Opinion
CASE NO. 98-2651-CIV-DIMITROULEAS
June 10, 2003
FINDINGS OF FACT AND CONCLUSIONS OF LAW
THIS CAUSE came on for non-jury trial on April 11, 14, 15, 16, 17, 21, 22, 23, 24, and 25, 2003. The Court has carefully considered the arguments of counsel, the evidence presented, and the testimony of the witnesses. The Court has also determined the credibility of witnesses and is otherwise fully advised in the premises.
Pursuant to Rule 52(a) of the Federal Rules of Civil Procedure, the Court makes the following Findings of Fact and Conclusions of Law. To the extent, if any, that the Findings of Fact as stated may be deemed Conclusions of Law, they shall be considered Conclusions of Law. Similarly, to the extent the matters expressed as Conclusions of Law may be deemed Findings of Fact, they shall be considered Findings of Fact.
I. INTRODUCTION
1. This is a copyright infringement action in which the Plaintiffs seek injunctive relief pursuant to 17 U.S.C. § 502 and § 119(a)(5)(B). Plaintiffs also seek costs and reasonable attorneys' fees pursuant to 17 U.S.C. § 505.
2. Plaintiff CBS Broadcasting, Inc. ("CBS") is a New York corporation with its principal place of business in New York, New York. Plaintiff Fox Broadcasting Company ("Fox") is a Delaware corporation with its principal place of business in Los Angeles, California. CBS and Fox operate the CBS and Fox television networks, respectively, which provide CBS and Fox network programming to television stations nationwide that are affiliated with CBS and Fox networks.
3. Plaintiffs ABC Televison Affiliates Association, CBS Televison Network Affiliates Association, FBC Television Affiliates Association and NBC Television Affiliates Association are voluntary membership trade associations representing network stations that are affiliates with their respective networks.
4. Defendant EchoStar Communications Corporation is a Nevada corporation. Defendants EchoStar Satellite Corporation and Satellite Communications Operating Corporation are Colorado corporations and subsidiaries of EchoStar Communications Corporation. The fourth defendant, DirectSat Corporation, a Delaware corporation, has merged with EchoStar Satellite Corporation. All defendants have their principal place of business in Littleton, Colorado. (The Court will refer to all the defendants collectively as "EchoStar.") EchoStar has 8.5 million customers. It has eight satellites orbiting the earth at 22,000 miles over the equator.
5. Plaintiffs claim that EchoStar's retransmission via satellite of copyrighted programming owned by Plaintiffs violates Plaintiffs' copyright in its network television broadcasts. The principal issue is whether EchoStar's actions are permitted by the Satellite Home Viewer Act ("SHVA"), as amended by the Satellite Home Viewer Improvement Act ("SHVIA"), which grants a limited statutory license to satellite carriers transmitting distant network signals to private homes if the subscribers are "unserved households." 17 U.S.C. § 119 (a)(2)(A), (B).
The Court has previously determined, in its Order dated March 31, 2003, that each of the plaintiffs has standing under the Copyright Act to pursue this copyright infringement lawsuit against EchoStar.
The Court will refer to both the Satellite Home Viewer Act and the Satellite Home Viewer Improvement Act collectively as SHVA unless otherwise indicated.
II. PROCEDURAL HISTORY
6. On October 19, 1998, EchoStar filed suit in Colorado against CBS, Fox, NBC and ABC seeking a declaratory judgment that EchoStar's method of qualifying subscribers for distant network programming complied with the law. On November 5, 1998, the Plaintiffs filed their complaint in this Court alleging that EchoStar was infringing Plaintiffs' copyrights by providing distant network programming to "served" households in violation of SHVA. The Colorado district court granted Plaintiffs' motion to transfer the suit filed by EchoStar to Florida and the Colorado litigation was consolidated with this action.
7. On September 29, 2000, the Court entered a preliminary injunction against EchoStar. On November 22, 2000, the Eleventh Circuit stayed the enforcement of the preliminary injunction. On September 17, 2001, the Eleventh Circuit vacated the preliminary injunction holding. The case was then remanded to this Court.
8. EchoStar filed a Motion for Summary Judgment on February 3, 2003. Plaintiffs filed a Motion for Summary Judgment on All Counts of Their Complaint and Count I of EchoStar's Counterclaim on February 4, 2003. Plaintiffs also filed a Motion for Summary Judgment on Counts II-V of EchoStar's Counterclaim on February 4, 2003. The Court granted Plaintiffs' Motion for Summary Judgment on Counts II-V of EchoStar's Counterclaim on March 24, 2003. The Court denied both EchoStar's Motion for Summary Judgment and Plaintiffs' Motion for Summary Judgment on All Counts of Their Complaint and Count I of EchoStar's Counterclaim on March 31, 2003. In that Order on Motions for Summary Judgment, the Court stated that all Plaintiffs had standing to proceed in this case. The Court further indicated that the issues to be decided at trial would be whether EchoStar is in violation of SHVA, whether a threat of future violation exists and whether EchoStar has engaged in a "willful or repeated pattern or practice."
9. Plaintiff ABC was dismissed from this case on April 15, 2002. Plaintiff NBC was dismissed on November 25, 2002.
III. FINDINGS OF FACT A. EchoStar and Distant Network Programming
10. EchoStar is a satellite carrier within the meaning of the Satellite Home Viewer Act ("SHVA"), 17 U.S.C. § 119, as amended by the Satellite Home Viewer Improvement Act ("SHVIA").
11. EchoStar operates a Direct Broadcast Satellite ("DBS") service called "DISH Network," which offers satellite television programming to subscribers who receive programming using small satellite dishes.
12. EchoStar retransmits the signals of network stations in two different ways: on a "local-to-local" basis and as distant signals. Local-to-local retransmission refers to the satellite delivery of network stations to a subscriber within the subscriber's own local market. EchoStar currently offers local-to-local programming in approximately 62 out of the 210 U.S. television markets.
13. Distant network signals are network stations from outside a subscriber's market area. For example, a person who lives in Fort Lauderdale but receives an ABC. CBS, Fox or NBC network station from New York City is receiving "distant network programming" or "distant network stations."
14. EchoStar sells distant ABC, CBS, Fox and NBC network programming to customers throughout the United States. As of April 2002, EchoStar had approximately 1,180,000 distant network subscribers.
B. EchoStar's Past Compliance Efforts 1. Subscribers signed up during PrimeTime 24 Period
15. From approximately March 1996 until approximately July 19, 1998, EchoStar offered distant network programming pursuant to an agreement with PrimeTime 24 Joint Venture ("PrimeTime 24"). During this period, subscribers were signed up using a subjective method which depended on the viewer's opinion concerning the quality of his or her picture.
16. In December 1996, CBS, Fox and other plaintiffs brought suit against PrimeTime 24. In 1998, this Court issued a preliminary injunction, and then a permanent injunction, requiring PrimeTime 24 to terminate the delivery of distant network signals to subscribers who had been signed up illegally. On July 19, 1998, days after the entry of the preliminary injunction, EchoStar terminated the delivery of the PrimeTime 24 distant network programming package to its subscribers and switched virtually all of the customers receiving the PrimeTime 24 network programming to its own, new distant network station packages, called DISHNets East and DISHNets West. On August 25, 1998, Michael Schwimmer, EchoStar's Vice President of Programming wrote affiliate station managers asking them to waive any interest in litigating EchoStar implementation of a red light/green light method instead of an ILLR method. The letter placed a burden on the station managers to object to this proposal within one (1) week.
2. Red Light/Green Light Method
17. EchoStar claims that after its break with PrimeTime 24, it implemented a "red light/green light" zip code system to determine which subscribers were eligible for distant network programming. A red light/green light system means that the satellite provider would refuse to sign up subscribers in a "red light" zip code for distant signals, unless they obtained a waiver from the station.
18. However, EchoStar presented no evidence that its red light/green light system was reasonably calculated to prevent signups of ineligible subscribers. In fact, Richard Biby, an outside vendor, testified that he was directed to use a 95% confidence factor in running Longley-Rice. This setting will shrink the predicted coverage areas of TV stations.
19. Further, Michael Hawkins, EchoStar's former employee on whom it relied to prepare four declarations submitted to this Court during 1999-2000 about SHVA compliance matters, testified that comparing EchoStar's red light zip codes to the area shown by the Individual Location Longley-Rice ("ILLR") model was like comparing a "golf ball to a softball." That is, the red light area defined by EchoStar was much smaller that the ILLR-predicted Grade B area.
20. Additionally, even if a Customer Service Representative ("CSR") was advised that a subscriber was in a red light zip code, the CSR was able to sign up the subscriber for distant network signals anyway.
21. Further, Plaintiffs' expert witness, Jules Cohen, P.E., asked Decisionmark to analyze EchoStar's signups during the period that EchoStar claims it used the red light/green light method, using the monthly signup lists and the red light zip code list provided by EchoStar. Mr. Cohen's testimony and expert reports support the conclusion that EchoStar signed up large numbers of subscribers in the forbidden red light zip codes. In January 1999, for example, EchoStar signed up 17,380 new subscribers for Fox distant stations in red light zip codes. See Cohen Supp. Expert Report, Table 4. That same month, EchoStar signed up 14,622 new subscribers for distant CBS stations in red light zip codes for that network. See Cohen Supp. Expert Report, Table 3. Overall, during the five months from November 1998 through March 1999, EchoStar signed up 62,374 subscribers in red light zip codes for CBS distant stations and 68,979 subscribers in red light zip codes for Fox distant stations.
The Court finds that the zip code list provided to Plaintiffs by EchoStar does constitute a list of red light zip codes used by EchoStar during the red light/green light period.
22. Additionally, an ILLR analysis of this same group of subscribers shows that EchoStar signed up 167,000 subscribers during these five months predicted to receive a signal of Grade B or better. For example, the ILLR analysis showed that 69% of EchoStar's customers signed up for CBS distant stations during this period are predicted by ILLR to receive a signal of at least Grade B intensity, and that 41% are predicted to receive a signal of Grade A intensity.
3. ILLR Methodology
23. The methodology was named Individual Location Longley-Rice ("ILLR") by the Federal Communications Commission ("FCC") in February 1999. EchoStar asserts that it has been using the ILLR methodology since 1999 to determine which new subscribers are eligible for distant signals. From the time EchoStar began using ILLR until October 2000, EchoStar ignored many stations that are predicted by the ILLR model to deliver Grade B or better signals to EchoStar subscribers by excluding all network stations that are not in the same Nielsen defined DMA as the subscriber. Also, EchoStar continued to include interference in its ILLR analysis until January 2002. Finally, EchoStar uses two vendors, Dataworld and Decisionmark, to verify the ILLR status of its subscribers.
24. In May 2002, EchoStar provided Plaintiffs with a complete list of its subscribers as of April 25, 2002.
On October 2, 2001, Plaintiffs served discovery requests on EchoStar seeking a complete list of EchoStar's subscribers as of October 2001. Plaintiffs filed a motion to compel the October 2001 subscriber list, which was granted. However, when asked for the requested copy of the October 2001 list, EchoStar stated that it had not retained a copy of the October 2001 list. The April 2002 list is the only complete subscriber list supplied by EchoStar to Plaintiffs (other than EchoStar's initial list of subscribers in October 1998).
25. Mr. Cohen, Plaintiffs' expert witness, supervised an Individual Location Longley-Rice ("ILLR") analysis by Decisionmark of the April 2002 subscriber list. Mr. Cohen is a qualified expert, and he was a credible witness. In May, 2000, the FCC relied on Mr. Cohen's work and published part of his study in an order on land use and land cover. Additionally, Mr. Cohen's reliance on Decisionmark, a competent ILLR vendor also used by EchoStar, to make the necessary ILLR predictions was reasonable.
26. The ILLR model takes into account terrain and, for UHF stations, land use and land cover, to determine the likely signal strength from any given television transmitter to a particular household.
27. In running the ILLR model, the first step is "geocoding," which employs a very large database of street names and addresses throughout the country to generate a latitude and longitude for each address. Geocoding produces accurate latitude and longitude for the vast majority of households, although it is not as precise for "Rural Route" or "General Delivery" addresses. Mr. Cohen testified that there were only approximately 43,000 of these rural route addresses in the database and that any deviations would have had a de minimus effect on his calculations and opinions. In theory, one could visit the location and use a Global Positioning System ("GPS") meter to get a more precise latitude and longitude for each location; however, that solution is not practical for large volumes of addresses.
28. Once the geocoding process had generated a latitude and longitude for a household, the Longley-Rice predictive model is used to predict the signal strength of the household. This is called the "Individual Location" Longley-Rice model because the prediction is made on a "point-to-point" basis from the televison transmitter to the household. Levels of intensity of over-the-air signals, such as "Grade A" or "Grade B," are objective measures of signal strength, expressed in a unit of measure called dBu.
29. Any errors in the geocoding process discussed by EchoStar's witnesses would have a de minimus effect on the outcome of the ILLR analysis supervised by Mr. Cohen.
30. The ILLR analysis supervised by Mr. Cohen shows that EchoStar has hundreds of thousands of illegal distant network programming subscribers for each of the four networks. On a network-by-network basis, the Decisionmark ILLR analysis supervised by Mr. Cohen showed the following:
(i) 457,584 (or 50.9%) of the 898,847 subscribers were predicted to receive at least a Grade B signal from an ABC station (excluding stations owned and operated by ABC, Inc.);
(ii) 481,659 (or 55.7%) of the 864,494 CBS subscribers were predicted to receive at least a Grade B signal from a CBS station;
(iii) 383,987 (or 38.7%) of the 993,490 Fox subscribers were predicted to receive at least a Grade B signal from a Fox station; and
(iv) 489,315 (or 56.4%) of the 867,240 NBC subscribers were predicted to receive a Grade B signal from an NBC station (excluding stations owned and operated by NBC).
31. For each network, hundreds of thousands of subscribers are predicted to receive the more powerful signal of Grade A intensity; 255,980 for ABC; 244,022 for CBS; 256,741 for Fox; and 256,503 for NBC.
The numbers in Paragraphs 30-33 reflect the corrected ILLR which includes the following changes: (1) for "General Delivery" and similar non-street addresses, this analysis uses street addresses from field "Address 2" where a usable address is present; (2) no rounding upwards of dBu.
32. Even when EchoStar's claims of waivers and grandfather status are treated as valid, the ILLR analysis shows high numbers of illegal subscribers to distant network programming for each of the four networks. Taking into account the waiver and grandfather claims, the ILLR analysis showed that:
(i) 238,048 (or 26.5%) of the 898,847 subscribers were predicted to receive at least a Grade B signal from an ABC station (excluding stations owned and operated by ABC, Inc.);
(ii) 232,699 (or 26.9%) of the 864,494 CBS subscribers were predicted to receive at least a Grade B signal from a CBS station;
(iii) 200,422 (or 20.2%) of the 993,490 Fox subscribers were predicted to receive at least a Grade B signal from a Fox station; and
(iv) 243,342 (or 28.1%) of the 867,240 NBC subscribers were predicted to receive a Grade B signal from an NBC station (excluding stations owned and operated by NBC).
33. For each network, a large number of subscribers are predicted to receive the more powerful signal of Grade A intensity, even when treating as valid EchoStar's waiver and grandfather claims; 164,409 for ABC; 152,140 for CBS; 156,906 for Fox; and 163,011 for NBC.
34. EchoStar attempted to rebut the ILLR analysis conducted by Decisionmark and supervised by Mr. Cohen. Noting that EchoStar has the burden of proving that each of its subscribers resides in an "unserved household" and that merely criticizing Plaintiffs' ILLR analysis would be insufficient to meet this burden, the Court finds that EchoStar's criticisms of the ILLR analysis presented by Plaintiffs are without substance.
35. First, the Court finds that Dr. Charles Jackson, EchoStar's primary witness presented to criticize the Cohen/Decisionmark ILLR analysis, while qualified in this field, lacked credibility based on the manner in which he answered questions asked by Plaintiffs' counsel.
36. Further, EchoStar argues that it was inappropriate to use the ILLR model (and current specifications for the locations and power of TV transmitters) for subscribers who were initially signed up at previous times. However, the Court finds that the effect on the ILLR analysis due to changes in the characteristics of TV transmitters is de minimus. Similarly, the effects of stations changing from one network to another network (for example, a CBS affiliate station changing to an NBC affiliate station) on the ILLR analysis would also be de minimus.
37. Also, the Court agrees with Mr. Cohen's testimony that the difference in the depression angle testified to by EchoStar witness, William Hammet, also had a de minimus effect on the ILLR analysis.
4. The Ergen promise and the 1999 Decisionmark ILLR analysis
38. In an effort to dissuade this Court from issuing a preliminary injunction, EchoStar CEO Charles Ergen made a formal pledge, under penalty of perjury, in September 1999 that each of the subscribers that EchoStar originally signed up using the unlawful "do you like your picture?" method would be checked by Decisionmark using the ILLR model and that all ineligible subscribers would be terminated.
39. EchoStar did send a list of its former PrimeTime 24 subscribers in September 1999 to Decisionmark for ILLR analysis. That list contained approximately 331,000 subscribers. In addition to sending Decisionmark a list of former PrimeTime 24 subscribers as of late September 1999, EchoStar also sent Decisionmark other lists created as of that time: (1) a list of all of EchoStar's roughly three million subscribers, whether or not they received distant network signals; and (2) a list of all of EchoStar's roughly 893,000 subscribers receiving distant network stations.
40. EchoStar received Decisionmark's ILLR analysis sometime in October 1999. Although it had promised the Court that it would turn off the former PrimeTime 24 subscribers shown to be Grade A or Grade B by the Decisionmark analysis (other than those that had waivers or signal tests), EchoStar made no report in any later court filing about the results of the Decisionmark ILLR analysis. In November, 1999, the copyright law was amended.
41. When Plaintiffs sought to obtain the Decisionmark analysis, EchoStar filed a motion to quash. Plaintiffs then filed a motion seeking production of the Decisionmark ILLR data, which was granted by the Magistrate Judge on April 9, 2003, two days before trial began. As a result of that Order, Decisionmark made the documents from its ILLR analysis available on a File Transfer Protocol ("FTP") Internet site.
42. Mr. Cohen, worked with two different database experts (one from Decisionmark and one from Plaintiffs' law firm) to ensure that the Decisionmark ILLR analyses were interpreted properly.
43. Among the total of 331,586 PrimeTime 24 subscribers, the ILLR analysis showed that the percentages of Grade A subscribers were 61% (ABC), 60% (CBS), 58% (Fox) and 60% (NBC). Further, the Decisionmark ILLR analysis showed that EchoStar was delivering distant network stations to more than 258,000 former PrimeTime 24 subscribers who were predicted to receive a Grade A signal from at least one of the four networks. In other words, if EchoStar had kept Mr. Ergen's promise to the Court in the September 1999 declaration, it would have terminated one or more distant network stations to more than 258,000 subscribers. And this figure does not include those predicted to be Grade B but not Grade A, who may later have been grandfathered after the enactment of SHVIA on November 29, 1999, if they had still been subscribers on October 31, 1999.
44. As a percentage of the total of 879,808 distant signal subscribers, those predicted to receive a Grade A signal were 53% for ABC, 51% for CBS, 48% for Fox and 52% for NBC. The percentage predicted to receive at least a Grade B signal, but not a Grade A signal, were 18% for ABC, 21% for CBS, 12% for Fox and 27% for NBC. Further, the Decisionmark ILLR analysis showed that EchoStar was delivering distant network stations to more than 630,000 subscribers who were predicted to receive a Grade A signal from at least one of the four networks. Thus, if EchoStar had turned off all of its Grade A subscribers, it would have terminated one or more distant network stations to more than 630,000 subscribers, amounting to 72% of its total distant subscribers and 21% of its total customer base of three million. Again, this does not include subscribers predicted to be Grade B but below Grade A.
45. No credible evidence was presented to the Court to support the contention that EchoStar turned off distant signals for compliance reasons to any of the more than 258,000 former PrimeTime 24 Grade A subscribers that Decisionmark told EchoStar about in October 1999. Nor is there any credible evidence that EchoStar turned off distant signals for compliance reasons to any of the more than 630,000 Grade A subscribers that Decisionmark told EchoStar about at that time.
46. It appears that EchoStar executives, including Mr. Ergen and David Moskowitz, when confronted with the prospect of cutting off network programming to hundreds of thousands of subscribers, elected instead to break Mr. Ergen's promise to the Court. During his testimony, Mr. Ergen stated that EchoStar intended to run its distant network programming subscribers through a Decisionmark ILLR analysis, but, as Mr. Ergen was aware of the possibility that the law regarding the subscribers could change (as he was lobbying for such a change), EchoStar decided to wait for the new law (anticipating the passing of SHVIA, which would grandfather certain subscribers). However, when Mr. Ergen made his promise to the Court, he did not qualify that promise by stating that he would not follow through if the new legislation was passed.
47. Further, the Court notes that when Mr. Moskowitz, an EchoStar executive who worked closely on SHVA compliance, was questioned during his deposition about the 1999 Decisionmark ILLR analysis, he paused for an unusually long period of time and then answered the questions concerning the ILLR analysis in a vague manner, unable or unwilling to give any details on the results of the analysis or EchoStar's actions following the analysis.
5. EchoStar's claims of mass turnoff's of ineligible distant network subscribers
48. Mr. Ergen and other EchoStar witnesses testified concerning alleged mass turnoffs of illegal distant network programming subscribers between the time that EchoStar ended its relationship with PrimeTime 24 and early 2002. However, EchoStar witnesses also admitted that EchoStar does not have a single list of subscribers whose distant network programming was terminated by EchoStar for compliance reasons nor other documentation of such mass turnoffs for compliance reasons. Instead, EchoStar points to its monthly lists of disconnected subscribers, submitted to the networks pursuant to SHVA reporting requirements, as proof that EchoStar conducted large-scale compliance terminations.
49. The Court rejects the monthly lists as proof of compliance-related terminations. As Mr. Ergen and Mr. Povenmire admitted, subscribers stop receiving distant network stations for a wide variety of reasons, including involuntary disconnections for failure to pay one's bills as well as voluntary disconnections because the subscriber has chosen to switch to cable, to switch to DirecTV, to switch to local-to-local service, to switch to C-band dish, to switch to over-the-air reception or to cancel EchoStar (or just their distant network subscription) because the household concludes in can no longer afford it. Both Mr. Ergen and Mr. Povenmire acknowledge that it is impossible to determine from EchoStar's monthly subscriber lists which subscribers had their service disconnected for any particular reason. Therefore, an increase in the numbers of subscribers on a monthly disconnect report is insufficient evidence that large numbers of subscribers were terminated for compliance reasons.
C. EchoStar's Compliance Efforts Today
50. The Court finds that, with the exception of its use of two vendors, EchoStar is currently taking reasonable steps to ensure that only those potential subscribers that are eligible under the law are signed up for distant network programming. Robert Lee of the CBS Affiliates Association, a witness for Plaintiffs, testified that EchoStar today is making legitimate efforts to qualify subscribers.
51. Mr. Povenmire described in detail and demonstrated for the Court EchoStar's current qualification process. EchoStar currently involves many people and departments in its efforts to ensure its compliance, including more than 7,000 Customer Service Representatives ("CSR"), each of whom is responsible for utilizing the ILLR databases to qualify new customers who wish to receive distant network channels. EchoStar provides each CSR with training and graphic screens designed to prevent misuse. Also, CSRs can no longer tinker with the system in order to approve ineligible households for distant network programming.
52. Additionally, in an effort to eliminate any inadvertent activation errors, EchoStar has instituted a back-up compliance system (the "WAC" or "Exemption Reporting" process). At the end of each month, EchoStar reviews all the people that were signed up in that previous month and reanalyzes them in order to disconnect any ineligible subscribers. EchoStar also "requalifies" subscribers anytime a subscriber attempts to modify network programming or moves to a new household.
53. At then end of each month, EchoStar goes through all the people who were signed up for networks in that previous month, and reanalyzes them. If any ineligible subscriber is identified, EchoStar promptly terminates distant network programming to such subscribers.
III. CONCLUSIONS OF LAW A. The Satellite Home Viewers Act
54. In 1988, Congress amended the Copyright Act to include the Satellite Home Viewers Act ("SHVA"). SHVA creates a narrow exception to a network's exclusive copyright over its programming in the form of a compulsory license allowing satellite carriers to retransmit copyrighted network programming for private home viewing to persons residing in "unserved households." 17 U.S.C. § 119 (a)(2)(B). Congress sought to achieve two goals in enacting this provision: (1) to provide network programming to the small number of households that otherwise lacked access to it; and (2) to preserve the existing network/affiliate television distribution system by preventing satellite delivery of network programming to other households. H.R. Rep. No. 100-187, Pt. 2 at 20 (1988); see ABC, Inc. v. PrimeTime 24 Joint Venture, 184 F.3d 348, 350-351 (4th Cir. 1999); CBS Broadcasting, Inc. v. PrimeTime 24 Joint Venture, 48 F. Supp.2d 1342, 1355 (S.D. Fla. 1998).
55. As stated above, the satellite carrier's license is limited to "unserved households." Under SHVA, an unserved household was described as a residence that:
(A) cannot receive, through the use of a conventional, stationary, outdoor rooftop receiving antenna, an over the air signal of a primary network station affiliated with that network of Grade B intensity (as defined by the Federal Communications Commission) of a primary network station affiliated with that network, and
(B) has not, within 90 days, before the date on which that household subscribes, . . . subscribed to a cable system that provides the signal of a primary network station affiliated with that network.17 U.S.C. § 119 (a)(2) (1996). The Satellite Home Viewers Improvement Act ("SHVIA"), enacted in November 1999, renewed SHVA's existing statutory copyright license and altered the definition of unserved household. CBS Broadcasting, Inc. v. EchoStar Communcations Corp., 265 F.3d 1193, 1198 (11th Cir. 2001). Now, under the SHVIA amendments, "unserved household" is defined in five different ways.
56. First, an unserved household means a household that "cannot receive, through use of conventional, stationary, outdoor rooftop receiving antenna, an over-the-air signal of a primary network station affiliated with that network of Grade B intensity as defined by the Federal Communications Commission under section 73.683(a) of title 47 of the Code of Federal Regulations, as in effect on January 1, 1999," 17 U.S.C. § 119 (d)(10)(A). Second, a household is unserved if it receives a waiver from each network station affiliated with a particular network that is predicted to deliver a Grade B or better signal to the subscriber's residence. 17 U.S.C. § 119 (d)(10)(B). Third, the term unserved household includes subscribers who receive a signal of less than Grade A intensity for a particular network and who received satellite service of that network's signals on October 31, 1999 or had such service terminated for SHVA ineligibility between July 11, 1998 and October 31, 1999 ("grandfathered subscribers"). 17 U.S.C. § 119 (d)(10)(C). Fourth, an unserved household includes subscribers who receive distant signals through a satellite dish located on a commercial truck or recreational vehicle, and who satisfy the strict documentation requirements. 17 U.S.C. § 119 (d)(10)(D). Finally, a exemption exists for secondary transmissions by C-band services of network stations that a subscriber to a C-band service received before any termination of such secondary transmissions before October 31, 1999. 17 U.S.C. § 119 (d)(10)(E).
The Grade B standard found in the first definition of unserved household has been a part of SHVA since its enactment in 1988.
C-band service is defined as "a service that is licensed by the Federal Communications Commission and operates in the Fixed Satellite Service under part 25 of title 47 of the Code of Federal Regulations." 17 U.S.C. § 119 (a)(2)(B)(iii)(II). This provision is not relevant to EchoStar's small-dish service.
57. A satellite carrier, such as EchoStar, has the burden at trial of proving that its transmission of distant network programming goes only to unserved households. 17 U.S.C. § 119 (a)(5)(D); CBS Broadcasting, Inc. v. EchoStar Communications, 265 F.3d at 1201; CBS Broadcasting, Inc. v. PrimeTime 24, 48 F. Supp.2d at 1356; ABC, Inc., 184 F.3d at 352.
58. Congress has provided two methods to be used by a Court in determining the signal intensity at a particular location: (1) predictions made using a computer model approved by the FCC called the Individual Location Longley-Rice ("ILLR") model ( 17 U.S.C. § 119 (a)(2)(B)(ii)(I)); and (2) signal intensity measurements, following FCC-prescribed procedures, in the vicinity of the subscriber's home ( 17 U.S.C. § 119 (a)(2)(B)(ii)(II)). The first method allows a satellite provider to establish presumptively that a subscriber qualifies as unserved where the ILLR model establishes that the residence cannot receive a Grade B over-the-air signal of a primary network station. 17 U.S.C. § 119 (a)(2)(B)(ii)(I); CBS Broadcasting, Inc. v. EchoStar Communications Corp., 265 F.3d at 1200.
B. EchoStar's Compliance with SHVA
59. The Court finds that EchoStar has failed to meet its burden of proving that its subscribers are "unserved households." First, EchoStar has failed to show that its distant signal subscribers cannot receive an over-the-air signal of Grade B intensity. EchoStar did not present at trial any ILLR analysis of its distant network signal subscribers nor did EchoStar present any results of signal intensity measurements at the homes of representative subscribers. Additionally, Mr. Cohen's testimony and reports, as discussed in the Court's Findings of Facts, support the conclusion that hundreds of thousands of EchoStar's distant network programming subscribers are not unserved households. The Court notes that Mr. Cohen's direction to Decionsmark to round dBu levels to 47 when the result was above 46.5, but below 47, was improper. However, Mr. Cohen corrected this error, and his conclusion that a large number of EchoStar's distant network subscribers are ineligible was not disturbed by the error. Finally, the Court finds that the use of NAD83 is appropriate (as opposed to NAD27) to calculate data points; however, the effect of any possible use of NAD27 by Decisionmark in the ILLR analysis of the April 2002 subscriber list would be de minimus and would not affect Mr. Cohen's conclusions regarding the large number of ineligible subscribers; the difference between NAD83 and NAD27 results in a difference of one hundred (100) feet or less in an actual location.
An engineering firm hired by EchoStar, Hammett Edison, did perform a small-scale ILLR analysis of 956 EchoStar subscriber locations. However, as William Hammett testified, the purpose of that analysis was not to present an audit of the eligibility of EchoStar distant network programming subscribers. Rather, the purpose of the study was to illustrate the fact that different vendors sometimes produce different results when performing an ILLR analysis.
A dBu level of 47 is the cutoff for determining Grade B eligibility. Approximately 12,000 customers had a dBu initially rounded up to 47 by Mr. Cohen. Similar de minimus amounts of customers had a dBu rounded up to 56 and 64 for higher numbered television channels.
60. Additionally, no basis exists in SHVA nor in the FCC's directives concerning the ILLR model to support the use of the "DMA rule." From the time EchoStar began using ILLR until October 2000, EchoStar, in running potential subscribers through the ILLR model, failed to consider stations that were predicted by the ILLR model to deliver a Grade B or better signal to its subscribers because these stations were not in the same Nielsen-defined DMA as the subscribers. The Court finds that use of the "DMA Rule" is improper and that EchoStar's ILLR methodology was flawed during the time that it was applying the "DMA Rule." Further, EchoStar presented no evidence that it turned off any ineligible subscribers it had previously signed up while using the "DMA Rule."
61. Next, the Court finds that EchoStar's use of interference after May 2000 was improper. In May 2000, the FCC issued a First Report and Order in which it made certain modifications to the ILLR model and published a detailed "cookbook" prescribing, on a step-by-step basis, how the model is to be employed. FCC First Report Order, May 2000 at A-2. The FCC's detailed directions explained how to calculate whether an individual household was unserved. Although the FCC had previously indicated that interference was to be included in the ILLR model, the FCC's May 2000 order changed the use of interference. The FCC reaffirmed the point when it later issued OET (Office of Engineering and Technology) Bulletin 72 in July 2002, which repeated the same "cookbook" directions, again excluding any consideration of interference.
62. Consideration of interference can only help EchoStar by treating a subscriber with a relatively weak (but above Grade B) signal as unserved because of the predicted presence of interfering signals from other stations. The FCC is aware of the issue of interference, since in another "cookbook" about how to use the Longley-Rice model for a different purpose, it provided specific directions on how to take interference into account. See OET Bulletin 69 ("Longley-Rice Methodology for Evaluating TV Coverage and Interference") (July 1997). Had the FCC wished to include interference in the current ILLR model, it would have done so.
63. Further, EchoStar was advised by Decisionmark of the FCC's May 2000 ruling that the ILLR model no longer included interference. In fact, Mark Castagneri of EchoStar sent an email dated July 27, 2000 to several other EchoStar employees, attaching a memo prepared by Jane Schlegel of Decisionmark. In that memo, Ms. Schlegel indicated that "[i]nterference calculations were eliminated." Additionally, in a follow-up email dated July 31, 2000, Mr. Castagneri indicated that "[b]asically, interference is out and LU/LC is in for calculating signal strength." Mr. Castagneri attached an earlier email from Ken Franken of Decisionmark making the same "cookbook" point discussed above — that the FCC-prescribed recipe for ILLR analysis precludes use of interference. Mr. Franken's email indicated that the FCC official most knowledgeable about the ILLR program, Bob Eckert, agreed that interference was no longer part of the ILLR model after the May 2000 Order.
64. Accordingly, EchoStar's use of interference at all times after August 2000 (at the latest) was in knowing violation of applicable legal standards.
65. Additionally, the Court finds that EchoStar's past and present use of two vendors, Dataworld and Decisionmark, to perform the ILLR analysis of potential subscribers is improper. EchoStar could have chosen either of these two competent vendors to check the ILLR status of its subscribers; however, EchoStar choose to use both vendors in order to exploit inaccuracies in the databases of either company to its own benefit. In every case in which the two vendors differ as to eligibility, EchoStar took the vendor that predicted the household to be unserved. Therefore, a subscriber was only turned down for distant network programming if found to be ineligible by both vendors. The Court finds that this particular practice of using two vendors, which always favors the satellite carrier and never favors the network station, is unlawful. Further, the Court finds that EchoStar's ILLR analysis of its subscribers during the period that it used two vendors (up to the present) is fundamentally flawed and that EchoStar signed up subscribers using the two vendor method who should not have been considered eligible for distant network programming. Certainly, there are times that one vendor's database will be more current; for example, after September 11, 2001, the New York City antenna for local stations was moved to a lower and different location, presumably affected the signal strength. Additionally, a tower in Scott's Bluff, Nebraska, recently collapsed. A translator recently went down in Oregon. Conversely, a higher tower was constructed in Salisbury, Maryland in 2001. Those changes were not shown to be captured more frequently by either vendor; therefore, any deviation would be de minimus. Finally, since 1999, Nielsen has moved 103 counties into different DMAs. This change also had a de minimus effect.
66. Further, the Court finds that EchoStar has not met its burden in showing that any of its distant signal subscribers meet the grandfathering requirements. Section 119(e) provides:
(e) Moratorium on copyright liability. Until December 31, 2004, a subscriber who does not receive a signal of Grade A intensity (as defined in the regulations of the Federal Communications Commission under section 73.683(a) of title 47 of the Code of Federal Regulations, as in effect on January 1, 1999, or predicted by the Federal Communications Commission using the Individual Location Longley-Rice methodology described by the Federal Communications Commission in Docket No. 98-201) of a local network television broadcast shall remain eligible to receive signals of network stations affiliated with the same network, if that subscriber had satellite service of such network signal terminated after July 11, 1998, and before October 31, 1999, as required by this section, or received such service on October 31, 1999.17 U.S.C. § 119 (e).
67. In essence, this grandfathering provision contains two requirements: (1) the subscriber either has to have been a customer on October 31, 1999 or has to have had distant network service terminated "as required by this section" after July 11, 1998 and October 31, 1999; and (2) the customer has to be below Grade A as predicted by the ILLR model.
68. First, EchoStar executive Rex Povenmire testified that EchoStar does not have any list of its distant signal subscribers as of October 31, 1999. Nor did EchoStar present a list of subscribers who were terminated after July 11, 1998 and before October 31, 1999.
69. Second, EchoStar has presented no ILLR study showing that its supposedly grandfathered subscribers are predicted to receive less than a Grade A signal.
70. Also, EchoStar continued to serve distant network subscribers who, even if they originally qualified, no longer qualified for grandfather status, according to the testimony of Rex Povenmire and Suzanne Beall.
71. in December 2000, EchoStar announced that it was going to terminate distant networks to any grandfathered subscriber who under the ILLR model existing in December 2000 was predicted to receive a Grade A signal from one or more networks. According to Mr. Povenmire, all existing distant network grandfathered subscribers were requalified under the ILLR model existing at that time and those subscribers who were predicted to receive a Grade A signal from one or more networks were identified for possible termination of such distant network programming. As a part of this process, the subscribers who were identified had their "grandfather" service code (or specialty flag) removed from the subscriber's account.
72. EchoStar received numerous complaints from subscribers who were targeted to be disconnected from one or more networks who believed that they still qualified to receive distant network programming because they had received a Grade B signal as of October 31, 1999. As a result of the complaints, EchoStar did not terminate distant network programming to any of these subscribers because it concluded that under SHVIA, a subscriber who received a Grade B signal of a network and who was a subscriber as of October 31, 1999 continued to be grandfathered and thus eligible for distant network programming. However, EchoStar's interpretation of the grandfathering provision of SHVIA is incorrect. That provision states that grandfather status applies only to a "subscriber who does not receive a signal of Grade A intensity." 17 U.S.C. § 119 (e). Thus, any subscriber who currently receives a Grade A signal does not qualify for grandfather status.
73. Mr. Povenmire further testified that in February, 2002, EchoStar again requalified all of its existing distant network subscribers using the ILLR model existing at the time, and those subscribers who were predicted to receive a Grade A signal had his or her grandfather service code (or specialty flag) removed from his or her account. However, there is no indication that such subscribers were terminated from distant network service.
74. Currently, EchoStar only re-qualifies a subscriber with a grandfather service code (or specialty flag) if the subscriber requests a change in programming or receives other services by EchoStar. If such a re-qualification shows that a subscriber is predicted to receive a Grade A signal from any network for which the subscriber had previously enjoyed grandfather status, the subscriber's distant network programming for such network is disconnected as part of EchoStar's monthly WAC process and the grandfather service code (or specialty flag) is removed from the subscriber's account.
75. Based on the above description of EchoStar's process of issuing grandfather status to its distant network subscribers, EchoStar has provided service to a large number of subscribers who have been incorrectly issued grandfather status.
76. Next, the Court finds that EchoStar has also failed to prove that any of its subscribers are eligible because they have obtained waivers from the relevant network stations. For a particular household to have a valid waiver with respect to a particular network, it is necessary to obtain a waiver from every station affiliated with that network that the ILLR model predicts to deliver a Grade B signal to the household.
77. EchoStar failed to obtain complete waivers for several years. Instead. EchoStar improperly considered viewers to be waived based on consent only from the station in the subscriber's own Nielsen-defined designated market area ("DMA"). As EchoStar Vice President Suzanne Beall testified:
Q: There was a time, wasn't there, when EchoStar was processing waivers manually; isn't that right?
A: Correct.
Q: And at that time when EchoStar was processing waivers manually, as a matter of company policy, EchoStar would only seek waivers from the station — from the DMA station, the station in the DMA, in which the subscribers resided; is that correct?
A: I believe, yes.
4/17/03 Aft. Tr. 123-24. This practice continued through at least early 2000.
78. Further, at trial EchoStar simply presented a massive list of subscribers as to which it claimed waivers. However, it did not present any credible testimony that it had obtained waivers from each station predicted to deliver a signal of Grade B intensity to each subscriber or to any particular subscriber. Accordingly, EchoStar has failed to prove that any of its 1.2 million distant network subscribers are "unserved households" because they have obtained waivers from stations. Additionally, EchoStar's methods both of sending bulk waivers without the customer's requesting one and of placing the burden on stations to object to a waiver are questionable practices.
EchoStar indicated during its opening statement that it would offer an expert witness, Dr. Rausser, to describe an "audit" of a fraction of EchoStar's waiver claims; however, Dr. Rausser was not called as a witness at trial. Instead, EchoStar asked Dr. Charles Jackson to summarize Dr. Rausser's report. Dr. Jackson is not a statistician and it appears that his only familiarity with Dr. Rausser's work is a reading of Dr. Rausser's report before testifying. As discussed above, the Court found that Dr. Jackson lacked credibility and his brief account of Dr. Rausser's report, which was not entered into evidence, does not assist EchoStar in meeting its burden of proof as to waivers.
79. Finally, EchoStar has defaulted in proving that any of its subscribers are eligible under the RV/commercial truck exception. This exception is limited to persons who have strictly complied with the documentation requirements of section 119(a)(11). See Conference Report, intellectual Property and Communications Omnibus Reform Act of 1999, H.R. Rep. 106-464 at 99 (1999) ("To prevent abuse of this provision, the exception for recreational vehicles and commercial trucks is limited to persons who have strictly complied with the documentation requirements set forth in section 119(a)(11)."). Those documentation requirements are as follows:
(B) Documentation Requirements. A recreational vehicle or commercial truck shall be deemed to be an unserved household beginning 10 days after the relevant satellite carrier provides to the network that owns or is affiliated with the network station that will be secondarily transmitted to the recreational vehicle or commercial truck the following documents:
(i) Declaration. A signed declaration by the operator of the recreational vehicle or commercial truck that the satellite dish is permanently attached to the recreational vehicle or commercial truck, and will not be used to receive satellite programming at any fixed dwelling.
(ii) Registration. In the case of a recreational vehicle, a copy of the current State vehicle registration for the recreational vehicle.
(iii) Registration and license. In the case of a commercial truck, a copy of
(I) the current State vehicle registration for the truck; and
(II) a copy of the valid, current commercial driver's license, as defined in regulations of the Secretary of Transportation under section 383 of title 49 of the Code of Federal Regulations, issued to the operator.17 U.S.C. § 119 (a)(11).
80. Again, EchoStar has simply produced a list of 12,000 customers that it claims are entitled to this exemption. The only witness who mentioned anything about the required documentation was Rex Povenmire, who said that he had looked at 12 of the subscribers allegedly falling under the RV/commercial truck exemption, but he stated that he did not know what to look for because he never reviewed the documentation provisions. Accordingly, EchoStar has failed to prove that any of its subscribers are eligible under the RV/commercial truck exemption.
C. Willful or Repeated Violations
81. Under SHVA, a copyright violation exists where a satellite carrier makes either "willful" or "repeated" transmissions of a network station's television programming to subscribers who do not reside in an unserved household. 17 U.S.C. § 119 (a)(5)(A). In order to prove willfulness, "it is necessary only to show that a person knew it was doing the acts in question, not that the person knew those acts were wrong." CBS Broadcasting, Inc. v. PrimeTime 24 Joint Venture, 48 F. Supp.2d at 1356. Under this standard, EchoStar's violations of the unserved household were clearly willful. Even if the "willful or repeated" standard requires a finding of gross negligence, EchoStar's violations also meet this standard. See ABC, Inc., 184 F.3d at 353 ("For unlicensed transmission to be repeated within the meaning of SHVA, the satellite carrier must have acted with gross or aggravated negligence.").
82. The grounds for finding that EchoStar's violations are "willful or repeated" include the following:
(1) EchoStar has failed to present credible evidence, either in the form of an ILLR analysis or signal intensity measurements, that any of its subscribers are unserved as defined under SHVA.
(2) Plaintiffs' evidence indicates that a significant percentage of EchoStar's distant network programming subscribers receive a signal of Grade B intensity or better.
(3) EchoStar incorrectly incorporated the DMA Rule into its ILLR analysis until October 2000.
(4) EchoStar's use of interference after May 2000 was unlawful.
(5) EchoStar improperly employs two vendors to perform ILLR analyses on potential subscribers and only refuses distant network programming to a subscriber if that subscriber is found ineligible by both vendors.
(6) EchoStar did not establish that any of its subscribers qualify for grandfather status; to the contrary, EchoStar was aware that many of these subscribers receiving service as grandfathered subscribers were predicted to receive a Grade A signal, yet such subscribers were not terminated.
(7) EchoStar failed to present credible evidence to support its claims that many of its subscribers have received waivers or have properly established they qualify for the RV/commercial truck exception.
D. Pattern or Practice
83. Under SHVA, if a satellite carrier is found to have engaged in a "willful or repeated pattern or practice" of transmission to households that are not unserved, and
(i) if the pattern or practice has been carried out on a substantially nationwide basis, the court shall order a permanent injunction barring the secondary transmission by the satellite carrier, for private home viewing, of the primary transmissions of any primary network station affiliated with the same network. . . .17 U.S.C. § 119 (a)(5)(B)(i).
84. The Court finds that, even if a pattern or practice existed at the time this case was filed, no such pattern or practice currently exists which would warrant such an extreme sanction. The current qualification system employed by EchoStar's CSRs and applied to every potential distant network subscriber is a reasonable system to prevent ineligible households from receiving distant network programming (other than EchoStar's use of two vendors). Additionally, EchoStar's WAC system further ensures that no new ineligible households are signed up. Finally. EchoStar makes royalty payments ($0.15 per subscriber per month) twice a year to the copyright office and has paid over $40,000,000 so far. Such efforts by EchoStar to comply with the law support the conclusion that no pattern or practice of willful or repeated violations exists that would warrant a nationwide injunction barring the transmission of distant network stations. Even Robert Lee of the CBS Affiliate Association noted that today EchoStar is making legitimate efforts to qualify subscribers.
E. Injunctive Relief
85. When a satellite carrier has committed individual violations, a court may award any of the remedies provided by sections 502 through 506 and 509 of the Copyright Act, with certain limitations concerning damages. 17 U.S.C. § 119 (a)(5)(A). Section 502(a) authorizes the Court to "grant temporary and final injunctions on such terms as it may deem reasonable to prevent or restrain infringement of a copyright." 17 U.S.C. § 502 (a). Therefore, the Copyright Act grants broad discretion to the court in fashioning a remedy that the Court deems appropriate under the circumstances of a particular case. CBS Broadcasting, Inc. v. PrimeTime 24 Joint Venture, 48 F. Supp.2d at 1360.
86. To be entitled to a permanent injunction under the Copyright Act, a copyright owner must show past infringement and the threat of future infringement by the defendant. CBS Broadcasting, Inc. v. PrimeTime 24 Joint Venture, 48 F. Supp.2d at 1358; Pacific Southern Co. v. Duncan, 744 F.2d 1490, 1499 (11th Cir. 1984); Walt Disney Co. v. Powell, 897 F.2d 565, 567-68 (D.C. Cir. 1990); Morley Music Co. v. Cafe Continental, Inc., 777 F. Supp. 1579, 1583 (S.D. Fla. 1991). Accordingly, there is a presumption of irreparable harm from copyright infringement. CBS Broadcasting, Inc. v. PrimeTime 24 Joint Venture, 48 F. Supp.2d at 1358.
87. However, even if the existence of harm were relevant to Plaintiffs' case, the Court finds that Plaintiffs have demonstrated that they suffer economic injury from EchoStar's illegal transmissions of network programming. Plaintiffs' witnesses testifying by deposition explained that local networks depend on the sale of advertising for their revenues. Further, it is likely that some illegal EchoStar distant signal subscribers will be Nielsen households and, as a result of the receipt of the distant signals, the local affiliates will have less Nielsen-reported viewing because these households will report (for example) viewing C.S.I. from the CBS station in New York City rather than from their local CBS station. Lower Nielsen ratings means lower advertising revenues. Therefore, although Plaintiffs need not show the existence of harm in order to obtain a permanent injunction in this case, Plaintiffs have sufficiently established that such harm exists.
88. As discussed above, past infringement of Plaintiffs' copyrighted works has been established in this case. Further, a threat of future infringement exists in this case as EchoStar continues to provide distant network programming to subscribers who were signed up under unlawful methods and who are ineligible for such service; additionally, EchoStar continues to use two vendors to perform ILLR analyses, which the Court has found unlawful. Therefore, a permanent injunction against EchoStar is appropriate in this case.
89. A separate Final Judgment will be entered herein consistent with the Court's Findings of Fact Conclusions of Law.