Opinion
Civil Action No. 3801-VCP.
Submitted: April 20, 2010.
Decided: August 13, 2010.
Denise Seastone Kraft, Esquire, Paul D. Brown, Esquire, K. Tyler O'Connell, Esquire, Aleine Porterfield, Esquire, EDWARDS ANGELL PALMER DODGE LLP, Wilmington, Delaware; Peter C. Schechter, Esquire, EDWARDS ANGELL PALMER DODGE LLP, New York, New York; Attorneys for Plaintiff.
Daniel A. Dreisbach, Esquire, Harry Tashjian, IV, Esquire, RICHARDS, LAYTON FINGER, P.A., Wilmington, Delaware; Faith E. Gay, Esquire, Robert Juman, Esquire Jehan Aslam, Esquire, QUINN EMANUEL URQUHART SULLIVAN, LLP, New York, New York; Attorneys for Defendant.
OPINION
This case tasks the Court with determining the amount of royalties owed to the designer of audio compression software under a technology licensing agreement. Plaintiff, Vianix Delaware LLC ("Vianix"), and Defendant, Nuance Communications, Inc. ("Nuance"), are parties to a Technology Licensing Agreement (the "TLA") that went into effect on January 20, 2003. Under the TLA, which is remarkable for its poor drafting, Vianix licensed its audio compression software, known as "MASC," to Nuance for use in Nuance's line of dictation and speech recognition products. In return, Nuance agreed to pay Vianix royalties whenever it licensed a product containing MASC to a customer. Eventually, the relationship between Vianix and Nuance soured because Vianix believed Nuance was underpaying royalties. Vianix hired an auditor to conduct a royalty audit of Nuance and, in February 2008, sent Nuance an invoice for $12.6 million for outstanding royalties, an amount far in excess of all amounts Nuance previously had paid under the TLA. Nuance refused to pay the invoice or any substantial amount related to it, and Vianix filed this suit alleging that Nuance breached the TLA by underpaying royalties. Nuance responded by filing two counterclaims, alleging that Vianix breached the TLA by sending the $12.6 million invoice in bad faith and seeking a declaration that Nuance complied with its obligations under the TLA.
Numerous issues regarding the amount of royalties Nuance owes Vianix remained in dispute when this action went to trial, the majority of which involve disputes over the interpretation of the TLA and whether certain Nuance products were royalty-bearing. This Opinion represents my post-trial findings of fact and conclusions of law on those issues.
For the reasons stated, I find that: (1) in addition to all products for which Nuance admits it owes royalties, Nuance owes Vianix a royalty for certain Powerscribe SDK version 3.0 units, but does not owe a royalty for any other disputed product; (2) Nuance does not owe a royalty for telephone dictators; (3) Nuance underpaid royalties for user licenses because it failed to track and pay a royalty for every End User of a MASC-bearing product, as required by the TLA; (4) the TLA does not require the payment of renewal fees; (5) Nuance does not owe any royalties for Additional Site fees beyond those already paid; (6) Vianix is not entitled to application of a growth factor to approximate 2008 iChart sales; and (7) a 6 percent service and maintenance agreement fee and interest of 1.5 percent per month must be applied to all royalties that Nuance owes Vianix. To determine the amount of damages Nuance owes, I direct the parties to use these findings to populate the Excel spreadsheet created by Vianix's expert witness, Douglas Ellis, and then calculate the royalties Vianix is owed. Finally, I deny both of Nuance's counterclaims and decline the requests of both Vianix and Nuance for their respective attorneys' fees and costs in connection with this litigation.
I. BACKGROUND A. The Parties
Plaintiff, Vianix, is a Delaware limited liability company with its principal place of business in Virginia Beach, Virginia. Vianix is a wholly-owned subsidiary of Vianix, LLC ("Vianix Virginia"), which is a Virginia limited liability company. Vianix developed audio compression software known as "MASC Technology," or simply "MASC," an acronym for Managed Audio Sound Compression. Vianix markets MASC as a means of compressing audio files to smaller sizes to allow for easier storage and transmission while retaining high audio perceptual quality.
Vianix Virginia entered into the TLA, but then assigned its rights and obligations under the TLA to Vianix in October 2005. JX 188 ("TLA") at Consent to Assignment and Waiver. For the sake of simplicity, unless otherwise specified, I use the term "Vianix" to refer to both Vianix Virginia and Vianix.
Defendant, Nuance, is a Delaware corporation with its headquarters in Burlington, Massachusetts. Nuance is a worldwide leader in providing speech and imaging solutions to businesses and consumers. On or about March 31, 2006, Nuance acquired Dictaphone Corporation ("Dictaphone"), which three years earlier had entered into the TLA with Vianix. Before this acquisition, Dictaphone, headquartered in Stratford, Connecticut, was a leading provider of dictation and speech recognition products for the healthcare industry.
For brevity, I use the term "Nuance" to refer to both Nuance and Dictaphone unless otherwise specified.
B. Facts
In 2002, Vianix and Nuance began negotiating an agreement whereby Vianix would license MASC to Nuance. Reza Hashampour, Vianix's President and CEO, served as its lead negotiator, while David Pearah, Nuance's director of product marketing, and Donna Heffernan, Nuance's licensing manager, were the lead negotiators for Nuance. The parties exchanged several drafts of the TLA during negotiations. Vianix and Nuance signed the TLA in late January 2003 and made it effective as of January 20, 2003.To say the TLA is poorly drafted is an understatement. The Agreement is replete with ambiguous and inconsistent provisions, typos, and disjointed, incomplete paragraphs. The TLA grants Nuance a license to use MASC and integrate MASC into its speech recognition and dictation products, which were used primarily by people working in the healthcare industry. In exchange for this license, Nuance undertook to pay Vianix site fee and user fee royalties pursuant to Exhibit E of the TLA "for each MASC Technology Sublicense transaction that [Nuance or its] Distributors enter into with any [Nuance] Clients and/or End-Users." Exhibit E, entitled "Licensing Fees," contains a table labeled "Price Structure," which is reproduced below without the accompanying notes. Nbr. Of Existing Initial Cost / Site Nbr. Of Cost / End [Nuance] Client Seats User Sites
Id. at 1-2.
Id. § 8.1.
The first part of the table specifies an Initial Cost per Site owed by Nuance that decreases as the number of Nuance Client Sites increases. Exhibit E also calls for payment of a Cost/Additional Site at a discount of 40 percent from the Initial Cost/Site whenever a Nuance customer who has paid the Initial Cost/Site adds a subsequent site. The right half of the table in Exhibit E relates to Cost/End User, which varies with the number of "Seats." Exhibit E states that: "The Cost per End User shall be calculated in accordance with the number of per seat licenses cumulatively sold by [Nuance] during each year of this this [sic] Agreement." Thus, for example, the table indicates that for the first 2000 per seat licenses Nuance sold in a year, the Cost/End User would be $12.00 each.
Id. at 24.
The TLA also contains a merger/nonwaiver clause that provides:
This Agreement constitutes the entire agreement between the parties concerning the subject matter hereof. The failure of either party to require performance of any provision of this Agreement shall not be construed as a waiver of its rights to insist on performance of that same provision, or any other provision, at some other time. Any waiver, variation or amendment, or modification, of any term or condition of this Agreement shall be effective only if signed by an authorized representative of both parties hereto. The waiver by either party of any right created by this Agreement in one or more instances shall not be construed as a further continuing waiver of such right or any other right created by this Agreement.
Id. § 17.
The TLA had an initial term of three years. In January 2006, the parties amended the TLA to extend its term by three years so that it would terminate on January 20, 2009.
Id. § 7.
JX 69 at VIA012818-19.
On October 22, 2004, the IVS Addendum to the TLA went into effect. This Addendum relates to Nuance's Integrated Voice Systems or IVS business unit, which sold products to customers outside of the healthcare industry, such as lawyers and insurance companies. Nuance apparently licensed all of its IVS products on a concurrent or capacity basis. That is, there was a limit on the number of End Users who could access the system at any one time. The IVS Addendum provided for a modified royalty structure whereby Nuance would pay Vianix a flat-fee or fixed royalty for each IVS license sold. The royalty consisted of two parts: the Initial Cost/Site and the Cost/User. Instead of providing for a Cost/End User as in the TLA, the IVS Addendum provided that Nuance would owe a Cost per User, with the User license fees being calculated according to a schedule that assigned a presumed number of Users per Site to each of the several types of concurrent user licenses Nuance offered. For example, if a customer purchased an EX20 license, the user license fee would be calculated on the basis of 5 Users per Site, regardless of the number of actual End Users there were. Similarly, if a customer purchased an EX250 license, which authorized many more concurrent users than an EX20 license, the calculation would be based on 100 Users per Site. Because Nuance had to pay Vianix $20 per user of an IVS product, it would owe, in addition to a site fee, $100 for every EX20 license and $2,000 for every EX250 license.
T. Tr. 1283 (Pearah).
JX 193 at NUAN3645-47.
In March 2005, in recognition of a dispute over the site fees associated with Nuance's Physician Direct product, the parties agreed to the Physician Direct Addendum. This Addendum provides for payment of a higher Cost per User than under the TLA or the IVS Addendum, but no site fees. The Cost per User was to be paid based on the "actual number of users per site."
TLA at Physician Direct Addendum.
Id. ¶ 3.
By the beginning of 2007, Vianix grew concerned about Nuance's consistent late payments of royalties and apparent discrepancies between the sales growth Nuance was touting in press releases and the meager, in Vianix's view, royalties, it was paying under the TLA. Vianix's concerns were heightened by comments made by Nuance personnel, especially Simon Howes, Nuance's Vice President of Engineering, indicating that MASC was in Nuance's "core technology." This prompted several telephone conversations in the spring and summer of 2007 where Nuance attempted to explain the role of MASC in its products and the parties debated which products contained MASC. Unbeknownst to Nuance, Vianix recorded these conversations and, in fact, recorded conversations with Nuance until shortly before it commenced this litigation. At the time of trial, three of Nuance's witnesses, Heffernan, Howes, and Raghu Vemula, had personally sued Hashampour and Vianix for making these recordings. During these recorded conversations, Vianix repeatedly threatened to audit Nuance.
T. Tr. 157-60 (Hashampour).
JX 285-a.
T. Tr. 112 (Hashampour).
Id. at 111 (Hashampour); JX 276 — JX 291.
JX 605; JX 606.
T. Tr. 172 (Hashampour).
On June 13, 2007, Vianix followed through on its threats and hired the Invotex Group ("Invotex") to conduct an audit of Nuance pursuant to a provision in the TLA. The objective of the audit "was to provide Vianix with a report on the degree of compliance by Nuance . . . with respect to the [TLA]." Debora Rose Stewart headed the audit for Invotex. In performing the audit, Invotex visited Nuance's offices on two occasions, in September and December 2007, for a few days each time. Invotex issued its final audit report in February 2008, in which it found that Nuance had not fully complied with the TLA and had underpaid Vianix by $2,488,218 ($2,009,604 in royalties plus $478,614 in accrued interest), a number that "would increase if it were determined that there is more than one end user per concurrent license." The nearly $2.5 million underpayment Invotex found did not include "amounts due Vianix for Nuance sales of (1) Power Scribe and EX Speech for the period of January 1, 2007 — December 31, 2007, (2) Enterprise, IVS, and OEM SDK sales for the period April 1, 2007 — December 31, 2007 and (3) I-Chart sales for the period December 13-21, 2007."
JX 168 at 2. Section 8.3 of the TLA provides: "Upon reasonable notice, during normal business hours, but no more than once annually, VIANIX shall have the right to audit the applicable systems, books and records of [Nuance] only as they relate to [Nuance's] obligations under this Agreement."
Id.
Id. at 3.
Id. at 4.
Id.
On February 22, 2008, Vianix issued its Invoice #215 to Nuance. Referring to the information made available by Nuance in the audit, the Invoice purported to charge Nuance over $12.6 million for allegedly outstanding royalties, service and maintenance agreement fees, license renewal fees, and interest. The $12.6 million figure also included the cost of the Invotex audit. On March 18, 2008, Vianix sent Nuance Invoice #215-R, which reduced the amount charged to Nuance to slightly over $12.4 million.
JX 56.
JX 58.
The parties agreed to terminate the TLA in December 2008. To date, Nuance has paid Vianix over $1.2 million in royalties under that Agreement.
T. Tr. 289 (Hashampour).
DX 1.
C. Procedural History
Vianix filed its Complaint on June 2, 2008. The Complaint contains two claims, one for breach of contract and the other for injunctive relief. On August 4, 2008, Nuance filed its Answer and Counterclaims, which asserted claims for breach of contract and a declaratory judgment that it has complied in full with its payment obligations under the TLA. Although Nuance filed a First Amended Answer and Counterclaims on December 15, 2008, which added counterclaims for fraud in the inducement and rescission based on an allegation that Vianix stole the source code for MASC, it effectively withdrew those additional counterclaims when it omitted them from its Second Amended Answer and Counterclaims filed on January 14, 2009. Vianix answered the remaining counterclaims on January 27, 2009.The Court conducted a six-day trial from November 12 to 19, 2009. After the parties filed extensive post-trial briefs and proposed findings of fact and conclusions of law, the Court heard post-trial oral argument on April 20, 2010.
D. Parties' Contentions
The parties make myriad contentions. Vianix contends that: (1) virtually all healthcare-related products that Nuance marketed contained MASC; (2) Nuance owed royalties for its telephone dictators; (3) the TLA required Nuance to pay a user fee royalty for every end user of a product containing MASC; (4) the TLA required Nuance to pay a three-year renewal fee on all user licenses; (5) Nuance owed an Additional Site fee royalty for every nonphysically contiguous facility operated by a Nuance customer; (6) Nuance's counterclaims are without merit and should be denied; and (7) it should be awarded its attorneys' fees and costs as the prevailing party. Assuming it succeeds on all of these points, Vianix claims that Nuance owes it more than $30 million in damages, interest, and attorneys' fees and costs.For its part, Nuance contends that: (1) many of its products that Vianix claims contain MASC do not, in fact, contain MASC; (2) no royalty is owed for telephone dictators because those dictators do not use MASC; (3) the TLA only requires payment of a royalty for every license Nuance sold, rather than every end user of its products; (4) the TLA does not require payment of a renewal fee; (5) Additional Site fee royalties are owed only when a customer purchases a second set of Nuance servers; (6) it has succeeded on its counterclaims for breach of contract and a declaratory judgment; and (7) it is the prevailing party and, thus, entitled to its attorneys' fees and costs.
II. ANALYSIS A. Overview of the Damages Award
The parties agree that no choice of law analysis is required here because the laws of the home states of the original parties to the TLA, Virginia for Vianix Virginia and Connecticut for Dictaphone, are substantially identical. Pl.'s Conclusion of Law 1; Def.'s Conclusion of Law 2. To the extent practicable, I have relied on relevant precedent from both Virginia and Connecticut.
To aid in my consideration of potential damages, I have referred to a damages calculation tool (the "Damages Spreadsheet") created by Vianix's expert witness, Douglas Ellis. The Damages Spreadsheet is a Microsoft Excel spreadsheet that contains data entry points relating to every disputed issue in this case. In ruling upon the various issues before me, I will be determining, in essence, how a cell or cells in the Damages Spreadsheet will be populated with data. Many of my decisions will relate to whether a specific cell should contain any value at all. For instance, one decision point will be whether EX Speech, a server that is a part of the Enterprise Express or EX system, contains MASC. If I determine that EX Speech contains MASC, the cell pertaining to EX Speech will be populated with the number of EX Speech licenses Nuance sold. If, on the other hand, I determine that EX Speech does not contain MASC, a zero value will be entered into that cell to signify that no royalty was owed for that product. Other decisions will affect the final calculation in a more complex manner and may involve the application of a multiplier to certain values to account for lapses in Nuance's recordkeeping. When all the relevant cells are populated with data, the Damages Spreadsheet should yield the total amount of damages that Vianix is owed.
Nuance objects to the use of the Damages Spreadsheet, but I find that its objections lack merit. While Nuance does not object to the structure of the Spreadsheet or the substance of any of its calculation formulas, it contends that the Spreadsheet contains several biases that work in Vianix's favor. For example, Nuance objects to the Spreadsheet's use of three multipliers, one to approximate growth in iChart product sales in 2008, one to estimate the number of Additional Sites, and one to estimate the number of End Users. According to Nuance, use of these multipliers will flood the Damages Spreadsheet with hypothetical data. Nuance's concerns are unfounded, however, and appear to relate more to the Spreadsheet's intended capability to accommodate both parties' views of the case. In terms of the multipliers, Nuance's complaints about the Damages Spreadsheet amount to no more than an argument that no multipliers should be applied here. If Nuance is correct in that regard, I will order that no multiplier be entered in the designated multiplier cells, and Nuance's objection will be mooted. If, however, I adopt Vianix's view of the case and determine, for instance, that Nuance breached the TLA by failing to collect data on End Users or noncontiguous facilities, it may be necessary to use estimated data, including multipliers, to make up for Nuance's inability to produce actual data. Thus, I find it appropriate to use the Damages Spreadsheet in this situation.
The parties also disagree as to which set of data I should use to populate the Damages Spreadsheet. Vianix urges use of a hybrid data set consisting of Nuance press-releases relating to Nuance's iChart product, the data from the Invotex audit, and a compilation of data created in November 2009 by Jeffrey Bourassa, the head of Nuance's compliance department (the "Bourassa Data"). Vianix urges the Court to look to the Invotex audit data as the primary data source and supplement it with a June 2007 Nuance press release for data on iChart users and the Bourassa Data for data on certain products for which Invotex apparently did not collect data during the audit. Nuance, on the other hand, argues that I should rely exclusively on the Bourassa Data.
JX 168.
JX 562. Bourassa compiled the Bourassa Data by combing through three separate Nuance databases to derive a complete set of Nuance's sales data for the relevant period. T. Tr. 1386-96.
PX 120.
Generally, I agree with Nuance on this issue. Specifically, I find the Bourassa Data to be the most reliable and complete data set in the record. Moreover, there is no reason to use the Invotex audit data over the Bourassa Data. Although the two data sets are similar, the Bourassa Data is more recent and more complete. Likewise, there is no reason to use the June 2007 iChart press release as a data source. This press release touts the number of healthcare providers who had used iChart up to that point in time, but it appears to be mere marketing puffery and to include iChart users who used the product before it contained MASC. As a result, Vianix has failed to prove a connection between the number of users cited in the June 2007 press release and the actual use of MASC. Therefore, I accord this document (PX 120) limited weight for purposes of determining the royalties Vianix is owed.
See JX 168 at 4.
PX 120.
Exclusive use of the Bourassa Data may not make Vianix entirely whole, however. As previously noted, Nuance did not track certain data that may be relevant in determining the royalties owed to Vianix. Consequently, the Court may need to apply a multiplier, for example, to the data that is available to obtain a responsible estimate of damages that is supported by the record. In any event, if at any point I decide not to use the Bourassa Data as the input for the Damages Spreadsheet, I will specify what exactly should be used. Unless so specified, the Damages Spreadsheet will be populated with the Bourassa Data.
Nuance also complains that Vianix has not submitted sufficient evidence to justify any award of damages. According to Nuance, nothing in the record provides the Court with a basis for awarding damages without improperly resorting to speculation, particularly with regard to the multipliers Vianix seeks for End Users and additional sites and Vianix's decision not to elicit testimony from Ellis or another expert as to the amount of the damages. I disagree. While a plaintiff cannot recover damages that are "merely speculative or conjectural," the law requires only that there be a sufficient evidentiary basis for making a fair and reasonable estimate of damages, rather than absolute certainty. In a context analogous to the situation here, the United States Supreme Court stated almost 80 years ago that:
Kronenberg v. Katz, 872 A.2d 568, 609 (Del. Ch. 2004) (quoting Henne v. Balick, 146 A.2d 394, 396 (Del. 1958)); see also S. New Eng. Tel. Co. v. Henkels McCoy, Inc., 1996 WL 456977, at *4 (Conn. Super. Ct. July 17, 1996) ("[D]amages may not be based on speculation, conjecture, or sympathy. While mathematical exactitude is often impossible when proving damages, the plaintiff bears the burden of providing sufficient evidence for the trier to make a fair and reasonable approximation."); Shepherd v. Davis, 574 S.E.2d 514, 524 (Va. 2003) (Plaintiffs had the "burden of proving with reasonable certainty the amount of damages and the cause from which they resulted; speculation and conjecture cannot form the basis of the recovery.").
Henne, 146 A.2d at 396.
Where the tort itself is of such a nature as to preclude the ascertainment of the amount of damages with certainty, it would be a perversion of fundamental principles of justice to deny all relief to the injured person, and thereby relieve the wrongdoer from making any amend for his acts. In such case, while the damages may not be determined by mere speculation or guess, it will be enough if the evidence show the extent of the damages as a matter of just and reasonable inference, although the result be only approximate. The wrongdoer is not entitled to complain that they cannot be measured with the exactness and precision that would be possible if the case, which he alone is responsible for making, were otherwise.
Story Parchment Co. v. Paterson Parchment Paper Co., 282 U.S. 555, 563 (1931) (citing Eastman Kodak Co. v. S. Photo Mat'ls Co., 273 U.S. 359, 379 (1927)). Although Story Parchment involved a tort case, as opposed to a breach of contract action, the same reasoning applies to the extent Nuance is responsible for the absence of necessary data.
Here, Nuance's failure to collect certain data arguably could prevent the Court from making a responsible estimate of damages. Before reaching such a conclusion, however, I need to consider the entire record before me. While the Court must be careful to avoid speculation, the absence of mathematical exactitude does not preclude an award of damages. As for the absence of any testimony on Vianix's behalf from a damages expert, I find plausible Vianix's protestations about the difficulty, if not the impossibility, of finding a damages expert who could have testified helpfully on the quantification issues in this case. While the absence of such evidence may limit the award Vianix ultimately receives, I find that there is sufficient evidence to support at least some damages award in this case. Thus, I reject Nuance's contention that any award of damages would be unduly speculative.
B. Which Nuance Products Contain MASC?
The TLA provides that Nuance "shall pay to VIANIX license fees as set forth in
Exhibit E for each MASC Technology Sublicense transaction" that Nuance enters into. As per the TLA, "SUBLICENSE: shall mean the granting by or on behalf of [Nuance] a sublicense to an End-User to use the MASC Technology; the Sublicense to include certain rights and limitations specified in this Agreement." Accordingly, whether a Nuance product contains MASC is the touchstone for determining whether Nuance owes Vianix a royalty when it grants a Sublicense of that product.
TLA § 8.1.
Id. § 1.18.
1. Products which Nuance concedes contain MASC
The parties agree that the following Nuance products contain MASC and are royalty-bearing: (1) EX Voice versions 6.5 and 7.0; (2) Transnet; (3) EWS version 2.0 (self-edit configuration) and version 3.0 (all configurations); (4) Powerscribe SDK versions 1.4 through 2.1; (5) JobLister; (6) Olympus Uploader; (7) IVS; and (8) iChart. Accordingly, the Damages Spreadsheet should be populated with the Bourassa Data on the number of licenses Nuance sold and the quarter in which those sales were made for each of these products.
T. Tr. 1231 (Howes).
Id. at 1192 (Howes).
Id. at 1227 (Howes).
JX 83 at VIA009976.
T. Tr. 1192 (Howes).
Id. at 1495 (Shroff).
JX 83 at VIA009976. I reject, however, Vianix's contention that the Court should find that Nuance upgraded every IVS customer to a MASC-bearing system as soon as MASC was incorporated into IVS. The Bourassa Data includes information on when IVS upgrades to MASC occurred, both pre-and post-audit. JX 562 at NUAN153358-64. While the number of upgrades to MASC shown in the Bourassa Data is small, Vianix has the burden of proving the actual number of upgrades, and it failed to show the Bourassa Data are incorrect. In that regard, I find the two documents Vianix cites on this topic inconclusive as to whether Nuance upgraded all IVS customers. See JX 259; JX 354.
JX 83 at VIA009976. iChart is a dictation solution targeted to smaller entities. iChart uses the EX servers, but the customers do not purchase these servers. Rather, the customer is billed on a per transaction basis, with a fee charged based on the number of lines dictated or transcribed. There were two configurations of iChart: on-site, where the EX servers were located at the customer's facility, and hosted, where customers used servers located at the Nuance data center. T. Tr. 1439-41 (Shroff).
2. Powerscribe Workstation Radiology
In its opening post-trial brief, Vianix concedes that Nuance's Powerscribe
Workstation Radiology product does not contain MASC. Thus, a value of zero should be entered into the Damages Spreadsheet for this product.
Pl.'s Opening Br. ("POB") 23.
3. EX Voice version 6.4
EX Voice is a server that is an essential part of the Enterprise Express, or EX, system, Nuance's primary healthcare dictation product. The EX system could be configured to contain multiple computer servers, including EX Text and EX Speech, but always contained EX Voice. EX Voice receives and stores voice files for later processing. The parties agree that EX Voice version 6.4 contained MASC, but Nuance argues that this product is not royalty-bearing because the MASC it contained not only was nonoperational, but had to be disabled to avoid causing the system to crash. This raises the question of whether Nuance owed a royalty whenever one of its products contained MASC, even if the MASC in that product could not be used.
T. Tr. 457 (Heffernan).
Id. at 1184-86, 1198-99 (Howes). Because a Nuance customer could never own EX Text or EX Speech without owning EX Voice, Nuance contends that it cannot owe separate royalties for EX Text or EX Speech on top of the royalties owed for EX Voice, based, apparently, on the view that Nuance sold a single license for the EX system as a whole, rather than individual licenses for each component of the EX system, such as EX Text and EX Speech. While the parties presented little evidence on the topic of how Nuance licensed its EX system and its individual components, I consider Nuance's argument at least plausible. In any event, I find, infra Parts II.B.4 and II.B.5, that neither EX Text nor EX Speech contained MASC; thus, neither product is royalty-bearing for that reason.
T. Tr. 1231-32 (Howes); Dep. of Christopher Adams ("Adams Dep.") 118; JX 330 at NUAN-22622 (EX Voice version 6.5 "can play Vianix without crashing.").
The TLA requires the payment of royalties "for each MASC Technology Sublicense transaction" Nuance entered into. As defined in the TLA, Sublicenses were granted "to an End-User to use the MASC Technology." Nuance knew before it released EX Voice version 6.4 that the MASC in that product did not function properly and disabled MASC to avoid use of it. Because MASC was disabled in EX Voice version 6.4 and customers could not use MASC in this product, the licenses Nuance granted for EX Voice version 6.4 arguably were not Sublicenses under the TLA. Yet, I need not decide that issue because Vianix conceded at argument that the EX Voice 6.4 issue was so inconsequential that the Court could ignore it in determining the amount owed to Vianix. Thus, I will assume for purposes of this Opinion that Nuance's sales of licenses for EX Voice version 6.4 were not MASC Technology Sublicense transactions and, hence, not royalty-bearing under the TLA. Therefore, a value of zero should be entered into the Damages Spreadsheet for EX Voice version 6.4.
TLA § 8.1.
Id. § 1.18.
T. Tr. 1231 (Howes) ("[J]ust prior to releasing it, we discovered a major problem between the known software and the MASC technology. And while we waited to resolve that, we basically disabled the MASC technology. So in 6.4, all our MASC was dormant in the system. It was not used, and all the features were turned off."); Adams Dep. 118.
Arg. Tr. 22, 33.
4. EX Speech
The EX Speech server performs automated speech recognition, converting an audio file into a draft text file of what was dictated. Vianix's argument that EX Speech contains MASC is based on a set of instructions found on Nuance's website for removal of MASC from something called the EXSpeech System. Nuance avers that the EXSpeech System referenced in the instructions is different from the EX Speech server and refers to the entire Enterprise Express system, which also includes EX Voice and, thus, indisputably contains MASC. While some uncertainty exists regarding the import of the instructions, the rest of the evidence cited by both parties makes it clear that the EX Speech server itself does not contain MASC. Aside from the MASC removal instructions, all the evidence cited by Vianix indicates that files destined for the EX Speech server were converted from MASC files to another type of file before they were transferred to EX Speech, which is consistent with Nuance's allegation that EX Speech could not accept MASC files. Because this evidence is credible and convincing, I find that EX Speech did not contain MASC. Accordingly, I mandate that a zero value be entered into the Damages Spreadsheet in the cell designated to record the number of licenses sold for EX Speech.
PX 110.
JX 136 ("Vianix files will be rejected in EXSpeech," so MASC files needed to be converted "to 64K mu-law for EXSpeech users only, so that they can go through EXS[peech]."). Nuance employees used Vianix and MASC interchangeably when referring to MASC-encoded files. E.g., JX 141 ("Record natively in desktop audio, compress to Vianix, send to EXV[oice], decompress from Vianix, send to EXS[peech].").
T. Tr. 1200-01 (Howes); JX 14; JX 136; JX 142.
5. EX Text
EX Text is a server that stores and transmits text files and patient data. An EX
Text server does not itself accept any voice files. EX Text generally works in tandem with Transnet or Integrated Transnet to send patient data to transcriptionists and receive final drafts of transcribed text from transcriptionists. Based on their close relationship to EX Text, Transnet, Integrated Transnet, and two elements of Integrated Transnet, Correction Client and Word Client, sometimes are referred to as components or clients of EX Text.
T. Tr. 1214-15 (Howes).
Demonstrative Ex. ("Dem. Ex.") 3 at 32. Nuance presented several Demonstrative Exhibits at trial that I refer to in this Opinion. Those Demonstratives indicate, at a general level, how certain of Nuance's products operate in relation to the MASC Technology. For the most part, the descriptions reflected in these Demonstratives are undisputed and supported by the testimony of one or more of Nuance's witnesses. Furthermore, Vianix, which has the burden of proof as to the allegedly inadequate royalty payments, did not submit any evidence supporting a different description with the exception of the disputes described infra, Part II.C, regarding the alleged use of MASC in connection with doctors who use telephone dictation. Thus, unless otherwise indicated, I have inferred that the various Nuance products operate as indicated in the Demonstrative Exhibits cited in this Opinion.
See Pl.'s Pretrial Br. 29; Adams Dep. 90 ("Word Client and Correction Client are part of [EX] Text.").
All of the documents cited by Vianix for the proposition that EX Text contains MASC aver that assorted EX Text clients or components ( i.e., Transnet and Integrated Transnet, as well as something colorfully called the "didgeridoo player") contain MASC. Vianix has not produced, however, any evidence that the EX Text server itself contains MASC, and, in light of the fact that the EX Text server never comes into contact with audio files, it is reasonable to infer that this product does not contain MASC. Accordingly, I so find and order that a zero value be entered into the Damages Spreadsheet in the cell designated to record the number of licenses sold for that product.
JX 10 at NUAN70795 ("Vianix expansion (i.e. playback) within the EX Text client."); JX 78 ("Vianix MASC is on all C[orrection] C[lient] 1.3 clients with EXT[ext] 6.x server software."); JX 134 at NUAN6585 (as a "Feature Specification" of EX Text, "[a]dopt Vianix codec [or technology] on all EX Text clients (including Transcription Editor, Correction Editor) that are expected to playback audio to a user"); Adams Dep. 109 (the didgeridoo player is "an audio player that the Enterprise Text team developed to play back audio," including MASC files).
6. EWS 2.0 Dictate and E-Sig
Enterprise Workstation ("EWS") is a "front-end browser-based solution that offers physicians flexible input and editing options, to reduce report turn-around time and the costs associated with transcription." In effect, EWS allows a doctor to perform various functions relating to dictation on a computer, depending on what configuration of EWS they own. Nuance offered three different configurations: (1) dictate; (2) self-edit; and (3) E-signature ("E-sig"). In the dictate configuration, a doctor can create a voice file through her computer (by, for example, speaking into a microphone) and send that voice file to the EX servers for further processing. The self-edit configuration includes front-end speech recognition software and allows a doctor to dictate directly into a computer and see, through the speech recognition function, text on the screen in front of her in real-time. E-sig allows a doctor to review text files that previously were transcribed and electronically sign these documents. While the parties agree that the self-edit configuration of EWS version 2.0 and all configurations of EWS version 3.0 contain MASC, they dispute whether the dictate and E-sig configurations of EWS version 2.0 contain MASC.
Pre-Trial Order 24.
Id.
Dem. Ex. 3 at 29.
Id. at 32.
Id. at 17.
Vianix bases its argument that the disputed products contain MASC on a number of internal Nuance communications in which various engineers made statements regarding the presence of MASC in EWS version 2.0. Nuance contends that any statement implying that MASC was in all configurations of EWS version 2.0 was based on a misunderstanding about the presence of MASC in the various configurations of EWS version 2.0. Nuance also introduced evidence of a clean install test a Nuance engineer ran on all configurations of EWS version 2.0, which found that the only configuration that contained MASC was self-edit. Multiple Nuance engineers attributed the allegedly mistaken beliefs about the presence of MASC in the dictate and E-sig configurations to tests that were run on machines on which MASC had been installed that also ran EWS version 2.0 self-edit.
JX 42; JX 76; JX 79; JX 147.
T. Tr. 1228 (Howes), 1320 (Vemula); JX 391.
T. Tr. 1229 (Howes); JX 391.
I find Nuance's evidence on this topic credible. There was clearly a great deal of confusion at Vianix over whether EWS version 2.0 contained MASC. Importantly, only one of the emails cited by Vianix contains a direct statement that all configurations of EWS version 2.0 contain MASC. The other emails merely include vague assertions about MASC being in EWS version 2.0 without any mention of its different configurations. Even the email from Howes, Nuance's Vice President of Engineering, that directly states that MASC is in all configurations of EWS version 2.0 is based on a statement from Adams, a senior director of engineering at Nuance, that MASC is "[a]lso in EWS 2.0 and later," another vague assertion that makes no mention of EWS's various configurations. As such, all of the statements upon which Vianix relies reasonably can be attributed to a lack of precision on the part of the speaker.
JX 386; JX 391.
JX 76.
JX 42; JX 79; JX 147.
JX 76. At trial, Howes testified that his statement in this email was incorrect and the result of his mistaken belief about the presence of MASC in EWS version 2.0. T. Tr. 1228.
The best evidence that the dictate and E-sig configurations did not contain MASC is the clean install test performed by Nuance in August 2007. Vianix dismisses the clean install as a hoax, but provides no evidence to back up its claim. More importantly, nothing prevented Vianix from conducting its own test and presenting competent expert testimony on this issue, but it failed to do so. Moreover, in a conversation with Vianix on July 2, 2007, two months after Howes's email was sent, Vemula, one of Nuance's chief engineers, explicitly stated that MASC was in only the self-edit configuration of EWS version 2.0. Having considered all the relevant evidence, I find that Nuance has made the stronger showing on this issue and that Vianix has not proven that either the dictate or the E-sig configuration of EWS version 2.0 contained MASC and, thus, was royalty-bearing. Accordingly, a zero value should be entered into the Damages Spreadsheet for these products.
JX 391.
JX 279-a at 22-24.
7. Grundig 9000 and Philips Walkabout
Vianix conclusorily asserts that two DSS uploaders that accept voice dictations in a digital format for transfer to a computer, the Grundig 9000 and the Philips Walkabout, contain MASC. At least two Nuance witnesses testified to the contrary. As Vianix cited no evidence tending to show that either the Grundig 9000 or the Philips Walkabout contains MASC, I find that neither of those products is royalty-bearing. Accordingly, a value of zero should be entered into the Damages Spreadsheet for those products.8. Personal JobLister
Nuance concedes that Personal JobLister, or "Lucid" as it is known internally at Nuance, contains MASC. Personal JobLister is sold exclusively as a part of Nuance's IVS product, however, for which there is a flat-fee royalty structure separate from the standard royalty structure for Nuance's healthcare products. Accordingly, royalties for sales of Personal JobLister are covered by the flat IVS royalty fee, and Vianix is not entitled to anything beyond that. Because there is no allegation that Nuance failed to pay the royalties due for its sales of Personal JobLister under the IVS Addendum, a value of zero should be inserted in the Damages Spreadsheet for this product.
Def.'s Opening Br. ("DOB") 19; T. Tr. 510 (Heffernan).
T. Tr. 509 (Heffernan); JX 193 at NUAN3647.
9. GoMD
GoMD, also known as iPaq, is a product that allows dictations recorded on a handheld personal digital assistant device to be uploaded onto a computer. Vianix contends that GoMD contains MASC based on several emails that discussed tentative plans to incorporate MASC into that product. Nuance avers that it never actually implemented these plans or incorporated MASC into GoMD, an allegation that the evidence corroborates. Anuj Shroff, a Nuance director of engineering, testified at trial that GoMD never contained MASC. In addition, the most recent email cited by Vianix for the converse of that proposition still refers to Nuance being "tentatively tasked with doing . . . Vianix compression" in GoMD and expresses concern about the compatibility of such a product with EX Voice and EX Speech. Based on this equivocal evidence and Vianix's failure to produce any direct evidence from an inspection of the product or otherwise that Nuance ever actually integrated MASC into GoMD, I find that Vianix has not shown that GoMD ever contained MASC. Accordingly, a value of zero should be entered into the Damages Spreadsheet for this product.
Howes Dep. 103.
JX 7 — JX 9; JX 353.
T. Tr. 1495.
JX 8.
10. Correction Client and Word Client
Correction Client and Word Client are PC applications used by transcriptionists. Correction Client allows a transcriptionist to proofread and correct draft transcriptions of voice files that have been created by speech recognition software. Word Client plays a voice file that has not gone through speech recognition to allow a transcriptionist to transcribe the file by listening to it. When Correction Client and Word Client were combined with Transnet, which contains MASC, in a product known as Integrated Transnet, that product became royalty-bearing because of the presence of MASC in Transnet. Otherwise, neither Correction Client nor Word Client contains MASC, as confirmed by one of the documents relied upon by Vianix. Vianix and Nuance even agreed at one point that Nuance was double-paying royalties to the extent it counted both Integrated Transnet and Correction Client as royalty-bearing because only the Transnet portion of Integrated Transnet contained MASC. Therefore, no royalty is owed for either Correction Client or Word Client beyond that which is due for the Integrated Transnet product. Accordingly, a zero value should be inserted into the Damages Spreadsheet for these products.11. Powerscribe SDK version 3.0
Powerscribe SDK is a software development kit Nuance sold to its Original Equipment Manufacturer ("OEM") customers that allowed them to create and sell products incorporating Powerscribe SDK and its speech recognition capability. The parties agree that Powerscribe SDK versions 1.4 through 2.1 contain MASC and version 3.0 does not. A dispute exists, however, as to whether a royalty accrued when Nuance shipped an optional disc containing MASC to Powerscribe SDK version 3.0 users who previously had purchased a MASC-containing version of SDK to make version 3.0 backwards compatible. Vianix contends that the shipment of this disc generates a royalty, while Nuance asserts that it does not.The TLA dictates that a royalty is owed whenever Nuance enters into a MASC Technology Sublicense transaction and defines a Sublicense as a license granted to an End-User "to use the MASC Technology." Here, Nuance shipped discs containing MASC to its OEM customers so these customers could use MASC, as the disc allowed the OEM customers to continue to incorporate MASC into their products. Thus, the shipping of the disc along with Powerscribe SDK version 3.0 constitutes a MASC Technology Sublicense transaction for which Nuance must pay a royalty. Accordingly, the Damages Spreadsheet should be populated with the Bourassa Data, to the extent it exists, for the number of optional MASC-containing discs that Nuance shipped with Powerscribe SDK version 3.0.
TLA § 8.1.
Id. § 1.18.
Vianix asserts that all but one of Nuance's preexisting OEM customers at the time of the release of Powerscribe SDK version 3.0 requested the optional disc, but cites no evidence in support of this assertion. Pl.'s Finding of Fact 26. To the extent the Bourassa Data does not include accurate and complete information on the number of MASC discs Nuance shipped, I will accept Vianix's assertion, as it was not challenged by Nuance, and mandate that the number of preexisting OEM customers at the time of version 3.0's release, less one, be used in the Damages Spreadsheet for Powerscribe SDK version 3.0.
On a related note, Vianix also complains that Nuance failed to properly track site and user fees for its OEM customers. Vianix's claim about the lack of sufficient tracking of site fees rings hollow. There is no evidence (1) that Nuance failed to collect site fees from OEM customers who purchased Powerscribe SDK and submit those fees to Vianix or (2) that Nuance was required to pay a site fee when an end user purchased a product that contained Powerscribe SDK from an OEM customer.
There is, however, at least some indication that Nuance failed to pay required user fees for Powerscribe SDK. Nuance admits that it was required to pay a user fee for at least each license that an OEM customer sold to an end user. Vianix only received these user fee royalties to the extent that OEM customers self-reported sales of products containing a MASC-bearing version of Powerscribe SDK to Nuance. Moreover, it appears that Nuance did not do an adequate job of obtaining this information from its OEM customers. To remedy this deficiency, Vianix seeks a judgment requiring use in the Damages Spreadsheet of the data it obtained through subpoenas to Nuance's OEM customers. I conclude this data should be used to the extent it accurately reflects royalties owed under the TLA. Also, to the extent it exists, the Bourassa Data on the number of licenses sold by OEM customers should be entered into the Damages Spreadsheet in addition to the data obtained via subpoena, provided it does not duplicate that data.
T. Tr. 514-15 (Heffernan).
See PX 206 — PX 210.
One subpoenaed company, Siemens, provided a count of End Users. The other subpoena responses provided only the number of user licenses issued. In this latter case or any other situation where information on End Users is not available, the number of user licenses should be used in the absence of any more reliable information.
C. Is Nuance Required to Pay a Royalty for Telephone Dictators?
The parties strenuously disputed the next issue: whether doctors who dictate by telephone use MASC and, thus, qualify as End Users such that Nuance would owe a royalty for each such doctor who uses a Nuance product. Vianix contends the answer is yes, based on three pieces of evidence. First, Vianix relies on a statement made in an April 14, 2006 email by Adams, Nuance's senior director of engineering, explaining the value of using MASC:
Vianix gets us around (I believe) 16K-18K (8KHz/11KHz). For the most part, [Nuance] has used 32K audio files, and Vianix effectively cuts that in half. This is nice for transmission, especially at the data center with all the transcribers downloading massive amounts of files every day. It's also nice on the web dictation side, because we have less to transmit to the center.
JX 36.
Vianix asserts that the massive amount of audio files Adams referred to must be coming from telephone dictators because, according to Nuance, the vast majority of its customers are telephone dictators. Therefore, the benefit referred to in Adams's statement would be realized only if MASC were used with files created by telephone dictators.
DOB 13, 23.
Nuance, on the other hand, alleges that telephone dictators cannot create MASC files and the EX system never converted a telephone dictation file to MASC. In that regard, Nuance emphasizes Adams's deposition testimony that his remarks about the advantages of MASC were purely theoretical in nature. At his deposition, Adams also stated repeatedly that telephone dictations were never converted to MASC.
T. Tr. 1223 (Howes), 1460-61 (Shroff).
Adams Dep. 143-44.
Id. at 37, 143-44.
I find Adams's testimony difficult to square with his April 14, 2006 email and, therefore, am skeptical of it. He might have been speaking hypothetically as he claims, but the language he used certainly does not indicate that. By the same token, the quoted passage does not state directly and unequivocally that Nuance products actually converted telephone dictations into MASC. In evaluating this evidentiary conflict, I am mindful that Vianix has the burden of proof on this issue. It knew about the purported conflict between Adams's email and his deposition testimony, but made no attempt to call him as a witness at trial. More importantly, the disputed fact is one that should be objectively verifiable through discovery. There is no reason to believe that Vianix could not have determined for itself from examining Nuance's systems or those of its customers whether MASC ever was used with files created by telephone dictators. Yet, Vianix failed to present such evidence or its own technical expert on the issue. In these circumstances, I find the statement in Adams's email to be ambiguous and that he might have been speaking theoretically. Thus, while that statement may provide some evidence that MASC was used on files created by telephone dictators, it is not enough, on its own, to support a finding that this was the case.
Vianix next points to a statement from Howes explaining why the EX servers contained MASC:
So the MASC CODEC on the server there is another — you don't see it on here because it's too low a level of detail, but there is a management client, right, where an administrator, for instance, could go into that particular system and look at all the work that's in the system. And if they decided to right click and listen to that job, we would then — we would then play that Vianix file out for an administrator to listen to. And that playing of the file or converting the file is done in the central system, so we need a copy of Vianix just to do that — to support that monitoring and management tool.
A "codec" is a computer program capable of encoding and decoding a digital data stream. The word is a portmanteau of compression-decompression. T. Tr. 27 (Hashampour). Thus, the MASC codec is capable of encoding a file into MASC or decoding it from MASC to another format.
Howes Dep. 152-53.
Vianix argues that Howes's statement refers to a process by which audio files recorded by telephone dictators are converted to MASC files that are then transmitted to administrators and transcriptionists. But, the statement makes no reference to telephone dictators. I must examine, therefore, the context in which it was made.
Howes made the statement quoted above in response to a line of questions regarding the ability of the EX servers to handle files created by computer users that were already in MASC format. In response to an earlier question, Howes stated that, for files being sent to transcriptionists, "if the voice originated through the telephone, the form of those files would be ADPCM." This statement affirmatively indicates that, at least initially, audio files created by telephone dictators were not MASC files. In addition, the presence of MASC in EX servers such as EX Voice can be explained by the need to convert MASC files to a different format in certain circumstances, such as when a file was destined for the EX Speech server, which could not handle MASC files. Accordingly, Howes's statement does not support Vianix's contention that telephone dictators used MASC.
Id. at 151-52.
Id. at 151. ADPCM is a similar sound compression technology to MASC. T. Tr. 1179-80 (Howes).
JX 136.
The third piece of evidence Vianix cites is a Nuance demonstrative exhibit that arguably shows an analog voice file going between a dictating doctor and the EX Voice server and then a MASC file going out of the EX Voice server to JobLister and Transnet. Vianix claims that this exhibit accurately depicts what happens to files created by telephone dictators: they are converted to MASC in the EX Voice server before being sent to a transcriptionist. The demonstrative exhibit, however, does not provide a detailed depiction of the path a voice file created by a telephone dictator takes in terms of whether or not it is ever converted to MASC at the EX server. Instead, the focus of the exhibit is on another input option, showing the path taken by a file created in MASC format by the self-edit configuration of EWS version 2.0.1. The bubbles numbered 1 through 6 on the exhibit obviously are intended to illustrate the path that data related to a MASC voice file take on the paths shown from the computer containing EWS self-edit to the SDK server and EX Voice server, then to JobLister and Transnet, and finally to the EX Text server. There is no numbered bubble on the path from the doctor dictating by telephone to the EX Voice server, which is consistent with the fact that the demonstrative exhibit does not focus on that path. A previous exhibit did illustrate the path taken by a file created by a telephone dictator. As Nuance's counsel who oversaw the creation of the exhibit conceded, the presence of the telephone dictator on Demonstrative Exhibit 34, page 2 is somewhat confusing. The exhibit, however, does not constitute persuasive evidence that audio files created by telephone dictators are ever converted to MASC.
Dem. Ex. 34 at 2.
See id. at 12, 17.
Arg. Tr. 46-47.
In sum, Vianix has failed to produce sufficient evidence to support a finding that telephone dictators used MASC. Accordingly, I find that Vianix has not met its burden of proving by a preponderance of the evidence that telephone dictations ever were converted to MASC and, therefore, were royalty-bearing. I, thus, order the entry of a zero value into any aspect of the Damages Spreadsheet that relates exclusively to telephone dictators.
Additionally, Nuance produced a number of pre-litigation statements that express the views of both parties that files created by telephone were never converted to MASC. JX 116 (statement in internal Nuance email that "[t]elephony users need not be licensed"); JX 278-a at 10 ("SIMON [Howes]: Just the ones on the phone don't use Vianix. REZA [Hashampour]: That's correct."); JX 285-a at 51-52.
D. How Are User Fees Calculated? 1. Interpretation of the Cost/End User provision
The parties dispute how Nuance was to calculate the royalties it owed for user fees
under the TLA. This dispute revolves largely around how Exhibit E's provisions regarding user fees are to be interpreted. Both Connecticut and Virginia follow the objective theory of contract interpretation. Under this theory, "[a] contract must be construed to effectuate the intent of the parties, which is determined from the language used interpreted in the light of the situation of the parties and the circumstances connected with the transaction." "Where the language of the contract is clear and unambiguous, the contract is to be given effect according to its terms." "In interpreting contract items, we have repeatedly stated that the intent of the parties is to be ascertained by a fair and reasonable construction of the written words and that the language used must be accorded its common, natural, and ordinary meaning and usage where it can be sensibly applied to the subject matter of the contract."
Petisi v. Strottman Int'l, Inc., 1998 WL 601236, at *3 (Conn. Super. Ct. Aug. 28, 1998); Adams v. Doughtie, 2003 WL 23140076, at *12 (Va. Cir. Ct. Dec. 31, 2003).
Dirienzo Mech. Contrs., Inc. v. Salce Contr. Assocs., Inc., 2010 WL 2520617, at *4 (Conn. App. Ct. June 29, 2010) (quoting Office of Labor Relations v. New Eng. Health Care Empls. Union, Dist. 1199, 951 A.2d 1249, 1254 (Conn. 2008)); see also Sw. Va. Hosps. v. Lipps, 68 S.E.2d 82, 90 (Va. 1952).
Dirienzo Mech. Contrs., 2010 WL 2520617, at *4 (quoting HLO Land Ownership Assocs. Ltd. P'ship v. Hartford, 727 A.2d 1260, 1264 (Conn. 1999)); see also City of Chesapeake v. States Self-Insurers Risk Retention Gp., Inc., 628 S.E.2d 539, 541 (Va. 2006).
Dirienzo Mech. Contrs., 2010 WL 2520617, at *4 (quoting Wolosoff v. Wolosoff, 880 A.2d 977, 983 (Conn. App. Ct. 2005)); see also Kemp v. Levinger, 174 S.E. 820, 823 (Va. 1934).
Turning to the provisions of Exhibit E of the TLA, I first note that because the TLA contains a merger clause, the parol evidence rule applies. Under that rule, the unambiguous terms of a written contract may not be varied or contradicted by extrinsic evidence. If a contract's terms are ambiguous, however, then extrinsic evidence may be admitted to explain the ambiguity.
TLA § 17.
Tallmadge Bros., Inc. v. Iroquois Gas Transmission Sys., L.P., 746 A.2d 1277, 1290-91 (Conn. 2000); Spotsylvania Cty. Sch. Bd. v. Seaboard Sur. Co., 415 S.E.2d 120, 126 (Va. 1992).
Tallmadge Bros., 746 A.2d at 1291; Prospect Dev. Co. v. Bershader, 515 S.E.2d 291, 296 (Va. 1999).
Exhibit E contains a table, the relevant portion of which reads as follows:Nbr. Of Seats Cost/End User
2000 $12.00 ea. 2001-5000 $9.00 ea. 5001+ $7.50 ea.Exhibit E also contains the following provision: " Cost/End User. The Cost per End User shall be calculated in accordance with the number of per seat licenses cumulatively sold by [Nuance] during each year of this this [sic] Agreement." The page following the page on which the table and definition of Cost/End User appear ("Page 25") contains a fragment of a paragraph which states, in its entirety: "with accompanying Sublicense to MASC TECHNOLOGY. Relevant `tracking' unit for these products would either be the Web Server Software License (per site) or the Dictate Module license (per user)."
TLA at 24.
Id. at 25.
Because this portion of the TLA is so poorly drafted, unpacking it to determine its meaning requires that I consider two separate questions. One is whether the parties intended to pay a royalty for every End User or, instead, only for every license Nuance sold without regard to the nature of that license. The other ancillary question is, even if the parties meant there to be a royalty for every End User, did they nonetheless agree, due to the practical difficulties of counting End Users, for example, to compute the royalties based solely on the number of licenses Nuance sold.
I start by finding that Exhibit E unambiguously requires Nuance to pay a royalty for every End User, as End User is defined in the TLA. Nuance argues for a different interpretation, namely, that royalties are owed for every license sold, rather than every End User. Although Nuance cites two portions of Exhibit E in support of its position, I find that neither passage casts any doubt on the clarity of Exhibit E's mandate. Nuance first cites the paragraph fragment on Page 25 which states that the relevant per user tracking unit is something called the "Dictate Module license." Seizing on this reference to the "tracking unit" being a license, Nuance contends that the TLA does not require it to pay a royalty for every End User, but rather only for every license it sold. This argument is unpersuasive, however, for the following reasons.
The TLA defines an End User as: "The final or ultimate individual authorized user of a [Nuance] Product and MASC TECHNOLOGY. End-User Sublicenses are based on the actual number of authorized system users of MASC TECHNOLOGY." TLA § 1.12. For purposes of clarification, because telephone dictators did not use MASC, see supra Part II.C, telephone dictators are not End Users.
Nuance sold both concurrent and nonconcurrent licenses. A concurrent license allows multiple individuals to use a single license, so long as only one person is using that license at any given time. For example, if a medical transcription company purchased ten concurrent licenses, an unlimited number of individuals could use these licenses, so long as no more than ten individuals used the licenses at any one time. T. Tr. 331 (Heffernan).
TLA at 25.
First, I find that the evidence compels the conclusion that the obviously incomplete language on Page 25 is merely a result of sloppy drafting. The language itself meshes poorly with the rest of Exhibit E. Page 25 speaks in terms of tracking units and Dictate Module licenses, terms not used anywhere else in the TLA. Moreover, at no point during the six-day trial did anyone even attempt to define what a Dictate Module license was. In these circumstances, the only reasonable inference I can draw is that neither party intended the language on Page 25 to be included in the final draft of the TLA. While courts generally construe contracts so as to give effect to all their provisions and not render any provision meaningless, the facts here indicate that the parties did not intend to include the fragmentary and inconsistent language on Page 25 as part of the TLA.
CitiSteel USA, Inc. v. Gen. Elec. Co., 78 F. App'x 832, 837 (3d Cir. 2003).
Second, Nuance argues that the definition of Cost/End User "speaks in terms of counting `licenses,'" asserting, in effect, that Cost/End User really means Cost per license. Yet, that is not a reasonable interpretation of this provision. The definition of Cost/End User in the final version of the TLA states that "The Cost per End User shall be calculated in accordance with the number of per seat licenses cumulatively sold by" Nuance. When viewed in tandem with the table in Exhibit E, this language indicates that Cost/End User decreases as the "Nbr. Of Seats" increases, i.e., there is a volume discount. The definition of Cost/End User, therefore, designates per seat licenses as the unit or metric by which to determine the applicability of any volume discount and ultimately calculate the royalty due. Moreover, the TLA clearly defines End User as "[t]he final or ultimate individual authorized user of a [Nuance] Product and MASC TECHNOLOGY," and this term cannot be written out of the TLA in the manner Nuance suggests. Accordingly, I find that the TLA unambiguously requires Nuance to pay Vianix a royalty for every End User of a MASC-bearing product.
DOB 33.
TLA at 24.
Id. § 1.12.
As to the ancillary question identified supra, however, I do find ambiguous the use of "Nbr. of Seats" in the table and the term "per seat licenses" in the definition of Cost/End User as they relate to the method of calculating the Cost for each End User. Neither "seats" nor "per seat licenses" are defined in the TLA, leaving the meaning of these terms unclear. Nuance asserts that a per seat license can refer to both a unique user license and a concurrent license and, thus, encompasses all licenses that Nuance sold. This view arguably makes the "per seat" part of per seat license meaningless. Yet, Nuance seems to suggest that a "seat" corresponds to a single End User in the case of an individual license or, for a customer having a concurrent license, the maximum number of users who can use the Nuance product at one time. Vianix, on the other hand, effectively contends that there is no difference between the phrase "per seat" and "per End User." Based on the inartful drafting of the TLA, I find that both these proposed constructions are reasonable, but mutually exclusive, thus rendering this language ambiguous.
T. Tr. 1272 (Pearah).
See Bijur v. Bijur, 831 A.2d 824, 829 (Conn. App. Ct. 2003) ("A word is ambiguous when it is capable of being interpreted by reasonably well-informed persons in either of two or more senses.") (internal quotations omitted); Courtney v. Courtney, 2006 WL 1675064, at *14 (Va. App. Ct. June 20, 2006) (an ambiguity exists when "competing contractual parties proffer two reasonable, but mutually exclusive, interpretations of the allegedly ambiguous provision").
One plausible interpretation of these terms is that "Nbr. Of Seats" is really just shorthand for per seat licenses as that term is used in the definition of Cost/End User and, as Nuance contends, simply means number of licenses. Defining the terms in this manner, however, is problematic, as the procedure by which user fee royalties are to be calculated breaks down if "Nbr. Of Seats" is read to mean number of licenses, without regard to the number of End Users covered by each license. Because Nuance sold concurrent licenses, there is no correlation between the number of licenses it sold and the number of End Users authorized to use a Nuance Product with MASC Technology under those licenses. Thus, were Exhibit E construed in this fashion, Nuance could almost unilaterally control the royalties it owed without regard to the number of End Users involved. On the other hand, Vianix's position that the number of seats equates to the number of End Users raises the question why the parties used the word "seats" instead of End Users. Thus, I find the terms "Nbr. Of Seats" in the table in Exhibit E and "per seat licenses" in the definition of Cost/End User to be ambiguous and look to extrinsic evidence to resolve this ambiguity.
Unfortunately, the parties provided little in the way of extrinsic evidence relating to the meaning of either "Nbr. Of Seats" or per seat licenses. They did, however, produce three preliminary drafts of the TLA, which I find helpful. All three drafts contain a table in Exhibit E identical to the one that appears in the signed version of the TLA. Thus, the parties consistently used the column heading "Nbr. Of Seats" throughout all versions of the TLA. The October 23, 2002 draft contains a provision entitled "End User Licenses" that states:
TLA at 24; JX 67 at 25; JX 68 at 26; JX 184 at 25.
In addition, [Nuance] will pay to VIANIX a fee based on the number of End-Users of the MASC TECHNOLOGY that is based on the cumulative number of End-Users covered by any Sublicense during the term of this Agreement. For example, for the first 2000 End-Users covered by a Sublicense Agreement, [Nuance] will pay VIANIX a one-time fee under the initial term of this Agreement of $12.00 per End-User. For the next End-Users added, up to and including 5000 End-Users, [Nuance] will pay $9.00 each for each such End-User. . . .
JX 67 at 25.
Neither the November 26, 2002 draft nor the January 13, 2003 draft of the TLA contains any relevant changes to this End User Licenses provision.
JX 68 at 26; JX 184 at 25.
In the drafts, therefore, the parties clearly used "Nbr. Of Seats" to mean number of End Users. The End User Licenses provision in the drafts unmistakably based the End User royalty owed to Vianix on the cumulative number of End Users. Because the description of the End User fees comports with the table, I am convinced "Nbr. Of Seats" in the table meant number of End Users in all three draft versions of the TLA.
JX 67 at 25; JX 68 at 26; JX 184 at 25.
But, the End User Licenses provision in the drafts did not make it into the final version of the TLA; it was replaced by the much less clear definition of Cost/End User, which introduced the reference to "the number of per seat licenses." Vianix treats this change as immaterial to the principle that Nuance would pay a user fee for each authorized End User. Nuance disagrees and contends that the fact that this provision was changed to require a fee based on the cumulative number of per seat licenses, instead of the cumulative number of End Users, shows that the parties considered and rejected using End Users as the basis for computing royalties. In a December 8, 2003 email, however, Nuance's Pearah stated that "the contract was kept intentionally vague to cover any type of end-user license." Pearah's remark shows that Nuance knew that the reference to a per seat license was vague and potentially ambiguous in terms of how it applied to concurrent licenses, but decided not to press for clarification on that point. This behavior on the part of Nuance undercuts its position, as there is no indication that Vianix ever understood that the change made to the TLA was to have the effect Nuance now claims it has. In these circumstances, it was incumbent on Nuance to make sure Vianix shared its understanding of the reference to per seat licenses in the Cost/End User definition. Otherwise, it would risk having the term construed in accordance with Vianix's understanding of it. Nuance's failure to attempt to clarify this vague language provides a further reason to hold that both "Nbr. Of Seats" and per seat licenses mean End Users for purposes of interpreting Exhibit E.
JX 345.
See Supermex Trading Co. v. Strategic Solutions Gp., 1998 WL 229530, at *9 (Del. Ch. May 1, 1998) (holding that "[s]ubjective understandings of a party to a contract which are not communicated to the other party are of no effect. They are irrelevant to the interpretation of the contract and should not and will not be given any weight."); cf. United Rentals, Inc. v. Ram Hldgs., Inc., 937 A.2d 810, 835-36 (Del. Ch. 2007) (describing the forthright negotiator doctrine: "[I]n cases where the extrinsic evidence does not lead to a single, commonly held understanding of a contract's meaning, a court may consider the subjective understanding of one party that has been objectively manifested and is known or should be known by the other party.").
This view finds some additional support in the draft versions of the TLA. In the drafts, the parties used "Nbr. Of Seats" to mean number of End Users; therefore, the continued use of "Nbr. Of Seats" in the final version of the TLA comports with Vianix's construction. Specifically, this fact suggests the parties still intended number of seats to mean End Users when they executed the TLA.
As a result of this finding, Nuance owes Vianix royalties for all End Users of Nuance's MASC-bearing products as follows: For each year the TLA was in effect, Nuance owes Vianix (1) $12.00 per End User for the first 2,000 new End Users in that year, (2) $9.00 per End User for the next 3,000 new End Users in that year, and (3) $7.50 per End User for each additional new End User after the 5,000th End User. To date, Nuance has paid some of those royalties, but not as much as it owes. In particular, because Nuance has admitted that it did not track End Users as required under the TLA, as properly construed, and only paid a royalty for each license it sold, rather than for every End User, I find that Nuance breached the TLA by underpaying royalties to Vianix throughout the life of the TLA.
I use the term "new" here because, as discussed, infra Part II.E, the TLA does not require Nuance to pay Vianix renewal fees on user license fees.
T. Tr. 337-38 (Heffernan).
2. Determination of an appropriate multiplier
Because Nuance failed to collect any data on the number of End Users, I cannot use actual data to determine the user fee royalties that Nuance owes Vianix. Accordingly, I must determine whether the Court can responsibly estimate the number of End Users Nuance added during each royalty reporting period based on the evidence of record in this action. I begin by examining the data Nuance collected on the number of licenses it sold for MASC. Nuance relies on the Bourassa Data. That data includes: (1) Pre-Audit data which has a count of "licenses" for each quarter through Q3 07 broken down by product and (2) Post-Audit data with "licenses" counts for each quarter from Q4 07 to Q4 08, the last quarter in which the TLA was in effect. The Bourassa Data on Enterprise and Enterprise Upgrades (Pre- and Post-Audit) contains separate break-outs for "User" and "Concurrent" licenses. A cursory review of this data suggests that the number of concurrent licenses is less than 15 percent.
The next question is whether the information available from Nuance provides a basis from which the Court responsibly could estimate the number of End Users Nuance added during a particular period. I conclude that it does in the sense that applying an appropriate multiplier to the Nuance data is likely to provide a reasonable estimate of the total number of End Users for the relevant period. Because Nuance sold concurrent licenses, the number of End Users of its products will be greater than the number of licenses it sold, as more than one End User could use each concurrent license. Moreover, because Nuance sold many concurrent licenses and, from the evidence presented, a large number of individuals potentially could use these licenses, I find that the number of licenses Nuance sold reflects only a fraction of the true number of End Users of these products. Thus, I consider it appropriate to apply a multiplier to the data Nuance collected on the number of concurrent licenses sold to make a responsible estimate of the number of End Users covered by such licenses.
Vianix urges the Court to apply a 20x multiplier to non-iChart user licenses and not to use a multiplier for iChart, but rather to estimate the number of iChart End Users using the number of iChart users Nuance claimed to have in a June 2007 press release, PX 120. As discussed supra Part II.A, I decline to use the June 2007 press release as a basis for calculating royalties. Instead, I will apply a multiplier that the evidence indicates is likely to yield a reasonable estimate of the number of users authorized to use a Nuance product, iChart or non-iChart, with MASC Technology and for whom a royalty would be due.
In light of all the evidence presented, I find that a 5x multiplier is appropriate to apply to the number of concurrent user licenses Nuance sold to approximate the number of End Users under those licenses. Vianix's requested 20x multiplier is based on information on or counts of the number of medical professionals and transcriptionists who work for Nuance's customers compiled by Don Buer, Vianix's Vice President of product management and licensing. Vianix asserts that these counts support using a multiplier anywhere from 10x to 50x, but the evidence shows that multipliers in this range are unreasonably high. Buer's counts are highly unreliable for a number of reasons. First, Buer apparently counted all "healthcare providers" at hospitals that used Nuance products. This metric undoubtedly overestimates End Users by including nurses, technicians, and others who would have no reason to use dictation products either alone or in combination with the MASC Technology. Not even all doctors who worked for a Nuance customer would have used Nuance products, let alone royalty-bearing products. This is especially true given that, even at the apex of Nuance's use of MASC, only about half of Nuance's healthcare products contained MASC, and there was no guarantee that customers who used earlier versions of those products would have upgraded to later versions that eventually did incorporate MASC. Furthermore, a number of doctors who worked for Nuance customers were undoubtedly telephone dictators who, as previously discussed, did not use MASC and, therefore, did not generate royalties.
Id. at 737, 791, 796-99 (Buer).
Id. at 856 (Buer).
In addition, Nuance did not sell concurrent licenses for all of its products. For example, while Nuance admits to selling concurrent licenses for JobLister and Transnet, it could not sell concurrent licenses for EWS because the functionality of that product was based on recognizing an individual's voice, thereby preventing an EWS license from being shared. For all of these reasons, I consider the 20x multiplier Vianix advocates based on Buer's counts to be inappropriate because it is unduly speculative and not sufficiently supported by the record.
JX 199; T. Tr. 331 (Heffernan); Heffernan Dep. 180-81.
In determining that a 5x multiplier is appropriate, I consider particularly relevant the IVS Addendum, which provides a pricing mechanism for concurrent licenses that appears to make the royalty for such a license a function of the total number of End Users likely to use the license. The IVS Addendum expressly provided for five different types of IVS concurrent licenses that Nuance could sell: EX20, EX50, EX125, EX250, and EXFULL. As Heffernan testified, the EX20 license, for instance, could handle a maximum of 20 total End Users. For each type of license, the IVS Addendum also specified a lesser imputed number of Users per Site for each license upon which the royalty calculation would be based. For example, the Users per Site for EX20 was five. The IVS Addendum then required Nuance to pay Vianix a $20 royalty for each User per Site.
JX 193 at NUAN3645-47.
T. Tr. 511. Similarly, there could be 50 total End Users on an EX50 license, and that number would increase to 125 and 250, respectively, on EX125 and EX250 licenses. There was apparently no limit on the number of End Users who could use the EXFULL license.
Heffernan further testified that the number of Users per Site represents the maximum number of concurrent users that the license allowed, or, in other words, the number of concurrent licenses associated with that product. Therefore, dividing the maximum number of End Users of an IVS license by the Users per Site provides a ratio that indicates the relation between the number of concurrent licenses associated with that product and the total number of authorized End Users. For EX20, this calculation yields a 4:1 ratio or 4x multiplier. For EX50, there were 15 Users per Site, yielding a 3.33x multiplier; for EX125, 60 Users per Site for a 2.08x multiplier; and for EX250, 100 Users per Site for a 2.5x multiplier. The fact that all of these multipliers are between 2x and 4x supports using a multiplier at the high end of that range, or even slightly higher to account for the fact that Nuance failed to keep track of the number of End Users of the concurrent licenses at issue here. Moreover, Vianix's expert witness, Ellis, said that if he were asked to determine a multiplier to approximate End Users from licenses sold, he would refer to the IVS Addendum, meaning he likely would choose a multiplier somewhere in the neighborhood of 3x to 4x.
Id. at 511-12.
Nov. 14, 2009 Ellis Dep. 93-94.
When it performed the audit, Invotex also assumed that a 3x multiplier was appropriate to estimate End Users. Furthermore, Nuance generally charges three times as much for concurrent licenses as noncurrent licenses. In fact, when attempting to reconcile its potential liability with the $12.6 million post-audit invoice sent by Vianix, Nuance assumed that, if its construction of the TLA was incorrect and royalties were to be based on the number of End Users, 3x was an appropriate multiplier to estimate End Users. Finally, a multiplier of at least 3x meshes with the typical practice of transcription companies to run three daily shifts of transcriptionists, meaning the ratio of End Users to users able to use the product simultaneously under a concurrent or capacity license would be at least three to one for products like JobLister and Transnet, which were used primarily by transcriptionists and for which Nuance admittedly sold concurrent licenses.
JX 289-a at 59. Stewart, the Invotex auditor, based her view that a 3x multiplier was appropriate on the fact that many hospitals run three shifts of workers a day. Id.
JX 62; JX 101.
JX 101 at 2.
Having considered all the relevant evidence and the fact that Nuance failed to keep the necessary data to enable the Court to determine precisely the number of End Users covered by the concurrent licenses Nuance granted under the TLA, I conclude that a 5x multiplier should be applied to the number of concurrent licenses Nuance sold. This multiplier is slightly above the 3x to 4x multiplier the IVS licenses support, but I find the increase to 5x appropriate based on the added uncertainty here attributable to Nuance's inadequate recordkeeping. Accordingly, I mandate entry of a 5x multiplier into the Damages Spreadsheet in a manner appropriate to convert the actual number of concurrent licenses Nuance sold into an estimate of total End Users covered by those licenses.
Ellis testified that the Damages Spreadsheet contains a cell for "Percent Concurrent Licenses." T. Tr. 1094. It appears the Bourassa Data includes an accurate and complete count of the actual number of concurrent licenses sold under the TLA. If that is correct, the parties should use that number in the Damages Spreadsheet. If the Bourassa Data cannot be used for that purpose, 15 percent should be entered in the "Percent Concurrent Licenses" cell in the Spreadsheet.
E. Does the TLA Require Payment of a Renewal Fee?
Vianix contends that the TLA requires Nuance to pay a renewal fee on End User fees after three years, the initial term of the TLA. Vianix bases its argument on the language of Exhibit E of the TLA, which refers to both the Initial Cost/Site and Cost/Additional Site as one-time fees, but does not refer to Cost/End User as a one-time or nonrecurring fee.The merger clause again comes into play here, and I can consider extrinsic evidence only if I determine that the language at issue is ambiguous. Vianix contends that the TLA calls for a renewal fee on user fees because the user fee is the only fee in the TLA that is not explicitly defined as a one-time fee. But, if the parties truly intended user fees to recur, they almost certainly would have spelled out the procedure by which the renewal fees would be incurred and how they were to be paid. For example, would the renewal fees accrue three years after the date of the TLA or after the first royalty was paid for a particular group of End Users? Vianix simply asserts that the parties intended for user fees to recur every three years consistent with the original term of the TLA, but Exhibit E, which governs licensing fees, makes no reference to the three-year term of the TLA. Under these circumstances, I find the maxim expressio unius est exclusio alterius (the express mention of one thing excludes all others) inapplicable, as it is not clear from the mere omission of the words "nonrecurring" or "one-time" from the definition of Cost/End User that the parties intended user fees to recur. Moreover, the absence of any terms that would govern the recurrence of user fees suggests that the parties did not intend those fees to be recurring. Thus, I consider the Cost/End User language at issue here, which does not specify that user fees are either recurring or nonrecurring, to be ambiguous and, therefore, look to the proffered extrinsic evidence to resolve the ambiguity.
See supra notes 121-123 and accompanying text.
The extrinsic evidence overwhelmingly shows that the parties intended user fees to be nonrecurring. In the October 23, 2002 draft of the TLA, the "End User Licenses" section of Exhibit E stated: "Such End-User payment will create a right for the covered End-User to use the MASC TECHNOLOGY for a period of three years from the date of issue of relevant Sublicense." A November 26, 2002 redlined draft of the TLA strikes the language cited in the previous sentence and contains the following comment by a Nuance representative:
JX 67 at 25.
We want a one-time fee, not a fee that recurs every 3 years. Please clarify that after 3 years, we are not required to pay renewal fee for sites sold during 3 year contract. The per user fees are a one-time fee, which is the reason we are paying a higher price.
JX 68 at 26.
By the time the TLA was signed, all language referring to a three-year term for user licenses had been removed from the Cost/End User section of Exhibit E. In addition, the parties modified the language of section 2(a) of the TLA to grant Nuance a "perpetual" license to use and distribute Vianix technology. Nuance avers that change was designed to ensure it would not have to pay renewal fees, and both of the people who negotiated the TLA for Nuance testified that the word "perpetual" was added for exactly that purpose. Indeed, the Nuance negotiators credibly testified that they "would have just walked away from the deal" had Vianix insisted on such a recurring fee.
TLA at 24.
Id. § 2(a).
T. Tr. 435 (Heffernan), 1288 (Pearah); see also JX 350 (describing the addition of the word "perpetual" as an "important change").
T. Tr. 1288 (Pearah); see also T. Tr. 430-36 (Heffernan); Pearah Dep. 148.
Nevertheless, Vianix downplays the use of "perpetual" in section 2(a) as irrelevant, arguing that section 2(a) grants Nuance a license to use MASC, while section 2(b) gives Nuance the ability to sublicense MASC to its customers. According to Vianix, if the parties had intended to make user fees nonrecurring, the word "perpetual" would have been inserted in section 2(b), rather than section 2(a). I see no merit in Vianix's argument and, therefore, find that the insertion of the word "perpetual" into the TLA shows the parties' intent that no fee would recur after three years. Thus, I find that the TLA did not require Nuance to pay a three-year renewal fee on user fees and order that a value of zero be entered into the Damages Spreadsheet in the cell designated to track renewal fees.
Vianix also cites the fact that Nuance paid renewal fees on at least six occasions as evidence that the parties intended user fees to be recurring. PX 104 — PX 109 Nuance presented credible evidence, however, that Heffernan challenged the renewal fees when they were invoiced and Nuance's payment of renewal fees was inadvertent. T. Tr. 517 (Heffernan); JX 70; Heffernan Dep. 128. In the context of all the evidence on this issue, I consider Nuance's mistaken payment of a handful of renewal fees inconsequential.
F. Site Fees
Vianix claims that Nuance has underpaid royalties owed for site fees, particularly
fees owed for Additional Sites. This dispute focuses largely on the proper construction of the provisions of the TLA defining Initial Cost/Site and Cost/Additional Site. Nuance contends that it owes an Initial Site fee when a customer buys a set of EX servers and an Additional Site fee only when that customer or one of its affiliates buys another set of EX servers. Accordingly, Nuance ties the payment of all site fees to the purchase of a set of EX servers. Vianix, on the other hand, relies on the TLA's definition of a Client Site as "a physically contiguous facility" and claims that Nuance owes a royalty for each additional site operated by its customers or its affiliates that is not a physically contiguous facility to the original site.
TLA at 24.
The relevant provisions in Exhibit E of the TLA read as follows:
Number of [Nuance] Client Sites shall be calculated cumulatively over the term of this Agreement. A Client Site is defined as a physically contiguous facility.
Initial Cost/Site. The Initial Cost per Site shall refer to a one-time fee paid by [Nuance] for Sublicensing the MASC Technology at a particular [Nuance] Client/Sublicensee place of business or at the Client/Sublicensee's Affiliate site. This is a non-recurring cost and shall apply only to the initial sublicense for any corporate client including that client's affiliates. This Initial Cost per Site is in addition to the Cost/Additional Sites and the Cost/User License fee. . . .
Cost/Additional Site. After a particular corporate client or its affiliates have paid the Initial Cost/Site, any subsequent sites for the same corporate client or its affiliates, shall pay a one-time fee equivalent to 40% less than the applicable Initial Cost/Site (ie. Initial Cost/Site = $200: Cost/Add'l Site = $120).
Id.
Preliminarily, I must determine whether any of the relevant language in these provisions is ambiguous. If it is, I may consider parol evidence.
Regrettably, the definition of Client Site on which Vianix relies so heavily is hopelessly vague. One can imagine, for example, two large hospital campuses. In one, all of the wings of the hospital are connected; in the other, the various functional components or centers may be in close proximity to one another, but separated by parking lots. Assuming that in both hospitals there are two EX Voice servers in two separate wings, does that mean that only an Initial Site fee is due for the first hospital because both wings with servers are part of a physically contiguous facility, while in the second hospital an Additional Site fee would be due for the second of the two servers? Such a construction would seem reasonable, but another entity also may view the components of the second hospital as part of one physically contiguous facility, focusing on the fact that they all are on a single property. Based on the language of the TLA, I cannot say the latter interpretation is unreasonable. Accordingly, this TLA language is ambiguous.
The definition of Initial Cost/Site conditions payment of a site fee on Nuance "Sublicensing MASC Technology at a particular [Nuance] Client/Sublicensee place of business or at the Client/Sublicensee's Affiliate site." The meaning of this language is not entirely clear, as MASC Technology includes both EX servers and all other MASC-bearing Nuance products, such as EWS and Powerscribe SDK. Based on the breadth of the term "MASC Technology," this definition arguably favors a position closer to Vianix's where a site fee would be owed for any location where a MASC-bearing product is used.
Id.
A Cost/Additional Site is owed only after the Initial Cost/Site has been paid and must be paid for "any subsequent sites for the same corporate client or its affiliates." This definition requires a fee to be paid for sites, rather than servers, which would seem to indicate that an Additional Site fee is owed whenever a customer adds a new facility, though this cannot be said with certainty because the TLA does not define "site" and the definition of Client Site is vague. Complicating matters further is a fragment of a paragraph that appears on Page 25 of the TLA which states: "Relevant `tracking' unit for these products would [] be the Web Server Software License (per site)." While this language helps Nuance in the sense that it talks about servers (and is, in fact, the only place in the TLA where servers are mentioned), nothing in the definition of either Initial Cost/Site or Cost/Additional Site mentions a tracking unit and, as previously discussed, the parties likely meant to omit Page 25 from the TLA entirely. All in all, the definitions relating to site fees are vague, internally inconsistent, and ambiguous. Therefore, I will consider extrinsic evidence as a possible source of information to resolve the ambiguity.
Id.
Id. at 25.
In contrast to the final, signed version of the TLA, all three drafts of the Agreement contain identical language requiring payment of a royalty for Cost/HQ Site, which is the "cost for Sublicensing the MASC Technology at a particular [Nuance] Client/Sublicensee's primary or headquarters site," rather than an Initial Cost/Site. All three drafts also contain an illustration which indicated that an Additional Site fee was owed for each of a customer's 100 "additional medical and administrative locations," but that illustration does not appear in the final version. Indeed, the language currently at issue first appeared in the version of the TLA the parties signed. The change in the site fee structure from the draft stage to the final version of the TLA favors Nuance due to the removal of both the "additional medical and administrative locations" language and the illustration implying that a royalty was owed for virtually every location a Nuance customer maintained. I infer that the removal of these provisions and their replacement with more vague provisions reflects Nuance's unwillingness to pay a site fee for every one of its customers' locations without regard to whether a MASC-bearing server was located at that site.
JX 67 at 25; JX 68 at 26; JX 184 at 25.
Furthermore, Nuance's negotiators, Heffernan and Pearah, testified credibly that Nuance would not have signed the TLA were it required to count every single one of its customers' facilities and that Vianix knew that sites were synonymous with servers for the purpose of calculating site fees. Other evidence indicates that Vianix did not devise its claim that a site fee royalty was owed for every facility run by a Nuance customer until after it commenced this litigation. Exhibit G of the TLA, a sample Nuance customer agreement that Vianix approved, contains no requirement that customers report to Nuance the number of facilities or medical/administrative locations where MASC is used, which would be necessary for Nuance to track site fees in the manner Vianix claims was required. Likewise, at no point while the TLA was in effect did Vianix ask Nuance if it was tracking the number of facilities or locations where MASC was used. Nor did Vianix ask Invotex to look into the issue of Additional Site fees during the audit, despite generally asking the auditors to look into every possible area where it felt Nuance might be underpaying royalties. And, perhaps most damaging to Vianix's claim is the fact that there was no charge for Additional Site fees in either of the $12 million invoices Vianix sent to Nuance in February and March 2008, even though Vianix plainly attempted to include every charge it possibly could think of in those invoices.
T. Tr. 423 (Heffernan), 1270 (Pearah).
TLA at 26-29.
T. Tr. 429 (Heffernan).
Id. at 932-33 (Buer).
JX 56; JX 58; T. Tr. 934 (Buer).
Vianix based its litigation-inspired claim that Nuance underpaid Additional Site fees almost exclusively on the incredibly vague definition of Client Site in the TLA, but it adduced no probative evidence that the Agreement required Nuance to pay a fee for every facility or location its customers operated. Based on the trial record, therefore, I find that the TLA required payment of a Cost/Additional Site only when a customer licensed a second (or third, fourth, etc.) set of EX servers from Nuance. As I understand the evidence, the Bourassa Data already accounts for all such instances. Thus, I find that Nuance adequately reported and paid site fees. Accordingly, I direct the parties to use the Bourassa Data for Additional or Affiliate Sites and either no multiplier or a multiplier of one in the Damages Spreadsheet, whichever will accurately indicate the absence of any additional Affiliate Sites beyond those indicated in that Data.
G. Growth Factors
Vianix seeks use of, and the Damages Spreadsheet contains a cell for, a growth factor that would be applied to Nuance's 2007 iChart sales to approximate 2008 iChart sales. In effect, Vianix distrusts the Bourassa Data, which shows little to no growth in iChart sales between 2007 and 2008. Instead of using the Bourassa Data, Vianix proposes that the Court estimate the 2008 sales by applying a growth factor. Vianix bases its argument on SEC filings and Nuance press releases, the latter of which claim "growth in the range of 30 to 40 percent" for iChart.I reject Vianix's request to use a growth factor to estimate sales of iChart or any other Nuance product. Vianix failed to show why the Bourassa Data for 2008 iChart sales are inadequate. Moreover, the sources of information Vianix urges the Court to use are too general to be useful. The press release talks of 30 to 40 percent growth, but does not specify what metric grew. According to Nuance, the press release refers to revenue growth, but it is unclear how that relates to the factor that matters for royalty purposes, growth in End Users of a Nuance product with MASC Technology. Moreover, even if the press release did refer to growth in users, it would be of limited value because many iChart users are non-royalty-bearing telephone dictators. Hence, Vianix has failed to prove that the growth claimed in the press release would equate to growth in royalty-bearing users.
The SEC filings are similarly flawed. These filings depict the cumulative revenue growth across all of Nuance's products, including those which do not contain MASC. As with the press releases, Vianix has not shown a correlation between the growth reported in the SEC filings and any growth in royalties owed for iChart users. Thus, the evidence adduced by Vianix regarding a growth factor provides no basis to doubt the accuracy of the Bourassa Data for 2008 iChart sales. Accordingly, I order that a value reflecting a zero growth rate be entered into all cells in the Damages Spreadsheet relating to growth factors.
H. SMA Fees
Vianix contends that the royalty calculation should include a 6 percent annual
service and maintenance agreement ("SMA") fee. Section 4.1(e) of the TLA obligates Vianix to "[p]rovide Maintenance and Support Services as set forth in Exhibit C." Exhibit C requires Nuance to pay Vianix "6% annually of per user royalty fees for the MASC's Technology Software [sic]" for maintenance of the MASC software. Nuance does not dispute that it owes SMA fees. Therefore, I order the parties to make appropriate entries in the Damages Spreadsheet to reflect a 6 percent SMA fee on each year's unpaid royalties.
TLA at 21.
I. Interest
The TLA also entitles Vianix to interest on all damages it is awarded. Section 8.2 of the TLA sets up a framework by which Nuance is to pay royalties and provides that if Nuance is late in making royalty payments, "interest may accrue on any overdue amounts at the rate of 1.5% per month or the highest rate allowable by law, whichever is less." As neither Connecticut nor Virginia law limits the amount of interest parties can agree to via contract, 1.5% per month is the applicable interest rate under the TLA. Moreover, because the TLA specifies that interest is due on "any overdue amounts," Vianix is entitled to compound interest. Thus, interest at a rate of 1.5% per month compounded quarterly will be applied to all damages awarded to Vianix.Under the framework prescribed by section 8.2 of the TLA, Nuance was to send Vianix a royalty report within twenty days of the end of each quarter. Vianix was then to use this royalty report as a basis for formulating an invoice for Nuance. section 8.2 specifies that interest begins to accrue on unpaid royalties thirty days after Nuance receives an invoice from Vianix. Accordingly, interest realistically could not begin to accrue on unpaid royalties until approximately sixty days after the end of a quarter. I thus find that interest should begin to accrue on the damages Vianix is owed sixty days after the end of the quarter in which the damages originally became due and direct the parties to use the Bourassa Data to determine when these damages became due.
Id.
To the extent Vianix claims that the accrual of royalties and related interest should be measured for each Nuance product from the date Nuance's product engineering groups released a newer, MASC-bearing version of that product, I find this claim without merit because it conflicts with section 8.2 of the TLA and the evidence of record, including the information contained in the Bourassa Data regarding the timing of upgrades.
J. Nuance's Counterclaims
Nuance asserts two counterclaims. The first is that Vianix breached the TLA by issuing false invoices that sought payments vastly in excess of the amounts that were properly due under the TLA. In this claim, Nuance takes umbrage with the $12.4 million invoice Vianix sent to it in the wake of the audit. Nuance claims that Invoice #215-R and several other invoices Vianix sent to Nuance violate section 8.2 of the TLA, as well as the covenant of good faith and fair dealing.Under Connecticut law, "every contract carries an implied duty requiring that neither party do anything that will injure the right of the other to receive the benefits of the agreement." To prove a breach of the implied covenant of good faith and fair dealing, the claimant must show: (1) the existence of a contract between the claimant and the counterclaimant; (2) that the counterclaimant engaged in conduct that injured the claimant's right to receive some benefit the claimant expected to receive; and (3) that the counterclaimant acted in bad faith. There is no dispute that a contract existed between Nuance and Vianix. I also will give Nuance the benefit of the doubt and assume that by sending inflated invoices Vianix injured Nuance's right not to be burdened by invoices that were far greater in amount than what Nuance actually owed. Therefore, my analysis of this counterclaim will focus on whether Vianix acted in bad faith in sending the invoices or sent these invoices fraudulently.
Avitable v. 1 Burr Rd. Operating Co. II, LLC, 2010 WL 2926242, at *5 (Conn. Super. Ct. June 4, 2010). Apparently, an implied covenant of good faith and fair dealing is not recognized under Virginia law. See US Airways, Inc. v. Commonwealth Ins. Co., 2004 WL 1637139, at *1 (Va. Cir. Ct. July 23, 2004) ("The law in Virginia supports the Court's finding that U.S. Airways cannot seek recovery for bad faith."); Spiller v. James River Corp., 1993 WL 946387, at *6 (Va. Cir. Ct. 1993) (Virginia law does not recognize a claim for breach of a covenant of good faith and fair dealing); Sneed v. Am. Bank Stationary Co., 764 F. Supp. 65, 67 (W.D. Va. 1991) (Virginia does not recognize an implied covenant of good faith and fair dealing); Ward's Equip., Inc. v. New Holland N. Am., Inc., 493 S.E.2d 516, 520 (Va. 1997) ("[I]n Virginia, when parties to a contract create valid and binding rights, an implied covenant of good faith and fair dealing is inapplicable to those rights."). In light of my holding that Nuance has failed to prove that Vianix breached the implied covenant of good faith and fair dealing, I find that any relevant distinctions between Connecticut and Virginia law are immaterial.
Jones v. H.N.S. Mgmt. Co., 2003 WL 22332837, at *4 (Conn. Super. Ct. Sept. 25, 2003).
Nuance has failed to show that Vianix acted either fraudulently or in bad faith. Nuance's counterclaim relies heavily on the fact that Vianix sent an invoice for more than $12 million days after receiving the Invotex audit, which found that Nuance had underpaid royalties by roughly $2.5 million. Based on all of the evidence, however, the discrepancy between the audit and the invoice is not indicative of bad faith on the part of Vianix. Rather, it reflects Vianix's misunderstanding of Nuance's technology and MASC's role in this technology, as well as its frustration over its inability to obtain information it considered necessary from Nuance and a certain amount of hubris regarding the importance of its MASC Technology. While Vianix's beliefs in this regard precipitated the parties' current situation and proved incorrect in many respects, these beliefs were honestly held and somewhat justified in light of what certain high-ranking Nuance employees were telling Vianix. Thus, I do not consider Nuance's actions to rise to the level of bad faith.
JX 56; JX 58; JX 168.
I first note that Vianix justifiably invoiced Nuance an amount in excess of what Invotex found it was owed. The audit expressly stated that it did not account for certain fees that Vianix believed it was due under the TLA, specifically renewal fees and fees for concurrent users. The audit also did not include any fees for certain disputed products for certain periods of time. Thus, it was reasonable for Vianix to attempt to estimate the full amount it believed it was owed, and the fact that it drastically overestimated this amount does not mean it prepared the invoice in bad faith.
JX 168 at 4 ("These numbers would increase if it were determined that there is more than one end user per concurrent license. . . . Nor does [the audit] include any renewal of licenses due upon the third anniversary of the initial sale.").
Id.
While Vianix incorrectly asserted that the TLA required payment of renewal fees, the evidence indicates that it honestly believed in its position. Hashampour, Vianix's President and CEO, testified credibly at trial about his belief that Vianix was owed a renewal fee and the reasons for that belief. The TLA is a poorly written document fraught with ambiguities, including ones that relate to a possible renewal fee. Based on these facts, I cannot say that Vianix acted in bad faith in pressing for payment of renewal fees.
T. Tr. 72.
Vianix also correctly asserted that Nuance failed to keep adequate records of End Users and, thus, was justified in attempting to estimate the true number of End Users to the best of its ability. Although Buer's efforts to estimate this number were flawed, I find that he made them in good faith and spent a significant amount of time compiling information on Nuance's customers for that purpose. Buer conceded that the approach he employed to derive a multiplier to estimate End Users was unscientific, and ultimately I found that the record supported a much lower multiplier than he suggested. Nevertheless, Buer's glaring overestimation of what Nuance owed Vianix appears to have resulted from a misunderstanding regarding Nuance's technology and the proper construction of certain aspects of the TLA. For example, Vianix's position that even telephone dictators' files were manipulated using MASC Technology, which found some support in certain internal Nuance documents, explains Buer's decision to include every doctor who worked for a Nuance customer in estimating the number of End Users. All of these facts further undermine Nuance's counterclaim for breach of the implied covenant of good faith and fair dealing.
Id. at 736-37, 796-97 (Buer).
Id. at 797 (Buer).
In addition, Nuance often sent conflicting signals to Vianix about which of its products contained MASC, leading to numerous communications where Nuance attempted to explain the implementation of MASC in its products to Vianix. Furthermore, there is ample evidence that Howes and, perhaps, other key Nuance employees were unclear about which of its products contained MASC until well after the commencement of this litigation.
See, e.g., JX 279-a.
JX 76; JX 78; JX 142; JX 391; JX 392.
Lastly, I consider it relevant that the MASC Technology represented Vianix's most important product and Nuance appeared to be a leader in the business of providing speech dictation solutions to businesses. These facts caused Vianix to expect a large and continuing royalty stream from the TLA with Nuance. In fact, that never happened. All told, for the almost six years the TLA was in effect, Nuance paid to or has agreed it owes Vianix a total of only $1,246,315. Although that is not an insignificant amount of money, it pales in comparison to Vianix's expectations and Nuance's reported commercial success over the same period. Therefore, it is not surprising that Vianix heard what it wanted to hear from Nuance. Vianix keyed on Nuance's statements spouting superlatives about MASC and noting how "the best implementation of MASC is across the board" and convinced itself that Nuance's integration of MASC into its products would be widespread, if not universal. Similarly, when Nuance told Vianix that MASC was in its "core technology," Vianix took this to mean that MASC was in all of Nuance's products and available to every user of these products. Now that all the evidence is in, I have found that Vianix's estimates of royalties were far too optimistic. But, Nuance has not shown that Vianix prepared the challenged invoices fraudulently or in bad faith. As such, Nuance has not met its burden of proof; thus, I will dismiss with prejudice its counterclaim alleging that Vianix breached the TLA.
JX 92; JX 104; JX 114.
JX 285-a.
Nuance's second counterclaim seeks a declaratory judgment that it has complied in full with its payment obligations to Vianix under the TLA. Because I have found that Nuance underreported the royalties it owed by failing to track End Users and refusing to recognize that a royalty was owed when it shipped a disc containing MASC to users of Powerscribe SDK version 3.0, I hold that Nuance has not complied fully with its payment obligations under the TLA. Therefore, I will dismiss this counterclaim with prejudice.
K. Attorneys' Fees
Both parties seek their attorneys' fees and costs pursuant to a provision in the TLA that states:
In the event that either VIANIX or [Nuance] brings suit against the other party for any matter arising out of or in connection with this Agreement, the prevailing party shall be entitled to the payment by the other party of the prevailing party's attorney's fees and court costs incurred in connection with such litigation.
TLA § 16.
While courts in both Virginia and Connecticut routinely enforce contractual fee-shifting provisions, neither Vianix nor Nuance fairly can be deemed the prevailing party here. Delaware courts have "typically looked to the substance of a litigation to determine which party predominated." Moreover, a party who is deemed a prevailing party under an attorneys' fees provision such as the one at issue here typically is entitled to recover all of its attorneys' fees, even if it does not win every disputed claim. A court can find, however, that no party may be regarded as having prevailed, and I find that is the situation here.
Total Recycling Servs. of Conn., Inc. v. Conn. Oil Recycling Servs., LLC, 970 A.2d 807, 813 (Conn. App. Ct. 2009); West Square, L.L.C. v. Commc'n Techs., Inc., 649 S.E.2d 698, 702 (Va. 2007).
W. Willow-Bay Ct., LLC v. Robino-Bay Ct. Plaza, LLC, 2009 WL 458779, at *8 (Del. Ch. Feb. 23, 2009) (citing Comrie v. Enterasys Networks, Inc., 2004 WL 936505, at *2-3 (Del. Ch. Apr. 27, 2004)). While the law of either Connecticut or Virginia governs here, both Vianix and Nuance cited the same three Delaware cases ( W. Willow-Bay, Comrie, and Brandin, infra note 200) for the proposition that they are the prevailing party. Accordingly, I will accede to the parties' request and apply Delaware law on this issue.
Brandin v. Gottlieb, 2000 WL 1005954, at *28 (Del. Ch. July 13, 2000).
Id. (citing AHS N.M. Hldgs., Inc. v. Healthsource, Inc., 2007 WL 431051, at *9-10 (Del. Ch. Feb. 2, 2007)).
Each side prevailed on a handful of issues. Vianix proved that Nuance breached the TLA by underpaying royalties owed for End Users and licenses of Powerscribe SDK version 3.0 and successfully defended against Nuance's counterclaims. Nuance succeeded in showing that it did not owe additional royalties for telephone dictators, renewal fees, or Additional Site fees and refuted Vianix's claims regarding the presence of MASC in a number of products. But, to be declared the prevailing party, a litigant must achieve "predominance in the litigation." Neither Nuance nor Vianix has managed that here. In these circumstances, with each party winning on a number of disputed issues, there is no clear-cut prevailing party and, therefore, no basis under the TLA for shifting fees entirely onto the other party.
Brandin, 2000 WL 1005954, at *28.
Nuance arguably won a plurality of the issues in dispute at trial. Even so, I cannot as a matter of equity award Nuance its attorneys' fees because its poor recordkeeping provided the primary impetus for Vianix to bring this suit and measurably increased the complexity of the litigation. Nuance was slow in recognizing when products were upgraded to contain MASC in several instances, failed to keep track of certain data needed to calculate royalties, and, in the words of the auditor from Invotex, kept "the worst records I've ever audited." These problems led Vianix to become distrustful of Nuance and send the inflated $12 million invoice. Nuance compounded the deficiencies in its recordkeeping by failing to produce accurate sales data until the day the parties' pretrial briefs were due. Based on these actions, I find that Nuance bears a greater degree of responsibility than Vianix for the substantial cost in time and money of this litigation. Therefore, I conclude that it would be inappropriate to award Nuance its attorneys' fees as the putative prevailing party on the relatively technical ground that it won more disputed issues than Vianix.
See Sci. Accessories Corp. v. Summagraphics Corp., 425 A.2d 957, 967 (Del. 1980) (upholding an award of attorneys' fees against a prevailing party when that party was "disloyal," even though that disloyalty did not amount to a breach of fiduciary duty).
T. Tr. 378 (Heffernan).
See supra Part II.D.
Stewart Dep. 66.
Vianix complained about the tardiness of the production of this data at trial, but did not formally object to its admission. The parties dispute whether Vianix ever actually requested this information in discovery.
In light of the outcome of this action, with both Vianix and Nuance winning on several claims and contentions and Vianix recovering what may be millions of dollars in damages, but far less than it claimed, I hold that there was no prevailing party and decline both parties' requests for their attorneys' fees and costs.
Nuance also seeks its attorneys' fees and costs on the ground that Vianix pursued its claims in bad faith. See Openwave Sys. Inc. v. Harbinger Capital P'rs Master Fund I, Ltd., 924 A.2d 228, 246 (Del. Ch. 2007) ("[W]here the losing party has acted in bad faith, vexatiously, wantonly, or for oppressive reasons, the court will award attorneys' fees to deter abusive litigation in the future, thereby avoiding harassment and protecting the integrity of the judicial process.") (internal quotations omitted). Because Vianix successfully proved that Nuance breached the TLA by underpaying royalties and presented a number of other colorable, although ultimately unsuccessful, arguments, I find that Nuance has not shown that Vianix pursued its claims in bad faith. Therefore, this argument provides no basis on which to award Nuance its attorneys' fees and costs.
III. CONCLUSION
For the reasons stated in this Opinion, I find that: (1) of all disputed products, Nuance owes royalties only for Powerscribe SDK version 3.0, and only to the extent that Nuance shipped optional discs containing MASC to its customers; (2) no royalties are owed for telephone dictators; (3) Nuance owes a royalty for every End User of a MASC-bearing product, and (a) the amount of that royalty varies with the number of authorized End Users licensed by Nuance in each calendar year, but Nuance did not keep track of that data because it counted licenses, not End Users, and some of those licenses were concurrent licenses; (b) therefore, a 5x multiplier must be applied to the number of concurrent user licenses Nuance sold to approximate the true number of End Users licensed; (4) Nuance does not owe a royalty for renewal fees; (5) Nuance does not owe any royalties for Additional Site fees beyond what it has already paid; (6) no growth factor should be applied to determine 2008 iChart sales; (7) a 6 percent annual SMA fee should be applied to all royalties that Nuance owes; and (8) interest should be applied to all amounts Nuance owes Vianix at a rate of 1.5 percent per month compounded quarterly as stated in this Opinion. I thus order Vianix to populate the Damages Spreadsheet created by its expert Ellis in accordance with the rulings made in this Opinion. I further dismiss with prejudice Nuance's counterclaims for breach of contract and a declaratory judgment to the extent they seek any relief contrary to the rulings summarized above. Finally, I reject the requests of both parties for an award of their attorneys' fees and costs.Counsel for Vianix shall submit, after conferring with opposing counsel, a fully-populated version of the Damages Spreadsheet containing a final calculation of damages owed for unpaid royalties, SMA fees, and interest, as well as a proposed form of judgment implementing the rulings set forth herein within twenty days of the date of this Opinion.