Tech. Innovations, LLC v. Nstein Techs., Inc.

Full title: TECHNOLOGY INNOVATIONS, LLC, Plaintiff, v. NSTEIN TECHNOLOGIES, INC. and…

Court: UNITED STATES DISTRICT COURT MIDDLE DISTRICT OF FLORIDA FORT MYERS DIVISION

Date published: Apr 27, 2015

Case No: 2:10-cv-341-FtM-38DNF (M.D. Fla. Apr. 27, 2015)

Opinion

Case No: 2:10-cv-341-FtM-38DNF

04-27-2015

TECHNOLOGY INNOVATIONS, LLC, Plaintiff, v. NSTEIN TECHNOLOGIES, INC. and OPEN TEXT CORPORATION, Defendants.

SHERI POLSTER CHAPPELL UNITED STATES DISTRICT JUDGE

ORDER ¹

1 Disclaimer: Documents filed in CM/ECF may contain hyperlinks to other documents or Web sites. These hyperlinks are provided only for users' convenience. Users are cautioned that hyperlinked documents in CM/ECF are subject to PACER fees. By allowing hyperlinks to other Web sites, this Court does not endorse, recommend, approve, or guarantee any third parties or the services or products they provide on their Web sites. Likewise, the Court has no agreements with any of these third parties or their Web sites. The Court accepts no responsibility for the availability or functionality of any hyperlink. Thus, the fact that a hyperlink ceases to work or directs the user to some other site does not affect the opinion of the Court.

This matter comes before the Court upon consideration of United States Magistrate Judge Douglas Frazier's Report and Recommendation (Doc. #172) filed on October 22, 2014. Judge Frazier recommends several claim constructions for the patent at issue. (Doc. #172 at 40-41). Plaintiff Technology Innovations, LLC, filed an Objection to the Report and Recommendation (Doc. #173) on November 5, 2014. Thereafter, Defendants Nstein Technologies, Inc. and Open Text Corporation² filed a Response to Plaintiff's Objection (Doc. #174) on November 20, 2014. The matter is now ripe for review.

2 Hereinafter referred to, collectively, as "Defendants."

Background

This action arose from an alleged infringement of United States Patent No. 6,829,613 ("the Patent" or "'613 patent"), entitled "Techniques for Controlling Distribution of Information from a Secure Domain." The Patent "describes a system, method, and computer program for controlling distribution of a message from a secure domain to a destination outside the secure domain." Stated in simpler terms, the Patent outlines technology that scans outgoing messages to see if they violate established security policies. If the outgoing message violates a security policy, the technology prevents the message from being sent. Plaintiff believes that two of Defendants' products - Digital Asset Management ("DAM") and Open Text Semantic Navigation ("OTSN") - infringe on the Patent. Defendants, however, dispute Plaintiff's infringement claim and believe there is a stark difference between the technology outlined in the Patent and their DAM and OTSN products. The Parties submitted Claim Construction Briefs (Doc. #139; 145; 146) and Judge Frazier issued a recommendation on those Briefs (Doc. #172). Plaintiff takes issue with nearly every recommended construction and filed an Objection to Judge Frazier's Report and Recommendation. (Doc. #173).

Standard of Review

After conducting a careful and complete review of the findings and recommendations, a district judge may accept, reject, or modify the magistrate judge's report and recommendation. 28 U.S.C. § 636(b)(1); see also Williams v. Wainwright , 681 F.2d 732 (11th Cir. 1982). When a party objects to proposed findings or recommendations made in the report, the district judge must conduct a de novo review. § 636(b)(1). Even in the absence of an objection, the district judge reviews legal conclusions de novo. See Cooper-Houston v. So. Ry. Co ., 37 F.3d 603, 604 (11th Cir. 1994).

Discussion

Having considered the Report and Recommendation and independently examined the file, the Court accepts the Report and Recommendation to the extent specified below.

A. Plaintiff's General Objections to Methodology

Plaintiff begins by objecting to "the general methodology" that Judge Frazier employed in reaching his recommended claim constructions. (Doc. #173 at 5-11). In essence, Plaintiff argues that including three preliminary sections in the Report and Recommendation before beginning to construct the disputed claims tainted the subsequent individual claim constructions. A thorough review of Judge Frazier's Report and Recommendation, however, reveals that this is simply not true. While Judge Frazier uses the word "construction" in these preliminary sections, he does not begin construing claims in these sections. The actual analysis accompanying the recommended claim constructions confirms this. Judge Frazier provides detailed individual claim constructions for each of the disputed claims without a single reference to his preliminary sections. Notably, Plaintiff fails to point the Court to any evidence indicating otherwise, and instead chooses to rely on a single conclusory sentence. But this will not suffice. Without any evidence illustrating that these preliminary sections "tainted" the subsequent analysis, Plaintiff's argument is unpersuasive. However, that being said, the Court will not adopt these preliminary sections and will instead focus on Judge Frazier's thorough and detailed individual recommended claim constructions.

B. Plaintiff's Objections to the Individual Claim Constructions

Plaintiff continues by objecting to each of Judge Frazier's recommended claim constructions. (Doc. #173 at 11-27). The Court will conduct a de novo review³ of each construction in turn.

3 In conducting the de novo review, the Court considered Plaintiff's Opening Claim Construction Brief (Doc. #139), Defendants' Responsive Claim Construction Brief (Doc. #145), Plaintiff's Reply Claim Construction Brief (Doc. #146), Plaintiff's Objections to the Report and Recommendation (Doc. #173), and Defendants' Response to Plaintiff's Objections (Doc. #174).
--------

1. Legal Standard for Constructing Claims

A "bedrock principle" of patent law is that "the claims of a patent define the invention to which the patentee is entitled the right to exclude." Phillips , 415 F.3d at 1312 (internal quotations and citations omitted). The Court must construe a claim "to accord [it] the meaning it would have to a person of ordinary skill in the art at the time of the invention." Innova/Pure Water , Inc. v. Safari Water Filtration Sys., Inc., 381 F.3d 1111, 1116 (Fed. Cir. 2004). To do so, the Court considers, among others, three sources: the claims themselves, the specification, and the prosecution history (if it is in evidence). Markman v. Westview Instruments , Inc., 52 F.3d 967, 979 (Fed. Cir. 1995), aff'd, 517 U.S. 370, 116 S.Ct. 1384, 134 L.Ed.2d 577 (1996).

The claim language itself provides the starting point, "for that is the language the patentee has chosen to particularly point out and distinctly claim the subject matter which the patentee regards as his invention." Innova/Pure Water , Inc., 381 F.3d at 1116 (internal citations, quotations, and alterations omitted). But this language must also "be read in view of the specification, of which [it is] a part." Markman , 52 F.3d at 979. The specification includes a written description of the invention that "may act as a sort of dictionary, [] explain[ing] the invention and [] defin[ing] terms used in the claims." Id. And, if it is in evidence, the Court must also consider the patent's prosecution history. See Id. ("Th[e] undisputed public record of proceedings in the Patent and Trademark Office is of primary significance in understanding the claims.") (internal citations and quotations omitted).

2. "controlling distribution of a message from a sender to a recipient"

No.	Term	Plaintiff's ProposedConstruction	Defendants' ProposedConstruction	Judge Frazier'sRecommendedConstruction
1.	"controllingdistribution ofa messagefrom a senderto a recipient"	No construction ofentire phrasenecessary after"distribution,""message," "sender,"and "recipient" areconstrued.	Controlling the deliveryof a MESSAGE at aboundary of a securedomain by determiningwhether delivery of theMESSAGE violates asecurity policy.	Controlling the deliveryof a MESSAGE at aboundary of a securedomain by determiningwhether delivery of theMESSAGE violates asecurity policy.

The Parties first dispute relates to the construction of the term "controlling distribution of a message from a sender to a recipient," found in Claims 1 and 23. '613 patent col. 17 ll. 5-6, col. 19 ll. 21-22. Plaintiff proposes that no construction of the entire term is necessary after the Court construes four other terms: "distribution," "message," "sender," and "recipient." Defendants propose, and Judge Frazier recommends, a construction that provides, "controlling the delivery of a message at a boundary of a secure domain by determining whether delivery of the message violates a security policy." The Court agrees with Judge Frazier and Defendants that, when read in light of the entire specification, the term requires a construction that denotes a security context.

Recently, in Virnetx , Inc. v. Cisco Sys., Inc., 767 F.3d 1308 (Fed. Cir. 2014), the Federal Circuit addressed a similar construction. There, the district court had construed the term "secure communication link" as "a direct communication link that provides data security." Virnetx , Inc., 767 F.3d at 1317. On appeal, the defendants argued that, when the term was read in light of the entire specification, the construction required the addition of the word "anonymity" at the end, despite the fact that "anonymity" did not appear once in the claim language. Id. The Federal Circuit agreed, explaining that "[t]he fact that anonymity [was] 'repeatedly and consistently' used to characterize the invention strongly suggest[ed] that it should be read as part of the claim." Id. at 1318. Notably, the court reached this conclusion despite the fact that "the specification mechanically preface[d] most passages with the phrase 'according to one aspect of the present invention.'" Id. The court also found it telling that the background and summary of the invention clearly identified anonymity "as a key part of the novel solution to the specific problem identified in the prior art." Id.

This same logic is applicable to the construction at issue. Plaintiff alleges it is improper to import the phrases "secure domain," "boundary," and "security policy" into the construction because these phrases are not found anywhere in the claim language itself. Indeed, no one disputes the absence of these phrases from the claim language. But only Plaintiff refuses to recognize that these phrases are repeatedly and consistently used to characterize the invention. And this repeat and consistent use strongly suggests that they should be read as part of the claim.

To exemplify this, the Court has to proceed no further than the Background of the Invention. The first sentence of that section provides, "[t]he present invention relates generally to the field of computer data security and content-based filtering of information." '613 patent col. 1 ll. 41-43. It goes on to explain that "increased connectivity has facilitated the free flow of information," but "has also created data security problems. . . ." '613 patent col. 1 ll. 53-55. It further explains that there are a number of "boundary controller" products available that attempt to solve these problems by "monitor[ing] the contents of outgoing messages passing from secure domains and flag[ging] those messages which violate security policies." '613 patent col. 2 ll. 24-26. But these boundary controllers operate using "keyword lists" or "dirty word lists," "plagued by errors and inconsistencies." '613 patent col. 2 ll. 32-34, col. 2 l. 47. Thus, "[t]here is a need for a system which can provide greater information security than that offered by prior art techniques." '613 patent col. 2 ll. 49-51.

Having framed the problem, the Summary of the Invention begins by providing a solution: "The present invention describes a system, method, and computer program for controlling distribution of a message from a secure domain to a destination outside the secure domain." '613 patent col. 2 ll. 54-57. The remainder of the Summary of the Invention is composed of five paragraphs all beginning with "according to an embodiment," or some variation thereof. Together, these paragraphs describe how the present invention provides superior security to that of prior art by utilizing "semantic modeling" to determine whether an outgoing message violates a security policy, rather than a "keyword" or "dirty word" list. See '613 patent col. 2 ll. 57-67, col. 3 ll. 1-38.

Similar to the plaintiff in Virnetx who attempted to hide behind "according to one aspect of the present invention" language, Plaintiff attempts to hide behind the Summary of the Invention's "according to an embodiment" language. Plaintiff adamantly asserts the Court should ignore this language because it is prefaced by "according to an embodiment." The Court disagrees. The Background and Summary of the Invention clearly identify the present invention as providing a key solution to the "errors and inconsistencies" experienced when using prior art to scan outgoing messages to determine whether they violate security policies. And this identification is significant to the construction at issue. See Virnetx , 767 F.3d at 1318 (finding "anonymity" must be added to a construction where it is part of the specific problem in prior art that the patent identified in its background and summary sections); see also CVI/Beta Ventures , Inc. v. Tura LP, 112 F.3d 1146, 1160 (Fed. Cir. 1997) ("In construing claims, the problem the inventor was attempting to solve, as discerned from the specification and the prosecution history, is a relevant consideration.").

It is also significant that the remainder of the specification continually references "secure domain," "boundary," and "security policy." Beyond the Background and Summary of the Invention, the specification contains twelve columns. Eight of these twelve columns mention or discuss the phrases that Plaintiff contends are improper to import into the construction at issue. See, e.g., '613 patent col. 4 ll. 21-30, col. 5 ll. 59-64, col. 7 l. 55, col. 9 ll. 59-65, col. 10 ll. 3-11, col. 12 ll. 61-67, col. 13 ll. 26-33, col. 14 ll. 32-51. Clearly, these phrases are "repeatedly and consistently used to characterize the invention." And this "strongly suggests that [they] should be read as part of the claim." Virnetx , 767 F.3d at 1318.

Plaintiff makes one final attempt at convincing the Court that Defendants' proposed construction is improper, arguing that this construction violates the doctrine of claim differentiation. The Court disagrees. The doctrine of claim differentiation "refers to the presumption that an independent claim should not be construed as requiring a limitation added by a dependent claim." Curtiss-Wright Flow Control Corp. v. Velan , Inc., 438 F.3d 1374, 1380 (Fed. Cir. 2006) (citations omitted). In other words, "a dependent claim must add a limitation to those recited in the independent claim." Id. Otherwise, "reading an additional limitation from a dependent claim into an independent claim would not only make the additional limitation superfluous, it might render the dependent claim invalid." Id.

Plaintiff argues that Defendants' proposed construction violates the doctrine because there are dependent claims - specifically, Claims 19 and 41 - that would be rendered superfluous if "security policy," "secure domain," and "boundary" were construed into the term at issue, found in independent Claims 1 and 23. This argument fails for several reasons. To begin, dependent Claims 19 and 41 have at least one additional limitation not found in either independent Claim 1 or 23: both "permit[] distribution of the message to the recipient if the security policy is not violated." '613 patent col. 18 ll. 61-62, col. 21 ll. 24-25. And the adoption of Defendants' proposed construction does not alter this fact. This, in and of itself, defeats Plaintiff's argument. Nevertheless, the Court would also like to point out that the doctrine of claim differentiation creates only a presumption. Regents of Univ. of Cal. v. Dakocytomation Cal., Inc., 517 F.3d 1364, 1375 (Fed. Cir. 2008). And as with all presumptions, it is rebuttable. Id. It is "not a hard and fast rule and will be overcome by a contrary construction dictated by the written description or prosecution history." Id. Even if Plaintiff successfully raised the presumption, the Court's analysis above easily rebutted it.

The Court, therefore, construes the term "controlling distribution of a message from a sender to a recipient" as "controlling the delivery of a message at a boundary of a secure domain by determining whether delivery of the message violates a security policy."

3. "distribution"

No.	Term	Plaintiff's ProposedConstruction	Defendants' ProposedConstruction	Judge Frazier'sRecommendedConstruction
2.	"distribution"	Electronictransmission orelectronic delivery	See No. 1.	See No. 1.

The Parties' second dispute relates to the construction of the term "distribution." Plaintiff proposes a construction that provides, "electronic transmission or electronic delivery." Defendants and Judge Frazier believe that no additional construction is necessary based on the first term's construction analysis above. To that end, Defendants and Judge Frazier both appear to construe "distribution" as "delivery." Plaintiff notes that it "would agree to construe 'distribution' to mean 'electronic delivery,' to accommodate how Defendants have interpreted that term." Because the Parties agree on construing the term at issue as "electronic delivery," the Court finds this construction appropriate.

The Court, therefore, construes "distribution" as "electronic delivery."

4. "message"

No.	Term	Plaintiff's ProposedConstruction	Defendants' ProposedConstruction	Judge Frazier'sRecommendedConstruction
3.	"message"	An electronic itemthat holds informationand is capable ofbeing transmittedelectronically, whichcan have a variety ofdifferent forms,including, forexample, adocument, a webpage, or an emailmessage.	An electroniccommunication createdby a SENDER for aspecified RECIPIENToutside the securedomain, which includes(1) text and (2) meta-information about thespecified RECIPIENT. AMESSAGE originates ina secure domain and isintended for a	An electroniccommunication createdby a SENDER for aspecified RECIPIENToutside the securedomain, which includes(1) text and (2) meta-information about thespecified RECIPIENT.A MESSAGE originatesin a secure domain andis intended for a

destination outside asecure domain.

destination outside asecure domain.

The Parties' third dispute relates to the construction of the term "message." Plaintiff proposes a construction that provides, "[a]n electronic item that holds information and is capable of being transmitted electronically, which can have a variety of different forms, including, for example, a document, a web page, or an email message." Defendants propose, and Judge Frazier recommends, a construction that provides, "[a]n electronic communication created by a sender for a specified recipient outside the secure domain, which includes (1) text and (2) meta-information about the specified recipient. A message originates in a secure domain and is intended for a destination outside a secure domain." The Court finds Defendants' proposed, and Judge Frazier's recommended, construction appropriate.

Plaintiff first objects to using the phrase "an electronic communication" rather than "an electronic item," but fails to provide any support for this argument. Nevertheless, as Judge Frazier correctly notes, the specification indicates that "communication," rather than "item," is the correct phrase. As discussed above, the Background of the Invention describes how "increased connectivity has facilitated the free flow of information," leading to an increased amount of information "being stored and communicated/distributed in electronic form." '613 patent col. 1 ll. 51-54 (emphasis added). It further describes how organizations traditionally tried to qualm the security risks associated with this increased communication/distribution by "employing security personnel who manually monitor[ed] the contents of the information carrying messages which originate[d] in a secure domain and whose destination lie[d] outside the secure domain." '613 patent col. 2 ll. 8-12. To be sure, the messages were being communicated. It next describes how prior art arrived that "automate[d] the task of controlling the dissemination of information from a secure domain." '613 patent col. 2 ll. 22-23. The messages were still being communicated.

Now, the present invention provides an even more secure, automated system/method/computer program for "controlling distribution of a message from a secure domain to a destination outside the secure domain." '613 patent col. 2 ll. 54-57. And the messages are still being communicated. Plaintiff prefers a construction that construes the term as "capable of being transmitted electronically." But there is no discussion in the specification that the present invention controls messages that are "capable of being transmitted." To the contrary, the specification illustrates that the present invention controls whether communicated messages reach their destination outside the secure domain. And the prosecution history independently confirms this. See (Doc. #145-5 at 13 (describing the present invention as technology that "identif[ies] and control[s] [] electronic communications"). Therefore, construing "message" as an electronic communication, rather than an electronic item, is appropriate.

Plaintiff next objects to importing the phrase "which includes (1) text and (2) meta-information about the specified recipient" into the construction. To be specific, Plaintiff takes no issue with importing first part of the phrase - "(1) text" - but adamantly disagrees with importing the second part - "(2) meta-information." In doing so, Plaintiff rebukes Judge Frazier for reading a "description of the details of [an] embodiment[]" into the construction. Remarkably, later in its argument, Plaintiff points the Court's attention to a different description of an embodiment in an attempt to support its own construction. But Plaintiff cannot have it both ways.

The specification provides that "the present invention parses the outgoing message to extract meta-information . . . from the outgoing message." '613 patent col. 7 ll. 61-63. It also notes that examples of meta-information include, among others, "message sender information" and "message recipient information." '613 patent col. 7 ll. 64-65. Plaintiff correctly notes that this language is located in the Description of the Specific Embodiments section of the specification. This location, however, does not render it immaterial. Indeed, "certain sections of the specification are more likely to contain statements that support a limiting definition of a claim term than other sections." C.R. Bard , Inc. v. U.S. Surgical Corp., 388 F.3d 858, 864 (Fed. Cir. 2004). But a statement's location is not determinative. Id. Rather, determining "what import to give language from the specification must, of course, be determined a case-by-case basis." Id.

In this circumstance, the Court finds that the statement should be imported into the construction at issue. The specification describes what action the present invention takes with messages - it "parses the outgoing message to extract meta-information." '613 col. 7 ll. 61-62. This indicates that the message must contain meta-information for the present invention to parse. Plaintiff has no argument to dispute this fact, and instead continues to rely on its argument that the Court should find this statement immaterial because of its location. But this argument alone is not enough to persuade the Court otherwise.

Plaintiff asserts one additional argument related to this part of the proposed construction, again bringing up the doctrine of claim differentiation. Plaintiff contends that importing the phrase "meta-information about the specified recipient" into the construction at issue would render dependent Claims 7 and 8 superfluous. The Court disagrees. Both Claims 7 and 8 contain additional limitations that prevent the claim differentiation presumption from arising. Claim 7 relates to "parsing the message to extract meta-information. . . ." '613 patent col. 17 ll. 54-55. Thus, specifying what information is contained in the meta-information does not alter this action. Moreover, while Claim 8 specifies that the meta-information contains "information about the recipient," it also contains three additional limitations. '613 patent col. 17 ll. 61-64. And this fact is fatal to Plaintiff's argument.

Plaintiff's final objection to Defendants' proposed construction relates to the phrase, "by a SENDER for a specified RECIPIENT." In two sentences, Plaintiff avers that this phrase is recited in the claim preamble, and this "strongly indicates" that the term "message" should not also include limitations related to a sender and recipient. Notably, Plaintiff provides no supporting citation for this argument. And the Court is not persuaded without such support. While Plaintiff correctly notes that the preamble contains limitations related to a sender and recipient in some fashion - "from a sender to a recipient" - it does not contain the additional limitation included in Defendants' proposed construction - "for a specified recipient." In the Court's view, this additional limitation rebuts any "indication" that the phrase should not be imported into the construction at issue.

The Court, therefore, construes the term "message" as "an electronic communication created by a sender for a specified recipient outside the secure domain, which includes (1) text and (2) meta-information about the specified recipient. A message originates in a secure domain and is intended for a destination outside a secure domain."

5. "sender"

No.	Term	Plaintiff's ProposedConstruction	Defendants' ProposedConstruction	Judge Frazier'sRecommendedConstruction
4.	"sender"	A source of amessage beingtransmitted	A person who is amember of a securedomain who sends aMESSAGE to aspecified RECIPIENT.	A person who is amember of a securedomain who sends aMESSAGE to aspecified RECIPIENT.

The Parties' fourth dispute relates to the construction of the term "sender." Plaintiff proposes a construction that provides, "a source of a message being transmitted." Defendants propose, and Judge Frazier recommends, a construction that provides, "a person who is a member of a secure domain who sends a message to a specified recipient." The Court finds Defendants' proposed, and Judge Frazier's recommended, construction appropriate.

Plaintiff first takes issue with the phrase "a person" in Defendants' proposed construction, contending that a "sender" is not necessarily a person. In support, Plaintiff focuses its attention on the prosecution history and the cited prior art; specifically on the '518 continuation application, where, as Plaintiff explains, the Patent Examiner compared "sender" in the claim to a keyboard in the Morita patent. The Court finds this argument unpersuasive. In the Court's view, the specification contains numerous indications that the present invention refers to a "sender" as a person. For instance, as Defendants and Judge Frazier point out, the specification explains that a sender has a "security clearance level." '613 col. 3 l. 14. Plaintiff fails to provide a single example where a keyboard would have a security clearance level. Another example is that the present invention controls messages from leaving the secure domain based on "natural language processing." '613 patent col. 1 ll. 42-48. This indicates that a person is sending a natural language message, rather than a source, such as a machine, sending a message in code. Again, Plaintiff is unable to provide the Court with a counter-example to this point.

Plaintiff also takes issue with the phrase "secure domain" in Defendants' proposed construction. To that end, Plaintiff argues that "Defendants' proposed construction incorrectly provides that a 'sender' must universally be 'a member of a secure domain,' even when not identified as having that characteristic." Notably, Plaintiff fails to provide any example of language from the Patent where the sender is not a member of a secure domain. Nevertheless, the first sentence of the Summary of the Invention alone rebuts Plaintiffs' argument: "[t]he present invention describes a system . . . for controlling distribution of a message from a secure domain to a destination outside the secure domain." '613 patent col. 2 ll. 54-57. If the message is coming "from a secure domain" and going "to a destination outside the secure domain," it logically follows that the sender is located in the secure domain when the message is sent.

The Court, therefore, construes "sender" as "a person who is a member of a secure domain who sends a message to a specified recipient."

6. "recipient"

No.	Term	Plaintiff's ProposedConstruction	Defendants' ProposedConstruction	Judge Frazier'sRecommendedConstruction
5.	"recipient"	A destination orprospective target fora message beingtransmitted.	A person who is not amember of the securedomain and is specifiedby the sender to receivethe MESSAGE.	A person who is not amember of the securedomain and is specifiedby the sender to receivethe MESSAGE.

The Parties' fifth dispute relates to the construction of the term "recipient." Plaintiff proposes a construction that provides, "a destination or prospective target for a message being transmitted." Defendants propose, and Judge Frazier recommends, a construction that provides, "a person who is not a member of the secure domain and is specified by the sender to receive the message." The Court finds Defendants' proposed, and Judge Frazier's recommended, construction appropriate.

Similar to its argument for the last construction, Plaintiff takes issue with including "secure domain" in the construction at issue. And once again, the Court finds that secure domain is a proper import. It has been repeatedly noted that the "present invention describes a system . . . for controlling distribution of a message from a secure domain to a destination outside the secure domain." '613 patent col. 2 ll. 54-57 (emphasis added). If a message is being sent to a destination outside the secure domain, it logically follows that the recipient is not a member of the secure domain. Plaintiff describes this as "circular logic," alleging that the "recipient need not be specified by the sender to receive the message" and that the "recipient can be a structure, not a person." But the Court disagrees. If there were not a recipient outside the secure domain, the present invention would be rendered futile: why would there be a need to "control distribution of a message from secure domain to a destination outside the secure domain?" Moreover, as noted above, the present invention uses natural language processing to control such distribution. See '613 patent col. 1 ll. 42-48. It therefore follows that the recipient is a person, rather than a structure that would receive code.

The Court, therefore, construes "recipient" as "a person who is not a member of the secure domain and is specified by the sender to receive the message."

7. "semantic model"

No.	Term	Plaintiff's ProposedConstruction	Defendants' ProposedConstruction	Judge Frazier'sRecommendedConstruction
6.	"semanticmodel"	A semanticrepresentation of amessage or messagecategory thatincludes text otherthan words containedin the message orthe messagecategory	A set of structuredknowledgerepresentations whichconsist of concepts andrelations that representthe meaning expressedby theMESSAGE/MESSAGECATEGORY.	A set of structuredknowledgerepresentations whichconsist of concepts andrelations that representthe meaning expressedby theMESSAGE/MESSAGECATEGORY.

The Parties' sixth dispute relates to the construction of the term "semantic model." Plaintiff proposes a construction that provides, "a semantic representation of a message or message category that includes text other than words contained in the message or the message category." Defendants propose, and Judge Frazier recommends, a construction that provides, "a set of structured knowledge representations which consist of concepts and relations that represent the meaning expressed by the MESSAGE/MESSAGE CATEGORY." The Court finds Defendants' proposed, and Judge Frazier's recommended, construction appropriate.

The Parties agree that "representation(s)" should be imported into the construction at issue. But their agreement ends there. Plaintiff asserts that "semantic representation" is more appropriate than "a set of structured knowledge representations" because "the word 'semantic' itself is an ordinary word that is not specific to the patented technology." Defendants, however, argue that "merely replacing 'model' with 'representation,' particularly without defining 'semantic,' does nothing to aid the jury in interpreting this claim term." The Court agrees with Defendants: simply replacing one word of the term at issue with a synonym does not provide any type of a construction. Not to mention, Defendants' proposed phrase - "a set of structured knowledge representations" - is found throughout the specification, describing "semantic model." See, e.g., '613 patent col. 14 ll. 35-36 ("The present invention constructs semantic knowledge representations. . . ."); id . at col. 11 ll. 49-52 ("Knowledge extraction system converts the text segments into a knowledge representation which . . . constitutes the semantic model for the message category."); see also id . at col. 7 ll. 28-29, col. 8 ll. 8-15. And this clearly provides a better construction than replacing a single word with a synonym.

The Parties also disagree on the remainder of the construction. Plaintiff avers that its remaining proposed construction - "text other than words contained in the message or the message category" - "stays true to the ordinary definition of 'semantic' and also corresponds to the intrinsic evidence." To this end, Plaintiff asserts that "constru[ing] the semantic model as consisting of both concepts and relationships would be contrary to the '613 [patent's] specification." The Court disagrees.

Indeed, the select portion of the specification that Plaintiff relies on appears, at first glance, to support its position. See '613 patent col. 10 ll. 59-65. A thorough reading of the specification, however, reveals that Plaintiff conveniently ignores the fact that the specification regularly discusses semantic model as consisting of both concepts and relations. See, e.g., '613 patent col. 7 ll. 34-36 ("These concepts and relations provide . . . . The concepts and relations are represented . . . ."), col. 11 ll. 63-65 ("The knowledge representation . . . consists of concept-relation-concept (CRC) triples and relation-concept (RC) tuples."). And Plaintiff's only other cited example, '613 patent col. 9 ll. 38-58, actually rebuts its own position and affirms this fact. See '613 patent col. 9 ll. 48-51 ("The semantic model for the message category may be represented by a network of semantic relations between these three concepts.") (emphasis added).

The Court, therefore, construes "semantic model" as "a set of structured knowledge representations which consist of concepts and relations that represent the meaning expressed by the message/message category."

8. "message categories"

No.	Term	Plaintiff's ProposedConstruction	Defendants' ProposedConstruction	Judge Frazier'sRecommendedConstruction
7.	"messagecategories"	Categories to whichmessages can beclassified asbelonging.	Categories to whichmessage can beclassified as belonging,where each category (1)is associated withdescriptive text, and (2)has an associatedsecurity clearance levelindependent of SENDERor RECIPIENT.	Categories to whichmessage can beclassified as belonging,where each category(1) is associated withdescriptive text, and (2)has an associatedsecurity clearance levelindependent ofSENDER orRECIPIENT.

The Parties' seventh disputes relates to the construction of the term "message categories." Plaintiff proposes a construction that provides, "categories to which messages can be classified as belonging." Defendants propose, and Judge Frazier recommends, a construction that provides, "categories to which messages can be classified as belonging, where each category (1) is associated with descriptive text, and (2) has an associated security clearance level independent of SENDER or RECIPIENT." The Court finds Defendants' proposed, and Judge Frazier's recommended, construction appropriate.

The Parties agree that the term at issue should be construed, at least in part, as "categories to which messages can be classified as belonging." But the agreement ends there. Plaintiff avers that including the remainder of Defendants' proposed construction "add[s] additional limitations, extracted from preferred embodiments, to the claim language." Plaintiff also contends that importing the remainder of Defendants' proposed construction would violate the doctrine of claim differentiation. The Court disagrees.

In support of its claim differentiation argument, Plaintiff contends that Defendants' proposed construction renders dependent claims 20, 21, 42, and 43 superfluous. A review of those dependent claims, however, reveals a fatal flaw to this argument: each of the dependent claims contains a limitation not found in Defendants' proposed construction. While the proposed construction imports language specifying that each message category has an associated security clearance level, the dependent claims reference "determining a security clearance level for the message category to which the message was classified." '613 col. 18 ll. 66-67, col. 19 ll. 7-8, col. 21 ll. 22-23, col. 21 ll. 31-32. In other words, the dependent claims' language references a method or computer program that determines what the associated security clearance level for the message category is, not that the method or computer program is assigning a security clearance level to the message category itself.

The remainder of Plaintiff's argument is equally unpersuasive. Plaintiff contends that Defendants' proposed construction imports sporadic and isolated portions of the specification, which is improper. But the Court views the proposed construction in a different light: it reflects the specification's repeat and consistent association of message categories with descriptive texts and security clearance levels. See, e.g., '613 patent col. 6 l. 24 ("message category descriptive text provided by the user"), col. 6 ll. 37-38 ("[t]he message category along with its associated description"), col. 6 l. 53 ("message category description"), col. 7 l. 23 ("message category description), col. 10 ll. 55-56 ("message categories along with their descriptive information"), col. 10 ll. 60-61 ("message category descriptive information"), col. 3 ll. 19-20 ("the security clearance level of the message category"), col. 6 ll. 46-50 ("message categories . . . and the associated security clearance level"), col. 12 ll. 58-59 ("the security clearance of the corresponding message category"), col. 13 ll. 19-20 ("security clearance levels of the message category"). And this repeat and consistent use "strongly suggests that [these phrases] should be read as part of the claim." Virnetx , 767 F.3d at 1318.

The Court, therefore, construes "message categories" as "categories to which message can be classified as belonging, where each category (1) is associated with descriptive text, and (2) has an associated security clearance level independent of sender or recipient."

9. "constructing semantic models for a plurality of message categories"

No.	Term	Plaintiff's ProposedConstruction	Defendants' ProposedConstruction	Judge Frazier'sRecommendedConstruction
8.	"constructingsemanticmodels for aplurality ofmessagecategories"	No constructionrequired. In thealternative, the termshould be construedas "making orforming semanticmodels for a pluralityof messagecategories bycombining orarranging parts orelements."	Extracting the meaningof each MESSAGECATEGORY from thedescriptive textassociated with eachMESSAGE CATEGORYby: (1) using naturallanguage processingand retrieving textsegments (different thanthe words of the text)that relate to themeaning of the text; and	Extracting the meaningof each MESSAGECATEGORY from thedescriptive textassociated with eachMESSAGECATEGORY by: (1)using natural languageprocessing andretrieving text segments(different than thewords of the text) thatrelate to the meaning of

then (2) extractingconcepts and relationsfrom the retrieved textsegments to representthe meaning of the textsegments.

the text; and then (2)extracting concepts andrelations from theretrieved text segmentsto represent themeaning of the textsegments.

The Parties' eighth dispute relates to the construction of the term "constructing semantic models for a plurality of message categories." Plaintiff proposes that no construction of this term is necessary "because the terms 'semantic model' and 'message categories' are defined elsewhere, and because the word 'constructing' has a plain meaning that would be easily understood and applied by the jury." In the alternative, should the Court construe the term, Plaintiff proposes a construction that provides, "making or forming semantic models for a plurality of message categories by combining or arranging parts or elements." Defendants propose, and Judge Frazier recommends, a construction that provides, "extracting the meaning of each message category from the descriptive text associated with each message category by: (1) using natural language processing and retrieving text segments (different than the words of the text) that relate to the meaning of the text; and then (2) extracting concepts and relationships from the retrieved text segments to represent the meaning of the text segments." The Court finds Defendants' proposed, and Judge Frazier's recommended, construction appropriate.

The specification explicitly describes the two "steps performed by the present invention for constructing a semantic model for a message category." '613 patent col. 6 ll. 15-17. "During the first stage, the present invention uses natural language processing and information retrieval techniques to generate a set of text segments that are closely related to the meaning of the message category as described by the message category descriptive text provided by the user." '613 patent col. 6 ll. 19-24. "During the second stage, the present invention extracts structured knowledge representation for the message category from the set of text segments retrieved in stage one." '613 patent col. 6 ll. 26-29. Plaintiff contends that relying on this explicit description of how the present invention constructs a semantic model for a message category is improper. To that end, Plaintiff highlights that the specification provides that constructing such a semantic model "may" be accomplished using these two steps, and, therefore, the Court should reject this language because it "certainly [is] not a requirement."

Plaintiff is correct - the specification provides that constructing such a semantic model "may" be accomplished through these two detailed steps. See '613 patent col. 6 l. 19 ("the step of constructing a semantic model for a message category may be accomplished in two stages"). But in the Court's view, including this detailed analysis of two steps that are typically taken to construct such a semantic model illustrates that a construction of the term at issue is necessary. And Defendants' proposed construction relies on explicit language from the specification addressing this exact term - constructing a semantic model for a message category. On the other hand, to support its proposed construction, Plaintiff directs the Court's attention to five locations where it avers that the specification uses "construct" and "build" interchangeably, and also to a section of the "earlier-filed '088 patent." Unlike Defendants' citation to the specification, however, none of Plaintiff's cites directly address the term at issue. While Plaintiff latches on to the word "may," Defendant directs the Court's attention to explicit support for its proposed construction. The Court disagrees that this single word - "may" - warrants ignoring the specification's detailed analysis of the term at issue. See Phillips , 415 F.3d at 1321 ("[T]he specification is the single best guide to the meaning of a disputed term. . . .") (internal quotations and citations omitted).

Plaintiff also argues that Defendants' proposed construction violates the doctrine of claim differentiation. In making this argument, Plaintiff contends that dependent claims 2 and 24 would be rendered superfluous if the Court adopts Defendants' proposed construction. The Court disagrees. Both Claims 2 and 24 contain limitations not found in Claim 1, even when construed with Defendants' proposed construction. For example, Claims 2 and 24 contain limitations that relate to "receiving descriptions for the plurality of message categories," a passive limitation, while Defendants' proposed construction relates to "extracting descriptions," an active limitation. '613 patent col. 17 ll. 21-22, col. 19 ll. 40-41. Nevertheless, the Court finds that the specification's detailed analysis relating to the term at issue overcomes any presumption raised (if any) by this doctrine. See Regents of Univ. of Cal ., 517 F.3d at 1375.

The Court, therefore, construes "constructing semantic models for a plurality of message categories" as "extracting the meaning of each message category from the descriptive text associated with each message category by: (1) using natural language processing and retrieving text segments (different than the words of the text) that relate to the meaning of the text; and then (2) extracting concepts and relations from the retrieved text segments to represent the meaning of the text segments."

10. "constructing a semantic model for the message"

No.	Term	Plaintiff'sProposedConstruction	Defendants'ProposedConstruction	Judge Frazier'sRecommendedConstruction
9.	"constructinga semanticmodel for	No constructionrequired. In thealternative, the	Extracting the meaningof the MESSAGE byextracting concepts	Extracting the meaning ofthe MESSAGE by extractingconcepts and relations from

themessage"

term should beconstrued as"making or forminga semantic modelfor the message bycombining orarranging parts orelements."

and relations from themeta-information of theMESSAGE torepresent the meaningof the MESSAGE. Theconstruction of theSEMANTIC MODEL ofthe MESSAGE triggersperformance of thecompare, classify, anddetermine steps.

the meta-information of theMESSAGE to represent themeaning of the MESSAGE.The construction of theSEMANTIC MODEL of theMESSAGE triggersperformance of thecompare, classify, anddetermine steps.

The Parties' ninth dispute relates to the construction of the term "constructing a semantic model for the message." Once again, Plaintiff proposes that no construction of the term is necessary. If, however, the Court finds that a construction is necessary, Plaintiff proposes a construction that provides, "making or forming a semantic model for the message by combining or arranging parts or elements." Defendants propose, and Judge Frazier recommends, a construction that provides, "extracting the meaning of the message by extracting concepts and relations from the meta-information of the message to represent the meaning of the message. The construction of the semantic model of the message triggers performance of the compare, classify, and determine steps." The Court finds Defendants' proposed, and Judge Frazier's recommended, construction appropriate.

Similar to the last term, the specification explicitly describes the "steps performed by the present invention for building a semantic model for an outgoing message." '613 patent col. 7 ll. 51-53. "[T]he present invention parses the outgoing message to extract meta-information (or functional fields) from the outgoing message." '613 patent col. 7 ll. 61-63. "The contents of the meta-information are then processed by a knowledge extraction system . . . to extract knowledge representations for the outgoing message." '613 patent col. 8 ll. 5-8. "The knowledge representations extracted . . . comprise the structured, meaning representation of the message and constitute the semantic model for the outgoing message." '613 patent col. 8 ll. 12-15. "After a semantic model for an outgoing message have [sic] been constructed, the semantic model of the going message is compared with the semantic models of the message categories to determine a degree of similarity between the semantic models." '613 patent col. 8 ll. 16-20.

Clearly, Defendants' proposed construction reflects the specification's explicit description of the process that the present invention uses to construct a semantic model of a message. Nevertheless, Plaintiff contends that Defendants' proposed construction is improper because "the term does not require construction." Rather, "the term 'semantic model' and 'message' are already defined elsewhere, and . . . the word 'constructing' has a plain meaning." The Court disagrees and finds that a construction of the term at issue is necessary. Beyond its "no-construction-is-necessary" argument, Plaintiff directs the Court's attention to its argument for the previous term, which the Court already dismissed as unpersuasive. To be clear, the Court finds Defendants' citation to the specification, where it explicitly addresses the term at issue, more persuasive than Plaintiff's citations to five areas where the specification "uses 'construct' and 'build' interchangeably" and its citation to a section of the "earlier-filed '088 patent."

The Court, therefore, construes "constructing a semantic model for the message" as "extracting the meaning of the message by extracting concepts and relations from the meta-information of the message to represent the meaning of the message. The construction of the semantic model of the message triggers performance of the compare, classify, and determine steps."

11. "comparing the semantic model of the message with the semantic models for the plurality of message categories"

No.	Term	Plaintiff's ProposedConstruction	Defendants' ProposedConstruction	Judge Frazier'sRecommendedConstruction
10.	"comparingthe semanticmodel of themessage withthe semanticmodels forthe pluralityof messagecategories"	No constructionrequired. In thealternative, the termshould be construedas "examining forsimilarities thesemantic model ofthe message and thesemantic models forthe plurality ofmessage categories."	Determining a degree ofsimilarity between theSEMANTIC MODEL forthe MESSAGE and eachpreviously constructedSEMANTIC MODEL forthe multiple MESSAGECATEGORIES.	Determining a degreeof similarity betweenthe SEMANTIC MODELfor the MESSAGE andeach previouslyconstructed SEMANTICMODEL for the multipleMESSAGECATEGORIES.

The Parties' tenth dispute relates to the construction of the term "comparing the semantic model of the message with the semantic models for the plurality of message categories." Plaintiff contends that no construction of the term is necessary because "most of its words are terms that are being construed separately," and the remaining words "have a plain meaning and need not be construed." If, however, the Court finds that a construction is needed, Plaintiff proposes a construction that provides, "examining for similarities the semantic model of the message and the semantic models for the plurality of message categories." Defendants propose, and Judge Frazier recommends, a construction that provides, "determining a degree of similarity between the semantic model for the message and each previously constructed semantic model for the multiple message categories." The Court finds Defendants' proposed, and Judge Frazier's recommended, construction appropriate.

Plaintiff insists the Defendants' proposed construction is improper because it "lacks support in either the intrinsic or the extrinsic evidence." Extrinsically, Plaintiff provides several dictionary definitions of "compare" that it believes support its alternate, proposed construction. Intrinsically, Plaintiff points to several portions of the specification that it believes illustrate "Defendants' proposed construction[] incorrectly include[s] extraneous limitations." The Court finds that no extrinsic evidence is needed in this instance. Rather, Plaintiff's intrinsic cites actually affirm that Defendants' proposed construction is appropriate because it mirrors the specification. For instance, Plaintiff cites to the Abstract: "The semantic model of an outgoing message is compared with the semantic models of the message categories to determine a degree of similarity between the semantic models." '613 patent; see also '613 patent col. 2 l. 66 - col. 3 l. 4 (noting that the present invention "compares" and "determines"); '613 patent col. 6 ll. 4-8 ("compared . . . to determine"). In the Court's view, this sentence illustrates that "comparing" entails "determining a degree of similarity between the semantic models." And this is exactly what Defendants' proposed construction provides.

Plaintiff also argues that Defendants' proposed construction violates the doctrine of claim differentiation. To that end, Plaintiff believes that the proposed construction renders dependent Claims 10 and 35 superfluous. Indeed, there does not appear to be an additional limitation in either dependent Claim 10 or Claim 35 that distinguishes these claims from independent Claim 1 or Claim 23 when read in light of Defendants' proposed construction. But this does not, as Plaintiff contends, automatically render Defendants' proposed construction improper. While addressing a similar situation, the Federal Circuit noted, as the Court does above, that "the doctrine of claim differentiation is not a conclusive basis for construing claims," and the specification can override the doctrine's effect. Am. Calcar , Inc. v. Am. Honda Motor Co., 651 F.3d 1318, 1337 (Fed. Cir. 2011) (citations omitted). Thus, whereas here, "[i]f a claim will bear only one interpretation, similarity will have to be tolerated." Laitram Corp. v. Rexnord , Inc., 939 F.2d 1533, 1538 (Fed. Cir. 1991) (citations omitted).

The Court, therefore, construes "comparing the semantic model of the message with the semantic models for the plurality of message categories" as "determining a degree of similarity between the semantic model for the message and each previously constructed semantic model for the multiple message categories."

12. "classifying the message based on the comparison"

No.	Term	Plaintiff's ProposedConstruction	Defendants' ProposedConstruction	Judge Frazier'sRecommendedConstruction
11.	"classifyingthe messagebased on thecomparison"	Identifying to whichmessage category, ifany, a messagebelongs, using thecomparison ofsemantic models(directly or indirectly).	After the comparison,identifying theMESSAGE as belongingto each MESSAGECATEGORY for whichthe degree of similaritysatisfies a presetthreshold for thatMESSAGECATEGORY.	After the comparison,identifying theMESSAGE asbelonging to eachMESSAGECATEGORY for whichthe degree of similaritysatisfies a presetthreshold for thatMESSAGECATEGORY.

The Parties' eleventh dispute relates to the construction of the term "classifying the message based on the comparison." Plaintiff proposes a construction that provides, "identifying to which message category, if any, a message belongs, using the comparison of semantic models (directly or indirectly)." Defendants propose, and Judge Frazier recommends, "after the comparison, identifying the message as belonging to each message category for which the degree of similarity satisfies a preset threshold for that message category." The Court finds Defendants' proposed, and Judge Frazier's recommended, construction appropriate.

Similar to other terms that the Court has construed, the specification explicitly addresses the term at issue. "A message is classified as belonging to a message category if the degree of similarity between the semantic model of the message and the semantic model of the message category exceeds a threshold degree of similarity." '613 patent col. 3 ll. 4-8; see also '613 patent col. 6 ll. 8-10, col. 8 ll. 22-30, col. 12 ll. 42-47. Clearly, Defendants' proposed construction mirrors the specification's multiple discussions of the term at issue. Nevertheless, Plaintiff contends that this proposed construction is improper because the word "classify" (or a variant thereof) is used throughout the specification and Defendants' construction "narrow[s] the plain and ordinary meaning of this phrase." But the Court finds this argument unpersuasive. While the Court does not dispute that "classify" (or some variant thereof) is used in several areas of the specification, the Court is not construing only the word "classify." The Court is tasked with construing the entire term - "classifying the message based on the comparison." The citations above illustrate that the specification explicitly describes what steps are taken in accordance with the entire term. And because Defendants' proposed construction adequately mirrors the specification's exact language, the Court finds this construction appropriate.

The Court, therefore, construes "classifying the message based on the comparison" as "after the comparison, identifying the message as belonging to each message category for which the degree of similarity satisfies a preset threshold for that message category."

13. "determining if the message can be distributed to the recipient based on the classification of the message"

No.	Term	Plaintiff's ProposedConstruction	Defendants' ProposedConstruction	Judge Frazier'sRecommendedConstruction
12.	"determiningif themessage canbe distributedto therecipientbased on theclassificationof themessage"	Determining if themessage is capableof being distributed tothe recipient based atleast in part on theclassification of themessage.	After the classification,determining if thedistribution of theMESSAGE to theRECIPIENT wouldviolate any securitypolicy for the securedomain using thesecurity clearance levelof the MESSAGECATEGORIESassociated with theMESSAGE as a result ofthe classification.	After the classification,determining if thedistribution of theMESSAGE to theRECIPIENT wouldviolate any securitypolicy for the securedomain using thesecurity clearance levelof the MESSAGECATEGORIESassociated with theMESSAGE as a resultof the classification.

The Parties' twelfth dispute relates to the construction of the term "determining if the message can be distributed to the recipient based on the classification of the message." Plaintiff proposes a construction that provides, "determining if the message is capable of being distributed to the recipient based at least in part on the classification of the message." Defendants propose, and Judge Frazier recommends, a construction that provides, "after the classification, determining if the distribution of the message to the recipient would violate any security policy for the secure domain using the security clearance level of the message." The Court finds Defendants' proposed, and Judge Frazier's recommended, construction appropriate.

Similar to the Parties' first dispute, Plaintiff again takes issue with Defendants' proposed construction importing security-oriented terms, such as "secure," "security clearance levels," and "security policy." And the Court, once again, finds this argument unpersuasive. To begin, the Court has already found that these terms are consistently and repeatedly used throughout the Patent, and this indicates that they should be read into the construction. See Virnetx , Inc., 767 F.3d at 1318. The Court has also already found that the present invention attempts to solve a problem related to security, further indicating that importing these security-oriented terms into the construction is proper. See CVI/Beta Ventures , Inc., 112 F.3d at 1160. Not to mention, Defendants' proposed construction mirrors the present invention's "determining" process: The "present invention determines if the message can be distributed to a recipient outside the secure domain by determining if the message violates a security policy." '613 patent col. 3 ll. 10-13; see also id . at col. 4 ll. 21-24. ("The present invention . . . controls the distribution of information . . . by automatically detecting outgoing messages which violate security policies. . . ."); id . at col. 9 ll. 59-65 (noting that the present invention determines whether a message can be distributed outside the secure domain based on whether it violates a security policy); id . at col. 12 ll. 50-54 ("The outgoing message classification information is then passed to security checker which determines whether the distribution of the outgoing message . . . violates a user-defined security policy."). Clearly, the specification supports Defendants' proposed construction.

The Court, therefore, construes "determining if the message can be distributed to the recipient based on the classification of the message" as "after the classification, determining if the distribution of the message to the recipient would violate any security policy for the secure domain using the security clearance level of the message categories associated with the message as a result of the classification."

14. "permitting distribution of the message to the recipient if the security policy is not violated"

No.	Term	Plaintiff's ProposedConstruction	Defendants' ProposedConstruction	Judge Frazier'sRecommendedConstruction
13.	"permittingdistribution ofthe messageto therecipient ifthe securitypolicy is notviolated"	No constructionrequired. In thealternative,"permitting" means"allowing."	If it is determined that asecurity policy is notviolated, forwarding theMESSAGE to theRECIPIENT.	If it is determined that asecurity policy is notviolated, the MESSAGEmay be forwarded tothe RECIPIENT.

The Parties' final dispute relates to the construction of the term "permitting distribution of the message to the recipient if the security policy is not violated." Plaintiff contends that no construction is necessary. If, however, the Court finds that a construction is necessary, Plaintiff proposes a construction that provides, "allowing distribution of the message to the recipient if the security policy is not violated." Defendant proposes a construction that provides, "if it is determined that a security policy is not violated, forwarding the message to the recipient." And Judge Frazier recommends a construction that provides, "if it is determined that a security policy is not violated, the message may be forwarded to the recipient." The Court finds that Judge Frazier's recommended construction is appropriate.

In explaining the term at issue, the specification notes, "[m]essages which do not violate any security policies may be forwarded to the recipient." '613 patent col. 3 ll. 22-23. Plaintiff contends that the Court should not consider this part of the specification because it relates to a specific embodiment. Defendants, however, believe that this portion of the specification, along with other cited portions, illustrate how a message is distributed if it does not violate a security policy. The Court agrees. The above-cited portion of the specification clearly denotes what happens when the present invention determines that a message does not violate a security policy and permits distribution. And to be sure, other sections of the specification confirm this. See '613 patent col. 10 ll. 15-18 ("if there is no security violation, the boundary controller may be configured to forward the message to a mail serve for distribution to the recipients"); id . at col. 12 ll. 66 - col. 13 l. 1 ("If no security policy is violated, the outgoing message may be forwarded . . . ."); id . at col. 13 ll. 31-32 ("Messages which do not violate any security policy may be forwarded . . . ."). Indeed, Defendants' proposed construction, for the most part, reflects these portions of the specification. But the Court agrees with Judge Frazier that excluding "may be" from the construction is improper and too limiting.

The Court, therefore, construes "permitting distribution of the message to the recipient if the security policy is not violated" as "if it is determined that a security policy is not violated, the message may be forwarded to the recipient."

Accordingly, it is now

ORDERED:

1. Plaintiff's Objection to the Report and Recommendation (Doc. #173) is OVERRULED to the extent stated herein.



2. The Report and Recommendation of United States Magistrate Judge Douglas Frazier (Doc. #172) is ADOPTED in part and the findings incorporated herein.



3. The disputed language contained in Claims 1, 19, 23, and 41 shall be constructed as follows:

Term No.:	Term:	Court's Construction:
1.	"controlling distribution of a messagefrom a sender to a recipient"	"controlling the delivery of a MESSAGE at aboundary of a secure domain by determiningwhether delivery of the MESSAGE violates asecurity policy"
2.	"distribution"	"electronic delivery"
3.	"message"	"An electronic communication created by aSENDER for a specified RECIPIENT outside thesecure domain, which includes (1) text and (2)meta-information about the specifiedRECIPIENT. A MESSAGE originates in asecure domain and is intended for a destinationoutside a secure domain."
4.	"sender"	"a person who is a member of a secure domainwho sends a MESSAGE to a specifiedRECIPIENT"
5.	"recipient"	"a person who is not a member of the securedomain and is specified by the SENDER toreceive the MESSAGE"
6.	"semantic model"	"a set of structured knowledge representationswhich consist of concepts and relations thatrepresent the meaning expressed by theMESSAGE/MESSAGE CATEGORY"
7.	"message categories"	"categories to which messages can be classifiedas belonging, where each category (1) isassociated with descriptive text, and (2) has anassociated security clearance level independentof SENDER or RECIPIENT."
8.	"constructing semantic models for aplurality of message categories"	"extracting the meaning of each MESSAGECATEGORY from the descriptive text associatedwith each MESSAGE CATEGORY by: (1) usingnatural language processing and retrieving textsegments (different than the words of the text)that relate to the meaning of the text; and then(2) extracting concepts and relations from theretrieved text segments to represent themeaning of the text segments"
9.	"constructing a semantic model forthe message"	"Extracting the meaning of the MESSAGE byextracting concepts and relations from the meta-information of the MESSAGE to represent themeaning of the MESSAGE. The construction ofthe SEMANTIC MODEL of the MESSAGEtriggers performance of the compare, classify,and determine steps."

10.	"comparing the semantic model ofthe message with the semanticmodels for the plurality of messagecategories"	"determining a degree of similarity between theSEMANTIC MODEL for the MESSAGE andeach previously constructed SEMANTICMODEL for the multiple MESSAGECATEGORIES"
11.	"classifying the message based onthe comparison"	"after the comparison, identifying the MESSAGEas belonging to each MESSAGE CATEGORYfor which the degree of similarity satisfies apreset threshold for that MESSAGECATEGORY"
12.	"determining if the message can bedistributed to the recipient based onthe classification of the message"	"after the classification, determining if thedistribution of the MESSAGE to the RECIPIENTwould violate any security policy for the securedomain using the security clearance level of theMESSAGE CATEGORIES associated with theMESSAGE as a result of the classification"
13.	"permitting distribution of themessage to the recipient if thesecurity policy is not violated"	"if it is determined that a security policy is notviolated, the MESSAGE may be forwarded to theRECIPIENT"

DONE and ORDERED in Fort Myers, Florida, this 27th day of April, 2015.

/s/ _________

SHERI POLSTER CHAPPELL

UNITED STATES DISTRICT JUDGE
Copies: All Parties of Record