CLAIM CONSTRUCTION ORDER

PHYLLIS J. HAMILTON, District Judge.

On March 25, 2015, the parties' claim construction hearing to construe the disputed terms of U.S. Patent Nos. 6,898,762 ("the `762 patent"), 7,065,709 ("the `709 patent), 7,103,906 ("the `906 patent"), 7,945,929 ("the `929 patent"), and 7,974,962 ("the `962 patent") (together, the "patents-in-suit") pursuant to Markman v. Westview Instruments, Inc., 517 U.S. 370 (1996), came on before this court. Plaintiff Netflix, Inc. ("plaintiff" or "Netflix") appeared through its counsel, Ashok Ramani, Tina Sessions, Ed Bayley, Michael Kwun, and Sharif Jacob. Defendants Rovi Corporation, Rovi Technologies Corporation, Rovi Guides, Inc., United Video Properties, Aptiv Digital Inc., and Starsight Telecast, Inc. (referred to collectively as "defendants" or "Rovi") appeared through their counsel, Yar Chaikovsky, Hong Lin, and Amol Parikh. Having read the parties' papers and carefully considered their arguments and the relevant legal authority, the court hereby rules as follows.

BACKGROUND

In this declaratory judgment action, defendants are the assignees/owners of the patents-in-suit. Four of the patents (the `709, `762, `929, and `962 patents) relate to interactive program guides for television shows and other visual media. Specifically, the `709 and `762 patents, which share a common specification, cover the use of a user's viewing history to generate personalized recommendations. The `929 patent covers the use of "combination categories" to organize programs. The `962 patent covers an invention that allows users to search for programs using categories, rather than just program names.

The fifth patent (the `906 patent) covers the creation of bookmarks for resuming playback of programs over different devices.

Netflix's original complaint was filed on December 21, 2011. In May 2012, the court stayed the case pending the outcome of an International Trade Commission ("ITC") investigation. In July 2014, after the ITC proceedings had concluded, the parties stipulated to lift the stay, and agreed to a schedule for claim construction. Shortly thereafter, Netflix filed a motion for summary judgment of invalidity under section 101, intending it to be heard before claim construction, but the court opted to consolidate the two hearings. While the court need not necessarily construe the claims before ruling on the invalidity motion, because the § 101 analysis is "often dependent on the scope and meaning of the claims," the court will, in this case, first construe the disputed claim terms. See Internet Patents Corp. v. Active Network, Inc., ___ F.3d ___, 2015 WL 3852975 (Fed. Cir. Jun 23, 2015).

DISCUSSION

A. Legal Standard

In construing claims, the court must begin with an examination of the claim language itself. The terms used in the claims are generally given their "ordinary and customary meaning." See Phillips v. AWH Corp., 415 F.3d 1303, 1312-13 (Fed. Cir. 2005); see also Renishaw PLC v. Marposs Societa' per Azioni, 158 F.3d 1243, 1248 (Fed. Cir. 1998) ("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."). This ordinary and customary meaning "is the meaning that the terms would have to a person of ordinary skill in the art in question at the time of the invention." Phillips, 415 F.3d at 1313. A patentee is presumed to have intended the ordinary meaning of a claim term in the absence of an express intent to the contrary. York Products, Inc. v. Central Tractor Farm & Family Ctr., 99 F.3d 1568, 1572 (Fed. Cir. 1996).

Generally speaking, the words in a claim are to be interpreted "in light of the intrinsic evidence of record, including the written description, the drawings, and the prosecution history, if in evidence." Teleflex, Inc. v. Ficosa North Am. Corp., 299 F.3d 1313, 1324-25 (Fed. Cir. 2002) (citations omitted); see also Medrad, Inc. v. MRI Devices Corp., 401 F.3d 1313, 1319 (Fed. Cir. 2005) (court looks at "the ordinary meaning in the context of the written description and the prosecution history"). "Such intrinsic evidence is the most significant source of the legally operative meaning of disputed claim language." Vitronics Corp. v. Conceptronic, Inc., 90 F.3d 1576, 1582 (Fed. Cir. 1996).

With regard to the intrinsic evidence, the court's examination begins, first, with the claim language. See id. Specifically, "the context in which a claim is used in the asserted claim can be highly instructive." Phillips, 415 F.3d at 1314. As part of that context, the court may also consider the other patent claims, both asserted and unasserted. Id. For example, as claim terms are normally used consistently throughout a patent, the usage of a term in one claim may illuminate the meaning of the same term in other claims. Id. The court may also consider differences between claims to guide in understanding the meaning of particular claim terms.

Second, the claims "must [also] be read in view of the specification, of which they are a part." 415 F.3d at 1315. When the specification reveals a special definition given to a claim term by the patentee that differs from the meaning it would otherwise possess, the inventor's lexicography governs. Id. at 1316. Indeed, the specification is to be viewed as the "best source" for understanding a technical term, informed as needed by the prosecution history. Id. at 1315. As the Federal Circuit stated in Phillips, the specification is "the single best guide to the meaning of a disputed term," and "acts as a dictionary when it expressly defines terms used in the claims or when it defines terms by implication." Id. at 1321.

Limitations from the specification, such as from the preferred embodiment, cannot be read into the claims absent an express intention to do so. Teleflex, 299 F.3d at 1326 ("The claims must be read in view of the specification, but limitations from the specification are not to be read into the claims.") (citations omitted); CCS Fitness, Inc. v. Brunswick Corp., 288 F.3d 1359, 1366 (Fed. Cir. 2002) ("a patentee need not describe in the specification every conceivable and possible future embodiment of his invention."); Altiris v. Symantec Corp., 318 F.3d 1363, 1372 (Fed. Cir. 2003) ("resort to the rest of the specification to define a claim term is only appropriate in limited circumstances"). To protect against this, the court should not consult the intrinsic evidence until after reviewing the claims in light of the ordinary meaning of the words themselves. Texas Digital Systems, Inc. v. Telegenix, Inc., 308 F.3d 1193, 1204-05 (Fed. Cir. 2002) (to act otherwise "invites a violation of our precedent counseling against importing limitations into the claims") (citations omitted).

Finally, as part of the intrinsic evidence analysis, the court "should also consider the patent's prosecution history, if it is in evidence." Phillips, 415 F.3d at 1317. The court should take into account, however, that the prosecution history "often lacks the clarity of the specification" and thus is of limited use for claim construction purposes. Id. In most cases, claims can be resolved based on intrinsic evidence. See Vitronics, 90 F.3d at 1583. Only if an analysis of the intrinsic evidence fails to resolve any ambiguity in the claim language may the court then rely on extrinsic evidence, such as expert and inventor testimony, dictionaries, and learned treatises. See Vitronics, 90 F.3d at 1583 ("In those cases where the public record unambiguously describes the scope of the patented invention, reliance on any extrinsic evidence is improper"). However, the court generally views extrinsic evidence as less reliable than the patent and its prosecution history in determining how to read claim terms, and its consideration is within the court's sound discretion. See Phillips, 415 F.3d at 1318-19.

B. Legal Analysis

The parties' briefs present ten terms/phrases for construction, but at the hearing, the court noted that, for three of the terms, the parties agree that the plain and ordinary meanings of those terms should control. Thus, as stated at the hearing, no construction will be issued for the "first database . . . based on the user's viewing history at the interactive program guide client" term, or for "client device" or "user." The court will address the remaining seven disputed terms/phrases in turn.

1. "program guide" and "program listing"

The term "program guide" is found in claims 1, 6, 13, 15, and 17 of the `762 patent and claims 13 and 17 of the `709 patent; and "program listing" is found in claims 13, 14, 17, and 18 of the `709 patent. Although the parties propose different constructions for each term, their briefs present the arguments together, as they involve related issues.

For "program guide," Rovi proposes that it be construed as "a guide that displays program listings," while Netflix proposes that it be construed as "an application that produces display screens identifying the channels and times on which television programs will air."

For "program listing," Rovi proposes that it be given its plain and ordinary meaning, and in the alternative, proposes that it be construed as "information about available programs." Netflix proposes that "program listing" be construed as "information identifying the channel and time on which a television program will air or previously has aired."

For both terms, the dispute boils down to Netflix's inclusion of the "channel and time" language. Rovi accuses Netflix of making an "improper attempt to limit the claims to the broadcast television context," even though the claims were drafted broadly, with the intent of covering more than just broadcast television. As an example, Rovi points out that the specification mentions video-on-demand programs, which can be viewed at any time (i.e., no "time" listings). See, e.g., `762 patent, 12:45-46 ("Program guide server 25 may obtain, for example, video-on-demand programs").1

Netflix responds by arguing that the patents were drafted "in the context of late 1990s technology," and thus, the invention was limited to a "television-specific solution" involving "scheduled programming" with a "specific channel and time for broadcast." Netflix further argues that the figures in the patents show times and channels. See `762 patent, Fig. 6-7, 8a-8c, 16a-16c, 20b.

Netflix also argues that the specification draws a distinction between "program listings" and "video-on-demand videos," suggesting that the claims' use of "program listings" should exclude video-on-demand. See, e.g., `762 patent at 22:4-8 ("program guide server 25 may search its database or databases at other facilities for program guide data (e.g., program listings, additional program information, etc.), other information (e.g., software, Internet links, etc.), or videos (e.g., video-on-demand videos)").

In its reply, Rovi points out that the specification describes "video-on-demand video" as a type of "program guide data," not as something distinct and separate. See `762 patent at 12:61-65 ("User selectable criteria may also include what program guide server 25 searches for such as, for example, program listings, program information, web sites, video-on-demand videos, software, or any other suitable program guide data, other information, or videos").

Netflix argues that Rovi "has lost this argument before," citing to a Federal Circuit case which, according to Netflix, involves a "Rovi TV Guide patent that shares two inventors and has a specification strikingly similar to the `709 and `762 patents." See United Video Properties, Inc. v. Amazon.com, Inc., 561 Fed. App'x 914 (Fed. Cir. 2014). In United Video, the Federal Circuit affirmed a construction of "interactive program guide" as "an application that produces interactive display screens that include television program schedules and channel information." Id. at 916.

Rovi responds by arguing that the United Video case involved a different patent with a crucial distinction — the lack of any disclosure of video-on-demand programs — making it irrelevant to the current dispute. Rovi argues that the `762/'709 specification's disclosure of video-on-demand programs expands the claims beyond just broadcast television.

Overall, the court finds that Netflix's proposed constructions (specifically, the "channels and times" language) would improperly narrow the claims to the area of broadcast television. The claims themselves contain no such limitations, and the specification refers, on multiple occasions, to types of media other than broadcast television. See `762 patent at 4:49-57 ("The program guide data transmitted by main facility 12 to interactive television program guide equipment 17 may include television programming data (e.g., program identifiers, times, channels, titles, descriptions, series identifiers, etc.) and other data for services other than television program listings (e.g., help text, pay-per-view information, weather information, sports information, music channel information, associated Internet web links, associated software, etc.).") (emphasis added); and at 12:45-49 ("Program guide server 25 may obtain, for example, video-on-demand programs, web site links, games, chat group links, merchandise information, or any other suitable information or programming from data sources 14 located at main facility 12 or other facilities.").

The court also agrees with Rovi that the United Video case is of limited value here, as it involved a different patent with a materially different specification. The `762 and `709 patents not only disclose video-on-demand in the specification, they also specifically describe "video-on-demand video" as a variety of "program guide data." Thus, the court finds that the claims cover more than just broadcast television programming, making any reference to "channels and times" unduly limiting.

That said, Rovi's proposed construction for "program guide" is circular ("a guide that displays program listings"), and would not aid a jury. Thus, the court finds that the plain and ordinary meaning of "program guide" and "program listings" should govern.

2. The order of steps for "determining from a program listing database a set of programs not yet watched" and "applying the at least one of the associated program criteria to the set of programs not yet watched to generate at least one personal viewing recommendation"

These terms appear in claims 13 of the `709 patent (a method claim) and claim 17 of the `709 patent (a system claim). The parties do not dispute the actual meaning of these terms; instead, the dispute is limited to the order in which these steps must be performed. Rovi argues that no specific order is required, while Netflix argues that the "determining" step must be performed before the "applying" step in the method claim, and that the processing circuitry must be operative to "determine" before it "applies" in the system claim, or else the claim is indefinite.

As an initial matter, both parties seem to agree that, for a system claim, no order of steps is required, because as Netflix puts it, "system claims, by their nature, do not require that any steps actually be performed." Dkt. 134 at 11, n.1. However, Netflix does argue that the system must be "operative to perform" the "determining" and "applying" steps in that specific order, otherwise the claims are invalid.

The court finds a number of issues with Netflix's argument regarding the "system" claim. First, the court notes that Netflix points to no binding authority holding that a system claim can be required to perform steps in a certain order, or to be "operative to perform" steps in a certain order. Second, even the non-binding authority on which Netflix does rely — the ITC opinion — is misleadingly presented by Netflix. Critically, the Commission held that "[s]ystem claims do not have method steps, and thus cannot perform steps in a specific order," and "[t]herefore, we find that an order-of-steps analysis does not apply to system claims." ITC opinion at 22. While that passage appears to put the issue to rest, Netflix directs the court to a different portion of the opinion, where the Commission applied an infringement-type analysis (referred to as the "technical prong of the domestic industry requirement") to the accused products. ITC opinion at 24-25. Netflix provides the court with no reason to apply this infringement-type analysis to the claim construction process, let alone any authority requiring it to do so. Thus, the court finds that no order of steps is required for claim 17, as it is a system claim.

The court further finds the indefiniteness analysis of IPXL Holdings, LLC v. Amazon.com, Inc. to be wholly inapplicable to this case. See 430 F.3d 1377 (Fed. Cir. 2005). The IPXL court held that a single claim which "recit[es] both an apparatus and a method of using that apparatus renders a claim indefinite." Id. at 1384. In this case, claim 17 of the `709 patent recites only a system claim, and it is only Netflix who is attempting to characterize the claim as method-like, by attempting to require an order of steps.

In short, the court finds no basis on which to require an order of steps for system claim 17, nor any basis on which to find the claim is indefinite. Because the only dispute relates to the order of steps, the court finds no construction to be necessary on claim 17.

Turning to the method claim, the court first looks at the language of claim 13 itself, with the relevant two steps underlined:

Rovi argues that the "determining" step and the "applying" step could be performed in either order. In other words, one possibility is to first filter between "watched" and "unwatched" shows (i.e., the "determining" step), and then to generate recommendations (i.e., the "applying" step) using only the "unwatched" subset of shows; but an equally valid possibility would be to first generate a list of recommendations (i.e., the "applying" step), and then to filter that list to include only shows that the user has not watched (i.e., the "determining" step).

Rovi argues that both embodiments are mentioned in the specification, and that Netflix's construction (which requires that "determining" comes before "applying") reads out one of the embodiments. See `709 patent, 20:24-38. In fact, Rovi argues that it may even be preferable to perform the "applying" step first. Because a user has likely watched only a small subset of all available shows, filtering out all of the "watched" shows would hardly have any effect on the overall list of shows.

Rovi also argues that the claim is drafted using the phrase "comprising," which makes it open-ended, and allows for other, unrecited steps to be performed.

Netflix responds by arguing that, even if the specification includes another embodiment where the "applying" step happens first, the claims exclude that embodiment. Netflix then argues that it is irrelevant whether there is an "advantage" to performing the applying step first, because the only question is what the claim language requires.

Although steps of a method claim ordinarily need not be performed in any particular order, in some cases, the claim language requires that "as a matter of logic or grammar, they must be performed in the order written." See Altiris, 318 F.3d at 1369 (citing Interactive Gift Express, Inc. v. Compuserve Inc., 256 F.3d 1323, 1343 (Fed. Cir. 2001)). The Altiris court cited two examples where such an order was required.

In Loral Fairchild Corp. v. Sony Corp., one step recited the "forming" of a first structure, and another step recited the aligning of that structure with a second structure. 181 F.3d 1313, 1321 (Fed. Cir. 1999). Because the structures could not be "aligned" unless the first structure had already been formed, the court held that the "literal language of the claims" required the "forming" step to be performed before the "aligning" step. Id.

Similarly, in Mantech Environmental Corp. v. Hudson Environmental Services, Inc., the Federal Circuit considered a method claim reciting four steps, where each of the steps referenced something indicating that the above step had already been performed. 152 F.3d 1368, 1375-76 (Fed. Cir. 1998). For example, one step recited the addition of acetic acid to groundwater, while another step referred to "said acidified groundwater," implying that the acetic acid had already been added. Id. at 1376. The court thus held that "the sequential nature of the claim steps is apparent from the plain meaning of the claim language." Id.

The court finds that the language of claim 13 does require the "determining" step to be performed before the "applying" step. The "determining" step speaks of "determining from a program listing database a set of programs not yet watched", and the "applying" step then covers "applying the at least one of the associated program criteria to the set of programs not yet watched to generate at least one personal viewing recommendation." The only way of applying criteria to "the set of programs not yet watched" is to first generate the "set of programs not yet watched," which occurs in the "determining" step.

The court agrees with Rovi that the specification discloses an alternative way of performing the steps, where the "applying" step occurs before the "determining" step, but if Rovi intended to claim that embodiment, then the drafter should have included an additional claim that reversed the order. The claim language speaks for itself, and the court cannot expand the plain meaning of the language in order to include alternative embodiments that the patentee intended to claim, but did not actually claim.

The discussion from the ITC proceedings is helpful on this point. The administrative law judge initially sided with Rovi, based on the disclosures in the specification. However, the Commission reversed, concluding that "finding that the limitations can be performed in the reverse order is accomplished only by impermissibly rewriting the claim language." ITC opinion at 21 (citing Helmsderfer v. Bobrick Washroom Equipment, Inc., 527 F.3d 1379, 1383 (Fed. Cir. 2008)).

Rovi argues that the claim was only drafted using the language of "a set of programs not yet watched" and "the set of programs not yet watched" in order to avoid a rejection for lack of antecedent basis. Regardless of the patentee's reason for choosing those words, the words must be interpreted according to their plain meaning. As mentioned above, if the patentee intended to claim an embodiment where the "applying" step was performed before the "determining" step, the application could have included a separate claim reversing the order of the steps.

Finally, to the extent that Rovi argues that claim 13 is a "comprising" claim, and thus could involve additional, unrecited steps, the court agrees that the claimed method could involve an additional "applying" step to be performed before the "determining" step. However, this order addresses only those steps that are actually recited in the claim, and the recited "applying" step is directed to "the set of programs not yet watched" which is the result of the "determining" step. Whether there may be an additional, unrecited "applying" step that occurs at some other time is not an issue before the court.

For the reasons stated, the court finds that no order of steps is required for system claim 17, and that the "determining" step must be performed before the "applying" step in the context of method claim 13.

3. "user preference profile"

The term "user preference profile" is found in claims 13, 16, 17, and 20 of the `709 patent. Rovi proposes that it be construed as "criteria relating to user interest or disinterest in programs and/or program attributes," while Netflix proposes that it be construed as "set of user-defined preferences for program criteria."

The central dispute here is whether a user can express preferences for programs themselves, or just "program criteria."

Rovi notes that its construction allows users to express a preference in two ways — interest/disinterest in programs, and interest/disinterest in program attributes. Rovi points to a figure from the patent where the user can express interest/disinterest in "Friends" or "Frasier," and another figure where the user can express interest/disinterest in "game shows" or "horror." `709 patent, Fig. 13a. 13b. Rovi argues that Netflix's proposal reads out the embodiment disclosed in Figure 13a.

Netflix argues that the claims themselves discuss "program criteria," not "programs." Netflix also seeks to explain the figure cited by Rovi as allowing a user to indicate a "preference level for the television series criteria `Frasier' — an attribute shared by all episodes of that television series — not any particular program (e.g., the pilot episode of Frasier)." At the hearing, Netflix further explained this theory, asserting in its presentation that "Frasier is an attribute, not a program."

The court finds Netflix's argument to be unconvincing. When a user expresses a preference for a series (such as "Frasier"), he is indeed expressing a preference for a program itself, not for a certain "attribute." Netflix's proposal requires a far-too-narrow interpretation of the word "program." Accordingly, the court adopts Rovi's proposed construction, and construes "user preference profile" as "criteria relating to user interest or disinterest in programs and/or program attributes."

4. "combination category"

The term "combination category" is found in claims 11, 12, 17, and 19 of the `929 patent. Rovi proposes that it be construed as "a programming category composed of multiple simple categories," while Netflix proposes that it be construed as "a programming category composed of multiple simple categories, such as comedy-movies, comedy-series, sports-baseball, or sports-comedy."

Essentially, the parties agree on the first part of the construction, but disagree over whether the examples provided by Netflix should be provided.

Rovi argues that the examples could be improperly perceived as limiting, and expresses concern that Netflix seeks to make the claimed "combination categories" look different from its own combination categories, such as "critically-acclaimed emotionally independent dramas."

Netflix argues that its examples are entirely consistent with the claims and specification.

The court shares the concern that Netflix's examples could be perceived as limiting, and moreover, finds that they do not add anything of value to the construction. Thus, the court adopts Rovi's proposal, and construes "combination category" as "a programming category composed of multiple simple categories."

5. "receiver"

The term "receiver" is found in claim 11 of the `929 patent. Rovi proposes that it be given its plain and ordinary meaning, or in the alternative, that it be construed as "a device or component of a device that receives." Netflix proposes that it be construed as "circuitry in user equipment for receiving television signals."

The key disputes here are whether the receiver must be located within the user's equipment, and whether it is limited to receiving television signals.

Rovi argues that the purpose of the "receiver" in the claim is to receive program listings, which are then used by a processor to generate combination categories. Rovi further argues that the specification discusses generating those combination categories at a "back office processor," at the "transmission system," or "any other such suitable location." See `929 patent, 8:29-32. Because a "back office processor" would first need to receive the program listings before generating the combination categories, it follows that the program listings could be received at that back office processor — i.e., the listings need not be received at the "user equipment."

Regarding the "television" element of Netflix's proposal, Rovi concedes that the specification states that "user television equipment has a receiver, which is typically a set-top box." `929 patent, 5:57-58. However, while the television equipment may indeed have a receiver that receives television signals, Rovi argues that the claim language discusses the "receiver" in a broader context, not limited to receiving "television signals."

Netflix responds by arguing that the claims do not define "receiver," making it necessary to look at the specification. The word "receiver" appears only three times, and each time, it is in the context of user television equipment. `929 patent, 5:57-63. Netflix also emphasizes the "set-top box" language mentioned by Rovi.

Netflix argues that Rovi is improperly relying on extrinsic evidence that defines "receiver" far more broadly. Netflix also argues that this term was construed in United Video in a manner almost identical to Netflix's proposal.

In its reply, Rovi argues that the United Video construction is irrelevant because that patent "explicitly recited `receiver' as an element of user equipment," whereas the claim here "does not specify the location of the receiver."

Overall, the court finds that the "receiver" recited in claim 11 is not limited to receiving television signals, and its location is not limited to within user equipment. According to the language of the claim itself, the function of the receiver is to "receive[] a plurality of program listings," not to receive television signals. In that sense, the word "receiver" is used as a generic term for something that receives, and is not necessarily the same as the set-top box "receiver" mentioned in the specification. The court finds no support for limiting the receiver's function to receiving television signals, and in fact, finds that such a construction would contradict the language of the claim. At the very least, the "receiver" must be capable of receiving program listings, and Netflix's proposed construction would read out that claimed function.

The court also finds that the "receiver" of claim 11 need not be located within the user's equipment. As mentioned above, the "receiver" mentioned in the claim is not necessarily the same as the set-top box "receiver" mentioned in the specification, because they perform different functions, so there is no basis for importing the location of the set-top box into the claim.

That said, the court finds Rovi's proposal to be purely circular. Thus, the court instead finds that the term "receiver" shall be governed by its plain and ordinary meaning.

6. "search engine application"

The term "search engine application" is found in claims 1, 14, and 27 of the `962 patent. Rovi proposes that it be construed as "hardware, software, and/or firmware, which receives search requests and interfaces with one or more databases to respond to search requests," while Netflix proposes that it be construed as "search engine application for a video recorder."

The main dispute here is whether to limit the claims to video recorders, as opposed to equipment that allows mere playback of videos.

Rovi argues that Netflix's proposed limitation comes from the title of the patent, even though claims are not to be limited by a patent's title. Rovi further argues that the specification discloses the use of a search engine that is unrelated to a video recorder, including an on-demand feature.

Netflix argues that the "video recorder" limitation appears in more than just the title, it also appears in the abstract, the background of the invention, the summary of the invention, and the detailed description of the invention. Netflix further argues that all embodiments disclosed in the patent, including the on-demand embodiment, involve a video recorder.

Netflix separately argues that Rovi's proposal is unnecessary, because laypeople know what a search engine application is, and the proposed construction only serves to complicate the claims with jargon.

Overall, the court finds that including the word "recorder" in the construction would read out the "on-demand" embodiment of the invention. While Netflix argues that even on-demand programming requires the use of a recorder, the court finds Netflix's interpretation of the term "recorder" to be too broad.

To be sure, there is merit in the concept that, if a show is saved, it must have been in some sense "recorded." However, that does not mean that the specific device on which the show is saved is itself a "recorder." By way of example, a DVD player is capable of playing movies/shows that are saved (or recorded) on a disc, but the DVD player is not necessarily a recorder. Similarly, any saved on-demand shows must have been "recorded" in some sense of the word at some point in time, but it does not necessarily follow that a database of saved on-demand shows is a "recorder."

Thus, the court rejects Netflix's inclusion of the word "recorder" in its proposed construction.

However, the court also finds problems with Rovi's proposed construction. In particular, the court finds that the reference to "hardware, software, and/or firmware" to be overly vague and unnecessarily confusing to a jury.

In lieu of adopting either party's proposed construction, the court construes "search engine application" as "an application which receives search requests and responds to search requests."

7. "selectable categories of shows"

The term "selectable categories of shows" is found in claims 1, 14, and 27 of the `962 patent. Rovi proposes that it be given its plain and ordinary meaning, or in the alternative, that it be construed as "categories of shows that a user may select." Netflix proposes that it be construed as "groupings created from database entries in response to a user's search."

The dispute here is whether the categories are created only after the user runs a search.

Rovi argues that the patent states only that a user runs a search, and receives search results that may include shows or selectable categories of shows. It does not require that the categories be dynamically generated, and allows for the possibility that they are pre-existing.

Rovi argues that the Netflix proposal suffers from a number of problems. First, as mentioned above, Netflix's proposal requires that the categories be created "in response to" a user's search.

Second, Rovi argues that the word "groupings" does not appear anywhere in the patent, and thus introduces ambiguity.

Third, Rovi argues that Netflix's proposal requires that the show listings correspond to database entries that match the user's search. This argument is not particularly well-developed in the briefs, but it appears that Rovi is arguing that the patent does not require the categories to match the actual search input (i.e., the user could search for a title of a movie, and one of the categories might be movies that star the same actor as the searched-for movie, even though the user did not actually search for that actor).

In its response, Netflix first argues that the specification "consistently and exclusively describes search-result categories as created in response to the search." Netflix also argues that the patentee made statements during prosecution that limited the scope of the claims. Specifically, the patentee distinguished the claimed invention over the prior art by emphasizing that the prior art used "pre-defined" categories, whereas the claimed invention used categories created as a result of the search.

Overall, while Netflix is correct that the specification certainly does discuss the dynamic generation of categories, there is nothing in the specification that limits the use of categories to only those that are dynamically generated. The possibility of having pre-defined categories is left open. In its most general description of the invention, the patent states that "the results of a successful search query are organized using categories," and then says "for instance, results may be obtained by creating folders of all shows where a certain actor or actress appears." `962 patent at 9:30-33. Thus, the use of dynamically generated categories is presented as a "for instance," not as the entire invention.

The prosecution history estoppel argument advanced by Netflix appears to have merit, but a close look at the actual prosecution documents shows something different. The patentee was distinguishing the prior art (a reference referred to as "Kim"), because in Kim, the only way that a user could access categories was to scroll through a list of pre-defined categories — in other words, there would be a list of categories (comedy, drama, etc.), and the user could navigate through the various lists. See Dkt. 134-2 at 14 ("Kim describes a `search mode by category' mode that classifies and displays broadcast programs by category," "each category has submenus, and a user can select a submenu item to find broadcast programs corresponding to the submenu item").

Critically, the patentee said that, in Kim, "the categories and the submenus are not the results of a search in which the user enters one or more characters in an alpha-numeric input area, as specified by the applicants' amended independent claims." Also, "in Kim, when a user enters a search term using text inputs, the system produces matching broadcast programs, and not categories of programs." Id. at 15.

Thus, the patentee distinguished Kim not by the use of pre-defined categories in general, but by the fact that the only way to access those categories was to navigate through category-based menus, rather than by running a search. The key difference between the `962 and Kim is that, in the `962, the user can run a search and receive a category as a search result — regardless of whether that category was pre-defined or created dynamically. To use an example, if a user is looking for shows starring actor Damian Lewis, the system could already have a category for "shows starring Damian Lewis" ready to go and still fall within the bounds of the `962 patent or within Kim. What makes the `962 patent different from Kim is that, in the `962, a user can find that category by running a search for "Damian Lewis," whereas in Kim, the user must navigate through a prepared list of pre-defined categories, looking for a "shows starring Damian Lewis" category.

Because the court finds that Netflix's proposal improperly requires categories to be "created . . . in response to a user's search," it rejects Netflix's proposal, and instead finds that plain and ordinary meaning governs the term "selectable categories of shows."

C. Conclusion

In accordance with the foregoing, and for the reasons discussed above, the court construes the parties' disputed terms as follows:

1. "program guide" and "program listing" are given their plain and ordinary meaning.

2. For the steps of "determining from a program listing database a set of programs not yet watched" and "applying the at least one of the associated program criteria to the set of programs not yet watched to generate at least one personal viewing recommendation," the "determining" step must be performed before the "applying" step as part of the method claim, but no order of steps is required for the system claim.

3. "user preference profile" means "criteria relating to user interest or disinterest in programs and/or program attribuutes."

4. "combination category" means "a programming category composed of multiple simple categories."

5. "receiver" is given its plain and ordinary meaning.

6. "search engine application" meaans "an application which receives search requests and responds to search requests."

7. "selectable categories of shows" is given its plain and ordinary meaning.

IT IS SO ORDERED.