further provides that the keyboard controller is "distinct from the keyboard" and "the keyboard and keyboard controller may be physically separate or housed together in the same unit." Texas Order at 4-5. Plaintiff purports to agree with the Texas Order, but suggests that "distinct from the keyboard" should be omitted as superfluous and potentially confusing in light of the (more accurate) "physically separate or housed together" caveat. Pl Br (Doc #199) at 8-9. Defendants protest that throughout the patent, the keyboard and the keyboard controller are referred to as separate entities that are "connected" or "coupled." Def Br (Doc #207) at 13. As such, defendants contend, the reference to they keyboard controller being "distinct from the keyboard" is the only way to give meaning to the claim language.
The crux of the dispute flows from the parties' agreed construction for "keyboard," which is "an input apparatus containing internal circuitry to output or generate data scan codes." In light of the court's construction of "data scan code," the parties' agreed-upon construction of "keyboard" implies that the process of "converting a pairing of a row signal and a column signal in the keyboard matrix" occurs within the functional unit of the keyboard.
There is no basis in the claim language for prohibiting the keyboard controller and the keyboard from being housed together. But by the same token, the patent is unambiguous in treating the keyboard and keyboard controller separately -- they are distinct in the schematics shown in the figures of the patent and they are distinct by the literal words of the claims. Moreover, the court's construction of "connected" (see below) would seem to require that the keyboard and the keyboard controller be electronically or functionally discrete components.
In this light, "distinct from the keyboard" is not superfluous; to the contrary, it dovetails with the court's construction of "scan code" and "connected to." Although the keyboard and keyboard controller are not (necessarily) physically distinct, they must be electronically or functionally distinct. The court adopts the Texas construction in that respect. The parties are free to remind the jury that, as the Texas Order put it, "the keyboard and keyboard controller may be physically separate or housed together in the same unit." Nothing in the court's construction says otherwise.
A word is appropriate about the much-contested figure 7, which depicts a keyboard matrix connected to a microprocessor (250) with IRQ and data lines coming off it. The figure is difficult to decipher, for the specification states that the keyboard controller (220) is illustrated in figure 7, yet no such object appears in the figure. There is a later reference to figure 7 to the keyboard controller (130), but 130 does not appear on figure 7, and where it does appear (figures 1-4 and 6), 130 is an interrupt controller. Setting aside figure 7, however, the text of specification is consistent with the court's conclusion that the keyboard controller and the keyboard (including the keyboard microprocessor) can be closely physically connected but electronically or functionally distinct. For example, the specification explains that "the keyboard controller 220 and the keyboard 210 [may be] combined as a single unit." Patent at 7:33-34. "Combined as a single unit" suggests physical integration, but the continuing reference to the keyboard controller and the keyboard as distinct entities suggests some electronic or functional distinction between them.
Accordingly, the court construes "keyboard controller" as "a component, electronically or functionally distinct from the keyboard, that activates interrupt signals in response to receipt of data scan codes from the keyboard and, upon request, transmits the data scan codes to the computer."
"connected to" (claim 1)
Plaintiff proposes to construe this seemingly simple term as "joined or linked together, including (1) two parts that have been physically joined or designed together (at least in part) through which electrical signals may flow, or (2) the contact between a device and one of its constituent components through which electrical signals may flow." Defendants propose "join or fasten one thing to another usually by something intervening."
The court intuits that the dispute here arises from the technological progress of integrating multiple functions onto a single microchip. Plaintiff, unsurprisingly, would construe "connected" as "broad enough to encompass the contact between a device and one of its constituent parts." Pl Br (Doc #199 at 15. Defendants contend that this lacks support in the intrinsic record or dictionary definitions.
By its plain meaning, "connected to" refers to the "joining or fastening of one thing to another," just as defendants propose. But in the context of the patent, which relies heavily on schematics to give content to "connected to" (see figs 1-4, 6-10), the connections in question are not the nuts-and-bolts linkages of mechanical engineering. See ACTV, Inc v Walt Disney Co, 346 F3d 1082, 1088 (Fed Cir 2003) ("While certain terms may be at the center of the claim construction debate, the context of the surrounding words of the claim also must be considered in determining the ordinary and customary meaning of those terms."). There is little if anything in the patent about the physical size or contiguity of the elements depicted in the figures or described in the specification. Plaintiff's proposed construction is a (flawed) attempt at capturing both the unimportance of physical integration and the importance of identifying electronically or functionally discrete units. A physically unified microchip with two electronically distinct circuits that are "connected" within the chip exemplifies one circuit "connected to" another.
The court is thus not entirely satisfied with either proposed construction: Plaintiff's construction ventures too far from the plain meaning of the term to be construed; defendants' proposal's use of "thing" is too tied to a realm of physically discrete objects. The patent is about electronics and the figures represent functions; rather than simply "things," the claims describe "electronically or functionally discrete things." The court therefore makes its own construction, on which it will entertain a motion to reconsider if necessary: "Connected to" refers to "joining or fastening one electronically or functionally discrete thing to another." With this language, the parties should be able to explain to the jury whether the products in suit do or do not meet the patent's claimed scope involving electronically or functionally discrete components.
"coupled to" (claim 5)
The court sees no reason to give this term a construction different from the construction given to "connected to," nor do the parties offer significantly different constructions for these two terms. Accordingly, the court construes "coupled to" to refer to "joining or fastening one electronically or functionally discrete thing to another."
"second interrupt signal" (claims 1 and 5)
The parties agree that the "first interrupt signal" must be IRQ1, and so they agree that the "second interrupt signal" must not be IRQ1. The parties dispute whether the second interrupt signal can be any other interrupt (as plaintiff contends), or whether it is limited to one of the ISA-standard interrupts (as defendants contend). The question boils down to whether one skilled in the art would, in 1990 and in the context of a patent directed toward ISA-compatible computers, understand "interrupt" to encompass interrupts other than the ISA-standard IRQ interrupts. Plaintiff's expert says yes. Wedig Decl (Doc #201) ¶40 Defendants' expert says no. Buscaino Decl (Doc #208) ¶38.
The dispute turns on whether the claims must be read strictly in the context of ISA-compatible systems. If so, it appears that there is no significant disagreement that the claim reads upon only IRQ interrupts other than IRQ1. If the claim may be interpreted beyond the context of ISA-compatible systems, then there is no such limitation. Defendants point out that the agreed-upon construction of "first interrupt signal" flows from the fact that ISA-compatible computers use IRQ1 as the standard interrupt for keyboard input; why, defendants ask, should not the same limitation be imposed on "second interrupt signal"?
Whether it was correct or not for the parties to stipulate to the construction of "first interrupt signal" as they did (apparently a decision made in light of the prosecution history), the court sees no basis in the literal claim language to limit "second interrupt signal" to ISA-standard interrupts. If anything, the choice of "interrupt" -- a generic, but well-understood engineering term, see Wedig Decl (Doc #201) ¶40 --rather than "IRQ" or "ISA-standard interrupt" evinces the patentee's choice to eschew such a limitation in favor of a more expansive claim. Indeed, although IRQs are referenced throughout the specification, the claims themselves eschew any reference to IRQs.
It may well be that the independent requirement in the claims that the invention function "in an ISA-compatible computer" will tend to exclude non-ISA-standard interrupts. But that is not a limitation introduced by "second interrupt signal," and so it would be improper to construe "second interrupt signal" more narrowly than its literal terms. Accordingly, the court adopts plaintiff's proposed construction in this respect.
Plaintiff also seeks to elaborate on this construction by specifying that the interrupt "is generated when the non-conventional function key is depressed either alone or in conjunction with a conventional key." This additional language, plaintiff asserts, is necessary to construe "second interrupt signal" in the context of claim 5. The court rejects plaintiff's invitation to add this language to the construction, although the court returns to this issue shortly. In this context, it is enough to say that "second interrupt signal" carries no germ of a requirement about how or when the signal is generated; other words in the claims carry that meaning.
Accordingly, the court construes "second interrupt signal" as "any interrupt, other than IRQ1."
"upon receipt of a scan code corresponding to said non-conventional function key" (claim 5)
As promised, the court now returns to the question whether claim 5 is limited to, or at least reads upon, generating the second interrupt signal "upon receipt of a scan code that is generated by the activation of a single non-conventional function key alone." There is little to say on this subject; the literal terms of the claim are that the second interrupt signal is generated "upon receipt of a scan code corresponding to said non-conventional function key." Receipt of that scan code alone is a sufficient condition for triggering the second interrupt signal. The specification acknowledges that this is a possible (albeit nonpreferred) embodiment. Patent at 7:1-3 ("[T]he [non-conventional] thirteenth function key Fn can be used to specify a particular function in a manner similar to the [conventional] function keys F1 through F12.").
By the same token, nothing in the claim language suggests that receipt of that scan code alone is a necessary condition; it may be that, in a proper embodiment, claim 5 could read on a function key in combination with a conventional key. The Texas Order is consistent with this view. See Texas Order at 11 ("While the broad language of claim 5 does not expressly call for a second interrupt signal in response to a scan code corresponding to the non-conventional key with another key, neither does that language foreclose the option of a scan code representing a combination of keys.").
Whatever the case, the claim language is so clear that the court need not look to the specification (or elsewhere) to construe it. Accordingly, being unable to improve on the literal claim language, the court declines to construe the language identified by defendants.
"indexes a first memory location pointer" (claim 4)
The parties agree that, in the activity of a CPU, "to index" is "to select or access." There is also no dispute over what is indexed; the parties agree that the "memory location pointer" is "an identifier that corresponds to the start of the first conventional interrupt handling routine" of claim 1 (upon which claim 4 is dependent). But the parties dispute how the indexing is accomplished. Plaintiff contends that there is no limitation on the type of pointer. Pl Br (Doc #199) at 23-24. Defendants contend that the claim should be limited to indexing performed through an interrupt vector, because ISA-compatible computers perform indexing using interrupt vectors. Defs Br (Doc #207) at 23.
The dispute appears to mirror the dispute over "second interrupt signal"; there, as here, defendants assert that the claims read only upon ISA-compatible computers and therefore claim terms should be limited to their meaning in the ISA computing environment. The court is again unpersuaded. If the patentee had meant to limit the claim to particular methods of indexing, such a limitation would appear in the literal terms of the claim language. And indeed, as plaintiff points out, claims 6 and 7 refer to "interrupt vectors." Claim 4 does not. Accordingly, it is not so limited, and the court adopts plaintiff's proposed construction.
Accordingly, the court construes "indexes a first memory location pointer" as "selects or access an identifier that corresponds to the start of the first conventional interrupt handling routine."
"indexing a second memory location pointer" (claim 4)
The construction of this phrase tracks the construction of "indexes a first memory location," mutatis mutandis. Accordingly, the court construes "indexing a second memory location" as "selects or accesses an identifier that corresponds to the start of the second non-conventional interrupt handling routine."
Finally, with breathless accusations of sandbagging on plaintiff's part, defendants move to strike portions of plaintiff's reply brief, or in the alternative, to consider a supplemental declaration from defendants in rebuttal. Doc #219. It is undeniable that plaintiff devotes considerably greater attention to the construction of "scan code" in its reply brief than in its opening brief. Compare Pl Br (Doc #199) at 22-23 (less than one page of briefing on "scan code") with Pl Reply (Doc #213) at 11-14 (over three pages of briefing on "scan code"). But such is the vice of three-step Markman briefing; a patentee who files an opening brief may not be fully aware of the arguments that its opponent will advance in its responsive brief. In any event, the court has found it unnecessary to refer to the material presented in the reply brief (and, by extension, the rebuttal material offered in defendants' supplemental declaration). Accordingly, defendants' motion (Doc #219) is DENIED as moot.
IT IS SO ORDERED.
© 1992-2005 VersusLaw Inc.