UNITED STATES DISTRICT COURT EASTERN DISTRICT OF VIRGINIA NORFOLK DIVISION I/P ENGINE, INC. Plaintiff, Civil Action No. 2:11-cv-512 v. AOL, INC., et al., Defendants. DEFENDANTS� OPENING CLAIM CONSTRUCTION BRIEF Introduction Plaintiff I/P Engine, Inc. (�Plaintiff�) seeks to use claim construction as an opportunity to redefine � and inappropriately broaden � the scope of its patents. For example, even though the Patents concern Internet search using a combination of content filtering with collaborative filtering � which the parties agree requires feedback from other users with similar interests or needs � Plaintiff seeks to eliminate this collaborative requirement from many of the claims. Further, even though the patents recite a particular �scanning� process that requires crawling link to link for information, Plaintiff conflates scanning with a generic search process that is appropriately referred to as �searching� elsewhere in the Patents. And even though each asserted system claim recites three separate systems that perform disparate tasks, Plaintiff argues that these three systems may somehow be one and the same. Plaintiff�s attempt to redefine and broaden its patents should be rejected, and Defendants� proposed constructions should be adopted. Factual Background I. OVERVIEW OF THE ASSERTED PATENTS Plaintiff alleges infringement of U.S. Patent No. 6,314,420 (�the �420 Patent�) and U.S. Patent No. 6,775,664 (�the �664 Patent�). Both patents are directed to the concept of filtering search results based partly on collaborative feedback, as expressly stated in their titles. The �420 Patent is entitled �Collaborative/Adaptive Search Engine.� The �664 Patent is entitled �Information Filter System and Method for Integrated Content-Based and Collaborative/Adaptive Feedback Queries.� (emphases added.) Plaintiff asserts infringement of claims 10, 14, 15, 25, 27, and 28 from the �420 Patent and claims 1, 5, 6, 21, 22, 26, 28, and 38 from the �664 Patent. The two Patents claim priority to the same parent application and share a common specification. The asserted claims involve �scanning� (i.e., crawling) a network to retrieve information responsive to a user query, and then filtering the retrieved information through both content-based filters and collaborative filters. For example, �420 claim 10 recites: A search engine system comprising: [a] a system for scanning a network to make a demand search for informons relevant to a query from an individual user; [b] a content-based filter system for receiving the informons from the scanning system and for filtering the informons on the basis of applicable content profile data for relevance to the query; and [c] a feedback system for receiving collaborative feedback data from system users relative to informons considered by such users; [d] the filter system combining pertaining feedback data from the feedback system with the content profile data in filtering each informon for relevance to the query.1 1 Throughout this brief, Defendants have added bracketed letters denoting the various claim steps or elements, for the Court's convenience. 2 �420 Claim 25 is substantially similar, but is a method claim. Similar to the �420 claims, claim 1 of the �664 Patent recites: A search system comprising: [a] a scanning system for searching for information relevant to a query associated with a first user in a plurality of users; [b] a feedback system for receiving information found to be relevant to the query by other users; [c] a content-based filter system for combining the information from the feedback system with the information from the scanning system and for filtering the combined information for relevance to at least one of the query and the first user. �664 Claim 26 is substantially similar, but is a method claim. II. THE SPECIFICATION RECITES USING CONTENT-BASED AND COLLABORATIVE FILTERING FOR INTERNET SEARCHING As the Asserted Patents� specification acknowledges, search engines were well-known in the art. (See 1:20-26.)2 The patents disclose a purported improvement to known search engines by using two types of filtering also known in the art � content-based filtering and collaborative filtering � to filter search results. (See 2:20-27.) For example, the Background of Invention states, �The present invention is directed to an information processing system . . . with collaborative feedback data and content-based data and adaptive filter structuring, being used in filtering operations to produce significantly improved results.� (2:20-27.) Elsewhere, the specification states that �[t]he present invention combines collaborative filtering with contentbased filtering in measuring informons [information entities] for relevancy.� (23:39-41.) The 2 While the Asserted Patents� specifications are substantively identical, their line numbers are slightly different � e.g., a passage that appears at column 2, line 1 of the '420 Patent may appear at column 2, line 9 of the '664 Patent. Unless otherwise noted, all specification citations are taken from the '420 Patent. 3 patentees call their invention �CASE,� for �collaborative, adaptive search engine,� emphasizing that the addition of collaborative filtering was the heart of the disclosed invention. (23:30-32.) The specification also explains that the preliminary step of �scanning� a network, which is used to gather candidate search results for filtering, entails the use of a spider system that crawls the network. (See 25:39-40 (�a spider system scans a network to find informons for a current demand search.�))3 The specification notes that �scanning� is a process generally used by search engines in the prior art. (See 1:23-24.) III. THE ACCUSED TECHNOLOGY Plaintiff accuses several Google services of infringing the Asserted Patents, including Google AdWords and AdSense for Search. Plaintiff accuses the other Defendants to the extent they use AdWords or AdSense for Search. Plaintiff also separately accuses Defendant AOL�s Sponsored Listings service and Defendant IAC Search & Media�s Ask Sponsored Listings service. In an attempt to ensnare these services, Plaintiff improperly tries to twist and broaden the claim language. For instance, even though the specification repeatedly emphasizes that the claims require collaborative feedback data � i.e., data from other users with similar interests and needs � Plaintiff tries to read this requirement out of at least the �664 Patent. Plaintiff presumably adopts this strategy because none of the accused services utilize feedback data from users with similar interests and needs. For instance, each accused advertising service employs �click-through rate� data as part of the process of ranking candidate ads that may be shown to 3 Much of the specification (and many of the patent claims) discusses the use of a �wire,� which the patents describe as a query submitted continuously that is stored and updated over time. (1:57-58.) During prosecution, the Examiner relied on the wire as the point of novelty that rendered all �420 claims patentable, even though some of the �420 claims, including those asserted here, omit the wire element altogether. (Sohn Decl., Ex. H at 2.) None of the �664 claims contain the wire element. 4 users. Plaintiff alleges that this click-through rate meets the collaborative feedback limitations from the claims. However, the accused clickthrough rate data gives no regard to whether these past users are similar to each other or whether they are similar to any current user of the advertising service. Further, even though the claims require a particular �scanning� process that requires an item-by-item web crawl, Plaintiff construes scanning merely as �searching� to get around the fact that the accused Internet advertising systems do not perform web crawling. Finally, even though each asserted system claim recites three different systems, each of which performs a different task in the information retrieval and filtering process, Plaintiff argues that these three systems may all be one and the same system � because Plaintiff cannot point to three different systems in Defendants� accused services. Argument I. COLLABORATIVE FEEDBACK AND RELATED TERMS Term/Phrase Defendants� Construction Plaintiff�s Construction �collaborative feedback data� (�420 Claim 10, 25) data from users with similar interests or needs regarding what informons such users found to be relevant information concerning what informons other users with similar interests or needs found to be relevant �[feedback system for] receiving information found to be relevant to the query by other users� (�664 Claim 1, 26) [System using a process of filtering information by] determining what information other users with similar interests or needs found to be relevant [feedback system for] receiving information concerning what other users found to be relevant to the query A. Defendants� construction of �collaborative feedback data� properly recognizes that �data� must come from similar users The parties� dispute as to �collaborative feedback data� is fairly narrow. The parties agree that this term refers to what other users with similar interests or needs found to be relevant. 5 They disagree, however, over the precise formulation of the construction. Defendants� construction recites that collaborative feedback data is �data� that comes from �users with similar interests or needs regarding what informons such users found to be relevant.� Plaintiff�s construction, however, needlessly replaces �data� with �information� and eliminates the requirement that this data comes from other users with similar interests or needs � instead requiring only that it �concern� (whatever that means) such users. 1. Defendants� construction is supported by the intrinsic evidence. Defendants� construction appropriately makes clear that the collaborative feedback data must come from the other users with similar interests or needs. For example, Claim 25[c] recites �receiving collaborative feedback data from system users relative to informons considered by such users.� Thus, according to the plain language of the claims, the collaborative feedback data must come from these users. See Phillips v. AWH Corp., 415 F.3d 1303, 1312 (Fed. Cir. 2005) (en banc) (holding that claim construction begins with the words of the claims themselves). The specification also explains that �[c]ollaborative filtering employs additional data from other users� (24:37-38) and that �making effective use of collaborative input (CI) from other users U is a difficult problem� that the Patents had to allegedly solve. (16:32-33) (emphases added). Thus, like the claims, the specification supports Defendants� position that �collaborative feedback data� must come from users with similar interests or needs. See Phillips, 415 F.3d at 1315 (�the specification �is always highly relevant to the claim construction analysis. Usually, it is dispositive; it is the single best guide to the meaning of a disputed term.��) Plaintiff�s construction, however, merely provides that the collaborative feedback data �concern� what informons other users with similar interests or needs found to be relevant. This makes it unclear whether the collaborative feedback data must come from the other users as required by the claims and specification. 6 2. Plaintiff�s construction improperly swaps �data� for �information.� Plaintiff�s construction also redefines �data� as �information.� But this merely swaps one commonplace word for another, which adds no clarity to the claim and is thus improper. See Static Control Components, Inc. v. Lexmark Int�l, Inc., 502 F. Supp. 2d 568, 576 (E.D. Ky. 2007) (�simply swapping words with synonyms is not construction.�) By contrast, Defendants� construction sensibly recognizes that the word �data� needs no further interpretation. B. Defendants� construction of �[feedback system for] receiving information found to be relevant to the query by other users� includes the collaborative element that is necessary to the claimed invention �664 claim 1 requires a �feedback system for receiving information found to be relevant to the query by other users,� while �664 claim 26 requires �receiving information found to be relevant to the query by other users.� The parties dispute as to these terms is whether the other users have �similar interests or needs� as the intrinsic evidence requires. 1. Defendants� construction is supported by the intrinsic evidence. There is no dispute that �collaborative filtering� and �collaborative feedback data� are defined by reference to other users with similar interests or needs. As noted above, the parties agree that �collaborative feedback data� requires data or information about what informons other users with similar interests or needs found to be relevant. This point is critical, because the patentees repeatedly explain that the claimed invention requires �collaborative feedback data� and �collaborative filtering.� Thus, the �feedback system for receiving information found to be relevant to the query by other users� in claim 1 must be construed as a feedback system for receiving collaborative information � i.e., information that other users with similar interests or needs found to be relevant, and the step of �receiving information found to be relevant to the query by other users� in claim 26 (a recast method claim of claim 1) must be interpreted as 7 requiring collaborative information. Only with these constructions would claims 1 and 26 contain the collaborative element of the claimed invention required by the intrinsic evidence. For example, the final sentence of the specification�s �Background of the Invention� section, which distinguishes the claimed invention from the prior art, explains that �the present invention is directed to an information processing system . . . with collaborative feedback data and content-based data . . .� (Id. at 2:28-33) (emphasis added). Elsewhere, the specification states: �The present invention combines collaborative filtering with content-based filtering in measuring informons for relevancy, and further preferably applies adaptive updating of the content-based filtering operation.� (Id. at 23:41-44 (emphasis added)). These general statements of �the present invention� are appropriately used to define and limit the claims. See Am. Piledriving Equip., Inc. v. Geoquip, Inc., 637 F.3d 1324, 1334 (Fed. Cir. 2011) (�a statement in a specification that describes the invention as a whole can support a limiting construction of a claim term�); Honeywell Int�l Inc. v. ITT Indus., Inc., 452 F.3d 1312, 1318-19 (Fed. Cir. 2006) (finding that use of �this invention� and �the present invention� established that the patentee intended to limit the meaning of a term to that disclosed in the specification); C.R. Bard, Inc. v. U.S. Surgical Corp., 388 F.3d 858, 864 (Fed. Cir. 2004) (construing a claimed plug as requiring pleats because �[i]n two places, the patent describes in general terms what it deems to be the invention. In both places, the patent unequivocally defines the claimed plug as having pleats.�) Likewise, the Abstract states that �[a] user feedback system provides collaborative feedback data for integration with content profile data in the operation of the collaborative/content-based filter.� �The abstract is not an explanation of the preferred embodiment, which cannot limit the claims, but a summary of the invention itself on which the claims are based.� Al-Site Corp. v. Cable Car Sunglasses, 911 F. Supp. 410, 415 (N.D. Cal. 8 1994) (emphasis added). Thus, defining the claims to conform with the Abstract�s description of the invention is appropriate as well. See Hill-Rom Co., Inc. v. Kinetic Concepts, Inc., 209 F.3d 1337, 1341 fn * (Fed. Cir. 2000) (�We have frequently looked to the abstract to determine the scope of the invention�); see also Genzyme Corp. v. Transkaryotic Therapies, Inc., 346 F.3d 1094, 1099 (Fed. Cir. 2003) (finding that statements in a patent�s Abstract and Summary of Invention limited the scope of a phrase in the claims to a specific technique). It is also telling that the specification states �the present invention combines collaborative filtering with content-based filtering in measuring informons for relevancy, and further preferably applies adaptive updating of the content-based filtering operation.� (id. at 23:41-44 (emphasis added)). While adaptive updating is merely a �preferable� feature of the claimed invention, collaborative filtering has no such qualifying language. Rather, collaborative filtering is a required feature of the claimed invention. Because the intrinsic evidence shows that the claimed �present invention� requires collaborative feedback data, the step of �receiving information found to be relevant to the query by other users� in �664 claim 26 is properly limited to receiving collaborative information � i.e., information that other users with similar interests or needs found to be relevant. Likewise, the �feedback system for receiving information found to be relevant to the query by other users� in �664 claim 1 must be construed as a feedback system for receiving collaborative information � information that other users with similar interests or needs found to be relevant. 2. Plaintiff�s construction ignores the relevant intrinsic evidence. Plaintiff�s construction, �receiving information concerning what other users found to be relevant to the query,� merely shuffles the words in the claim. See Wi-Lan, Inc. v. Acer, Inc., 712 F. Supp. 2d 549, 574 (E.D. Tex. 2010) (rejecting proposed construction that �merely rearrange[s] the words of the term to be construed.�) And it makes no reference to other users 9 with similar interests or needs as necessary to capture the collaborative filtering required by the �664 Patent. Notably, Plaintiff has never provided any intrinsic evidence support for its construction.4 Given the specification�s repeated statements that collaborative feedback and filtering is an essential element of the claimed invention, Plaintiff�s attempt to read this element out of the claims is improper and Plaintiff�s construction should be rejected. Further, Plaintiff does not even propose a construction for the �feedback system� portion of the term �feedback system for receiving information found to be relevant to the query by other users;� it just repeats the phrase. This too shows that Plaintiff's construction should be rejected. II. �SCANNING� TERMS Term/Phrase Defendants� Construction Plaintiff�s Construction �scan[ning] a network� (�420 Claims 10, 25) Spider[ing] or crawl[ing] a network looking for items in a network �a scanning system� (�664 Claim 1) a system used to scan a network a system used to search for information Defendants construe �scanning a network� as �spidering or crawling a network,� in accordance with the specification�s definition of this term, and construe �a scanning system� in the �664 Patent as �a system used to scan a network,� in accordance with the meaning of that same term in the �420 Patent.5 By contrast, Plaintiff improperly equates �scanning� with the more generic term �searching.� 4 As explained at length in Defendants� Opposition to Plaintiff�s Motion to Compel Compliance with the Court�s Scheduling Order, Plaintiff�s failure to provide intrinsic evidence � and, in many cases, constructions themselves � has been consistent throughout the claim construction process. (See Dkt. 109 at 5-6.) 5 At Plaintiff's request, Defendants agreed to construe �scanning a network� and a �scanning system� rather than the larger phrases in which they appear, such as �scanning a network to make a demand search.� Defendants continue to maintain that the broader phrases are invalid as indefinite and/or for lack of enablement under 35 U.S.C. § 112. However, to narrow the issues at the claim construction hearing, Defendants have agreed raise such 10 A. Defendants� construction of �scanning a network� is compelled by the intrinsic evidence. Claims 10 and 25 of the �420 Patent recite �scanning a network to make a demand search.� The specification discloses that �scanning a network� to make a demand search involves spidering:6 �a spider system scans a network to find informons for a current demand search.� (25:39-40.) And in their discussion of the prior art, the patentees repeatedly link �scanning� to the operation of a spider. (See 1:23-24 (�Thereafter, the search site typically employs a �spider� scanning system . . .�)) The parent application to the �420 Patent also defines �scanning a network� to require the operation of a spider. (See 1:45-61 (�In the patent application which is parent to this continuation-in-part application . . . A continuously operating �spider� scans the network.�)). Because every instance of �scanning� in the specification is tied to the operation of a spider, the term �scanning� is properly construed as requiring such a spider. Kinetic Concepts, Inc. v. Blue Sky Med. Group, Inc., 554 F.3d 1010, 1018-19 (Fed. Cir. 2009) (construing the term �wound� as limited to skin wounds because �[a]ll the examples in the specification involve skin wounds�). Thus, �scanning a network� should be construed as �spidering or crawling a network� � with �crawling� being used a synonym for �spidering� given that spiders operate by crawling. This meaning is consistent with the relevant extrinsic evidence as well, which makes clear that �spiders� do indeed operate through crawling and can be considered �crawlers.� See, e.g., Ira S. Nathanson, �Internet Infoglut and Invisible Ink: Spamdexing Search Engines with indefiniteness and enablement issues apart from claim construction of portions of these phrases, which will not resolve the 112 issues. Halliburton Energy Servs., Inc. v. M-I LLC, 514 F.3d 1244, 1251 (Fed. Cir. 2008) (�Even if a claim term�s definition can be reduced to words, the claim is still indefinite if a person of ordinary skill in the art cannot translate the definition into meaningfully precise claim scope.�). 6 As explained below, �spidering� connotes a process of crawling from link to link in a network like the Internet. 11 Metatags,� 12 HARV. J. L & TECH. 43, 61 (1998) (�A spider (also known as a robot, crawler or indexer) is a program that scans the Web, crawling from link to link�); MICROSOFT COMPUTER DICTIONARY (5th Ed. 2002) at 493 (using �crawler� as a synonym for �spider�). Even the extrinsic evidence that Plaintiff cited along with its Preliminary Constructions supports this conclusion, since Plaintiff�s evidence defines �scanning� to require a sequential, item-by-item crawl. (See Preliminary Constructions (Sohn Decl., Ex. C) at 6 (�Scan -- . . . Computer Technology. 1. to examine sequentially each item in a list, each record in a file . . .�)) B. �A scanning system� is used to scan a network. �A scanning system� in �664 claim 1 should be construed simply as �a system used to scan a network,� with �scan a network� defined as �spider or crawl a network� as explained above. By way of illustration, claim 25 of the �420 Patent recites a first step of �scanning a network to make a demand search for informons� and a second step of �receiving the informons . . . from the scanning system.� Thus, in the �420 Patent, the �scanning system� is used to scan a network (for informons). The same conclusion applies to the �scanning system� of �664 claim 1, since there is no suggestion that the �scanning systems� of the �420 and �664 Patents differ in any way. See NTP, Inc. v. Research In Motion, Ltd., 418 F.3d 1282, 1293 (Fed. Cir. 2005) (�Because NTP�s patents all derive from the same parent application and share many common terms, we must interpret the claims consistently across all asserted patents.�) C. Plaintiff improperly equates �scanning� with �searching� Plaintiff seeks to construe scanning a network as �looking for items in a network.� Plaintiff�s overbroad construction is unsupported by the specification, which repeatedly ties �scanning� to the operation of a web crawler or spider system. Ironically, the improper nature of Plaintiff's construction is shown by the only intrinsic evidence Plaintiff points to in purported 12 support of its construction. (Preliminary Constructions at 6 (quoting 25:39-40 (�spider system 46C scans a network 44C to find informons for a demand search.�) (emphasis added))). Instead, Plaintiff�s construction of �scanning� as �looking for items� equates �scanning� a network with �searching� a network. The everyday meaning of �searching� is �looking for items.� See RANDOM HOUSE UNABRIDGED DICTIONARY (2nd Ed. 1993) at 1727 (search: 4. �to look at, read, or examine (a record, writing, collection, repository, etc.) for information�). Similarly, Plaintiff has construed �a scanning system� as �a system used to search for information.� The claims, however, make clear that �scanning� is not the same as �searching.� For example, �664 claim 26 requires �searching for information relevant to a query associated with a first user in a plurality of users.� Dependant claim 38 recites �[t]he method of claim 26 wherein the searching step comprises scanning a network . . .� The fact that the patentees used �searching� in claim 26 and �scanning� in claim 38 implies that these terms are different from each other, not synonyms as Plaintiff�s construction would provide. Moreover, claim 38 depends from claim 26 and recites that claim 26�s searching step comprises scanning, which implies that �scanning� is a narrower subset of �searching� rather than a synonym. See Enzo Biochem., Inc. v. Applera Corp., 599 F.3d 1325, 1334 (Fed. Cir. 2010) (�dependant claims are presumed to be of narrower scope than the independent claims from which they depend�); see also Sta-Rite Industries, LLC v. ITT Corp., 682 F. Supp. 2d 738, 743 (E.D. Tex. 2010) (�Courts presume a difference in meaning and scope when a patentee uses different phrases in separate claims.�). 13 III. �COMBINING� (�420 CLAIMS 10, 25; �664 CLAIMS 1, 26) Defendants� Construction Plaintiff�s Construction Plain meaning; alternatively, bringing together uniting into a single number or expression The parties dispute whether �combining� should have its plain and ordinary meaning, as Defendants assert, or whether it should be inappropriately limited to Plaintiff�s incorrect interpretation of one of the embodiments in the specification. A. Defendants� construction accords with the intrinsic and extrinsic evidence. When the asserted claims invoke the language of �combining,� they simply refer to various pieces of information being somehow brought together. For example, claim 26 of the �664 Patent requites �combining the information found to be relevant to the query by other users with the searched information.� Nothing in the claim language specifies how that combination occurs; it merely requires that �information found to be relevant to the query by other users� � i.e., the collaborative feedback data � be combined, in some way, with �the searched information.�7 Because there is no limitation on how this �combining� must occur, the term �combining� should be given its broad plain meaning of �bringing together.� See, e.g., RANDOM HOUSE WEBSTER�S COLLEGE DICTIONARY (1999) (combine: �to bring into or join in a close union or whole.�). 7 Defendants had originally proposed that terms such as �combining the information from the feedback system with the information from the scanning system� are indefinite. At Plaintiff's request, Defendants agreed to construe just the word �combining.� As with the longer phrases containing �scanning a network� or �a scanning system,� Defendants continue to maintain that the broader phrases containing �combining� or �combined� are invalid as indefinite and/or for lack of enablement under 35 U.S.C. § 112. Here, too, Defendants have agreed to raise such indefiniteness and enablement issues apart from claim construction, to narrow the issues for the Claim Construction Hearing. 14 The specification supports this plain meaning. The preferred embodiment indicates that the claimed combination occurs through a sequential application of filters. After a �query processor� receives a query from an individual user (25:9-10), it determines whether it should execute a �wire search� or a �demand search.� (25:10-18; see also Fig. 8.) A content-based filter structure performs either the wire search or demand search (25:27-30), then passes the resultant �demand search informons or wire search informons��i.e., the �searched information��to a �search return processor.� (25:53-56.) The search return processor has its own informon rating system that combines the already content-based filtered data with �collaborative feedback rating data.� (25:57-61.) The search results are then returned to the user. (26:1-7.) Thus, the specification teaches that �combining� content-based and collaborative filters involves bringing these two types of filters together in some manner and using them to sequentially filter information. B. Plaintiff�s construction of �combining� is contrary to the plain meaning and the intrinsic evidence. 1. Plaintiff�s �uniting into a single number or expression� construction is contrary to the claims Plaintiff instead proposes a construction of combining in a mathematical context, such that it must involve uniting �into a single number or expression.� Yet, the claim language does not support such a crabbed and narrow construction. For example, �664 claim 1 requires: [a] �a scanning system for searching for information relevant to a query�; [b] �a feedback system for receiving information found to be relevant to the query by other users�; and [c] �a content-based filter system for combining the information found to be relevant to the query by other users with the searched information and for filtering the combined information.� It is difficult to see how the disparate pieces of information from Steps [a] and [b] could be �united into a single expression,� much less �united into a single 15 number.� Rather, �combining� these pieces of information more naturally means bringing them together in some way (albeit unspecified), such as in a list of search results. Moreover, Step [c] requires the �combined information� be filtered. Plaintiff�s construction would thus require that a �single number of expression� (the output of the �combining� step under Plaintiff�s construction) be �filtered.� Yet there is no disclosure in the Patents of �numbers� or �expressions� being filtered, nor would it make sense to do so. Finally, dependent claim 5 recites �[t]he search system of claim 1 wherein the filtered information is an advertisement.� Since the �filtered information� is also the �combined information� under claim 1[c], claim 5 essentially requires that an advertisement be the �combined information.� Under Plaintiff�s construction, an advertisement would be what is �united into a single number or expression.� But this makes no sense, as an advertisement is not �a single number or expression.� Thus, Plaintiff�s position that �combining� is limited to �uniting into a single number or expression� is demonstrably incorrect. 2. Plaintiff�s construction reads out a preferred embodiment. As detailed above, a preferred embodiment first uses �content-based filtering� and then passes the filtered informons to a �search return processor� that performs the requisite collaborative filtering. The specification � none of which actually uses the words in Plaintiff's construction � describes that the �search return processor receives [] informons passed by the content-based filter structure� (25:53-55), i.e. the web pages themselves. There is no disclosure of the �search return processor� receiving the content-based filtering data employed by the content-based filter structure. Plaintiff�s requirement that the �combining� be performed by �uniting into a single number or expression� rather than by using two filters sequentially would thus exclude the preferred embodiment � a result the Federal Circuit has cautioned is �rarely, if 16 ever, correct.� Globetrotter Software, Inc. v. Elan Computer Group, Inc.. 362 F.3d 1367, 1381 (Fed. Cir. 2004) (citation omitted). Plaintiff�s remaining intrinsic evidence is similarly unavailing. Plaintiff cites Fig. 6 of the Patents to support its �combining� construction (Preliminary Constructions at 7). But the embodiment in Fig. 6 is simply a general rating system with an option of incorporating contentbased data and collaborative feedback data, depending on the system in which it is used. It is not required to do so. For example, the content-based filtering used to perform the wire search or demand search similarly uses the system described by Fig. 6. (25:27-30.) Yet at this point in its execution, the search engine embodiment of the specification has no collaborative feedback data to incorporate. (Cf. 25:53-61.) Accordingly, that the Fig. 6 embodiment has the ability to use a certain type of data does not mean that it must use that type of data�particularly when there is no evidence that Fig. 6 has access to that type of data. IV. THE SEPARATENESS OF THE CLAIMED SYSTEMS Defendants� Construction Plaintiff�s Construction The claimed �system for scanning a network,� �content-based filter system,� and �feedback system� of �420 Claim 10 must be different systems The claim language does not require the claimed system for scanning, content-based filter system, and feedback system of claim 10 of the �420 patent to be the same or different �systems� The claimed �scanning system,� �feedback system,� and �content-based filter system� of �664 Claim 1 must be different systems The claim language does not require the scanning system, content-based filter system, and feedback system of claim 1 of the �664 patent to be the same or different �systems� The parties dispute whether the claimed scanning systems, content-based filter systems, and feedback systems must be separate and distinct from the other systems recited in the claims (Defendants), or whether each of these disparately claimed systems can be the same system 17 (Plaintiff). This issue is important because Plaintiff�s infringement allegations focus on a single system, not the disparate systems recited in the claims. A. Defendants� construction is compelled by the claims The claims recite the separate nature of the claimed systems; each system is given a unique task. In �664 claim 1, the scanning system is tasked with �searching for information relevant to a query,� the feedback system is tasked with �receiving information found to be relevant to the query by other users,� and the content-based filter system is tasked with �combining the information from the feedback system with the information from the scanning system and [] filtering the combined information for relevance to at least one of the query and the first user.� �420 claim 10 recites a similar division of labor for the claimed systems. If the patentees intended that a single system could handle all these tasks, then there would be no need to recite three separate systems, all of which bear different names and are given different tasks. Rather, the patentees could have recited a single, unitary system that performed all these tasks. They did not. Thus, these systems are separate and different from each other, not one system. See Bristol Co. Ltd. P�ship v. Bosch Rexroth Inc., 684 F. Supp. 2d 1245, 1294 (D. Colo. 2010) (ruling that two claimed �signals� must be separate from each other given that they serve different functions in the claims). B. Plaintiff provides no support for its constructions Plaintiff has no support for its position that �the claim language does not require the scanning system, content-based filter system, and feedback system . . . to be the same or different �systems.�� Plaintiff has never provided any basis for why each of these systems are claimed separately if they could be the same system. Plaintiff�s position is also inconsistent with its own validity contentions. While Plaintiff now says the claims are agnostic as to whether the scanning system, content-based filter system, 18 and feedback system must be the same or different systems, Plaintiff stated in its validity contentions that these systems cannot be different systems. (See, e.g., Sohn Decl., Ex. D (Plaintiff�s Responses and Objections to Google�s Third Set of Interrogatories) at 11 (�Lashkari does not �combine� the features that Defendants allege meet each of the two components [scanning system and feedback system] because the cited features are parts of different systems.�) (emphasis added)). In a meet-and-confer of April 4, 2012, Defendants pointed out this inconsistency in Plaintiff�s position and asked for an explanation. Plaintiff did not provide one. Plaintiff�s inconsistent interpretations should be rejected and Defendants� construction adopted. V. �DEMAND SEARCH� (�420 Claims 10, 25) Defendants� Construction search engine query Plaintiff�s Construction a one-time search performed upon a user request The dispute here is whether Defendants� simple construction of �search engine query� should be adopted over Plaintiff�s cumbersome and ambiguous construction of �a one-time search performed upon a user request.� A. Defendants� construction is supported by the specification and would be readily understandable by a jury The Patents make clear that a �demand search� is just a regular search engine query � i.e., the type of query that is performed by regular search engines such as those in the prior art. Specifically, the Patents explain that the claimed invention can operate in one of two modes: �demand search� or �wire search� mode. (See 25:19-20; 25:27-28.) Unlike �wire� searches, which are continuous searches that rely on a stored query, �demand� searches are performed through the normal operation of regular search engines: �an integrated collaborative/content19 based filter is operated to provide ongoing or continuous searching for selected user queries, with a �wire� being established for each query. On the other hand, a regular search engine is operated to make immediate or short-term �demand� searches for other user queries.� (23:45-51) (emphasis added.) The specification elsewhere explains that �[w]hen a user makes a query for which a wire already exists, wire search results are preferably returned instead of regular search engine results . . . Otherwise, block 28C commands a demand search by a regular query engine.� (23:65-24:8 (emphasis added).) As shown by these passages, the Patents repeatedly state that a �demand search� is a normal search engine query made with a regular search engine (as opposed to the specialized, continuous queries that require a �wire�). Because every example of a �demand search� in the specification is a normal search query made by a normal search engine, Defendants� simple construction of �demand search� as �search engine query� accurately captures what the patentees meant by this term. See Kinetic Concepts, 554 F.3d at 1018-19 (construing �wound� as limited to skin wounds because �[a]ll the examples in the specification involve skin wounds.�) Moreover, �search engine query� is a simple construction that would be readily understood by a jury � and the goal of claim construction is to reduce claim language to simple language that a jury could understand. See, e.g., Curtiss-Wright Flow Control Corp. v. Z & J Tech. GmbH, 563 F. Supp. 2d 1109, 1115 (C.D. Cal. 2007) (�Claim construction is intended to clarify the legal meaning of claim language in the context of a particular case so that a jury can make a simple factual determination as to infringement or invalidity.�) B. Plaintiff�s construction would be confusing and ambiguous for a jury to apply Plaintiff�s construction of �demand search� � �a one-time search performed upon a user request� � is confusing and ambiguous. For example, if a demand search for �cars� was issued 20 by two different users at two different times, do these searches cease being �demand searches� because they are no longer �one-time�? If a single user performs two demand searches, do these searches cease being �demand searches� under the same logic? Plaintiff�s construction provides no answers to these questions and no clarity to a jury. It should be rejected. Plaintiff seeks to support its construction with the following sentence of the Abstract: �[t]he search engine system employs a regular search engine to make one-shot or demand searches for information entities which provide at least threshold matches to user queries.� (Preliminary Constructions at 6-7.) Yet Plaintiff�s construction of �demand search� has virtually no words in common with this Abstract sentence. The Abstract sentence does not include �performed upon a user request,� as does Plaintiff�s construction, and it includes the concept of threshold matching, which Plaintiff�s construction does not. If anything, this sentence supports Defendants� construction by emphasizing, as does the remainder of the specification, that a demand search is made by a �regular search engine� and thus should be construed simply as a �search engine query.� Moreover, while Plaintiff may argue that the Abstract�s description of a demand search as �one-shot� requires Plaintiff�s use of the term �one-time,� this argument would be misplaced. The Abstract uses the term �one-shot� simply to distinguish �demand searches� from the continuous �wire searches.� (See id. (�The search engine system employs a regular search engine to make one-shot or demand searches . . . The search engine system also employs a collaborative/content-based filter to make continuing searches for information entities which match existing wire queries.�)) Because none of the wire claims are at issue in this litigation, there is no need to distinguish demand searches from continuing wire searches by using the 21 adjective �one-shot� or �one-time� to describe the demand searches, or anything else. Rather, for the reasons explained above, such language would only serve to confuse the jury. VI. �INDIVIDUAL USER� / �FIRST USER� Defendants� Construction particular user Plaintiff�s Construction No further construction necessary beyond other terms The parties agree that �individual user� and �first user� mean the same thing. Defendants� construction of these terms comes from the intrinsic evidence. Plaintiff contends that no construction of these terms is necessary because the parties agreed to construe �user� as �an individual in communication with [a/the] network.� (Appendix A; Sohn Decl., Ex. E.) A. Defendants� constructions are faithful to the specification and claim language As for the construction of �first user,� the �664 claim step of �filtering the combined information for relevance to at least one of the query and the first user� demonstrates that the relevance determination is made as to the first user � i.e. the particular user. Because different users would find different items to be relevant, the relevance-filtering step makes sense only if the filtering is done for relevance to the particular user. As for the construction of �individual user,� the specification states that �[a] query processor receives queries from an individual user and other users.� (25:11-12; accord 26:1112.) By distinguishing the �individual user� from all the other users, the specification suggests that this particular user is different from the other users for purposes of the claimed invention. Defendants� construction properly acknowledges this �particularity� requirement. B. Plaintiff has declined to provide a construction for these terms As with most terms, Plaintiff initially provided no construction for these terms. Then, on April 5, Plaintiff proposed that �individual user� and �first user� both should be construed as 22 �a/the individual user.� (Sohn Decl., Ex. G.) On April 11, the day before Opening Briefs were due, Plaintiff said that no further construction was necessary for these terms because the parties had already agreed on a construction for �user.� (Sohn Decl., Ex. E.) But the parties have not agreed on a construction for �individual user� and �first user.� Specifically, Plaintiff has not confirmed an agreement that these terms refer to the �particular user� as to whom the claimed method is performed, or whether, as Plaintiff�s lack of agreement would seem to imply, the �individual user� and �first user� could be any user at all. As the proper construction of �individual user� and �first user� remains in dispute, the Court should resolve this dispute. See O2 Micro Int�l Ltd. v. Beyond Innovation Tech. Co., Ltd., 521 F.3d 1351, 1362-3 (Fed. Cir. 2008) (�When the parties present a fundamental dispute regarding the scope of a claim term, it is the court�s duty to resolve it.�) VII. RELEVANT/RELEVANCE TERMS Term/Phrase Defendants� Construction Plaintiff�s Construction �relevance to at least one of the query and the first user� (�664 Claims 1, 26) how well information satisfies the information need of at least one of the query and the first user No further construction necessary beyond other terms �[informons/information] relevant to a query� (�420 Claims 10, 25; �664 Claims 1, 26) [informons/information] that satisfy the individual user�s information need expressed in the query [informons/information] having relevance to the [individual/first] user�s information need in the query A. Defendants� constructions flow from the specification and the claims The specification states that �the �relevance� of a particular informon broadly describes how well it satisfies the user�s information need.� (4:5-6). In keeping with this definition, the parties agree that �relevance to a query� means �how well an informon satisfies the individual 23 user�s information need in the query.� (See Appendix A.) Defendants� constructions flow from this agreed definition in the specification. For example, Defendants� construction of �relevance to at least one of the query and the first user� takes this specification definition and applies it to the �query and first user� context. If �relevance� is defined as how well an informon satisfies the user�s information need, then �relevance to at least one of the query and the first user� must be �how well information satisfies the information need of at least one of the query and the first user.� Defendants� constructions for �informons relevant to a query� and �information relevant to a query� take the specification definition and apply it to the �query� context. If �relevance� is defined as how well an informon satisfies the user�s information need, then �informons relevant to a query� are informons that do in fact satisfy the individual user�s information need expressed in the query and �information relevant to a query� is information that does in fact satisfy the first user�s information need expressed in the query. Because Defendants� constructions take the specification�s definition of �relevance� and faithfully apply it to the various claim terms, Defendants� constructions should be adopted. B. See Phillips, 415 F.3d at 1315. Plaintiff�s constructions are grammatically nonsensical Plaintiff�s constructions of �informons relevant to a query� and �information relevant to a query� are grammatically nonsensical. Plaintiff construes these terms as �[informons/information] having relevance to the [individual/first] user�s information need in the query.� Again, however, the parties agree that �relevance to a query� should be construed as �how well an informon satisfies the individual user�s information need in the query.� Applying this agreed �relevance� construction to Plaintiff�s proposed constructions, the phrase �having relevance� in Plaintiff�s constructions would mean �having how well an informon satisfies . . .�, 24 which is grammatically nonsensical. Defendants pointed this out to Plaintiff, and Plaintiff had no response, merely stating that the parties would need to �agree to disagree.� Plaintiff has not offered any construction for �relevance to at least one of the query and the first user.� In an email of April 11, Plaintiff justified its refusal to provide a construction by stating that �the parties have already agreed to constructions for each of the components of this limitation.� (Sohn Decl., Ex. E). While it is true that the parties have agreed to a construction for �relevance to a query,� which has some overlap with the longer phrase �relevance to at least one of the query and the first user,� construing the shorter phrase does not eliminate the need to construe the latter. To the contrary: because the agreed construction of �relevance to a query� includes the words �query� and �user,� the jury might be confused about what additional meaning the additional phrase �relevance to at least one of the query and the first user� has, absent a construction for this phrase. Specifically, Defendants� construction clarifies that this phrase refers to how well the information satisfies the information need of either the query issued by the first user or the first user himself. VIII. ORDER OF STEPS (�420 CLAIM 25; �664 CLAIM 26) Defendants� Construction Plaintiff�s Construction �420 Claim 25: Step [a] must be performed before Step [b]; Steps [b] and [c] must be performed before Step [d] No construction necessary �664 Claim 26: Steps [a] and [b] must be performed before Step [c]; Step [c1] must be performed before Step [c2] 1. Defendants� constructions acknowledge the logical order of steps in the claims It is well-settled that method steps must be performed in the recited order �when the method steps implicitly require that they be performed in the order written,� such as when �each 25 subsequent step reference[s] something logically indicating the prior step has been performed.� Interactive Gift Exp., Inc. v. Compuserve Inc., 256 F.3d 1323, 1342 (Fed. Cir. 2001). (a) �420 claim 25 Under these principles, and as shown in the annotations below, Step [a] of �420 Claim 25 must be performed before Step [b] because Step [b] �references something logically indicating the prior step has been performed.� Specifically, Step [a] recites �scanning a network . . . for informons,� and Step [b] involves �receiving the informons in a content-based filter system from the scanning system.� Assuming that �the scanning system� is what �scans the network for informons,�8 it is impossible for the content-based filter system to receive informons from the scanning system (as required in Step [b]) until after the scanning system has scanned the network for these informons (as required in Step [a]). Accordingly, Step [a] must be performed before Step [b]. For similar reasons, Steps [b] and [c] must be performed before Step [d]. Step [b] requires filtering on the basis of content profile data, Step [c] requires receiving collaborative feedback data, and Step [d] requires �combining� the feedback data from Step [c] with the content profile data from Step [b]. Because Step [d] thus combines the output of Steps [b] and [c], Step [d] cannot be performed until after Steps [b] and [c] are completed. 8 See Section II(B), supra (explaining why �a scanning system� should be construed as "a system used to scan a network.") 26 (b) �664 claim 26 For �664 claim 26, as shown in the annotations below, Steps [a] and [b] must be performed before Step [c] because Step [c] references something logically indicating that Steps [a] and [b] have been performed. Specifically, Step [c] requires �combining the information found to be relevant to the query by other users with the searched information.� The �information found to be relevant to the query by other users� in Step [c] refers to the output of Step [b], since Step [b] involves �receiving information found to be relevant to the query by other users.� The �searched information� in Step [c] refers to the output of Step [a], since Step [a] requires �searching for information.� Thus, Step [c] requites �combining� the output of Step [a] and Step [a], meaning that Step [c] must be performed after Steps [a] and [b] are completed. For similar reasons, Step [c1] must be performed before Step [c2]. Step [c1] recites �combining the information . . .�, while Step [c2] recites �filtering the combined information.� Because the information must be combined before this combined information may be filtered, Step [c1] must be performed before Step [c2]. 2. Plaintiff improperly seeks to avoid an order of steps construction Plaintiff tacitly acknowledges that some steps from �420 claim 25 and �664 claim 26 must be performed in order. However, Plaintiff refuses to take a position on what these claim steps are, recently stating: �To the extent that some limitations presuppose that one of the earlier limitations be performed, that (limited) required ordering is clear from the claim language itself 27 and no construction is necessary.� (Sohn Decl., Ex. F.) Plaintiff argues that no order of steps construction is necessary � not because the claim steps may be performed in any order, but because order of steps �is an issue that will be addressed by the experts.� (Id.). Plaintiff�s apparent position that order of steps is a matter for expert discovery instead of claim construction is incorrect, as the Federal Circuit has held. Loral Fairchild Corp. v. Sony Corp., 181 F.3d 1313, 1322 (Fed. Cir. 1999) (deciding the requisite order of steps as a matter of claim construction); Interactive Gift, 256 F.3d at 1342 (same). Moreover, because Plaintiff has refused to agree to Defendants� proposed order of steps, or even say which steps must happen in order, the Court has a duty to resolve this dispute. See O2 Micro, 521 F.3d at 1362-3. Thus, the Court should reject Plaintiff�s attempt to avoid an order of steps construction and adopt Defendants� order of steps for �420 claim 25 and �664 claim 26. IX. ANTECEDENT BASIS TERMS Term/Phrase Defendants� Construction Plaintiff�s Construction �informons� / �the informons� (�420 Claims 10, 25) �informons� and �the informons� are the same informons �informons� provides antecedent basis for �the informons� �users� / �such users� (�420 Claims 10, 25) �users� and �such users� are the same users �users� provides antecedent basis for �such users� �a query� /�the query� (�420 Claims 10, 25; �664 Claims 1, 26) �a query� and �the query� are the same query �a query� provides antecedent basis for �the query� �a feedback system� / �the feedback system� (�420 Claim 10; �664 Claim 1) �a feedback system� and �the feedback system� are the same feedback system �a feedback system� provides antecedent basis for �the feedback system� �a scanning system� / �the scanning system� (�664 Claim 1) �a scanning system� and �the scanning system� are the same scanning system� �a scanning system� provides antecedent basis for �the scanning system� 28 �a first user� / �the first user� �a first user� and �the first user� �a first user� provides (�664 Claims 1, 26) are the same first user antecedent basis for �the first user� �a content-based filter system� / �the content-based filter system� (�664 Claims 1, 21) �a content-based filter system� and �the content-based filter system� are the same contentbased filter system �a content-based filter system� provides antecedent basis for �the content-based filter system� The seven term dyads recited above are instances in which a term is first introduced in a claim and then referred to with a definite article (�the� or �such�) later in the claim. For instance, �664 claim 1 introduces �a scanning system� and later refers to �the scanning system�; �420 claim 10 introduces �users� and later refers to �such users�; etc.9 Under the basic canon of �antecedent basis,� the second term in each dyad must be the same as the first term in the dyad. In other words, �a scanning system� in �664 claim 1 is the same scanning system as �the scanning system� recited later in the claim, �users� in �420 claim 10 are the same users as �the users� recited later in the claim, etc. SSL Servs., LLC v. Citrix Sys., Inc., 816 F. Supp. 2d 364, 388 (E.D. Tex. 2011) (�The use of the definite article �the� requires that �session key� have an antecedent basis, which is necessarily a session key referenced earlier in the claim steps�) (emphasis added); Zenith Elecs. Corp. v. Exzec, Inc., No. 93-5041, 1995 WL 275591, *5 (N.D. Ill. May 8, 1995) (�An indefinite article (�a� or �an�) is used to introduce an element. A definite article (�the� or �said�) is used when a term has already been introduced, thereby making mention of the earlier recitation of the element�) (emphasis added). Plaintiff does not dispute that terms such as �the scanning system� would be indefinite without a corresponding antecedent basis. See Halliburton Energy Servs., Inc. v. M-I LLC, 514 9 In the final dyad from this list (�a content-based filter system� / �the content-based filter system�), �a content-based filter system� is introduced in '664 claim 1 and �the contentbased filter system� appears in '664 claim 21 (which depends from claim 1). For all the other dyads, both terms in the dyad appear in the same claim. 29 F.3d 1244, 1249 (Fed. Cir. 2008) (explaining a claim could be indefinite �if a term does not have proper antecedent basis.�) Plaintiff thus asserts that, for example, �a scanning system� provides antecedent basis for �the scanning system� so as to preserve the validity of the claims. Yet Plaintiff steadfastly refuses to acknowledge the consequence of the antecedent basis canon � namely, that the second term in each dyad must be the same as the first. Plaintiff initially refused to provide any construction for these dyads at all. (Preliminary Constructions at 8). Plaintiff now states that the first item in each dyad �provides antecedent basis for� the second. Plaintiff thus would leave open the possibility that �the scanning system� in �664 claim 1 is a different scanning system than �a scanning system� recited earlier in the claim, etc. But the antecedent basis canon forecloses this possibility. See Tuna Processors, Inc. v. Hawaii Intern. Seafood, Inc., Nos. 2008-1410, 2008-1435, 2009 WL 1084197, at *6 (Fed. Cir. Apr. 23, 2009) (�[T]he introduction of a new element is accomplished through the use of an indefinite article, not through the use of a definite article.�); SSL Servs., 816 F. Supp. 2d at 388. Furthermore, Plaintiff�s construction would only serve to confuse a lay jury. It would provide no benefit to say, for example, �a first user provides the antecedent basis to the first user.� The jury will not know what to do with that. Only Defendants� constructions for these term dyads, which specify that the second term in each dyad must be the same as the first term in the dyad, would provide any useful meaning for the jury. Conclusion For the foregoing reasons, Defendants� constructions should be adopted. Dated: April 12, 2012 By: /s/ Stephen E. Noona Stephen E. Noona KAUFMAN & CANOLES, P.C. 150 West Main Street Post Office Box 3037 Norfolk, VA 23514 30 Telephone: (757) 624.3000 Facsimile: (757) 624.3169 By: /s/ David A. Perlson David A. Perlson QUINN EMANUEL URQUHART & SULLIVAN LLP 50 California Street, 22nd Floor San Francisco, CA 94111 Telephone: (415) 875-6600 Facsimile: (415) 875-6700 Counsel for Defendants GOOGLE INC., IAC SEARCH & MEDIA, INC., TARGET CORP., and GANNETT COMPANY, INC. By: /s/ Stephen E. Noona Stephen E. Noona KAUFMAN & CANOLES, P.C. 150 West Main Street Post Office Box 3037 Norfolk, VA 23514 Telephone: (757) 624.3000 Facsimile: (757) 624.3169 Robert L. Burns FINNEGAN, HENDERSON, FARABOW, GARRETT & DUNNER, LLP Two Freedom Square 11955 Freedom Drive Reston, VA 20190 Telephone: (571) 203-2700 Facsimile: (202) 408-4400 Cortney S. Alexander FINNEGAN, HENDERSON, FARABOW, GARRETT & DUNNER, LLP 3500 SunTrust Plaza 303 Peachtree Street, NE Atlanta, GA 94111 Telephone: (404) 653-6400 Facsimile: (415) 653-6444 Counsel for Defendant AOL, INC. 31 CERTIFICATE OF SERVICE I hereby certify that on April 12, 2012, I will electronically file the foregoing with the Clerk of Court using the CM/ECF system, which will send a notification of such filing (NEF) to the following: Jeffrey K. Sherwood Kenneth W. Brothers DICKSTEIN SHAPIRO LLP 1825 Eye Street NW Washington, DC 20006 Telephone: (202) 420-2200 Facsimile: (202) 420-2201 sherwoodj@dicksteinshapiro.com brothersk@dicksteinshapiro.com Counsel for Plaintiff, I/P Engine, Inc. Stephen E. Noona Virginia State Bar No. 25367 KAUFMAN & CANOLES, P.C. 150 W. Main Street, Suite 2100 Norfolk, VA 23510-1665 Telephone: (757) 624-3239 Facsimile: (757) 624-3169 senoona@kaufcan.com Counsel for AOL Inc., Google, Inc., Gannett Co., Inc., Target Corporation and IAC Search & Media, Inc. /s/ Stephen E. Noona Stephen E. Noona Virginia State Bar No. 25367 KAUFMAN & CANOLES, P.C. 150 West Main Street, Suite 2100 Norfolk, VA 23510 Telephone: (757) 624-3000 Facsimile: (757) 624-3169 senoona@kaufcan.com 11649500_1.DOC 32