PAULINE NEWMAN, Circuit Judge.
Wang Laboratories appeals the final judgment of the United States District Court for the Eastern District of Virginia,
granting summary judgment that America Online and Netscape Communications Corporation did not infringe, literally or under the doctrine of equivalents, the asserted claims of Wang’s United States Patent No. 4,751,669, issued June 14, 1984 (the ’669 patent). The judgment is affirmed.
BACKGROUND
The ’669 patent, entitled “Videotex Frame Processing,” is directed to an online information system developed by Wang engineers in 1983 and 1984 for use with the known videotex system. The claimed invention provides users with textual and graphical information from computer-controlled databases via interactive two-way communication over a telephone network. Using the videotex system it is possible to review, retrieve, and store pages or “frames” of data from many information suppliers. The overall system is illustrated in Figure 1 of the ’669 patent, which shows the subscriber’s equipment (12) connected through a two-way channel (16) to a switched telephone network (18), which in turn is connected to the equipment of suppliers (20) and (24) via two-way channels (22) and (23):
[[Image here]]
Wang asserted claims 1, 8, 9, 20, 22-28, and 38, against AOL and Netscape. On appeal Wang stresses infringement of claims 20 and 38. Claim 20 is directed to a keyword feature that allows the user to assign a name to a particular page or frame for ease in retrieval, and is said by Wang to read on AOL’s “favorite places” and Netscape’s “bookmark” features:
20. Apparatus for retrieving selected frames of information from a central videotex supplier of the information frames, each information frame having an associated unique identifier assigned by the supplier for retrieving the frame, the apparatus comprising
a display device connected to display the information frames,
storage means connected to store the identifier and a unique keyword selected by an operator and associated with the identifier,
menu means connected to display on the display device a menu frame containing the keyword,
data entry means connected to enter into the apparatus a request for retriev
al of a selected information frame by moving a cursor to the keyword associated with the selected information frame, and
a processor connected to be responsive to the data entry means for retrieving the information frame in response to the entry of the request by transmitting the associated identifier to the supplier.
Claim 38, whose structural components are presented in means-plus-function form, is directed to the feature of tagging a stored page or frame in order to identify and activate the decoding protocol when the information is retrieved:
38. Apparatus for locally processing frames of information received from central videotex suppliers, different frames being encoded in accordance with different protocols, comprising
means connected to locally store the information frames,
means connected to locally display the frames,
means connected to decode, the locally stored frames as they are displayed, and
means connected to tag each stored frame with a header indicating one of said different protocols as having been used for encoding the frame,
the means connected to decode being arranged to decode each frame in accordance with the protocol indicated by the header on the frame.
The summary judgment turned on interpretation of the asserted claims, specifically, whether the claims are limited to the specific technology by which the frames of information are processed and displayed. At the time the ’669 invention was made, several protocols existed for processing and displaying computer-generated data. They were of two general types: character-based protocols and bit-mapped protocols. In a character-based system the screen display is divided into a grid wherein a character, such as a letter or a number, is placed in each cell of the grid. A typical character-based system sends frames of digital information to a subscriber as a stream of encoded bytes identifying the characters to be displayed and containing instructions about the attributes of the displayed frame such as the size of the characters or their background color. The system also has a limited capability to display graphics by way of mosaic characters. As opposed to the grid-based characters-only display of character-based protocols, a bit-mapped protocol encodes an image with reference to the individual pixels of the display monitor. It is not disputed that the accused AOL and Netscape systems utilize bit-mapped protocols, which are today used for most Internet interactions.
Following a hearing at which both sides presented expert testimony, the district court construed all of the claims in suit as limited to character-based systems. This construction followed from the court’s definition of the term “frame,” which appears throughout the claims, as a “page of information assembled prior to display which is encoded in a character-based protocol ... to then be displayed on the screen representing a fixed full screen arrangement, such as rows and columns, of alphanumeric and graphic characters.” The court also held that the suppliers of data accessed by the claimed system must provide information frames encoded in a character-based protocol. This claim construction was appropriate, reasoned the court, because the specification of the ’669 patent is limited to character-based protocols, and because the prosecution history reinforces this limitation. Based on the claims as construed, the district court granted summary judgment of non-infringement as to literal infringement including infringement under 35 U.S.C. § 112 ¶ 6, and also under the doctrine of equivalents.
Claim Construction
Determination of infringement is a two-step analytic procedure, whereby first the claims are construed as a matter of law, and then are applied to the accused device, a question of fact.
See EMI Group
North America, Inc. v. Intel Corp.,
157 F.3d 887, 891, 48 USPQ2d 1181, 1184 (Fed. Cir.1998);
Multiform Desiccants Inc. v. Medzam Ltd.,
133 F.3d 1473, 1476, 45 USPQ2d 1429, 1431 (Fed. Cir.1998). As often occurs, the factual questions of infringement were resolved by the district court’s construction of the claims.
See Markman v. Westview Instruments, Inc.,
52 F.3d 967, 989, 34 USPQ2d 1321, 1337 (Fed.Cir.1995)
(en
banc) (Mayer, J., concurring) (“to decide what the claims mean is nearly always to decide the case”).
Wang argues on appeal that the narrow definition of “frame” adopted by the district court is not warranted by the specification, is not required by the prosecution history, and is negated by the overall structure of the claims. However, based on the record before us, we conclude that the district court’s claim construction was appropriate.
The Specification
The parties agreed before the district court that the term “frame” can in general usage be applied to bit-mapped display systems as well as to character-based systems; experts for both sides so testified. The disagreement was as to whether the term “frame” in the ’669 claims embraced this general usage, or whether the term would be understood by persons of skill in this field as limited to the character-based systems described in the ’669 patent. The district court started its analysis with the specification.
See Slimfold Mfg. Co. v. Kinkead Indus.,
810 F.2d 1113, 1116, 1 USPQ2d 1563, 1566 (Fed.Cir.1987) (claims are construed in light of the specification, of which they are a part).
A typical example of the ’669 patent’s usage and definition of the term frame is seen in the paragraph in the specification describing the videotex system utilized in the invention:
The videotex supplier electronically stores the information supplied by the different sources in the form of hundreds of thousands of pages (frames) each representing a collection (arranged, for example, in rows and columns) of alphanumeric and graphic characters to be displayed on a cathode ray tube (CRT) at the subscriber’s location.
’669 patent, col. 1, lines 16-22. This text is specific to “characters,” as are all similar descriptions in the specification. Wang argues that the term “frame” as used in the patent refers broadly to a unit of digital data that can be selected and displayed, and is not limited to use of a character-based protocol. Thus for the paragraph quoted
supra,
Wang contends that the display of “alphanumeric and graphic characters” can include both character-based and bit-mapped displays. Wang also argues that the parenthetical usage “(arranged, for example, in rows and columns)” means that any character-based limitation to “frame” gleaned from the above passage is merely exemplary, and that the display need not be the grid-like pattern of a character-based display system. Wang presented expert witness testimony to this meaning before the district court.
AOL and Netscape dispute Wang’s position, arguing that the usage of “frame” in conjunction with “characters” limits the invention to a character-based protocol. AOL and Netscape presented expert testimony to this meaning, pointing out that the ’669 specification consistently illustrates “frame” only with respect to a character-based protocol, The defendants also point out that in those patent drawings that contain details of the display, such as Figure 6 showing a software flow chart of the decoder module, only a character-based protocol is shown as being received, decoded, stored, and displayed:
[[Image here]]
According to the ’669 specification, “Whenever a stored frame is to be displayed, [the decoder module in Figure 6] must decode the characters and attributes” and deliver the decoded characters to a character-based display module. The end result, the specification continues, is “a displayable alphanumeric or graphic character for each of the row-and-column positions of the display.”
The only system that is described and enabled in the ’669 specification and drawings uses a character-based protocol. The specification mentions non-character-based protocols, for example, in the “Background of the Invention” statement that “Different videotex protocols have been developed in different countries, for example Prestel [a character-based protocol operated by British Telecom], Telidon [a bit-mapped graphics protocol] and NAPLPS [North American Presentation-Level Protocol Syntax, a bit-mapped protocol].” The district court viewed the references to bit-mapped protocols as acknowledgments of the state of the art, and not as an enlargement of the invention described in the patent. We agree, and conclude that the references to other known protocols do not describe them as included in the applicant’s invention, and that the specification would not be so understood by a person skilled in the field of the invention.
Reinforcing this conclusion, it is not disputed that Wang had not been able to implement a bit-mapped protocol in the claimed system; the inventors testified that they had been unable to develop a NAPLPS-based decoder. Wang argues that it is irrelevant to the construction of the claims whether the specification con
tains an enabling description of any bit-mapped decoder, stating that “enablement is a requirement for validity, not a factor in claim construction.” However, the claims are not properly construed to have a meaning or scope that would lead to their invalidity for failure to satisfy the requirements of patentability.
See Eastman Kodak Co. v. Goodyear Tire & Rubber Co.,
114 F.3d 1547, 1556, 42 USPQ2d 1737, 1743 (Fed.Cir.1997) (the court “seeks to interpret claims to preserve, rather than defeat, their validity”). Although Wang is correct that a claim is not invalid simply because it embraces subject matter that is not specifically illustrated, in order to be covered by the claims that subject matter must be sufficiently described as the applicant’s invention to meet the requirements of section 112. This requirement was not met as to protocols other than character-based.
See Modine Mfg. Co. v. United States Int’l Trade Comm’n,
75 F.3d 1545, 1557, 37 USPQ2d 1609, 1617 (Fed.Cir. 1996) (“when claims are amenable to more than one construction, they should when reasonably possible be interpreted so as to preserve their validity”) (citations omitted).
Wang also argues that even if the specification is deemed to describe only the character-based protocol the claims should not be so limited, as the interchangeability of character-based and bit-mapped information protocols was known at the time the ’669 invention was made. According to Wang, it is the user interface features of claims 20 and 38 that is the invention of the ’669 patent, and not the choice of protocol for the frames of information. The defendants respond that the entire ’669 specification is directed to, and is limited to, interactions based on character-based information frames, that this limitation underlies the subsidiary features of keywords and the like, and that the claims so state. We agree that the claims are not directed solely to the user interface, but to the electronic system that is described as implementing this interface.
Wang states that the character-based protocol is simply a “preferred embodiment,” and that the embodiment described in the specification does not set the boundaries of the claims, citing
Comark Communications, Inc. v. Harris Corp.,
156 F.3d 1182, 1186, 48 USPQ2d 1001, 1004 (Fed.Cir.1998), for its statement that “limitations from the specification are not to be read into the claims.” AOL and Netscape respond that when the subject matter that is claimed is the only subject matter that is described and enabled in the specification, that is the invention itself, and not simply a “preferred” example of a broader invention that is not described and enabled, citing
Modine,
75 F.3d at 1551, 37 USPQ2d at 1612 (when the “preferred embodiment” is described as the invention itself, the claims are not entitled to a broader scope than that embodiment). Although precedent offers assorted quotations in support of differing conclusions concerning the scope of the specification, these cases must be viewed in the factual context in which they arose. Whether an invention is fairly claimed more broadly than the “preferred embodiment” in the specification is a question specific to the content of the specification, the context in which the embodiment is described, the prosecution history, and if appropriate the prior art, for claims should be construed, when feasible, to sustain their validity. The usage “preferred” does not of itself broaden the claims beyond their support in the specification.
See id.; General American Transportation Corp. v. Cryo-Trans, Inc.,
93 F.3d 766, 770, 772, 39 USPQ2d 1801, 1803, 1805-06 (Fed.Cir.1996) (the teaching in the specification was “not just the preferred embodiment of the invention; it is the only one described”). The only embodiment described in the ’669 patent specification is the character-based protocol, and the claims were correctly interpreted as limited thereto.
Prosecution History
The prosecution history also supports the limitation to character-based
frames. In an Information Disclosure Statement filed by Wang during the prosecution of the parent application, Wang distinguished a reference (Fleming) describing the NAPLPS system by stating that the reference “encodes pictorial information ... on the pel [picture element] level, rather than on the character level.” Wang argues that this statement was in the parent application and does not apply to the continuation-in-part that is the ’669 patent. However, this subject matter is common to the continuation-in-part application, and argument concerning the Fleming reference was correctly viewed as applying to the common subject matter.
See Jonsson v. Stanley Works,
903 F.2d 812, 818, 14 USPQ2d 1863, 1869 (Fed.Cir. 1990) (applying arguments made in related applications). This history reinforces the conclusions that the inventors focused their invention, and the description and claims directed to that invention, on a character-based system, and that a person of skill in this field reading this history would so understand the explanation of the Fleming reference.
Claim Differentiation
Wang also argues that the overall claim structure supports and requires the conclusion that a character-based limitation should not be read into claims that do not explicitly recite it, under the doctrine of claim differentiation. However, all of the claims recite “frames,” which are shown in the specification as only character-based. The claim differentiation argument of itself does not support interpreting “frames” as unlimited to any protocol. As explained in
Multiform Desiccants, Inc. v. Medzam, Ltd.,
133 F.3d 1473, 1480, 45 USPQ2d 1429, 1434 (Fed.Cir.1998), “the doctrine of claim differentiation can not broaden claims beyond their correct scope, determined in light of the specification and the prosecution history and any relevant extrinsic evidence.”
On review of all of the arguments of both sides, with benefit of the summary judgment record including the views of experts for both sides, we conclude that the terms “frame” and “information frame” are directed to and limited to a character-based protocol. The district court’s claim construction as to this issue is affirmed.
The Means-Plus-Function Terms
Wang invokes 35 U.S.C. § 112 ¶ 6 in connection with the “means” clauses in the claims. In claim 23, for example, an apparatus for storing and retrieving frames of digital information on a subscriber’s equipment includes a processor comprising “means” whereby the subscriber associates a name with an information frame:
23. Apparatus for locally storing frames of digital information received from a central videotex supplier, comprising
a port connectable to the central supplier for receiving at least one of the information frames from the central supplier,
a display device connected to locally display at least one of the information frames received from the central supplier,
a storage device connected to locally store, after it has been displayed, as a stored frame for later redisplay, at least one of the information frames received from the central supplier,
a data entry device connected to enable the enable [sic] of data into the apparatus, and
a processor comprising
means connected to accept from the data entry device a name specified by a subscriber to be associated with a particular displayed information frame,
means connected to store the particular information frame in the storage device together with the associated name,
and
means connected to retrieve the particular information frame from
the storage device
in response to entry-through the data entry device of a command corresponding to the name specified to that frame.
(Emphases added.) The “means” elements at issue are directed to software structure that is described in the specification as a character-based decoder performing the function of storing and retrieving information frames. The specification teaches that in order to retrieve frames of information from memory for display, these frames are loaded into the frame table and decoded by the character-based decoder module described in the specification. Similarly for claim 38,
see supra,
the “means connected to locally store the information frames” and “means connected to decode the locally stored frames as they are displayed” are described in the specification as character-based structures, writing received characters into the frame table for storage and decoding a stored character-based frame for display.
Wang states that the “means” is not limited to the character-based protocol described in the specification, and that the known bit-mapped protocol is an equivalent means interchangeable with the character-based protocol. When the claimed function is adequately supported by structure in the specification, equivalent structure under § 112 ¶ 6 is not required to be described in the specification. Thus Wang argues that the known bit-mapped technology need not be included in the specification in order to be equivalent under § 112 ¶ 6. AOL and Netscape respond that although character-based and bit-mapped modules were known to perform similar functions, they had different capabilities and operated in ways that were so different that the Wang scientists were unable to adapt the bit-mapped protocol to their system. The defendants argue that Wang’s inability to implement bit-mapped technology should not be rewarded with a judgment of equivalency after others later succeeded.
The evidence in the record before us supports the district court’s ruling of non-equivalency under § 112 ¶ 6. The expert testimony was to the general effect that although these protocols are today interchangeable as modules when placed in self-contained “black boxes” and inserted into the appropriate software, they function in accordance with markedly different principles, have greatly different capabilities, and generally do not operate in substantially the same way. The ’669 inventors testified that Wang had discontinued development of the bit-mapped module in this system based, at least in part, on technologic difficulties. Witnesses also testified that Wang chose to work with the character-based Prestel protocol because it had already achieved popularity and was less complex than the bit-mapped NAPLPS system, although Wang knew that Prestel had limited graphics capabilities. Wang’s statement in the prosecution of the parent application that the Fleming reference to the NAPLPS system was not applicable because it did not describe a character-based system weighs heavily against Wang’s entitlement to reach this system under § 112 ¶ 6. Thus the specification, the prosecution history, and the testimony of witnesses support the conclusion that the bit-mapped protocol is not an equivalent under § 112 ¶ 6.
The district court’s summary judgment that there is not literal infringement is affirmed.
The Doctrine of Equivalents
The district court also summarily decided the issue of equivalency under the doctrine of equivalents. Summary judgment when factual issues are raised may nonetheless be appropriate when, on the non-movant’s version of the facts and drawing all reasonable factual inferences in favor of the non-movant, no reasonable trier of fact could find liability.
Anderson v. Liberty Lobby, Inc.,
477 U.S. 242, 250, 106 S.Ct. 2505, 91 L.Ed.2d 202 (1986). On appellate
review, we apply the same standard.
EMI,
157 F.3d at 891, 48 USPQ2d at 1184;
Strattec Security Corp. v. General Automotive Specialty Co.,
126 F.3d 1411, 1419, 44 USPQ2d 1030, 1037 (Fed.Cir.1997).
Wang correctly states that a known equivalent need not be described in the specification, citing
Wamer-Jenkinson Co. v. Hilton Davis Chemical Co.,
520 U.S. 17, 37, 117 S.Ct. 1040, 137 L.Ed.2d 146, 41 USPQ2d 1865, 1874 (1997) (“rejecting the more severe proposition that equivalents must not only be known, but must also be actually disclosed in the patent”). Wang argues that even on the district court’s claim construction the doctrine of equivalents applies to the character-based and bit-mapped protocols, for these protocols are interchangeable.
See id.
at 25, 36, 117 S.Ct. 1040 (“known interchangeability” is a useful objective standard of equivalency). The defendants respond that the prosecution history negates the use of a bit-mapped protocol in this system, citing Wang’s distinction of the Fleming reference to the NAPLPS system. It was not disputed that there are substantial differences between these technologies; Wang’s technical expert called them “huge, huge differences.” It was also undisputed that “AOL’s system permits graphic displays, the detail and complexity of which could not be produced by the character-based system disclosed in the ’669 patent.” The differences in operation, structure, and capabilities of these systems are so extensive that we must agree with the district court that no reasonable trier of fact could find only insubstantial differences, or substantially the same function, way, and result, between the character-based protocol of the ’669 patent and the accused graphical image systems. The judgment of non-infringement under the doctrine of equivalents is affirmed.
Costs
In accordance with Fed. R.App. P. 39(a), costs are taxed against the appellant.
AFFIRMED