JOHN A. WOODCOCK, JR., Chief Judge.
SurfCast, Inc. (SurfCast) has filed a patent infringement lawsuit against Microsoft
"Infringement analysis involves a two-step process: the court first determines the meaning of disputed claim terms and then compares the accused device to the claims as construed." Wavetronix LLC v. EIS Elec. Integrated Sys., 573 F.3d 1343, 1354 (Fed.Cir.2009) (citing Markman v. Westview Instruments, Inc., 52 F.3d 967, 976 (Fed.Cir.1995), aff'd, 517 U.S. 370, 116 S.Ct. 1384, 134 L.Ed.2d 577 (1996)). The construction of language in patent claims must be performed by a judge before any jury trial on the issue of infringement. Markman, 517 U.S. at 391, 116 S.Ct. 1384.
The primary source of evidence for claim construction is intrinsic evidence; that is, "the patent itself, including the claims, the specification and, if in evidence, the prosecution history." Vitronics Corp. v. Conceptronic, Inc., 90 F.3d 1576, 1582 (Fed.Cir.1996). The words used in a claim are generally given their ordinary and customary meaning. Phillips v. AWH Corp., 415 F.3d 1303, 1313-14 (Fed.Cir.2005) (en banc). However, "the `ordinary meaning' of a claim term is its meaning to the ordinary artisan after reading the entire patent." Id. at 1321. The correct construction of a claim term is one "that stays true to the claim language and most naturally aligns with the patent's description of the invention." Id. at 1316 (quoting Renishaw PLC v. Marposs Societa' per Azioni, 158 F.3d 1243, 1250 (Fed.Cir.1998)).
The patent specification supplies the primary basis for understanding the meaning of claim language. "Usually, it is dispositive; it is the single best guide to the meaning of a disputed term." Id. at 1315. In other words, the claim terms are to be understood in the context of the specification. Netword, LLC v. Centraal Corp., 242 F.3d 1347, 1352 (Fed.Cir.2001). A claim construction "that excludes a preferred embodiment from the scope of the claim is rarely, if ever, correct." Accent Packaging, Inc. v. Leggett & Platt, Inc., 707 F.3d 1318, 1326 (Fed.Cir.2013). Nonetheless, details from the specification may not be imported into the claim through the guise of claim interpretation. SuperGuide Corp. v. DirecTV Enters., Inc., 358 F.3d 870, 875 (Fed.Cir.2004). "[T]he line between construing terms and importing limitations can be discerned with reasonable certainty and predictability if the court's focus remains on understanding how a person of ordinary skill in the art would understand the claim terms." Phillips, 415 F.3d at 1323.
The ordinary meaning of a word to one skilled in the art at the time of the patent issue is not evaluated in isolation, however. "[T]he specification may
The prosecution history can also play a role in claim interpretation. Id. at 1317. However, because the prosecution history "represents an ongoing negotiation between the [United States Patent and Trademark Office (PTO)] and the applicant," it can lack the clarity of the specification and be less useful in claim construction. Id. A court may also consider extrinsic evidence, such as dictionaries, treatises, and expert testimony to the extent it is consistent with the intrinsic evidence. Id. at 1319. However, extrinsic evidence may not contradict the language of the claims or the teachings of the specification. Helmsderfer v. Bobrick Washroom Equip., Inc., 527 F.3d 1379, 1382 (Fed.Cir.2008).
Two further rules of construction aid the Court. First, "the presence of a dependent claim that adds a particular limitation gives rise to a presumption that the limitation in question is not present in the independent claim." Phillips, 415 F.3d at 1315 (citing Liebel-Flarsheim Co. v. Medrad, Inc., 358 F.3d 898, 910 (Fed. Cir.2004)). This is sometimes called the doctrine of "claim differentiation." E.g., Seachange Int'l, Inc. v. C-COR, Inc., 413 F.3d 1361, 1369 (Fed.Cir.2005). The presumption may be rebutted by "a contrary construction dictated by the written description or prosecution history." Id. Second, if after applying these rules of construction a claim term remains susceptible to more than one reasonable interpretation, the term is to be construed in a way that preserves the validity of the patent. Phillips, 415 F.3d at 1327. This will usually be a narrower construction. Id.
In 1998, Ovid Santoro and Klaus Lagermann
'403 Patent, at [57]. The `403 Patent has three independent claims — claims 1, 22, and 46 — and forty-nine dependent claims.
SurfCast commenced this patent infringement suit against Microsoft on October 30, 2012. Compl. (ECF No. 1). The parties submitted a joint claim construction statement on May 7, 2013. Joint Claim Construction Statement (ECF No. 57). With the statement came a claim construction chart showing the disputed terms and the parties' proposed constructions.
The Court originally scheduled a claim construction hearing on June 28, 2013 (ECF No. 75), but rescheduled for August 30, 2013 to accommodate the schedules of the parties' counsel. Notice Cancelling Claim Construction Hearing (ECF No. 76), Minute Entry (ECF No. 113). At the claim construction hearing, counsel for both sides indicated there might be some terms on which they might compromise and reach an agreement on construction. Consequently, the Court ordered the parties to attempt to negotiate on these terms and report back by September 6, 2013. Minute Entry (ECF No. 113). On September 6, 2013, both parties reported back that no compromise had been reached, but both parties also submitted revised claim constructions. Def. Microsoft Corporation's Notice of Proposed Compromise Claim Constructions (ECF No. 115); Def. Microsoft Corporation's Notice of Proposed Compromise Claim Constructions Attach. 1 Microsoft's Proposed Compromise Constructions (Def.'s Revised Constructions); Pl. SurfCast's Notice of Revised Claim Construction (ECF No. 116) (Pl.'s Revised Constructions).
On November 19, 2013, while claim construction was under advisement before the Court, the Patent Trial and Appeal Board of the PTO granted inter partes review of the `403 Patent under 37 C.F.R. § 42.108. Microsoft Corp. v. SurfCast, Inc., No. IPR2013-00292 (P.T.A.B. Nov. 19, 2013) (PTO Op.). Inter partes review is an administrative proceeding in which the PTO reviews an issued patent to determine if some claims should be canceled because they are unpatentable in light of the prior art. 35 U.S.C § 311(b); 37 C.F.R. § 42.108(a), (b). For the purpose of granting inter partes review, the PTO construed some of the claim terms in the `403 Patent currently before the Court. PTO Op. at 8-23. The PTO's construction is persuasive but not binding on this Court. See Def. Microsoft Corporation's Notice of Institution of Inter Partes Review of the Patent-In-Suit at 2 (ECF No. 146) (Nov. 21, 2013).
On December 20, 2013, SurfCast moved to stay proceedings pending the inter partes review. Pl.'s Mot. to Stay Pending Inter Partes Review (ECF No. 150) (Pl.'s Mot. to Stay). Microsoft opposed this motion on January 13, 2014. Def. Microsoft Corporation's Resp. to Pl.'s Mot. to Stay Pending Inter Partes Review (ECF No. 153) (Def.'s Opp'n to Stay). SurfCast replied to Microsoft's opposition on January 27, 2014. Pl.'s Reply in Support of Its Mot. to Stay Pending Inter Partes Review (ECF No. 154) (Pl.'s Reply to Stay).
SurfCast argues that the Court should stay proceedings in this matter because Microsoft will not suffer any undue prejudice from a stay, granting a stay during the inter partes review (IPR) will simplify the issues for trial, a stay will conserve judicial resources, and proceeding without a stay may result in inconsistent decisions. Pl.'s Mot. to Stay. Microsoft replies that a stay will not simplify any issues because there is an appellate process that will follow IPR, and, in any event, Microsoft will continue to pursue its inequitable conduct and unclean hands
The Court accepts SurfCast's formulation of the factors it should consider in deciding whether to grant a stay:
Pl.'s Mot. to Stay at 5 (citing Pi-Net Int'l, Inc. v. The Hertz Corp., No. CV 12-10012 PSG, 2013 WL 7158011, at *1, 2013 U.S. Dist. LEXIS 81570, at *3 (C.D. Cal. Jun 5, 2013)). On the first factor, the Court accepts Microsoft's assertions that it "expects that important testimony will be provided by long-time employees who were intimately involved in Microsoft's prior software development. But the technology industry often evolves at a staggering pace, and valued technology employees frequently move on." Def.'s Opp'n to Stay at 6-7. Microsoft is also concerned that, absent the Court's claim construction decision, it might not be able to modify its allegedly-infringing products to cut off potential ongoing damages. Id. at 7-8.
The Court also does not conclude that a stay will simplify the issues for trial. The outcome of the IPR process is too uncertain to predict with confidence that it will result in a different patent underlying the suit; indeed, waiting for the entire process to proceed through the PTO channels (including the likely appeals to the Federal Circuit) will likely add confusion and complexity. Furthermore, SurfCast has asked the Court to impose a stay only "pending the Board's final written determination of the IPR," Pl.'s Mot. to Stay at 10; it would have the Court lift the stay after the Board rules, but before the Federal Circuit resolves any appeals. Given that the Court disagrees with the PTO's construction of many of the disputed claim terms, see infra, it would make little sense to proceed to trial on whatever patent emerges from the IPR process. If the parties proceed briskly to dispositive motions and trial, the Federal Circuit will have the advantage of the outcome in this Court when considering the result of the IPR process.
Finally, on the third factor, the case has been underway for over a year and fact discovery is complete. A trial date has not yet been set, but a scheduling order will issue shortly after this claim construction opinion.
The Court concludes that the three factors weigh in favor of denying a stay at this time.
The words "tile" or "tiles" appear in every independent claim and many of the dependent claims. E.g., '403 Patent at claims 1, 22, 46. They are also elements of several other disputed terms. "Tiles" are fundamental to the entire patent, and so the meaning of this word merits careful and exacting analysis.
SurfCast proposes that "tile(s)" be construed as a "graphical representation of an associated information source capable of displaying refreshed content, the graphical representation being persistent and selectable to provide access to underlying information of the associated information source." Joint Claim Construction Chart
SurfCast argues that the word "tile" should mean "graphical representation of an associated information source capable of displaying refreshed content, the graphical representation being persistent and selectable to provide access to underlying information of the associated information source." Pl.'s Br. at 10. SurfCast starts with the premise, based on its expert's affidavit, that the term "tile" had no plain and ordinary meaning in the art at the time of the `403 Patent. Id. at 11. It also notes that the specification indicates that it is defining the word: "In the present invention, a third graphical representation of programs and files [distinct from windows and icons], herein called a tile, is introduced." '403 Patent at 8:29-30.
SurfCast observes that the specification states that "a tile presents content from any information source," while the claim language specifies that "each tile ... is associated with an information source." Pl.'s Br. at 11 (quoting `403 Patent at 7:64-65; `403 Patent at claim 1, 24:20-21). Furthermore, each independent claim associates a "refresh rate" or a "retrieval rate" — also disputed terms — with each "tile." Id. at 11 (quoting `403 Patent at claim 1, 24:23-28). SurfCast concludes that a person of ordinary skill in the art, reading the patent, would understand that a tile is "a graphical representation of an associated information source capable of display refreshed content." Id.
SurfCast also finds support for the phrase "the graphical representation being persistent" in the specification. Id. at 12. It points to the specification's description of one embodiment of the invention in which "[t]iles ... are created, saved, and restored via the metabase."
Next, SurfCast offers more language from the specification for the proposition that tiles are "`selectable' so as to `instantly provide[] the user with access to the underlying information' when the tile is selected." Id. (quoting `403 Patent at 9:25-27). SurfCast asserts that a person of ordinary skill in the art would appreciate that the "selectable" feature of the "tile" distinguishes it from a conventional window." Id. at 13.
Finally, SurfCast disputes Microsoft's proposed construction, because it fails to
Microsoft's preferred construction of "tile" was originally "[a]n area of a display which presents content from an information source."
Microsoft disputes SurfCast's construction, first on the grounds that "[t]he patent specification does not provide any special definition for `tile.'" Id. In Microsoft's view, the `403 Patent's use of "tile" throughout is consistent with the usage of the term in prior art, and particularly in United States Patent No. 5,321,750. Microsoft contends that SurfCast's construction improperly reads limitations from the specification into the claim language.
For instance, Microsoft asserts that nothing in the patent requires that "tiles" be "persistent." Id. at 8. The claims do not use the word "persistence," Microsoft points out, and several dependent claims speak of downloading tiles from a second computer onto the user's computer. Id. (citing `403 Patent at claims 15, 16). It also points to an embodiment in which the tiles are downloaded to the user's system when the user logs onto a separate server computer, `403 Patent at 22:1-15, which suggests to Microsoft that the user's computer does not have the tiles when the user's computer is started. This, Microsoft maintains, is inconsistent with the notion of "persistence" put forth by SurfCast.
Microsoft also argues that the only embodiment that uses the word "persistent" does so in the context of describing "items" stored in the "metabase." Def.'s Br. at 8 (quoting `403 Patent at 14:27-28). Even this storage is optional; "[a]ctual tiles that a user visualizes can be spawned from the metabase." Id. (quoting `403 Patent at 14:42-43) (emphasis added by Microsoft). Microsoft reads the phrase "can be" to indicate that even in the metabase embodiment, persistence of the tiles is optional. Microsoft further points out that the "metabase" is not mentioned in any claim.
As to "selectability," Microsoft relies primarily on the doctrine of claim differentiation
Microsoft also disputes that a "tile" must be a "graphical representation of an associated information source," as advanced by SurfCast. Id. at 9-10. Even the specification, Microsoft points out, only states that the content of a tile "is typically a miniaturized representation of a graphic or still-frame from a datastream." Id. at 9 (quoting `403 Patent at 9:3-5). Microsoft concludes that some tiles do not represent data in this way. Id. at 10.
As to "capable of displaying refreshed content," Microsoft objects because only independent claim 1 and its descendants — not independent claims 22 and 46 — have a reference to "refresh."
Finally, Microsoft vigorously disputes SurfCast's claim that Microsoft's preferred construction ignores the specification's distinctions between a "tile" and a "window." Id. at 10-11. Microsoft claims that prior art "windows" were both "persistent" and "selectable." Id. at 10 (citing id. at Ex. 1, Decl. of David R. Karger on Claim Construction for U.S. Patent No. 6,724,403 at ¶ 23 (ECF No. 64-1) (Karger Declaration)). Furthermore, Microsoft claims that SurfCast never argued to the patent examiner that its "tiles" were distinct from "windows" because of persistence or selectability; rather, SurfCast made other arguments distinguishing "windows" from "tiles." Id. at 11 (citing id. at Ex. 4, Amendment and Resp. Under 37 C.F.R. § 1.111 at 12 (ECF Nos. 60-4, 60-5)
As a preliminary matter, the Court must determine whether the patentee intended to "act as its own lexicographer." Thorner v. Sony Computer Entm't Am. LLC, 669 F.3d 1362, 1365 (Fed.Cir. 2012). The patentee seeking to redefine the plain and ordinary meaning of a term must "`clearly express an intent' to redefine the term." Id. (quoting Helmsderfer, 527 F.3d at 1381). If the patentee in this case intended to redefine "tile," the Court must give much heavier weight to the specification than to any extrinsic evidence of the plain and ordinary meaning of the word. On the other hand, if it appears that the patentee was using "tile" without any special definition local to the patent, the Court can look to extrinsic evidence of the plain and ordinary meaning at the time.
In Thorner, the Federal Circuit held that a district court wrongly redefined the word "attached" to mean "affixed to an exterior surface." 669 F.3d at 1367-68. The patent had used the word "attached" without explicitly redefining it, but all the
Here, by contrast, the Court concludes that the patentee intended to redefine the term. This intent is naked on the face of the patent. The specification devotes an entire section to "Tile Objects," which begins with this language: "In the ensuing discussion, tile objects are introduced and described and contrasted with existing elements of graphical user interfaces." `403 Patent at 7:62-64. The patentee clearly states that he is not using "tile" in any sense other than that which he is preparing to "introduce" and "describe." After criticizing the current art, consisting (in his view) of icons and windows, the patentee states: "In the present invention, a third graphical representation of programs and files, herein called a tile, is introduced." Id. at 8:29-30. Again, the patentee is "introduc[ing]" something called a "tile," which is different than the prior art. Id.
No clearer statement of intent to redefine is needed. This is not a case in which the Court must imply intent by analyzing the usage of the term in the specification, as the district court wrongly did in Thorner. The `403 Patent, expressly intends to redefine the word "tile." That the intent to redefine occurs in the specification is no barrier. See 3M Innovative Props. Co. v. Avery Dennison Corp., 350 F.3d 1365, 1369, 1371 (Fed.Cir.2003) (permitting the patent specification to redefine a term).
Consequently, the Court rejects Microsoft's contention that the patent uses "tile" in a manner to be understood by reference to the prior art. Its meaning must be found in the patent itself. The Court now turns to the parties' proposed constructions.
First, the Court does not accept Microsoft's proposed construction: "An area of a display which presents content from an information source." This is a much broader definition of "tile" than appears in the patent. The whole point of a "tile" in the invention is to combine features of "windows" and "icons" in a way that balances the advantages of each while discarding some features not desirable in the new "tile" interface paradigm.
The patent defines "tile" largely by reference to "window" and "icon"; that is, it explains what a "tile" is by adopting some elements of "window" and "icon" while discarding others, and also adding what it claims are some new features not present in either. To this end, it provides its own specific definitions of "window" and "icon," along with the requisite intent to define these terms locally. `403 Patent at 7:66-8:14. The patent makes quite clear what the patentee considers to be the defining characteristics of both "windows" and "tiles," and which of those characteristics are adopted by "tiles."
For instance, the patent states that a "tile" contains a "view[] of a single information source," Id. at 8:32, and that an icon does not. See id. at 8:2-14 (describing icons as the method by which one accesses information from a data source, distinct from the "window" in which one
Microsoft's proposed construction disregards this specificity in the patentee's description of a tile. Instead, Microsoft's definition would apply equally to both a "window" and an "icon" as the patentee presents both those terms. "An area of a display which presents content from an information source" describes virtually every visual element of the interface between a computer and a human being, a construction too broad to be correct.
SurfCast's proposed construction of "tile" begins with "graphical representation of an associated information source," Pl.'s Br. at 10, and adds specificity. Each independent claim speaks of "partitioning" or "arranging" a "display." Id. claim 1, 24:19-20; id. at claim 22, 25:28-29; id. at claim 46, 26:58. This language addresses the human interface purpose of the invention, making "graphical representation" an appropriate abstraction. The specification supports this interpretation when it refers to a tile as a "graphical representation of programs and files." Id. at 8:29-30. As for "an associated information source," the specification states that "a tile presents content from any information source," `403 Patent at 7:64-65, and the language of the first independent claim instructs that "each tile ... is associated with an information source." Id. claim 1, 24:19-21. Likewise, independent claim 22 speaks of "associat[ing] a first information source... to a first tile and a second information source ... to a second tile," id. at claim 22, 25:30-34, while independent claim 46 uses similar language. Id. claim 46, 26:59-63.
Microsoft's objections to this clause of SurfCast's proposed construction do not address the claim language itself, nor do they speak to the plain language of the specification cited above. Def.'s Br. at 9-10. The Court concludes that the construction of "tile" should begin with "graphical representation of an associated information source."
SurfCast next proposes to add the phrase "the graphical representation being persistent." Microsoft objects in substance only to the word "persistent." As a starting point, the Court notes that both parties apparently agree on what "persistent" means: "tiles are not lost when a computer is restarted." Pl.'s Br. at 12; Def.'s Br. at 8. The PTO concluded, after considering material from columns 4 and 15 of the `403 Patent, that construing a "tile" to be "persistent" would improperly import a limitation from the specification into the claims. PTO Op. at 11. The Court respectfully disagrees.
In support of its construction, SurfCast offers the phrase "preserved from session to session," found in reference to one preferred embodiment of the invention. `403 Patent at 15:54-64. This embodiment makes use of the "metabase" data store concept to cause tiles to persist across user sessions. Id. SurfCast also points to another embodiment that "replaces the functionality of the computer `desktop'" and "understands the interests of the user and acts as a repository for password and identifiers." Id. at 11:32-33, 46-47. But these are two phrases selected from two embodiments; the text does not mandate that
From the Court's perspective, the Summary of the Invention presents a more compelling case for persistence than Microsoft or the PTO credits. SurfCast points to this phrase to show that tiles are persistent: "The present invention comprises a grid of tiles that resides on the user's computer desktop." Id. at 4:37-38. The most relevant definition of "reside" is "to abide, lie, or be present habitually," THE RANDOM HOUSE DICTIONARY OF THE ENGLISH LANGUAGE 1638 (Stuart Berg Flexner & Leonore Crary Hauck eds., 2d ed. unabridged 1987), and this does strongly suggest persistence. At the same time, the sentence is ambiguous as to what, exactly, resides. Clearly something "resides" on the desktop, but the sentence could either mean that the grid resides there or that the tiles do. Nonetheless, a "grid" is a grid "of tiles"; if the grid abides, so must the tiles within. `403 Patent at 4:37-38.
The language of the independent claims provides no clear indication that tiles are persistent. Granted, the independent claims speak repeatedly of "assigning" refresh rates or information sources to tiles. But persistence is more than merely the ability to be assigned a value and hold it for some measurable period of time; according to SurfCast, "persistence" means that those assignments are necessarily stored and restored when the user terminates one session and starts another. See Pl.'s Br. at 12 ("the concept of permanence... mean[s] that tiles are not lost when a computer is restarted"). It is true that some of the dependent claims speak of "storing" the array of tiles on a "device" — which might implicate persistence. But the fact that some dependent claims make use of a particular method to implement persistence does not prove that the independent claims from which they descend necessarily have persistence. Just the opposite presumption applies under the doctrine of claim differentiation, and this presumption prevents the dependent claims from definitively imputing persistence to tiles.
In Thorner, the Federal Circuit faulted a district court for implying the limitation that "attached" meant "externally attached" from the fact that the embodiments only used "attached" to speak of external attachments. Thorner, 669 F.3d at 1368. But here the patent gives the reader more than mere consistent usage; at least arguably it introduces the idea of persistence as a feature of "tile" in the Summary of the Invention. The doctrine of claim differentiation creates a presumption that the persistence feature suggested in the dependent claims is not present in the independent claims, but that presumption can be rebutted by "a contrary construction dictated by the written description or prosecution history." Seachange, 413 F.3d at 1369. In this case, neither of these legal tools resolves the dispute; there is strong tension between the Summary and the claims. This leaves the Court at the fine "line between construing terms and importing limitations," which the Federal Circuit instructs "can be discerned with reasonable certainty and predictability if the court's focus remains on understanding how a person of ordinary skill in the art would understand the claim terms." Phillips, 415 F.3d at 1323.
Turning to a "big picture" conceptualization of the patent is not appropriate for every claim term. In some cases this would run afoul of the Federal Circuit's commandment not to import limitations of the specification into the claims. See, e.g., Section IV.L.2, infra (construing "grid"). If the rules of claim construction lead to a clear legal result, no further conceptualization is needed. Here, though, where the minutiae of the words do not lead to an impregnable legal answer in either direction, the big picture reveals what one skilled in the art would understand after reading the patent.
The Court will include "the graphical representation being persistent" in its construction of "tile(s)."
The specification, in the section defining a "tile," states that "[t]iles are selectable" so as to "instantly provide[] the user with access to the underlying information." `403 Patent at 9:25-27. This text occurs before any description of specific embodiments, and so gives rise to a strong inference that it applies to all tiles.
Against this inference, Microsoft raises the doctrine of claim differentiation, pointing out that some of the dependent claims mention selectability. Def.'s Br. at 9 (citing `403 Patent at claims 5, 32). As a preliminary matter, claims 5 and 32 do not actually describe selectability. Instead, they speak of "attributing a selected state or an unselected state to said first tile and said second tile" and "selectively assign[ing] a selected or unselected state to
However, even assuming that the language in some dependent claims does speak to the concept of "selectability," this would only give rise to a presumption that the "tiles" of the independent claims are not selectable. In this case the plain language of the specification, defining "tile," rebuts that presumption because it is a "contrary construction dictated by the written description." Seachange, 413 F.3d at 1369; see also Marine Polymer Techs., Inc. v. HemCon, Inc., 672 F.3d 1350, 1358-59 (Fed.Cir.2012) (rebutting the presumption of claim differentiation using descriptive material from the specification). In other words, where the patentee has defined a "tile" to be "selectable," tiles are still selectable even if an independent claim does not mention selectability and a dependent claim does.
Microsoft's stated objection to this phrase stems from its position that "refresh rate" and "retrieval rate" mean different things in the context of the patent. Def.'s Br. at 10. It points out, correctly, that the word "refresh rate" only appears in independent claim 1, while independent claims 22 and 46 use "retrieval rate" in its place. Id. This, in Microsoft's view, means that only the tiles of claim 1 are capable of displaying refreshed content, and therefore a "tile" generally is not.
The Court later addresses the respective meanings of "refresh rate" and "retrieval rate." However, even if "refresh rate" and "retrieval rate" are given Microsoft's construction, a "tile" is still "capable of displaying refreshed content." First, Microsoft concedes that the tiles of independent claim 1 are capable of displaying "refreshed" content, Def.'s Br. at 10; the dispute is only whether the tiles of independent claims 22 and 46 have the same characteristic. Microsoft's preferred construction of "retrieval rate," the phrase in those claims, is "a recurring time interval at which information is retrieved." Id.
Independent claim 22 speaks of a collection of tiles and a collection of information sources, a set of instructions for associating the tiles with the information sources, a set of instructions for retrieving information from the information sources "in accordance with a ... retrieval rate," and a set of instructions to "present information" to the tiles "in accordance with [the] ... retrieval rate." '403 Patent at claim 22, 25:25-42. The sum of this process — adopting for the moment Microsoft's construction of "retrieval rate" — is to repeatedly obtain information from some source at a "recurring time interval" and present it to the tile. Claim 22 also states that the display is "arrange[d] [into] an array of tiles." Id. claim 22, 25:28-29. If the display is divided up into tiles, a tile is a "graphical representation of an associated information source," and the tiles are periodically given new information from the information source, it must be the case that a tile is "capable of displaying refreshed content." This is true even if one accepts Microsoft's definition of "retrieval rate."
The relevant portions of independent claim 46 are nearly identical. That claim
Because all three independent claims logically require that tiles are capable of displaying refreshed content, the Court includes "capable of displaying refreshed content" in its construction of "tile."
It is impermissible to import details of the specification into the claims in the guise of claim construction. Super-Guide, 358 F.3d at 875. However, the Court takes as a corollary to this principle that it is equally impermissible to ignore the specification altogether to broaden a claim into invalidity. After careful consideration, the Court concludes that within the '403 Patent, a "tile" is "a graphical representation of an associated information source capable of displaying refreshed content, the graphical representation being persistent and selectable to provide access to underlying information of the associated information source."
SurfCast proposes that "array of tiles" be construed as "multiple tiles displayed in an orderly fashion." Joint Claim Construction Chart 1. Microsoft proposes that the phrase mean "[a] collection of tiles. The tiles may, but need not, be arranged in a grid." Id. The PTO construed "array of tiles" to mean "an ordered set of two or more tiles." PTO Op. at 11.
SurfCast argues that an "array of tiles" is "multiple tiles displayed in an orderly fashion." Pl.'s Br. at 14. Its argument rests mainly on portions of the specification that use the phrase "array of tiles." For instance, the Summary of the Invention states that the invention's method includes steps of "partitioning a visual display of a computer into an array of tiles in a non-overlapping configuration." Id. at 14 (citing '403 Patent at 4:57-59). The section defining "tile objects" states that a representative tile can contain a secondary grid of tiles; the representative tile "displays a further array of tiles that may be displayed in full by expanding [the representative tile] to occupy the full area of the display." Id. (citing '403 Patent at 9:8-11). It also cites several embodiments with similar language. Id. (citing '403 Patent at 12:6-8 ("FIG. 8 shows one embodiment in which all tiles are the same size and are presented in an array comprising M rows and N columns")); id. (citing '403 Patent at 13:59-60 ("When a tile is partitioned into a further array of tiles, the grid configuration program can also be used"). In SurfCast's view, this language establishes that when the patent uses the phrase "array of tiles" it means "multiple tiles displayed in an orderly fashion." Id.
Microsoft's preferred construction of "array of tiles" is: "A collection of tiles. The tiles may, but need not, be arranged in a grid." Def.'s Br. at 11. Microsoft begins with the proposition that "[a]n `array' is a common term in computer science with a well-understood meaning: a collection of items." Id. (citing DICTIONARY OF COMPUTER AND INTERNET TERMS 24 (Douglas Downing, Michael Covington, & Melody Mauldin Covington eds., 6th ed. 1998) (DICTIONARY OF COMPUTER TERMS) (defining "array" as "a collection of data items that
Microsoft disputes that an "array of tiles" should be inherently ordered, as SurfCast proposes, because "displayed in an orderly fashion" is "thoroughly subjective." Id. at 12. As a demonstration, Microsoft presents four small boxes inside a larger box, each smaller box placed into one of four equal quadrants of the larger box. Id. Although each small box is inside its own quadrant, its position relative to the top left of the quadrant varies slightly from quadrant to quadrant. Id. Microsoft proposes that "a juror who believes this arrangement is `orderly' is no less wrong or right than one who believes it is not." Id. The Court understands Microsoft to mean that it would be reasonable for a juror to find it orderly because the small boxes are all within one of four equal quadrants, but that it would also be reasonable for a juror to find it disorderly because each small box is at a slightly different offset within its quadrant. Microsoft cites Datamize, LLC v. Plumtree Software, Inc., 417 F.3d 1342, 1350-51 (Fed.Cir.2005) for the proposition that subjective claim requirements have in the past been held to render the claim invalid as indefinite.
Finally, Microsoft attacks the specification passages presented by SurfCast, claiming that they do not show that an "array of tiles" must necessarily display in an "orderly fashion." Def.'s Br. at 12. It claims that the passages show instead that an array of tiles may be ordered, but that this is not a required characteristic of an array of tiles. Id. at 12-13. Consequently, Microsoft concludes that SurfCast's construction improperly reads limitations from the specification into the claims.
First, the Court is not convinced by Microsoft's attempt to conflate the computer science definition of "array" with the patent's use of "array." It is true that "array" is a venerable term in computer science. E.g., Dennis M. Ritchie, The Development of the C Language, in 2 HISTORY OF PROGRAMMING LANGUAGES 688 (Thomas J. Bergin & Richard G. Gibson eds., 1996) (tracing the development of the early programming languages BCPL, B, and C, and noting the use of "array" to denote collections of data in memory). It is equally true that the word "array," as used by computer programmers in the year 2000 while writing computer programs, referred to a collection of data items. DICTIONARY OF COMPUTER TERMS, supra, at 24. But "array" as used in this sense refers to data stored in computer memory, not to a group of elements in a human interface. Microsoft's own documentation of the "CArray" class in Microsoft Foundation Classes 6.0, Microsoft's then-current software development kit in 2000, demonstrates that an "array" is a collection of instantiated objects residing in computer memory. CArray, Microsoft Developer Network, http://msdn.microsoft.com/en-us/library/aa235465(v=vs.60).aspx (last visited Mar. 12, 2014) (CArray Documentation) (describing a programming language construct used to store and refer to data in memory).
But when the '403 Patent speaks of an "array of tiles," it consistently refers to
SurfCast's proposed construction is more convincing than Microsoft's because it reflects, not only the consistent usage of "array of tiles" in the specification, but the overall purpose of the invention. The Court is mindful that in Thorner, the Federal Circuit reversed a district court for reading "attached" as "affixed to an exterior surface" merely because of consistent usage in the specification. 669 F.3d at 1368. Under Thorner, were consistent usage the only consideration here, it might be improper to read "array of tiles" as implying an orderly presentation.
First, "orderly" is not, as Microsoft claims, necessarily subjective. The most apt definition of "orderly" in this context is "arranged or disposed in a neat, tidy manner or in a regular sequence." Flexner & Hauck, supra, at 1363. The Court grants that "neat" and "tidy" are subjective. However, whether things are "arranged in a ... regular sequence" is mathematically objective. Although a "regular sequence" might be quite complex and governed by conditional rules, it is ultimately objective and predictable. The Court understands SurfCast to use "orderly" in this second, objective manner because it is the only usage that makes sense in the context of the '403 Patent.
Next, the overwhelming majority of references to "array of tiles" in the specification and claims are consistent with orderliness in the objective sense. The Summary of the Invention makes three broad statements about visual collections of tiles: (1) "[t]he present invention comprises a grid of tiles," '403 Patent at 4:37-39; (2) "the present invention comprises a method executed by a computer ... comprising the steps of: partitioning a visual display ... into an array of tiles in a non-overlapping configuration," id. at 4:55-61; and (3) "the present invention additionally includes ... a ... set of instructions to partition a display into an array of tiles." Id. at 4:65-5:3. If the invention comprises a grid of tiles, and also a method and instructions for rendering an array of tiles, it stands to reason that the array of tiles will be rendered with reference to a grid. This is not to say that "array" and "grid" mean the same things; it only means that the Summary imputes the orderly properties of a grid to the usage of "array."
The preceding discussion goes to consistent usage, and the Thorner Court made clear that consistent usage alone is not enough to impute a limitation to the claims. 669 F.3d at 1368. But in this case, the purpose of the invention and the description of its purported novelty confirm that an "array of tiles" is orderly.
The human interface concept described in the '403 Patent is meant to allow a computer user to keep track of multiple information sources in a more rational way than the prior art. It criticizes the icons and windows paradigm:
'403 Patent at 2:54-63. Likewise, it states that "[t]he user must contend with a wide range of icons and program windows that may occupy space on a user's display screen ... An effort to standardize the ways in which different types of information are presented to the user would be advantageous." Id. at 3:13-15, 16-19. Criticizing the then-present method of bookmarking internet locations and other information, it claims that "[t]he missing capability is a visual categorization in which an area of the display unit itself becomes the bookmark and the arrangement on the display becomes the categorization, independent of the type of content." Id. at 4:3-7.
This is the context in which the claims state that the invention is, for instance, "[a] method ... comprising ... partitioning a visual display of the device into an array of tiles." Id. claim 1, 24:15-20. The point of the invention, from the patent's perspective, is to bring order to a human interface paradigm that had grown chaotic as evolving technologies increased the number of information sources that computer users wished to view.
The parties initially agreed that these two phrases — the first found in independent claim 1, the second found in independent claims 22 and 46 — should be given the same construction. Joint Claim Construction Chart 1. Microsoft proposed that they both be construed as "depicting/depict an array of tiles on some or all of a display," and SurfCast proposed "dividing/divide some or all of a display into an array of tiles." Id. Microsoft later amended its proposed construction and asked that "arrange a display into an array of tiles" not be given any particular construction. Def.'s Revised Constructions at 1. The PTO construed the phrase to mean "dividing a display or window into two or more tiles." PTO Op. at 13.
SurfCast contends that the plain meanings of the claim terms "partitioning" and "arrange," as used in the patent, are more consistent with "dividing/divide some or all of a display" than "depicting/depict ... on... a display." Pl.'s Br. at 15. It claims that "partitioning" means to divide space, and here the space is a display. Id. It cites the American Heritage Dictionary for the common meaning of "partition": "to divide into parts, pieces, or sections." Id. (citing THE AMERICAN HERITAGE DICTIONARY OF THE ENGLISH LANGUAGE 1320 (3d ed.1992)). It insists that Microsoft's definition "operates to remove any meaning imputed by the term `partitioning' from the claim." Id. SurfCast cites Innova/Pure Water, Inc. v. Safari Water Filtration Systems, 381 F.3d 1111, 1119 (Fed. Cir.2004), for the proposition that "all claim terms are presumed to have meaning in a claim," and that consequently it would be improper to construe these phrases without the meaning of "partition." SurfCast also disagrees with Microsoft's construction because it "flips the focus of the claim language from the display to the tiles and only requires showing a collection of tiles on the display with no particular order or arrangement." Pl.'s Br. at 16.
Microsoft argues that the specification refers to the functionality described in the independent claims in terms of "displayed," "presented," and "rendered." Def.'s Br. at 13 (citing '403 Patent at 9:5-67, 12:6-8, 19:34-36, 22:51-56). It takes from this language that the array of tiles should be "depicted," its chosen verb. Microsoft also contends that SurfCast's "divided" will be confusing to the jury because it is not consistent with presentational language of the specification. It also claims that dependent claim 3 speaks of partitioning an array into non-overlapping tiles; in Microsoft's view, if claim 1 requires "dividing" then partitioning into non-overlapping tiles is superfluous. Thus, Microsoft contends, under the doctrine of claim differentiation, "dividing" cannot be the correct verb.
The dispute about these terms stems in large part from the parties' different understanding of "array of tiles." If, as Microsoft requested, an array of tiles is given its computer science meaning — a collection of virtual objects in memory — then it makes sense to "depict an array of tiles on some or all of a display." This is because the array itself, a blob of data in memory, has no inherent visual characteristics; the tiles within the array must be "depicted," or made perceivable by a human being. But if, as SurfCast requested, an array of tiles means "multiple tiles displayed in an orderly fashion," then it makes sense to "divide some or all of a display into an array of tiles." The array already has some visual characteristics because it is a human interface element (it is "displayed"); it remains only to present the array on screen by allocating some visual space to it.
The Court already determined that an "array of tiles" is "multiple tiles displayed in an orderly fashion," and that conclusion weighs heavily in favor of SurfCast's construction here. "Dividing/divide some or all of a display into an array of tiles" would be consistent with the Court's construction of "array of tiles," and therefore would make more sense to a jury. Microsoft's assertion of the doctrine of claim differentiation does not defeat SurfCast's construction because the "partitioning" of the visual display into an array of tiles, found in claim 1, does not require the "non-overlapping configuration of tiles ... [of a] uniform size and shape" found in claim 3. Partitioning the display — making room for the array of tiles — is distinct from laying out tiles within the display space allocated to the array of tiles.
Finally, the Court rejects Microsoft's amended proposal that "arrange a display into an array of tiles" be given its plain and ordinary meaning. The two phrases serve nearly identical functions in all three independent claims. Neither party has advanced any explanation of why claim 1 uses slightly different language to describe this process, and the Court can perceive none. Because both phrases appear to mean the same thing, the Court gives them the same construction.
In sum, both the phrase "partitioning a visual display of the device into an array of tiles" and "arrange a display into an array of tiles" mean "dividing/divide some or all of a display into an array of tiles." This follows naturally from the conclusion that "array of tiles" means "multiple tiles displayed in an orderly fashion."
SurfCast proposes that "user-defined array size" be construed as "the number and arrangement of tile positions in the array as specified by the user." Joint Claim Construction Chart 1. Microsoft proposes that "user-defined array size" be construed as "the number of elements
SurfCast proposes that "user-defined array size" means "the number and arrangement of tile positions in the array as specified by the user." Pl.'s Br. at 16. In support of this construction, SurfCast turns first to the prosecution history of the patent, in which the applicant addressed a rejection by the examiner based on similarity to prior art, "McFedries." Id. at 16 (citing '403 File History 14). This passage reads:
Id. SurfCast takes from this passage that "`user-defined array size' therefore means more than just a user selecting the number of tiles in the array"; rather, it also embraces "the number and arrangement of tile positions." Pl.'s Br. at 16-17.
SurfCast also points to the specification to show that a user may specify more than simply the number of tiles that are part of the array. Pl.'s Reply at 5. For instance, "`a user may specify a presentation of the grid, consisting of its dimensions (i.e., the number of tiles to display and their arrangement).'" Id. (quoting '403 Patent at 11:9-11). Likewise, it identifies "`one embodiment in which all tiles ... are presented in an array comprising M rows and N columns. There is no particular requirement that the arrangement consists of more than one row and more than one column.'" Id. (quoting '403 Patent at 12:7-10). SurfCast concludes that one skilled in the art would understand that the user may specify the number of rows and columns in a grid-like array, defining the size of the array. Id.
Consistent with its construction of "array of tiles" in the technical, computer science sense, Microsoft construes "user-defined array size" to mean "the number of elements in an array, as specified by a user." Microsoft begins with the proposition that "size," as a term, does not include the concept of arrangement or position. Def.'s Br. at 14. It presents the example of a living room; "[t]he size of one's living room says nothing of how it is arranged." Id. Microsoft also disputes SurfCast's interpretation of the prosecution history. Id. at 15. It reads the passage from the '403 File History to show that the applicant distinguished McFedries on the grounds that the prior art system did not allow a user to define "the number of panels" displayed. Id. (quoting '403 File History at 14) (emphasis in original). Microsoft interprets this response to mean that the applicant was not distinguishing McFedries on the ground that it could not let the user control the arrangement of panels. Id. Microsoft also points out that the response acknowledges that the McFedries invention could, in fact, control
The '403 File History is incomplete without understanding the "McFedries" prior art. This is a reference to a book by Paul McFedries, The Complete Idiot's Guide to Windows 95 (2d ed.1997). '403 File History at 11. From what the Court can discern,
Understanding that the prosecution history is of less interpretational value than the patent itself, the Court concludes that this passage supports SurfCast's construction. The applicants distinguished McFedries on the ground that their invention permitted the user to specify the number of positions available for tiles. While the distinction does not speak to the arrangement of those positions, it at least speaks to the number of positions. This is closer to SurfCast's proposed construction, "the number and arrangement of tile positions in the array as specified by the user," than it is to Microsoft's proposed construction, "the number of elements in an array, as specified by a user."
It remains to determine whether a "user-defined array size" also speaks of the arrangement of tile positions. This dispute, like several others already described, stems in part from the parties' divergent constructions of "array of tiles." If an "array of tiles" is a one-dimensional block of memory storing a collection of virtual objects, its "size" is simply a non-negative integer value. If, on the other hand, an "array of tiles" is "multiple tiles displayed in an orderly fashion," then the concept of "size" is more complex.
The Court has already construed "array of tiles" to mean "multiple tiles displayed in an orderly fashion." When speaking of the size of this orderly display, it is not sufficient merely to address the number of positions. For instance, if the orderly display was a grid, saying that there are sixteen positions would not fully describe the "size" of the array; it would also be necessary to specify that the array is 4×4, or 2×8, or 1×16. These dimensional measurements are arrangements of positions, not merely quantities of positions. It follows that "size," when applied to an "array of tiles," must include the concept of arrangement.
Consequently, the Court construes "user-defined array size" to mean "the
The phrase "refresh rate" appears in independent claim 1, while "retrieval rate" appears in independent claims 22 and 46. SurfCast proposes the same construction for both, while Microsoft proposes different constructions. The Court addresses the two phrases together.
Microsoft proposes that "refresh rate" be construed as "a recurring time interval at which information displayed in a tile is refreshed." Joint Claim Construction Chart at 1. It proposes that "retrieval rate" be construed as "a recurring time interval at which information is retrieved." Id. SurfCast proposes that both terms be construed as "criteria for updating." Id. Microsoft's revised construction proposes, as an alternative, that "refresh rate" mean "a rate at which information displayed in a tile is refreshed" and that "retrieval rate" mean "a rate at which information is retrieved." Def.'s Revised Constructions at 1. The PTO construed the terms to mean "a recurring time interval at which information displayed in a tile is refreshed or retrieved." PTO Op. at 14.
SurfCast proposes that both "refresh rate" and "retrieval rate" be construed as "criteria for updating." In support of its construction, SurfCast relies heavily on a preferred embodiment in which tiles are updated in response to two events that are not defined by a specific time interval: first, when a locally stored file is updated, and second, at a certain time of the day.
Next, SurfCast claims that the plain and ordinary meaning of "rate" is not limited to "a recurring time interval," as Microsoft proposes. It cites the dictionary definition of "rate": "a quantity measured with respect to another [measured] quantity." Id. at 18-19 (quoting THE AMERICAN HERITAGE DICTIONARY OF THE ENGLISH LANGUAGE at 1500).
Next, SurfCast contends that Microsoft's proposed construction adds ambiguity that will confuse the jury. Id. It points out that Microsoft's construction will result in some redundant verbiage if it replaces, word for word, either of the disputed claim terms. Id. For instance, claim 1 would become "updating information ... presented to said first tile in accordance with said recurring time interval at which information displayed in a tile is refreshed." See id.
Microsoft proposes that "refresh rate" mean "[a] recurring time interval at which information displayed in a tile is refreshed," and that "retrieval rate" mean "[a] recurring time interval at which information is retrieved." Def.'s Br. at 15. Microsoft's analysis begins from the principle that different claim words are presumed to have different scope. Id. (citing Applied Med. Res. Corp. v. U.S. Surgical Corp., 448 F.3d 1324, 1333 (Fed.Cir.2006)). Thus, in Microsoft's view, the two terms presumptively have different meanings. Id. Furthermore, Microsoft points out, the phrase "criteria for updating" does not appear in the patent. Id. In Microsoft's view, "criteria for updating" is so broad as to read "rate" out of the claims. Id. at 15-16.
Microsoft also disputes that the preferred embodiment cited by SurfCast supports the construction "criteria for updating." Id. at 16. Microsoft reads the language about updating a tile in response to changes in a word processing file on the disk, not as suggesting a time-independent refresh, but rather as describing a situation where the refresh functionality is turned off. Id. In other words, because the update is not time-bound, it must be that the user has elected not to refresh in accordance with a "rate" at all. Id. In a footnote, Microsoft expresses a fallback theory: that the embodiment describes a periodic check of the hard disk, at a very short time interval, to see if the file has changed; if it has, the process of the invention updates the tile. Id. at 16 n. 7.
In support of its own proposed construction, Microsoft claims that its proposal captures the ordinary sense of "refresh rate" and "retrieval rate." Id. at 16-17.
Microsoft is correct that different words in the patent presumptively mean different things. Applied Med., 448 F.3d at 1333 n. 3 (citing CAE Screenplates Inc. v. Heinrich Fiedler GmbH & Co. KG., 224 F.3d 1308, 1317 (Fed.Cir.2000)). However, the Federal Circuit expressed this concept with an additional subtlety: "[T]he use of two terms in a claim requires that they connote different meanings, not that they necessarily refer to two different structures." Id. (emphasis in original). The Federal Circuit added that "[t]he prosecution history, specification, comparison with other claims in the patent, and other evidence may require that two terms in a claim refer to different structures." Id. (citing Phillips, 415 F.3d at 1312-19). The '403 Patent describes software, not a tangible thing; however, software certainly has virtual characteristics that are conceptually akin to "structures."
The phrase "refresh rate" appears in independent claim 1 in this context:
'403 Patent claim 1, 24:15-33 (emphasis added). Each tile has a refresh rate; each tile has an information source; and the method updates information from each information source in accordance with the refresh rate. The Court has already construed "tile" to mean, in part, a "graphical representation of an associated information source capable of displaying refreshed content"; it follows that when the method updates the information source, it also updates the visual display of the tile with "refreshed content." Thus, besides being a property of a tile, the "refresh rate" governs two things: (1) under what circumstances the method retrieves new information from the information source, and (2) under what circumstances the method updates the associated tile's visual display.
The phrase "retrieval rate" appears in both independent claims 22 and 46. Claim 22 describes an electronic readable memory with five sets of instructions: (1) to control simultaneous communication with a plurality of information sources; (2) to arrange a display into an array of tiles; (3) to associate information sources with tiles; (4) to "retrieve information from ... [the] information source[s] in accordance with... [their] retrieval rate[s]"; and (5) to "present information to ... [the] tile[s] in accordance with ... [their] retrieval rate[s]." Id. claim 22, 25:25-44. Claim 46 is similar; it describes a device with a processor that executes instructions identical in substance to those of claim 22. Id. claim 46, 26:47-27:4. Notably absent from either claim 22 or 46 is any language about assigning a retrieval rate to a tile; dependent claims 41 and 42 add this feature to independent claim 22, and dependent claims 48 and 49 add it to independent claim 46. Id. at claims 41-42, 26:34-38; id. at claims 48-49, 27:7-11. Absent those specializations, the retrieval rate exists independently of the tile in claims 22 and 46. Nonetheless, the "retrieval rate" in these claims still governs (1) under what circumstances the instructions retrieve information from the information source, and (2) under what circumstances the instructions update the associated tile's visual display.
Under this analysis, there is no barrier to "refresh rate" and "retrieval rate" being given the same construction. The fact that the phrases use slightly different terms does not change the fact that they perform identical functions. The difference in nomenclature likely stems from the fact that in independent claim 1 the "refresh rate" is always assigned to tile; the tile is by definition "refreshable." By contrast, in claims 22 and 46, the "retrieval rate" is independent of the tiles, and so it is more appropriate to refer to the rate in terms of the information source, from which information is "retrieved." The patentee likely chose slightly different words because of the different context of the independent claims, but the meanings of both phrases can be captured by the same construction, contrary to Microsoft's position. The Court addresses shortly whether construing the terms identically is the correct result.
The parties' dispute turns largely on whether the embodiment described in column 12 of the '403 Patent mandates that
First, the words of the patent squarely rebut Microsoft's principle argument. The patent says:
'403 Patent at 25:50-57. The patent says nothing about the user disabling the refresh rate; rather it gives examples of a refresh rate. The Court does not conclude that in its first two examples of "managing" the refresh rate, the patent actually describes the user disabling it altogether.
Microsoft's fallback argument is equally unavailing. In its footnote 7, it claims that the embodiment describes a situation where the user has defined the refresh rate relative to a very short time period, and that the grid is checking the underlying file repeatedly over that very short period. Def.'s Br. at 16 n. 7. But that is not what the patent describes; it says "the user might configure the tiles to refresh only when the underlying data is written to the local hard drive." '403 Patent at 12:53-55. From the user's perspective — the one that counts in this sentence — the tile refresh is triggered by an event that is not bound to a discrete time interval; rather, it is a conditional event. Likewise, the example in which the refresh rate defines a time of day at which to update HTML data also describes a rate that is not bound to a specific time interval from the user's perspective.
Furthermore, nothing about the embodiment requires that it be implemented in the way that Microsoft describes in its footnote. Although repeatedly checking the underlying file is certainly one method to implement the embodiment, others are possible. For instance, the underlying file could be configured to broadcast notice to any listeners whenever any process made changes to the file; among those listeners could be the grid that is subject to the "refresh rate." This technique does not require periodically polling the file, and was well known in 2000. Observer, Microsoft Developer Network, http://msdn.microsoft.com/en-us/library/ff649896.aspx (last visited Mar. 12, 2014) (describing the Observer pattern, in which a "subject" object is programmed to automatically notify an "observer" object whenever the subject changes) (citing ERICH GAMMA et al., DESIGN PATTERNS: ELEMENTS OF REUSABLE OBJECT-ORIENTED SOFTWARE (1995)). This further demonstrates that the refresh rate need not be a specific interval of time.
The Court concludes that a "rate," as used in the claims, is not necessarily measured temporally. A construction limited to temporal rates would improperly exclude the preferred embodiment described in column 12, Accent Packaging, 707 F.3d at 1326, and nothing in the claim language or specification suggests that the term should be so limited.
Although Microsoft's main arguments against "criteria for updating" as the construction of "refresh rate" and "retrieval rate" do not prevail, SurfCast's terse, general phrasing could mislead the jury. Microsoft's proposals — that "refresh rate" be "[a] recurring time interval at which information displayed in a tile is refreshed" and "retrieval rate" be "[a] recurring time interval at which information is retrieved" — are more specific and precise in the context of the claims. However, as noted above, they improperly limit "rate" to only temporal periods. A better construction would use the best elements of both proposals. Consequently, the Court construes "refresh rate" to mean "the criteria for updating information displayed in a tile." Likewise, the Court construes "retrieval rate" to mean "the criteria for retrieving information."
This is the first of four disputed claim terms whose meanings are tightly related to each other. The Court treats them one at a time, understanding that the construction of each term influences the constructions of the others. Some of the dispute between the parties as to these terms hinges on their disagreement over "refresh rate" and "retrieval rate," which the Court has already resolved. Supra Section IV.E. Much of the remainder turns on a dispute over the meaning of "simultaneous." See infra, Sections IV.H to IV.I.
However, before reaching that issue, the Court must first address two precursor terms, of which the first is "updating information from a [first/second] information source in said plurality of information sources presented to said [first/second] tile." This phrase appears in independent claim 1. SurfCast contends that the term can be given its plain and ordinary meaning. Pl.'s Br. at 20. Microsoft proposes that it mean "updating the information displayed in the tile." Def.'s Br. at 18.
SurfCast proposes that "updating information from a first information source in said plurality of information sources presented to said first tile" be given its plain and ordinary meaning. Pl.'s Br. at 20. However, SurfCast devotes no briefing to this particular term, focusing instead on its preferred construction for the related terms containing "simultaneously." See id. at 20-26.
Microsoft proposes that the phrase means "updating the information displayed in the tile." Def.'s Br. at 19. In support of this construction, Microsoft offers that the "updating described in the patent uniformly refers to the updating of the tile image on the display." Id. (citing '403
First, SurfCast has not conceded the point. SurfCast's brief refers to "data transfer for a single update of the second tile," Pl.'s Br. at 21; data transfer is not necessarily the same thing as updating information on a display. However, Microsoft is correct that all three independent claims speak of updating a tile's visual display at the "rate" at which the various processes retrieve information from an information source. '403 Patent at claim 1, 24:26-28; id. at claim 22, 25:41-44; id. at claim 46, 27:1-4. Microsoft is also correct that "updating the information displayed in the tiles themselves is the very point of the invention." Def.'s Br. at 19 (citing '403 Patent at 7:55-59). The Court agrees that the language "updating information from a [first/second] information source in said plurality of information sources presented to said [first/second] tile," even in the context of independent claim 1, is somewhat ambiguous as to whether the operation includes updating the visual display of the tile. Consequently, it would be helpful to remind the jury that claim 1 also embraces this display-updating characteristic.
However, Microsoft's proposed construction is too narrow; it speaks only of "updating the information displayed in the tile." Def.'s Br. at 18. This construction could mislead the jury to conclude that the term refers only to updating a visual display, when in fact it also refers to obtaining information from an information source. Instead, the Court construes the phrase "updating information from a [first/second] information source in said plurality of information sources presented to said [first/second] tile" to mean "obtaining refreshed information from the information source and updating the information displayed in the tile." This captures both the notion of obtaining fresh information and of updating the visual display of the tile.
The second term that the Court must resolve before reaching the question of "simultaneously" is "in accordance with said [first/second] [refresh/retrieval] rate." This phrase appears, in its variations, in all three independent claims. '403 Patent at claim 1, 24:26-33; id. at claim 22, 25:36-40; id. at claim 46, 26:62-27:4. SurfCast proposes that the term be given its plain and ordinary meaning, Pl.'s Br. at 20; Microsoft proposes "at the interval set by the refresh/retrieval rate." Def.'s Br. at 18.
SurfCast proposes that this term should also be given its plain and ordinary meaning. Pl.'s Br. at 20. Again, SurfCast does not offer any argument defending this position. See id. at 20-26.
Microsoft proposes that the term be construed to mean "at the interval set by the refresh/retrieval rate." Def.'s Br. at 19-20. Microsoft argues that the phrase "in
First, the Court again disagrees with Microsoft's contention that SurfCast's brief contains such a concession. SurfCast's proposed construction for the "simultaneously updating ..." term is consistent with its theory that a "rate" is not necessarily a temporally-bound period. It says nothing about an "interval"; it simply describes "waiting for the next time at which the update is to occur as set by the refresh rate." Pl.'s Br. at 21. This interpretation is compatible with SurfCast's broader view of the meaning of "rate," and does not concede that the "in accordance with ..." term embraces a time interval. See Section IV.E.2.b, supra.
Microsoft's proposed construction is consistent with its view that a "rate" is a time interval, but the Court has already considered and rejected that proposition. Given that the Court has already construed "refresh rate" and "retrieval rate," little would be added by giving an additional construction to "in accordance with said... [refresh/retrieval] rate." The jury will be able to apply the correct meaning without assistance. The Court will give "in accordance with said [first/second] [refresh/retrieval] rate" its plain and ordinary meaning, qualified by its construction of "refresh rate" and "retrieval rate."
This phrase appears only in independent claim 1. '403 Patent at claim 1, 24:29-33. However, it has close analogs in independent claims 22 and 46 that the Court addresses later. Id. at claim 22, 25:27-28; id. at claim 46, 26:54-55.
Pl.'s Br. at 21. SurfCast revised its position, suggesting instead that the term "simultaneously" means that "the ongoing processes of updating information presented to the two tiles occur at the same time, even if the updating of the two tiles occurs at different rates." Pl.'s Revised Constructions at 1. Microsoft, by contrast, proposes that the whole phrase be construed as "simultaneously updating the information displayed in the second tile at the interval set by the second refresh rate." Def.'s Br. at 21. The PTO adopted SurfCast's proposal. PTO Op. at 15.
SurfCast's basic theory of the meaning of "simultaneously" is that the individual data transfers for the first and second tiles
In support of this interpretation, SurfCast offers the embodiment in which one tile is set to update when a document in the file system is changed, and a second tile is set to update HTML data from the internet at a particular time of day. Id. at 22-23. It observes that there will be some times during which the first process is actively updating, usually in response to the user saving work on an open document, and other times when it is not updating. Id. Likewise, there will be a particular time of day when the second process is downloading HTML data from the internet; at other times, the second process will not be updating. Id. SurfCast submits that these two processes are nonetheless simultaneous because they are both running at the same time, though not necessarily actively updating from the information source. Id. at 23.
Microsoft vigorously disputes that claim 1 refers to a "process" of any kind: "SurfCast once again ignores the claim language, which never uses the word `process.'" Def.'s Br. at 20 (emphasis in original). Microsoft submits that the claim language requires "only the updating of information, which may or may not require some type of data transfer and is never described as taking place over a `[period] of time."' Id. at 20. Addressing SurfCast's example, Microsoft claims that the specification only explains that the tile is refreshed "at a certain time of day," not over a period of time. Id. at 20 (quoting '403 Patent at 12:56-58).
Microsoft also disputes that the use of "simultaneously" in claim 1 permits anything but the updating of information at the same time — not, as SurfCast claims, the operation of two parallel processes that update independently of each other but which run at the same time. Id. at 21. Microsoft points to numerous instances of the word "simultaneously" in the specification that, in its view, show that the authors used the word to mean occurring in the same span of time. Id. (citing '403 Patent at 1:44-46, '403 Patent at 1:66-2:1, '403 Patent at 2:57-61, '403 Patent at 8:38-41, and '403 Patent at 11:12-14). Microsoft insists that because the ordinary meaning of "simultaneously" is "at the same time," SurfCast's proposal — that it means two simultaneous processes that may not update
Finally, Microsoft asserts that SurfCast's interpretation would read "simultaneously" out of the patent because it could apply to any time period. Id. at 22. Under this interpretation, an update that occurred on Day 1 and an update that occurred on Day 2 would be "simultaneous" because they both occurred in the same week. See id. At oral argument, Microsoft offered a clear metaphor; in its view, SurfCast's interpretation of "simultaneously" would mean that George W. Bush and Barack Obama both occupied the office of President of the United States "simultaneously" because they were both President in the first decade of the twenty-first century. Tr. of Proceedings, Markman Hr'g 72-73 (ECF No. 120) (Markman Hr'g).
The core of the disagreement over this term is the meaning of "simultaneously" in independent claim 1.
The analysis begins with the Court's construction of the subsidiary terms. The Court has construed "refresh rate" to mean "the criteria for updating information displayed in a tile." It has construed "updating information from a [first/second] information source in said plurality of information sources presented to said [first/second] tile" to mean "obtaining refreshed information from the information source and updating the information displayed in the tile." It has given the phrase "in accordance with said [first/second] [refresh/retrieval] rate" its plain and ordinary meaning, conditioned by the Court's construction of "refresh rate." These are all the elements of the action that is said to happen "simultaneously" in claim 1. Putting the pieces together, what is "simultaneous" is obtaining refreshed information from the information source and updating the information displayed in the tile, in accordance with the criteria for updating that information.
The key element that weighs in favor of SurfCast's basic concept of simultaneity is "in accordance with the criteria for updating." One single data transfer to update the tile can hardly be said to be "in accordance" with the criteria for updating that information, when the specification teaches that those criteria can include a repetitive time interval, a time of day, or an unpredictable condition such as the user saving a document. '403 Patent at 12:49-64. The whole clause to which "simultaneously" refers describes a continuous, iterative process, not a point in time.
Language in the specification strongly supports this conclusion as well, even setting aside the preferred embodiment that gives examples of refresh rates. For instance, in one embodiment, "[w]hen a given tile is refreshed, the refresh operation is completed before refreshing the next tile
Linguistically, Microsoft's construction would sever "in accordance with said second refresh rate" from the rest of the sentence and only apply "simultaneously" to the first half. If Microsoft were correct, and "simultaneously" only referred to updating the visual display of two tiles, Def.'s Br. at 22, then "in accordance with said second refresh rate" would not appear in the sub-clause that is governed by "simultaneously" at all. But this is not how the patent was drafted; the "in accordance with" language appears in both of the sub-clauses that are modified by "simultaneously":
'403 Patent at claim 1, 24:26-33 (emphasis added).
Microsoft argues that claim 1 never mentions "individual data transfers," and speaks only of updating the visual display of two tiles. Id. This is textually correct, but unduly narrow. A single iteration of "updating" a tile — within the overall repetitive process defined by the criteria for updating ("in accordance with ... [the] refresh rate") — necessarily involves a data transfer. The patent consistently associates updating the visual display of the tile with fetching refreshed information from some information source. E.g., '403 Patent at 8:29-66. Indeed, as Microsoft concedes, "updating the information displayed in the tiles themselves is the very point of the invention." Def.'s Br. at 19. It would be impossible to update the information displayed in the tile without first obtaining some refreshed information; this involves an individual data transfer.
To explain this point by using Microsoft's presidential analogy, the forty-third and forty-fourth Presidents are more like two individual updates within either of the two update processes of claim 1. They were both selected "in accordance with said first refresh rate" of a four year election cycle; they were not selected simultaneously, even under SurfCast's construction. It would be more apt to say that the Presidency of the United States and the Governorship of the state of Maine are "simultaneous" under this construction. They are two separate "tiles," or political offices, that are continuous but repetitively updated under independent criteria.
The phrase "simultaneous communication with a plurality of information sources" occurs in independent claims 22 and 46, and plays a functional role equivalent to that of the "simultaneously updating..." language discussed above. Claim 22 speaks of "[a]n electronic readable memory to direct an electronic device to function in a specified manner, comprising... a first set of instructions to control simultaneous communication with a plurality of information sources." '403 Patent at claim 22, 25:25-28. Claim 46 describes "[a] system for facilitating the organization and management of multiple data sources, comprising: a device that includes a processor configured to execute instructions... to ... control simultaneous communication with a plurality of information sources." Id. claim 46, 26:47-55. The descriptions of the actual instructions are nearly identical between claims 22 and 46. Compare id. at claim 22, 25:27-44 with id. at claim 46, 26:54-27:4.
SurfCast originally proposed that "simultaneous communication" be construed to mean "repetitive series of individual communications with each of a plurality of information sources which occur during the same period of time." Pl.'s Br. at 24. Microsoft asks that the term be given its plain and ordinary meaning. Def.'s Br. at 22. SurfCast later revised its proposal, suggesting instead that "simultaneous communication" mean that "the ongoing processes of communicating with multiple information sources occur at the same time, even if the communications with the information sources occur at different rates." Pl.'s Revised Constructions at 1.
SurfCast relies initially on its theory of the meaning of "simultaneously" discussed above. Pl.'s Br. at 25. It also contends that the ordinary meaning of "simultaneous communications," in the context of computer networks, embraces a repetitive series of individual communications with the information sources. Id. This is so because even if the retrieval of information for two tiles from an external web site were scheduled to commence at the same time, the communications with the external site would not literally occur at the same instant. Id. One of skill in the art would, SurfCast contends, understand that the communications, even if apparently simultaneous to the user, are actually broken up into "packets" and transmitted from the two external servers to the user's computer in an interleaved fashion. Id. SurfCast concludes that although the data transfers for those communications are technically happening in small segments at different times, the overall communication is still "simultaneous." Id. at 26. This, in SurfCast's view, supports its construction of "simultaneous communications." Id.
Microsoft incorporates its opposition to SurfCast's definition of "simultaneously" described previously. Def.'s Br. at 22. It
There are two points of contention within this disputed term: first, whether "simultaneous communication" must be construed as a "repetitive series" of communication; and, second, whether that repetitive series must "occur during the same period of time," as advanced by SurfCast, or simply "at the same time," as requested by Microsoft.
First, SurfCast's network packet example begs rather than answers the question. It does little good to claim that requests for HTML data initiated "at the same time" result in interleaved packets, and then assert that the interleaving means that "simultaneous" must therefore not be "in the same instant." The net result is that SurfCast claims "at the same time" does not mean "at the same instant"; it amounts to a circular definition. Microsoft's response is necessarily, and forgivably, merely a counter-assertion of definition rather than a useful explanation of the functional meaning of the words.
The Court has already considered the problem of the term "simultaneously" in the context of claim 1, and that result affects its view of claims 22 and 46. All three claims describe, with a few variations, the process of associating an information source with a tile and then periodically obtaining fresh information from the information source and presenting it to the tile. It would make little sense to adopt a construction of "simultaneous" in claims 22 and 46 contrary to the conceptual result in claim 1. However, the differences and relationships between the claims must inform the way in which that concept is expressed.
Claim 1 describes "[a] method executed by a device under the control of a program, said device including a memory for storing said program." '403 Patent at claim 1, 24:15-17. The substantive steps discussed above are said to be part of the method. Id. claim 1, 24:18-33. Claim 22, by contrast, describes "[a]n electronic readable memory to direct an electronic device to function in a specified manner," id. im 22, 25:25-26; the instructions (including "simultaneous communication") are part of the memory. Id. claim 22, 25:27-44. Finally, claim 46 describes "a system for facilitating the organization and management of multiple data sources, comprising: a device that includes a processor configured to execute instructions, a memory connected to said processor to store at least one program ... wherein said processor executes instructions." Id. claim 46, 26:47-53. The instructions of the claim perform the same substantive work as in claim 22. Id. at 26:54-65, 27:1-4. As noted previously, claim 1 includes the concept of assigning a refresh rate to a tile, which claims 22 and 46 lack (though that concept is added in dependent claims). And in claim 1, the "simultaneously" language addresses the update processes, whereas in claims 22 and 46 the "simultaneous" language merely addresses "communications with ... information sources." However, in both claims 22 and 46, the claim language makes clear that the "communications" are retrieval and presentation of information "in accordance with a ... retrieval rate." Id. claim 22, 25:36-44; id. at claim 46, 26:62-27:4. In other words, what is "simultaneous" in claims 22 and 46 is
In sum, apart from assigning a refresh rate to a tile, claims 22 and 46 do exactly the same basic work of communication that claim 1 describes. Claim 1 is a "method" executed by a device with a memory; claim 22 is a "memory"; and claim 46 is a "system ... comprising ... a device" with a memory and a processor. There is nothing about the fundamental nature of either claim 22 or 46 that makes the idea of "communications" any different than what appears in claim 1, and the Court already ruled that claim 1 describes two parallel processes, running simultaneously, that sometimes (in accordance with a refresh rate) update information from an information source. This is, precisely, a "repetitive series of individual communications with each of a plurality of information sources which occur during the same period of time," as SurfCast originally proposed.
However, the Court agrees that SurfCast's revised construction also captures the meaning of the term and is phrased in clearer language. The Court construes "simultaneous communication" to mean that "the ongoing processes of communicating with multiple information sources occur at the same time, even if the communications with the information sources occur at different rates."
This phrase occurs only in independent claim 1, as one of the elements of the method the claim describes. '403 Patent at claim 1, 24:23-25. SurfCast proposes that it be given its plain and ordinary meaning, Pl.'s Br. at 26, while Microsoft proposes that it be construed as "setting the rate at which a tile will display new information." Def.'s Br. at 23.
SurfCast argues that the phrase "assigning a [first/second] refresh rate to a [first/second] tile of said array of tiles" is not ambiguous, nor used in any way inconsistent with its plain and ordinary meaning. Pl.'s Br. at 26. In opposition to Microsoft's construction, SurfCast first claims that Microsoft's construction introduces an inconsistent definition of "refresh rate." Id. at 26-27. Second, SurfCast views Microsoft's construction as itself ambiguous, since it repeats some phrases of the claim language without clarifying them. Id. at 27. Third, SurfCast disagrees with the use of the term "new information." Id.
On this last point, SurfCast suggests that "new" is misleading because an update operation could result in the same data being displayed in the tile. Id. For instance, it points out that a tile associated with a weather application might present the same temperature to the user after an update if the temperature has not changed. Id. This, SurfCast claims, is within the scope of the invention but is excluded by using "new information" in the construction of "assigning a [first/second] refresh rate to a [first/second] tile of said array of tiles." Id.
Microsoft finds support for its construction of the term in the specification. It points out that the specification "explains that the disclosed system stores a data structure for each tile, which ... includes a `refresh function [] that handles updates to the tile image.'" Def.'s Br. at 23 (quoting '403 Patent at 10:4-5). Microsoft equates "updating the tile image" with "displaying new information." Id. It finds further support in the patent's description of different ways to implement the timing
Microsoft rejects SurfCast's objection that its construction redefines "refresh rate," and disputes the claim that "new information" is incompatible with the patent's update language. Id. Microsoft cites the definition of "update" in WEBSTER'S THIRD NEW INTERNATIONAL DICTIONARY 2517 (1986): "to bring up to date." It views "not updating the tile" as "inconsistent with the word `update.'" Id.
First, Microsoft offers no explanation of why the phrase "assigning a ... refresh rate to a ... tile" is ambiguous, and the Court perceives none. Even assuming, for the purpose of argument, that Microsoft's proposed construction is a fair restatement of the phrase, unambiguous claim language needs no further construction. See Phillips, 415 F.3d at 1314 ("In some cases, the ordinary meaning of claim language as understood by a person of skill in the art may be readily apparent even to lay judges, and claim construction in such cases involves little more than the application of the widely accepted meaning of commonly understood words"). The Court has already construed "tile," "refresh rate," and "array of tiles"; some justification ought to be offered for impressing on the jury a particular meaning of "assigning." There appears to be no reason to add further complexity to the phrase.
The Court construes "assigning a [first/second] refresh rate to a [first/second] tile of said array of tiles" to have its plain and ordinary meaning, qualified by the constructions the Court has already given to "tile," "array of tiles," and "refresh rate."
This phrase appears in dependent claims 11, 43, and 50 — one for each of the three independent claims. '403 Patent at claim 11, 24:61-62; id. at claim 43, 26:39-40; id. at claim 50, 28:1-2. SurfCast proposes that it be construed to mean that the "[refresh/retrieval] rate is assigned without direct user input," Pl.'s Br. at 28, while Microsoft proposes "assigning a refresh/retrieval rate without direct input from an end user indicating the rate." Def.'s Br. at 24.
SurfCast objects to Microsoft's construction on the grounds that it is improper to read "automatically" to mean "without input from an end user indicating the rate"; SurfCast preferred simply "without direct user input." Pl.'s Br. at 28. SurfCast reasons that the plain and ordinary meaning of "automatic" is "acting in a manner essentially independent of external influence or control," id. (citing THE AMERICAN HERITAGE DICTIONARY OF THE ENGLISH LANGUAGE at 125), and that Microsoft's construction leaves open the possibility that the user could assign the rate indirectly without explicitly indicating it. Id. Indeed, at oral argument Microsoft proposed precisely such a scenario. Markman Hr'g at 76.
In defense of its construction, Microsoft cites an embodiment in which the specification
The parties' dispute boils down to whether the "automatic" assignment happens without any direct user input at all, or whether it merely happens without any user input indicating the rate. The Court agrees with SurfCast that "automatically," in this context, means essentially independent of external influence or control. However, Microsoft's construction describes a process that is essentially independent. When the specification speaks of "automatic" assignment of a refresh rate, it is, as Microsoft points out, in connection with some form of human input that is ultimately converted into a refresh rate — just not human input specifying the precise refresh rate that will be adopted. See '403 Patent at 13:52-14:24. The assignment of a refresh rate would not, in fact, be very useful if it were not in some way influenced by what the human user tells the computer. The benefit claimed by this aspect invention is that it converts simplified user input into a reasonable refresh rate without the user having to think about it. Thus, the Court views Microsoft as presenting a clearer view of the term.
Consequently, the Court construes the phrase "[refresh/retrieval] rate is assigned automatically" to mean "[refresh/retrieval] rate is assigned without direct user input indicating the rate."
The term "grid" appears in claim 12: "[t]he method of claim 1 wherein said array comprises a grid wherein each of said tiles occupies a fixed position on said grid." '403 Patent at claim 12, 24:63-65. SurfCast proposes that "grid" be construed as "a regular layout of rows and columns, wherein a single tile may occupy more than one row and/or column." Pl.'s Br. at 28. Microsoft prefers simply "a regular arrangement of rows and columns." Def.'s Br. at 25. In its revised constructions, Microsoft offered to change "arrangement" to "layout" to move closer to SurfCast's interpretation, resulting in "a regular layout of rows and columns." Def.'s Revised Constructions at 2.
During oral argument, the parties nearly agreed that "grid" could be construed as "a regular layout of rows and columns wherein a single tile may, but need not, occupy more than one row and/or column." See Markman Hr'g at 79-80. However, Microsoft later withdrew that concession, concerned that the definition did not include a regular layout of rows and columns that did not support any "spanning" at all. Id. at 82-84.
The PTO construed "grid" to mean "any presentation of tiles in a row and column orientation." PTO Op. at 19.
SurfCast cites the specification's usage of "grid" in support of its position. Pl.'s Br. at 28-29 (citing '403 Patent at 11:62-64).
Microsoft views SurfCast's construction as an attempt to import an optional feature of a grid into the definition of the term. Def.'s Br. at 26. Microsoft's basic point is that while some grids could support spanning, not every grid must support spanning. Id. Some arrangements of rows and columns could still fall within the definition of "grid" but only permit each tile to take up one column/row combination. Id. Under SurfCast's interpretation, according to Microsoft, a regular arrangement of rows and columns that did not allow spanning would not be a grid. Id. Microsoft asserts that there is no basis for limiting the claim in this fashion. Id.
Just as with "tile," Section IV.A.2.a, supra, the patentee here shows intent to act as his own lexicographer with respect to "grid." The specification provides a section entitled "Grid Object" that explains that it will teach what a "grid" is in the context of the patent. '403 Patent at 10:40-43. This section of the specification then goes on to give numerous details of a "grid." For instance, it is a "matrix of tiles," id. at 10:49; it "controls the layout and priorities of the tiles," id. at 10:53-54;
However, the spanning feature on which SurfCast relies to support its construction comes from just one preferred embodiment, and is never mentioned again. See '403 Patent at 11:62-12:29. This suggests, not that all "grids" support spanning, but that one particular kind of "grid" supports it. If the patentee had meant that every grid supported spanning, the spanning feature would have occurred, not in an embodiment, but in the definition of a grid. Alternatively, had every instance of a grid in the specification mentioned or obviously supported spanning — or had spanning been inherently necessary for the invention to be useful, as with the "persistence" of "tiles" — then it might be appropriate to read it in. But this is not such a case as the persistence of tiles, where excluding the feature would defeat the entire purpose of the invention. A grid of tiles is still a meaningful feature even without spanning. Given the singularity of the reference, it would be improper to import this feature from one embodiment into the claim term. Thorner, 669 F.3d at 1368.
At oral argument, Microsoft suggested that construing "grid" to necessarily include spanning would be like claiming that "an automobile is a four-wheel vehicle, but has a top that may be retracted." Markman Hr'g at 77. This is a strong simile. Some cars have retractable roofs, but that does not mean that all cars have retractable roofs. In the same way, some grids support spanning — as taught by the preferred embodiment — but nothing about the specification compels the conclusion that all grids support spanning.
Nonetheless, it should be made clear to the jury that spanning is a possible feature of grids. The Court construes "grid" to mean "a regular arrangement of rows and columns, which may, but need not, allow a single tile to occupy more than one row and/or column." This satisfies Microsoft's objection that an arrangement of rows and columns that does not support spanning is still a "grid," and also preserves the preferred embodiment of columns 11-12.
This language appears in claim 12: "[t]he method of claim 1 wherein said array comprises a grid wherein each of said tiles occupies a fixed position on said grid." '403 Patent at claim 12, 24:63-65. SurfCast proposes that the term be construed as "wherein each tile occupies a particular position in the regular layout of rows and columns," Pl.'s Br. at 30, while Microsoft initially proposed "wherein each tile occupies a specific row and column position in the grid." Def.'s Br. at 26. Microsoft later revised its proposal to: "wherein each tile occupies a row and column position on the grid that cannot be changed by the end-user." Def.'s Revised Constructions at 2. The PTO construed the term to mean "each tile occupies a particular cell or a particular plurality of cells of said grid." PTO Op. at 20.
SurfCast argues that its construction accounts for the possibility that a tile could occupy more than one row/column position
Microsoft argues that its construction does not exclude a grid that contained tiles occupying more than one row/column position. Def.'s Br. at 26. In support of its construction, it cites a portion of the specification that describes the grid as storing the position of a tile by its row and column number. Id. at 26-27 (citing '403 Patent at 13:12-14). Microsoft also proposed, as an alternative, that the Court give the term its plain and ordinary meaning. Id. at 27.
At oral argument, Microsoft changed the nature of its objection. It now began to object that SurfCast's interpretation of "fixed" is incorrect because it allows for the tile to be moved from one position to another. Markman Hr'g at 80-81. Microsoft now claims that under SurfCast's interpretation, even if a tile is moved by the user it is still "fixed"; this is improper, according to Microsoft, because the classic concept of fixation in space implies that the tile can only appear in a single location.
First, the Court is not convinced by Microsoft's view that the term "fixed" means that the user can never move the tile. Nothing in the patent suggests this limitation. In fact, the definition of "tile" states that one preferred embodiment permits the user to adjust the properties of an already-created tile; it describes "a toolbar function ... which may permit an array of special buttons to appear on or adjacent to the tile for the purposes of adjusting properties of the tile." '403 Patent at 10:14-16. The specification also states that the column and row number is one of the attributes of a tile. Id. at 13:12-14 ("The grid also stores other attributes of tiles such as their respective positions on the grid as given by their column and row number"). Putting these elements together, the specification teaches that in some circumstances the user can move the tile from one "fixed" position to another. While the tile may indeed occupy only one position on the grid at a time, "fixed" does not mean "immutable." Microsoft's alternative construction reads in a limitation not supported by either the specification or the claim language.
After careful review of the two initial proposed constructions, the Court agrees with Microsoft that there is very little difference between them, and that both are susceptible to a reading that would either limit or not limit spanning. Nevertheless, the Court's construction of "grid" should eliminate any confusion. Because a "grid" permits spanning but does not require it, there is little to be gained by attempting to give a specialized meaning to "fixed position." The jury should readily appreciate, from the construction of "grid," that a tile within such a grid may or may not occupy more than one row and/or column. There is also little danger that the jury will infer from the word "fixed" that the tile can never be moved from its initial location on the grid. The
Both SurfCast and Microsoft approached the task of claim construction with clear, internally consistent theories of what the '403 Patent describes. That the Court has by and large agreed with SurfCast's constructions reflects in part the strength of that internal consistency on both sides. The construction of one term often impacts the construction of others, both at a lexicographic and a conceptual level. Nowhere is this more apparent than in the construction of "tile(s)," which, in one way or another, influences almost every other claim term. The parties have obviously considered carefully both the terms that they have selected for the Court to construe and their positions on those terms, and both parties have been exceptionally well represented, pressing their positions cogently and thoughtfully.
Claim construction is not an exercise performed in a void. As counsel for both sides readily admitted at oral argument, the construction of these disputed claim terms is central both to the validity of the patent and to the issue of infringement. The Court understands that its constructions of the disputed terms will return to the microscope of exacting analysis in later dispositive motions, jury instructions, and, at some point, de novo appellate review before the Federal Circuit.
The Court adopts the following claim constructions for the '403 Patent:
The term "tile(s)" will mean "a graphical representation of an associated information source capable of displaying refreshed content, the graphical representation being persistent and selectable to provide access to underlying information of the associated information source."
The term "array of tiles" will mean "multiple tiles displayed in an orderly fashion."
The terms "partitioning a visual display of the device into an array of tiles" and "arrange a display into an array of tiles" will both mean "dividing/divide some or all of a display into an array of tiles."
The term "user-defined array size" will mean "the number and arrangement of the positions in the array as specified by the user."
The term "refresh rate" will mean "the criteria for updating information displayed in a tile."
The term "retrieval rate" will mean "the criteria for retrieving information."
The term "updating information from a [first/second] information source in said plurality of information sources presented to said [first/second] tile" will mean "obtaining refreshed information from the information source and updating the information displayed in the tile."
The term "in accordance with said [first/second] [refresh/retrieval] rate" will be given its plain and ordinary meaning, qualified by the Court's construction of "refresh rate" and "retrieval rate."
The term "simultaneously" in claim 1 will mean that "the ongoing processes of updating information presented in the two tiles occur at the same time, even if the updating of the two tiles occurs at different times."
The term "simultaneous communications" in claims 22 and 46 will mean that "the ongoing processes of communicating with multiple information sources occur at the same time, even if the communications with the information sources occur at different rates."
The term "[refresh/retrieval] rate is assigned automatically" will mean "[refresh/retrieval] rate is assigned without direct user input indicating the rate."
The term "grid" will mean "a regular arrangement of rows and columns, which may, but need not, allow a single tile to occupy more than one row and/or column."
The term "wherein each of said tiles occupies a fixed position on said grid" will be given its plain and ordinary meaning, qualified by the Court's construction of "tile(s)" and "grid."
The Court DENIES SurfCast's Motion to Stay Pending Inter Partes Review (ECF No. 150).
SO ORDERED.
