Opinion
No. 08 C 6446.
August 29, 2011
MEMORANDUM OPINION AND ORDER
Plaintiff Nielsen Company produces and licenses advertising and demographic information for television and other media. Nielsen presents the information it collects in maps, referred to as "Designated Market Area" or "DMA" maps, which divide the United States into approximately 210 television market regions. Each county, or portion thereof, in the country is assigned to one of those regions on the basis of the television market that attracts the largest share of viewers from that county. Nielsen has created DMA maps since the 1960s. In this lawsuit, Nielsen charges Defendant Truck Ads, LLC, with unlawfully copying and infringing Nielsen's copyright in the DMA maps for the years 2002-03, 2003-04, and 2004-05 by displaying a map on the Truck Ads website and distributing similar maps to Truck Ads customers. Truck Ads defends the case by challenging the validity of Nielsen's copyright on several grounds: Truck Ads contends that because the maps depict facts or ideas, have been wholly or partially copied from others, or are in the public domain, they lack the originality required to merit copyright protection. Nielsen insists it does have a valid copyright, that the boundaries drawn on the maps are protectable elements, and that Truck Ads copied the maps.
Both sides have moved for summary judgment. For the reasons explained here, the court concludes that Nielsen does have a valid copyright, and that the DMA region boundaries are protectable elements. Whether Truck Ads in fact copied Nielsen's maps is a jury question, however. Accordingly, Truck Ads' motion is denied and Nielsen's motion is granted in part and denied in part.
BACKGROUND
The Nielsen Company (US), LLC, is a Schaumburg, Illinois-based company organized under the laws of Delaware. Defendant Truck Ads, LLC, is based in McLean, Virginia. (Def.'s 56.1(a) ¶ 1; Pl.'s 56.1(a) ¶ A.) Truck Ads' primary business is operation of an "affiliate system" of third parties who sell advertising on the sides of trucks and mobile billboards throughout the United States and numerous other countries. ( Id.) Essentially, Truck Ads provides a toll-free phone number (now 1-800-TRUCK WRAPPER, previously, 1-800-TRUCKADS), to its affiliates; when individuals interested in mobile advertising call that number, they are automatically redirected to the Truck Ads affiliate located in the market region linked to the area code and prefix of the caller's phone number. ( Id.; Deposition of Rod Harris, Truck Ads CEO, Def.'s Ex. 1 at 33-34 ["Harris Deposition"].") Truck Ads also operates a web site, truckads.com, on which it advertises a wide variety of other services not at issue in this case. In connection with its business, Truck Ads displays a "Designated Market Area" map that displays media markets throughout the United States. See "DMA Map of United States," online at http://www.truckads.com/licensed_affiliates1.asp#usamap (visited August 24, 2011). Whether these DMA maps are entitled to copyright protection, and whether Truck Ads has infringed, are hotly disputed.
I. Copyrightability of Nielsen's DMA Maps
For several decades, Nielsen has produced and licensed television ratings and audience estimates, and also analyzed advertising and demographic information for television, radio, print and the Internet. (Pl.'s 56.1(a) ¶ A.) Among the products that Nielsen annually creates and licenses are "Designated Market Area" ("DMA") maps. Nielsen first created these maps, depicting "geographically distinct marketing regions within the United States," in the 1960s. (Pl.'s 56.1(a) ¶ 1.) The parties agree that a DMA region "consists of all counties whose largest viewing share is given to stations of [the] same market area." (Def.'s Supp. ¶ 1; Pl.'s Supp. Resp. ¶ 1.) Nielsen explains that it draws the boundaries for its DMA regions by grouping counties together according to "its own estimates, projections and ratings of television program audiences." (Pl.'s 56.1(a) ¶ 1.) The DMA maps in this lawsuit each contain 210 market areas. ( Id. ¶ 3.) Nielsen holds copyrights for each of the three DMA maps at issue: 2002-03 (No. VA0001671073); 2003-04 (No. VA0001671036); and 2004-05 (No. VA0001671036). ( Id. ¶ 12.)
At its most basic, Nielsen's DMA process involves assigning each county or portion of a county to a single, exclusive region. (Pl.'s 56.1(a) ¶ 10(a).) Nielsen determines which counties fall into which DMA regions each year using data collected from "Local People Meters" and/or "measurement diaries." ( Id. ¶¶ 5, 6, 7; Pl.'s Ex. P, Deposition of Bruce Hoynoski, Senior Vice President Measurement Science Client Service Executive for Global Media Operations, at 80 ["Hoynoski Deposition"].) Local People Meters are electronic devices that constantly monitor the viewing behavior of residents in the 25 largest DMA regions to measure "both household tuning as well as persons' viewing behavior from those homes." (Hoynoski Dep. at 80.) Measurement diaries, used in smaller markets, are paper records distributed to a random sample of households based on a two-stage matching process. (Pl.'s 56.1(a) ¶ 8(a).) In extrapolating from this sample, Nielsen uses proprietary population estimates licensed from a third-party, Acxiom Corporation, and historical projections based on the U.S. Census and its own telephone survey data. ( Id. ¶ 8(b)-(f).) The measurement diaries, which Nielsen itself designs, consist of a grid on which families record "information about their television activity, such as when is the TV on, what channel are they tuned to, what is the name of the program they're watching, and then who specifically in the family is in the audience." (Hoynoski Dep. at 81.) In the 26th to 60th largest regions, Nielsen uses a combination of devices similar to the Local People Meters as well as measurement diaries. ( Id. at 80.) The meters installed in these homes measure "when the [television] set is on and what channel it is tuned to," but not the specific programs being viewed. ( Id.)
The system Nielsen uses to generate its sample is relatively complicated. After determining the appropriate sample size for a given region, Nielsen uses a random numerical digit to select an area from which it will draw samples. Nielsen then uses other random digits to select phone numbers (all within a particular area code and exchange) that are linked to households within those areas. (Pl.'s Ex. O at N000526-528.) A third party, Acxiom Corporation, furnishes Nielsen with full contact information for any household whose phone number is generated by this process. ( Id.) Nielsen then runs another random number generator based on its desired sample size to select a sample from the list returned by Acxiom. ( Id.) It also screens out business addresses and takes additional steps to maintain its desired sample size, geographic proportionality, and to accommodate other factors it deems important to its analysis.
Truck Ads does not dispute that Nielsen uses these particular methods, but argues that Nielsen has not shown that "use of diary data affected the assignment of counties to market areas," or that measurement diaries have any "demonstrated effect on any county assignments." (Def.'s Resp. ¶ 6, 7, 8(a)-(j).) As Hoynoski explained, however, Nielsen does in fact use the Local People Meters and measurement diaries to assist in determining DMA boundaries. The DMA boundaries are set based on the origination of the majority of channels and programs being viewed in a given area; that information is measured by the diaries and the Local People Meters:
The information collected from the diaries and from the set meters in all of our markets is used in determining the boundaries by examining the information as to what television stations were being watched so that we can determine the market location where their preferences are going. In diary-only markets, all examinations and determinations are made [based] on the diary data, and when I say the diary data, I'm talking about the data pertaining to household level activity, not to demographic usage. In the set meter markets where we have the diaries also, we examine the diary data first. If information comes back from the diaries that passes the testing rules for reassigning a county from its current DMA to a different one, we will then take a look at the set metered homes in that county, and if we have at least 40 homes sampled in that county, we will only reassign that county if the set data, set meter data, agrees with the diary. If it [the meter data] disagrees with the diary, then we overrule and we will not move that county. In the people meter markets where there are no diaries, all of our decisions on county assignments are based on the meter data.
"Household level activity" refers to the channels and programs that are being watched by someone in the household being monitored. "Demographic usage" refers to information about the demographic characteristics of those watching particular programs, for example, an 18-year-old male in a household watching a particular program.
(Hoynoski Dep. at 89-90.) Nielsen has also submitted to the court several "Local Reference Supplements," which describe the process of measuring, creating, terminating, retaining, and testing DMAs in some additional detail. (Pl.'s Ex. O at N000454-N000460; N000953-N000962; N000628-N000637.) Nielsen explains that it does have procedures to split a county into multiple segments, describes its process for determining whether to create a new DMA boundary, and, in support of its contention that it uses independent judgment in creating DMAs, notes that it may choose not to create a DMA boundary "that would otherwise qualify if Nielsen determines there is a lack of financial support for the market area." (Pl.'s 56.1(a) ¶ 10(b)-(d).)
The record does not explain to whom Nielsen distributes these Local Reference Supplements, but they appear to be directed towards Nielsen's licensees. In the foreword to one of the supplements, Nielsen explains that the supplements "sets forth in detail the methodology, techniques, policies and procedures" involved in the DMA process as well as other activities Nielsen engages in, which are not the subject of this lawsuit. Nielsen explains that "[a]n understanding of these matters will enable the local market user to better understand the meaning of the data, including limitations, and how the data may be used most effectively." (Pl.'s Ex. O at N000452.)
Without citing any direct evidence on the issue, Truck Ads disputes that Nielsen actually follows these procedures. Truck Ads also asserts that Nielsen's processes do not translate into actual changes to DMAs. As evidence for this assertion, Truck Ads notes that Nielsen's maps do not reflect any new split counties, any new DMA areas, or any other significant changes during the period from between 2002 through 2005. (Def.'s Resp. ¶ 10(b)-(c).)
Nielsen and Truck Ads also disagree about whether Nielsen creates new maps each year, or whether Nielsen makes only minor, incremental changes that are (in Truck Ads' view) insufficient to bestow copyright protection on the entire DMA maps. Nielsen claims that it "tests every county in the United States each year," and points to its 2002-2003 "Local Reference Supplement" that describes the testing procedures for each county. (Pl.'s Supp. Resp. ¶ 16; Pl.'s Ex. O at N000453-54.) Truck Ads maintains that Nielsen evaluates no more than 90 to 120 "remainder" counties for reassignment, and actually reassigns only 20 to 25 each year. (Def.'s Supp. ¶ 16.) Plaintiff challenges Truck Ads' assertion that Nielsen evaluates just 90 to 120 counties each year, but agrees that it reassigns only a small number of counties each year, explaining that it does so because in many cases reassignment is not warranted. (Pl.'s Supp. Resp. ¶ 16.)
No formal definition of "remainder" counties is offered by either party, but Hoynoski explains in his deposition that a "remainder" county is a county that is not assigned to a "metro or central area of a DMA." (Hoynoski Dep. at 174.) It appears these are counties that are at the periphery of a DMA, or counties that have been or could be assigned to a DMA non-contiguous to their physical locations. For example, Clay County, Georgia, was assigned to the Columbus, Georgia DMA, but "[d]ue to satellite viewing and/or cable carriage issues," was reevaluated for reassignment to the New York DMA. (Def.'s Ex. 20 at N000857, N000860.) (Whether the evaluation in fact resulted in reassignment is not clear from the record, but is not material to this dispute.)
In addition to its argument that Nielsen's creative work is absent or insufficient to warrant protection, Truck Ads contends that Nielsen's DMA maps were originally created by another consumer research company, Arbitron. As a result, Truck Ads contends, Nielsen can claim a copyright, at most, to the limited changes it has made to the year-to-year changes in its maps. (Def.'s Reply at 8.) In the 1960s, Arbitron, then known as the American Ratings Bureau, created maps dividing the United States into approximately 200 market regions referred to as "Areas of Dominant Influence," or ADIs. Arbitron's ADIs also assigned each county exclusively to one ADI, defined as "an area that consists of one or more Arbitron sampling units [normally one county or independent city] in which the commercial stations home to the ADI . . . received the preponderance of total viewing hours." (Def.'s Supp. ¶ 2, quoting Def.'s Ex. 10, "Arbitron Description of Methodology" at 1.)
According to Truck Ads, it was Arbitron that originated the idea of exclusive market areas, and Nielsen has not offered its own "original map, or original methods and criteria" that would support a finding of copyrightability. (Def.'s Supp. ¶¶ 4, 5.) Plaintiff's own expert acknowledges that "there are only so many ways that you can go about taking audience estimates and using them to make a decision about which — which market has the preponderance of viewing in that county." (Def.'s Supp. ¶ 10, citing Pl.'s Supp. Ex. C, Deposition of former Arbitron senior executive David Lapovsky at 167 ["Lapovsky Deposition"].) Nielsen does not otherwise address the charge that Arbitron first come up with the idea of exclusive market regions, instead noting that it is not "claiming protection for the `idea' of mutually exclusive marketing regions based on television audience estimates," nor arguing that Truck Ads has infringed on such an idea. (Pl.'s Supp. Resp. ¶ 3.) Moreover, Nielsen notes, it is not seeking in this lawsuit to enforce a copyright in the first map it created. ( Id. ¶ 5.)
Truck Ads emphasizes the similarities between Nielsen's 2004-05 map and Arbitron's 1989-90 regional maps: 205 market areas on both maps contain one or more of the same counties, and of those, 61 contain the exact same county assignments; 58 differ by only one county; and 41 more differ by just two counties. (Def.'s Supp. ¶ 14.) Of 3,090 counties that appear in both maps, only 190 are assigned to different market areas in the 2004-05 Nielsen map than in the 1989-90 Arbitron maps. ( Id.) David Lapovsky, a former Arbitron senior executive, noted that the fact that some of the market definitions are the same under both Arbitron's system and Nielsen's system should not be surprising; "there are some markets . . . where signals are such that the market would be very clearly defined." ( Id. ¶ 12, citing Lapovsky Dep. at 87-88.) In any case, Nielsen insists, Truck Ads' county-by-county comparison of the two maps is misleading, because while many of the county assignments may not differ from one map to the other, 75 percent of the DMA boundaries on Nielsen's 2004-05 map do differ from those that appear on Arbitron's 1989-90 maps. (Pl.'s Supp. Resp. ¶ 14; Pl.'s Supp. Ex. D, Ex. 1 at 7 ["Lapovsky Report"].)
II. Evidence of Infringement by Truck Ads
In addition to challenging Nielsen's claims to copyright protection for its maps, Truck Ads contends that it did not copy those maps at all and therefore cannot be liable for infringement. Nielsen contends that Truck Ads did in fact access and copy Nielsen's DMA maps. Specifically, according to Nielsen, Truck Ads copied DMA maps that Nielsen licensed to 500Plus, LLC, which operates TollFree500.com. (Pl.'s 56.1(a) ¶¶ 15-24.) Nielsen notes evidence that in 2003, Alex Eucare, the owner of 500Plus, contacted Truck Ads CEO Rod Harris about establishing a business relationship. ( Id. ¶ 17, citing Pl.'s Ex. R, Deposition of Truck Ads CEO Rod Harris at 215-16 ["Harris Deposition"].) Harris recalled that Eucare suggested that Harris's customers might be interested in purchasing toll-free vanity numbers though 500Plus. (Harris Dep. at 216.) After several discussions, Eucare testified, he discovered that someone from Truck Ads had downloaded 500Plus's DMA maps from the servers on which he hosted his web site. (Pl.'s 56.1(a) ¶ 18; Pl.'s Ex. S, Deposition of 500 Plus CEO Alex Eucare at 41-42 ["Eucare Deposition"].) According to Eucare, when confronted about the downloading, Harris responded by saying, "sorry, it's just business, nothing personal." (Pl.'s 56.1(a) ¶ 20; Eucare Dep. at 42.)
Neither party describes the nature of TollFree500's business, nor does the portion of 500Plus CEO Alex Eucare's deposition that is included in the record. Neither the 500Plus nor TollFree500 web sites appear to be currently active. Based on three pages that were printed out from the 500Plus web site and included in the record, it appears that the business provided a range of marketing tools and specialized in providing toll-free numbers. (Pl.'s Ex. G at 1-3.) In his deposition, Rod Harris refers to these as "vanity numbers," without any further elaboration. (Harris Dep. at 215.) Eucare suggests that he thought a customer of Harris's who was selling hot tubs could "put `500 hot tubs' on their trucks." (Eucare Dep. at 39.) It is not clear whether these vanity numbers were actually toll-free numbers with the prefix 500, or for what reason the business was called TollFree500 or 500Plus.
Nielsen does not explain how Harris would have had access to the servers hosting the TollFree500 web site. Eucare testified, "I got a call from our hosting guy that said, somebody has been downloading our stuff off our Website. And I said . . . so what do you need to do? He said, well, we need to lock things down on it. Okay, fine. And then they subsequently tracked it to Truck Ads." (Eucare Dep. at 42.)
Truck Ads disputes that these events occurred, and although Harris recalls conversations with Eucare, he disputes that there was any mention of copying of the DMA maps. (Harris Dep. at 217-20.) Harris also denies that he downloaded information from the TollFree500 website. (Def.'s Resp. ¶ 18; Harris Dep. at 122.) Truck Ads also challenges Nielsen's assertion that it is Nielsen's own maps that appear on the TollFree500 website pursuant to a license that dates from 2001; according to Truck Ads, the TollFree500 maps differ from Nielsen's maps in numerous respects. (Pl.'s 56.1(a) ¶ 15, 16; Def.'s Resp. ¶ 16.) The evidence the parties cite on this score is not enlightening. The only evidence of a TollFree500/Nielsen license appears to be Eucare's testimony that on an unstated date, he asked the (unnamed) company he had contracted with for his web site to obtain a license, and that company assured him it had done so. (Eucare Dep. at 24-26.) Truck Ads, for its part, supports its contention that the TollFree500 maps differ from Nielsen's maps merely by directing the court to Plaintiff's Exhibits C, D, E, and G-which contain several hundreds of pages of Nielsen's maps and the 500Plus maps. (Def.'s Resp. ¶ 16.)
Here is the entirety of the testimony on the record regarding the purported license:
Though he denies having downloaded maps from TollFree500, Harris does admit that TollFree500 was one of the sources he consulted in creating Truck Ads' DMA maps, which were first posted in late 2004. (Harris Dep. at 193.) Thus, he explains that TollFree500's site was a "foundation of information" he relied on when he "first started building these maps." ( Id.) Harris explained that he prepared maps by taking screen shots of maps from a program called MapPoint 2000, and then used another program, Paint Shop Pro, to modify the maps to reflect information presented on TollFree500 and "other sources." (Def.'s Supp. ¶ 36; Harris Dep. 188-92.)
MapPoint 2000 was a Microsoft software program that allowed the user to create custom maps depicting areas of the United States as well as other countries. It also contained demographic data provided by Claritas, Inc., so, for example, "you could . . . display average household income in Seattle broken down by census tract." "Review: MapPoint 2000 raises the bar for desktop mapping," InfoWorld (April 12, 1999).
Nielsen contends that there is no evidence of "other sources" aside from Nielsen's own maps. (Pl.'s Supp. Resp. ¶ 36.) Nielsen notes the remarkable similarities between Nielsen's 2004-05 DMA map and Truck Ads' November 15, 2006 map: both maps display 210 market areas, and of those, 108 have identical boundaries; 51 differ by the assignment of a single county; and 28 differ by the assignment of just two counties. (Pl.'s 56.1(a) ¶ 32(a)-(d).) Truck Ads challenges this tally, but apart from asserting that Nielsen did not review the Alaska and Hawaii maps, Truck Ads' assessment of the two sets of maps is very similar to Nielsen's: Truck Ads contends that 107 of the 206 DMAs reviewed have identical boundaries; that 50 of the 206 differ by only one county; and that 29 of the 206 differ by exactly two counties. (Def.'s Resp. ¶ 32(a)-(d).)
Harris acknowledged that his sources included, in addition to the TollFree500 maps, "[a] lot of newspaper publications [that] would have DMA maps and media market maps. . . ." (Def.'s Supp. ¶ 37; Harris Dep. at 114.) Harris explained that he "more or less" used the county groupings that he saw on the TollFree500 web site, and that he settled upon the 210 DMA Areas because "[i]t seemed to be the number that everybody else was using." (Harris Dep. at 117-18.) Harris believed that he would not infringe any copyright so long as he "used a different font, different county line thicknesses and different colors" than those used by the maps he was using as a basis for the Truck Ads' maps. (Pl.'s 56.1(a) ¶ 25; Harris Dep. at 192.)
In any event, Truck Ads contends, Harris had never seen one of Nielsen's maps when he created the Truck Ads maps. (Def.'s Supp. ¶ 37, citing Def.'s Supp. Ex. 13, December 29, 2006 letter from Rod Harris to Attorney Antony J. McShane ("I have never ever seen your client's maps.").) Nielsen disputes this and identifies additional circumstantial evidence that Truck Ads did indeed use Nielsen's maps: Nielsen notes, first, that in 2005 the Truck Ads web site contained a disclaimer explaining that "Truck Ads DMA maps may or may not match DMA maps produced by Nielsen Media Research and others." (Pl.'s Supp. Resp. ¶ 37.) Nielsen suggests that Harris accessed Nielsen's maps from a 2002 publication entitled "Out-of-Home Advertising Source," in which Truck Ads had placed an ad. (Pl.'s 56.1(a) ¶ 26, 28.) Truck Ads did receive a copy of that publication, but it contends that the 2002 Advertising Source did not contain maps from the time frame at issue here. In any event, Truck Ads observes, Harris "did not recall [the publication] and had put it in storage." (Def.'s Resp. ¶¶ 26-28.)
Nielsen has submitted excerpts from this publication as Plaintiff's Exhibit H. Nielsen does not discuss the particulars of any licensing arrangement it had with this publication.
III. Nielsen's Complaint
On November 10, 2008, Nielsen filed its initial complaint in this action. (Dkt. 1.) Nielsen's second amended complaint alleges a single cause of action for copyright infringement in violation of 17 U.S.C. § 106. (Dkt. 38.) Nielsen contends that Truck Ads infringed the copyright in three of Nielsen's DMA maps by using the copied DMA maps to define geographic regions and set prices in dealing with its affiliates, and including those maps in its contracts with its affiliates/licensees. (Pl.'s 56.1(a) ¶¶ 34-35.) Both parties now move for summary judgment. (Dkt. 120, 122.)
DISCUSSION
The Copyright Act, 17 U.S.C. § 106, provides the owner of a copyright with the exclusive rights to reproduce the work, to prepare derivative works based upon it, and to display the copyrighted work publicly. 17 U.S.C. § 106(1), (2), (5). In order to prevail on a claim of copyright infringement, a plaintiff must establish: "(1) ownership of a valid copyright and (2) copying of constituent elements of the work that are original." Janky v. Lake County Convention and Visitors Bureau, 576 F.3d 356, 361 (7th Cir. 2009) (quoting Feist Publications, Inc. v. Rural Tel. Serv. Co., 499 U.S. 340, 361 (1991)).
The court grants summary judgment "if the movant shows that there is no genuine dispute as to any material fact and the movant is entitled to judgment as a matter of law." FED. R. CIV. P. 56(a). In deciding a motion for summary judgment, the court will view all facts in the light most favorable to the non-moving party and draw all reasonable inferences in the non-moving party's favor. Ault v. Speicher, 634 F.3d 942, 945 (7th Cir. 2011). Where, as in this case, the parties have filed cross-motions for summary judgment, "the court is required to adopt a `Janus-like perspective,' viewing the facts for purposes of each motion through the lens most favorable to the non-moving party." Moore v. Watson, 738 F. Supp. 2d 817, 827 (N.D. Ill. 2010).
I. Ownership of a Valid Copyright
It is undisputed that Nielsen has obtained valid copyright registrations from the U.S. Copyright Office in DMA maps from 2002-03, 2003-04, and 2004-05. (Pl.'s 56.1(a) ¶ 12.) A valid copyright registration "constitute[s] prima facie evidence of the validity of the copyright and of the facts stated in the certificate." 17 U.S.C. § 410(c); JCW Investments, Inc. v. Novelty, Inc., 482 F.3d 910, 914-15 (7th Cir. 2007). Once the plaintiff presents the registration, "[t]he evidentiary weight to be accorded the certificate of registration made thereafter shall be within the discretion of the court." 17 U.S.C. § 410(c). The validity of a copyright based on issuance of a certificate of registration "is simply a rebuttable presumption and [] `the burden of proof in the sense of the risk of nonpersuasion . . . remains throughout the trial upon the party on whom it was originally cast.'" Mid America Title Co. v. Kirk, 59 F.3d 719, 721 (7th Cir. 1995) (quoting FED. R. EVID. 301). In this Circuit, copyrightability is an issue of law for the court to decide. Schrock v. Learning Curve Int'l, 586 F.3d 513, 517 (7th Cir. 2009); Gaiman v. McFarlane, 360 F.3d 644, 648-49 (7th Cir. 2004).
Truck Ads challenges the presumption of Nielsen's copyright validity on several grounds. As the court understands those arguments, Truck Ads contends, first, that Nielsen did not independently create the copyrighted work, but rather copied Arbitron's maps; second, that Nielsen's maps do not have the requisite level of creativity to support a valid copyright; third, that the FCC's use of Nielsen's maps strips them of copyright protection; and, finally, that whatever copyright protection might apply is limited only to the year-to-year changes in DMA regions, and not to the maps in their entirety. The court addresses these arguments in turn.
A. Independent Creation
According to Truck Ads, Nielsen essentially copied its maps from Arbitron. As a result, Truck Ads urges, Nielsen's maps lack the originality required to support a valid copyright. "As a constitutional and statutory matter, `the sine qua non of copyright is originality.'" Schrock, 586 F.3d at 518-19 (quoting Feist, 499 U.S. at 345). The level of originality required, however, "is extremely low; even a slight amount will suffice." Feist, 499 U.S. at 345. In addition, "a work may be original event though it closely resembles other works." Id. The work need only involve "independent creation plus a modicum of creativity." Id. The Seventh Circuit has explained that a work based on another work is copyrightable if there "is enough expressive variation from public-domain or other existing works to enable the new work to be readily distinguished from its predecessors." Bucklew v. Hawkins, Ash, Baptie Co., 329 F.3d 923, 929 (7th Cir. 2003).
The court can first dispense with Truck Ads' argument that Nielsen's maps do not cross the originality threshold because "Arbitron did the same things as Nielsen, and did them first." (Def.'s Resp. at 4.) Specifically, Truck Ads alleges that Arbitron came up with the idea of television market areas based on the largest portion of a county's viewing share. ( Id.) That Arbitron may have done this first does not defeat Nielsen's copyright, so long as Nielsen made an original contribution. It is a mistake to "equate[] originality with novelty; the law is clear that a work can be original even if it is not novel." Kelley v. Chicago Park District, 635 F.3d 290, 302-03 (7th Cir. 2011).
Truck Ads nevertheless insists that even the minimal Feist test is not met here. Truck Ads points to several pieces of evidence it believes establishes that Nielsen has not "independently created" anything sufficient to warrant copyright protection. Truck Ads first notes what it calls the "[p]articularly striking" evidence that Nielsen admitted it copied one of Arbitron's split counties. (Def.'s Resp. at 5.) Specifically, in a February 27, 1998 e-mail message, Nielsen stated that "one of the reasons [Nielsen identified a split county] was because [Arbitron] had split the county. Nielsen used the [Arbitron] map to determine the line of the split on the . . . map." (Def.'s Supp. ¶ 20.) Nielsen responds that it first split the county at issue, Oneida County, New York, in 1973, and that the reference to Arbitron confirms that "Nielsen made several observations" about the county in creating its own map. (Pl.'s Supp. Resp. ¶¶ 20, 21.) That a single county is split on both maps, however, does not by itself defeat a finding of originality. As Nielsen has acknowledged, the possibilities of arrangement in DMA maps are not infinite. If there were only one possible arrangement of the information, there could be no copyright in that arrangement, but, in some cases, there may be few options for presenting information. The court is unwilling to conclude that because Arbitron and Nielsen chose similar arrangements of county data, DMAs are devoid of copyright protection, nor that Nielsen must have copied Arbitron's work.
This statement came in what appears to be an internal e-mail exchange between Nielsen employees. The e-mail itself contains no identifying information, such as job titles, of those involved in the exchange, aside from their names. (Def.'s Supp. Ex. 5 at 1.) The e-mail chain indicates that the general manager of a television station in Syracuse, New York, asked a Nielsen employee why Oneida County was originally split. ( Id.) That employee then contacted another employee (again, no titles are offered), who sent a message to several other employees asking for the history of the split, evidently generating the response Truck Ads quotes. ( Id.)
The e-mail message explains that, in addition to the Arbitron split, Nielsen made its decision to divide the county "based on a change order issued May 18, 1973, [after] Stat Research had reviewed home by home viewing data and found `significant differences in viewing patterns between extreme eastern and western districts in Oneida County.' It was `split along postal zones.'" (Def.'s Ex. 5 at 1.) No further context is offered by either party. It is unclear whether "Stat Research" is an internal Nielsen division or a third-party entity, nor has Nielsen elaborated on the "significant differences in viewing patterns" or explained what a "change order" is.
Truck Ads notes that approximately 94 percent of the counties on Nielsen's 2004-05 map are assigned to the same television market areas as the ones to which Arbitron assigned them in a 1989 map. (Def.'s Supp. ¶ 14; Def.'s Resp. at 6.) Nielsen contends, however, that "no valid comparisons can be drawn" between the 1989-90 Arbitron map and Nielsen's 2004-05 map because "both Nielsen and Arbitron made annual county reassignments," and because Arbitron stopped producing its maps in 1993. (Pl.'s Resp. at 7.) This response is not entirely satisfying; as the court understands Truck Ads' argument, Truck Ads believes that Nielsen does not actually make annual reassignments, but instead simply copied Arbitron's work, and that the 6 percent difference that has evolved reflects an attempt to mask its duplication.
Nielsen's other arguments, however, are more persuasive. Nielsen first notes the analysis of its expert, David Lapovsky, a former Arbitron senior executive, who reviewed map data and concluded that approximately 75 percent of the boundaries of Nielsen's DMA maps are different from Arbitron's ADI maps, even if that difference accounts for just 6 percent of the total counties. (Pl.'s Resp. at 7, citing Lapovsky Report at 7.) Lapovsky also concluded that approximately 10 percent of the common market areas had differences of five counties or more, and that one, Denver, has a difference of twenty counties. ( Id.) Lapovsky noted that there are five market areas that Arbitron and Nielsen do not share at all. ( Id.) Truck Ads does not meet these assertions head-on. Instead, it argues that "[t]he fact that designated market areas may differ from one Nielsen map to another, or from Arbitron's 1989-90 map to Nielsen's 2004-05 map, demonstrates only that market areas may change over time as populations move and viewing habits shift, not that anything `original' exists about those county assignments in the copyright sense." (Def.'s Reply at 8.) The court understands this statement as a concession that Nielsen's maps are indeed more than simply copies of Arbitron's maps, and an argument that because the maps reflect factual material, they are not eligible for copyright protection. The court examines this argument below.
B. Factual Material and the Merger Doctrine
Facts are not copyrightable. Feist, 499 U.S. at 344. Factual compilations created with a "minimal degree of creativity" may be copyrighted, but copyright protection does not extend, for example, to "a garden-variety white pages directory, devoid of even the slightest trace of creativity," in which a list of telephone subscribers is arranged alphabetically by surname. Feist, 499 U.S. at 362.
Truck Ads urges that Nielsen's DMA maps are akin to the white pages directory because "only one arrangement would `consist[] of all counties whose largest viewing share is given to stations of that same market area'- i.e., the very definition of a television `DMA.'" (Def.'s Reply at 8.) This court addressed a similar argument when it ruled earlier on Truck Ads' copyright misuse claim. Nielsen Co. (US), LLC v. Truck Ads, LLC, No. 08 C 6446, 2011 WL 221838, at *8-10 (N.D. Ill. Jan. 24, 2011). Truck Ads argued that Nielsen has appropriated facts within the public domain and asserted copyright protection over them. Id. at *9. The court concluded that Defendant's own reference to various versions of the DMA maps defeats the suggestion that only one possible arrangement of DMAs exists. Id. at *10. The court also compared Nielsen's DMA maps to the maps discussed in Sparaco v. Lawler, Matusky, Skelly, Engineers LLP, 303 F.3d 460 (2d Cir. 2002), where the Second Circuit denied copyright protection to maps that "employ[ed] standard cartographic features without originality." Id. at 467. This court distinguished Sparaco, observing that, "[i]t may not be possible to portray a street of a definite slope in two different ways using standard cartographic features; but it is possible to portray a DMA with shifting, manmade, and artificial boundaries in myriad ways." Nielsen, 2011 WL 221838, at *10.
Truck Ads now renews this argument, presenting further support. First, Truck Ads points to Lapovsky's statement that only a finite number of ways exist to portray the same market data. (Def.'s Supp. ¶ 10.) Truck Ads also argues that, though Nielsen and others may have different methods of arriving at the same data, "Nielsen has not identified any demonstrable effect of such differences." (Reply at 8.) Truck Ads points to the case of Mid America Title Co. v. Kirk, 867 F.Supp. 673 (N.D. Ill. 1994), aff'd 59 F.3d 719 (7th Cir. 1995), in which this court examined whether Mid America could assert copyright protection over title commitments. The court noted, in a passage highlighted by Truck Ads, that "the fact that Mid America used a private, rather than a public source to gain information available to the public at large does not bear on the issue of originality; the decision to use one of two possible sources for a piece of data cannot legitimately be characterized as `original.'" Mid America, 867 F.Supp. at 685. This statement came in the context of a larger examination of title documents, which, the court ultimately concluded, contained essentially the same basic information, all of which was factual. Id. Mid-America is distinguishable from this case because here not only did Nielsen and Arbitron utilize different sources, but the sources contained different data and relied on different methods of obtaining that data.
Nielsen points out that one difference between what it does and what Arbitron did is that Nielsen used population estimates provided by Claritas, Inc., a company providing demographic data, which is now owned by Nielsen, while Arbitron used population estimates from Market Statistics, Inc. (Lapovsky Report at 4.) This difference alone may not suffice to provide the requisite originality to warrant copyright protection, but Nielsen contends it is just one example of the ways that Nielsen's methods of collecting and analyzing data differ from those used by Arbitron. Lapovsky explains, further, that the two companies use different viewership diaries, different methods to select who will receive those diaries, and different methods of projecting population estimates from their population data sets. (Lapovsky Report at 4-5.) Once a data set has been created, each company used different criteria and methods to determine the assignment of counties to DMAs, whether to split counties into DMAs, and whether to create new DMAs. ( Id. at 5-6.) These different methods may not have translated into significant differences between the maps produced by the two companies, but the similarities between the two firms' maps does not mean that there is only one possible outcome of this process — indeed, that there are differences supported by data demonstrates the contrary.
The record does not contain any further information about "Market Statistics, Inc.," and because it is a common name, and was utilized by Arbitron, which has not produced DMA maps in nearly twenty years, it is difficult to track down additional details about this company. Based on the information attributed to Market Statistics, Inc., it presumably is or was another market research company offering demographic information.
Notably, Truck Ads does not identify the origin of any factual data that it contends must be the actual source from which Nielsen and Arbitron created their maps. Instead, Truck Ads relies on the fact that Nielsen refers to "measurement diaries" and Local People Meters that "measure the programs being viewed"; this language, Truck Ads concludes, shows that "viewing share, like census population, is an ascertainable fact." (Def.'s Resp. at 8.) The Feist Court explained that, "Census takers . . . do not `create' the population figures that emerge from their efforts," 499 U.S. at 347, and the court agrees with Truck Ads that unadorned data cannot be protected by copyright. Here, however, there is no Census figure or other raw data generally available that measures the viewership statistics of individuals within a county, nor any current data (aside from Nielsen's) that links any county to its television market. The court is not aware of any sort of measurement device within a television, for example, that tracks signal strength reception and reports out such data. Census data is collected by actually counting each individual. Nielsen and any other companies that estimate viewership, in contrast, necessarily do so by sampling and extrapolation. In the process of sampling and extrapolating, Nielsen ascertains something that might approximate actual viewership, but is not in and of itself a "fact."
Truck Ads' related claim that the "merger doctrine" bars copyright protection for Nielsen's maps fails for the same reason. Under that doctrine, if "there is only one feasible way of expressing an idea, so that if the expression were copyrightable it would mean that the idea was copyrightable," the expression of that idea cannot be subject to copyright protection. Bucklew, 329 F.3d at 928. As explained previously, because different data sets underlie different DMA or market region arrangements, and different criteria are used by different companies to create different DMAs (or ADIs, or similar market regions), a DMA is susceptible of more than one form of expression, and not barred from copyright protection by the merger doctrine. As further support for its contention, Nielsen points to a Federal Communication Commission report which examined the implications of changing the basis of its market definitions from ADIs to DMAs. The report, issued in conjunction with the implementation of the Telecommunication Act of 1996, explains that:
The FCC uses television market regions to determine which local television channels fall into the "must carry" provisions for a given market. In re: Definition of Markets for Purposes of the Cable Television Mandatory Television Broadcast Signal Carriage Rules, FCC 96-197 at 7 (May 24, 1996) (Def.'s Ex. 14 at 1-3.). The Cable Television Consumer Protection and Competition Act of 1992, Pub.L. No. 102-385, 106 Stat. 1460 (1992), requires a cable operator to carry local broadcast commercial television stations within its market. For example, Comcast, which provides cable in areas including the Chicago DMA, is required to carry the local network affiliates and other commercial stations, should those stations elect to exercise "must-carry" rights. (Def.'s Ex. 14 at 3.) In order to determine the contours of these must-carry rights, the FCC originally relied upon Arbitron's ADIs, but switched to Nielsen's DMAs after Arbitron stopped producing ADIs. ( Id. at 1.) The FCC also uses market regions for several other purposes, discussed infra Section I.C.
Conceptually, their market designations — ADIs and DMA — are the same. They both use audience surveys of cable and noncable households to determine the assignment of counties to television markets based on the market whose stations receive the largest share of viewing in the county. The differences in their assignments of specific counties to particular markets reflect a number of factors, including slightly different methodologies and criteria as well as normal sampling and statistical variations.In re: Definition of Markets for Purposes of the Cable Television Mandatory Television Broadcast Signal Carriage Rules, FCC 96-197 at 7 (May 24, 1996) (Def.'s Ex. 14). The report went on to explain that substituting DMAs for ADIs would lead to changes in 126 of the 211 market areas. Id. at 8-9.
As explained previously, the threshold issue of whether Nielsen owns a valid copyright is an issue of law for the court to decide. Nielsen has demonstrated that television market areas are not "facts" devoid of copyright protection or "ideas" incapable of more than one form of expression, and are therefore not barred from copyright protection.
C. Use by the FCC
The court turns to Truck Ads' remaining challenge to the validity of Nielsen's copyright: that the FCC's use of Nielsen's DMAs precludes copyright protection. (Def.'s Br. at 10.) Specifically, the FCC uses DMAs for licensing satellite carriers, defining "must carry rights," defining limitations on exclusive rights under 17 U.S.C. § 122, and auctioning digital spectrum rights. Truck Ads cites two cases that, it contends, demonstrate that these uses by the FCC bar Nielsen's copyright claim. In Veeck v. Southern Building Code Congress Int'l Inc., 293 F.3d 791 (5th Cir. 2002), an individual who published model building codes on his web site sued the company that had authored the codes for a declaratory judgment of non-infringement. The version of the codes published on plaintiff's web site were those that had been adopted as law by two small towns in Texas. The court observed that "[t]he codes are `facts' under copyright law. They are the unique, unalterable expression of the `idea' that constitutes local law. . . . It should be obvious that for copyright purposes, laws are `facts.'" Id. at 801. Similarly, in Ocean Atlantic Woodland Corp. v. DRH Cambridge Homes, Inc., No. 02 C 2523, 2003 WL 1720073 (N.D. Ill. March 31, 2003) (Guzman, J.), the court recognized, as an affirmative defense, the allegation that housing development plans that the defendant home developer had copied from the plaintiff developer were in the public domain because they had been adopted by a village and "are, by law, required to be followed by the developer of the property." Id. at *7.
As the Veeck court explained, "copyrighted works do not `become law' merely because a statute refers to them." 293 F.3d at 805. Veeck goes on to identify several factors that counseled in favor of recognizing that the building codes at issue in that case are in the public domain. For example, Veeck pointed out that a work would not be in the public domain if it had been authored "by private groups for reasons other than incorporation into law." Id. The model code at issue in Veeck, by contrast, "serves no other purpose than to become law. [Defendant SBCCI] operates with the sole motive and purpose of creating codes that will become obligatory in law." Id. In this case, Nielsen points out, it is apparent that DMAs were not created with an eye toward legislative enactment. Indeed, the FCC license to use Nielsen's DMAs contains an express disclaimer of any such intention: "Nothing contained in this Agreement shall be construed to require Nielsen Media to grant a license to use the Licensed Material to prospective or FCC-licensed . . . operators, or to grant such a license on any certain or specific terms except upon such terms as Nielsen Media agrees to in its sole and unlimited discretion." (Pl.'s Supp. ¶¶ 1, 2.)
A case arguably more closely analogous than Veeck is CCC Information Services, Inc. v. Maclean Hunter Market Reports, Inc., 44 F.3d 61 (2d Cir. 1994). There, the publisher of a compilation of projected used car values called the "Red Book" sued another company, alleging that defendant had copied the projections into a database and offered the information to customers. The statutes of several states made reference to the Red Book as a source of used car values for insurance purposes, and defendant argued that, as a result, the Red Book had passed into the public domain. Id. at 73. Policy considerations supported that argument, the court acknowledged, but "a state's reference to a copyrighted work as a legal standard . . . [does not] result[] in loss of the copyright." Id. at 74. The court analogized the Red Book to copyrighted books assigned to be read by students in public schools — such an assignment does not result in a determination that the books enter the public domain. Id. A contrary conclusion might implicate the Takings Clause, the court observed. Id.
No authority establishes that the FCC's references to Nielsen's maps place them within the public domain. Those references instead suggest that the maps have value and apparently cannot be replicated without effort or cost by that agency. The FCC would presumably not welcome the conclusion that its use of the maps destroys Nielsen's incentive to produce them. Truck Ads' objection on this basis is overruled.
D. Limited Scope of Copyright
If Nielsen is entitled to copyright protection for its maps at all, Truck Ads argues, any such protection must extend only to the year-to-year changes in its maps. (Reply at 8-9.) Truck Ads has not been explicit about the legal doctrine at issue, but the court presumes this is a reference either to the copyright laws governing derivative works or to a theory that Nielsen's maps are nothing more than a compilation. The statutory language law that applies to both compilations and derivative works provides:
Truck Ads' briefs do not effectively separate this argument from an argument regarding protectable elements, addressed below together with the issue of substantial similarity.
The copyright in a compilation or derivative work extends only to the material contributed by the author of such work, as distinguished from the preexisting material employed in the work, and does not imply any exclusive right in the preexisting material. The copyright in such work is independent of, and does not affect or enlarge the scope, duration, ownership, or subsistence of, any copyright protection in the preexisting material.17 U.S.C. § 103(b). Under this rationale, Nielsen's protected right is minimal, Truck Ads contends, for the same reasons Truck Ads has challenged Nielsen's copyright: in Truck Ads' view, Nielsen did not originate its DMA maps. Instead, each year Nielsen merely reassigns and reexamines a small number of counties. Any assertion that Nielsen "originates each and every market, each and every year," is, in Truck Ads' view, "untenable." (Def.'s Reply at 9.)
Both parties point to Christopher Phelps Associates, LLC v. Galloway, 492 F.3d 532 (4th Cir. 2007) in support of their positions. In Phelps, where plaintiff charged defendant with copying copyrighted architectural plans, the defendant contended that the copyright protected only certain changes in the design from a previous version. Id. at 538. Because the plaintiff had authored the prior design, as well, however, the court concluded that plaintiff's copyright covered the entirety of the design. Id. "[W]hen the author of the derivative work also has a copyright on the underlying work, there is no need to protect the public domain or the author of the underlying work, as the entire work is that of the single author." Id. The fact that plaintiff had registered the entire design rather than registering the new design as a derivative work made no difference to the court: "The scope of registration need not precisely trace the scope of the copyright for the holder to sue." Id. at 539. Nielsen cites Phelps for the proposition that Nielsen's failure to register previous versions of its DMA map does not preclude its effort to enforce its copyright at this stage. (Pl.'s Resp. at 8.)
Truck Ads responds that Phelps defeats Nielsen's claims here because Nielsen cannot show authorship of the prior work — that is, Truck Ads repeats its contention that the similarities between Nielsen's maps and Arbitron's maps require the conclusion that Nielsen copied Arbitron's maps. The court has already rejected that argument, however. The record in this case contains significant evidence that Nielsen undertook a review process each year in creating its DMA maps. For example, Nielsen has submitted "Local Reference Supplements," which explain the procedures for determining DMAs (Pl.'s Ex. O), procedures that would be superfluous unless they did in fact influence the presentation of the DMA maps. In addition, Nielsen has presented e-mail messages to its employees, explaining the process for testing and annual reassignment of counties and DMAs. (Pl.'s Supp. Ex. G.) The court is unaware of any basis Nielsen would have for conducting these annual tests if it did not operate with some baseline data regarding these DMAs — in other words, the only logical explanation, based on the evidence in the record, is that Nielsen does in fact evaluate each county and DMA each year, and determines which are in need of further testing. The results of that annual testing process leads to county reassignments and, albeit rarely, to changes in an entire DMA region.
Truck Ads invites the court to assume that these supplements are a ruse, created in case a lawsuit such as this were to occur. The court declines the invitation. Nielsen has in fact presented evidence of independent creation, and Truck Ads' response to that evidence amounts to speculation. Truck Ads notes, for example, that Lapovsky said he had "no knowledge of how Nielsen developed its system" (Def.'s Br. at 7; Def.'s 56.1(a) ¶ 27) — but the court is uncertain why Lapovsky, a thirty-year employee of Arbitron (Pl.'s Supp. Ex. D ¶ 1), would have knowledge of the specifics that lead to Nielsen's creation of its first DMA map. Lapovsky was retained to respond to Truck Ads' expert testimony "regarding the similarity of the Arbitron and Nielsen methods and criteria and to articulate the methods and criteria that Arbitron used to determine its ADIs and to compare and contrast them to those in use by Nielsen to determine its DMAs." (Lapovsky Report at 1.) In short, Lapovsky was retained to examine the differences between the methodologies, not to determine how they were developed in the first place. Indeed, as Lapovsky testified, the suggestion that Nielsen simply copied Arbitron, or that the two collaborated, is illogical. "[H]aving been in that business for — for a number of years, that I know that Nielsen and Arbitron did not sit down — we — we were competitors. We couldn't sit down at a table and jointly arrive at procedures and methodology. So we — we had to arrive at them independently." (Lapovsky Dep. at 166.) Truck Ads offers no reason for the court to doubt this testimony.
Truck Ads also points to the affidavit of Bruce Hoynoski, a Nielsen senior vice president, who describes Nielsen's DMA methodology and history and explains that he believes DMAs were introduced in the 1960s, and has "general knowledge" of why, but not how, the maps were created. (Def.'s Ex. 23 at 58.) Again, however, Truck Ads does not explain why the court should infer from an expert witness's lack of knowledge of events of half a century ago that Nielsen did not originate or independently create its maps. Nor has Truck Ads offered a reason that the court should reject the record testimony concerning the annual DMA review process.
The court notes that Truck Ads has objected to Hoynoski's affidavit on a number of grounds. (Def.'s Resp. at 2-3.) Because the court resolves these motions without relying upon the affidavit, save for this single instance where Truck Ads itself offers it as evidence, the court need not rule on Truck Ads' objections at this point. Truck Ads does not object to the use of Hoynoski's deposition testimony, which was referenced in the background section of this opinion.
That Truck Ads could not track down the first DMA map itself, or any knowledgeable witness who could answer its questions about that map, does not create an issue of fact as to whether Nielsen independently created the maps at issue in this lawsuit. Truck Ads has not produced any non-speculative evidence that Nielsen copied Arbitron's maps, and the fact that significant similarities exist between ADIs and DMAs does not suffice to create an issue of fact for trial. Based on the ample unrebutted evidence of Nielsen's annual review process, the court concludes that Nielsen holds a valid copyright on the entirety of the boundaries representing DMA regions within each of the three maps at issue. The court therefore grants Nielsen's motion for partial summary judgment on its ownership of a valid copyright.
II. Copying
The second part of the infringement analysis is whether Defendant copied protectable elements of Plaintiff's work. Janky, 576 F.3d at 361. Plaintiff may establish copying through direct or circumstantial evidence — that is, evidence that "the defendant had access to the copyrighted work and the accused work is substantially similar to the copyrighted work." Tillman v. New Line Cinema Corp., 295 Fed. Appx. 840, 842 (7th Cir. 2008) (quoting Susan Wakeen Doll Co. v. Ashton-Drake Galleries, 272 F.3d 441, 450 (7th Cir. 2001)). Substantial similarity is determined by the "ordinary observer" test: "whether the accused work is so similar to the plaintiff's work that an ordinary reasonable person would conclude that the defendant unlawfully appropriated the plaintiff's protectable expression by taking material of substance and value." Incredible Technologies, Inc. v. Virtual Technologies, Inc., 400 F.3d 1007, 1011 (7th Cir. 2005). Defendant may rebut an inference of copying that arises from access and substantial similarity by showing that she independently created the work, for example, without copying anything, or by copying something other than plaintiff's copyrighted work. JCW Investments, 482 F.3d at 915.
A. Access
To establish copying by direct or circumstantial evidence, Plaintiff Nielsen must show that Truck Ads had access to the DNA maps. This test is readily met here, Nielsen contends: Truck Ads' CEO Harris admits to copying the maps from publicly available web sites, and Harris had communication with TollFree500.com, which Nielsen claims is a licensee of its DMA maps. As outlined above, Alex Eucare, CEO of 500Plus, says he had business dealings with Harris, and that he discovered Truck Ads had downloaded maps from its servers. Harris denies having downloaded any maps, but he admits that he "used the TollFree500 website to see how to make many of my maps" and used it as a "foundation." (Pl.'s 56.1(a) ¶ 23.)
Truck Ads discounts this evidence of access and insists that there is no direct evidence of copying. Because the TollFree500 Maps show a "1999-2001" copyright, "Truck Ads can not have logically infringed the original elements of Nielsen 2002-2005 maps." (Def.'s Resp. at 13.) The court is less certain. Discussions between Truck Ads and TollFree500 evidently did not begin until 2003, but the year or years of the maps that appeared on TollFree500's web site do not appear to be in the record. The fact that TollFree500's web site bears a "1999-2001" copyright symbol on TollFree does not mean that the maps themselves necessarily dated from that time period as it is possible that the copyright symbol for the web site was not updated. Nor has Nielsen made a compelling showing: Nielsen asserts that 500Plus documents provided by Defendant "include an identifiable cutout of Nielsen's 2002-2003 DMA Map" (Reply at 11), but it is not clear to the court to which documents Nielsen is referring or where they originated. (Plaintiff's Exhibit G, which purportedly includes these maps, contains approximately 100 pages.) A genuine dispute of material fact exists as to whether the TollFree500 web site maps are, in fact, Nielsen's maps, and whether they are the maps whose copyrights are at issue in this lawsuit. Therefore, on the issue of direct copying, both Nielsen's and Truck Ads' motions for summary judgment must be denied.
B. Substantial Similarity
Though the direct evidence of copying is insufficient to support summary judgment, Plaintiff may also proceed by presenting circumstantial evidence of copying, and substantial similarity. As described above, the evidence of copying is disputed, and the court concludes that there are disputes concerning the similarity issues as well.
1. Protectible Elements
To assess this issue, the court must determine the "protectable elements" or "protectable expression" of Plaintiff's maps. The court has already conducted some of this analysis above, determining that Nielsen does create its maps anew every year, and that the DMA assignments in the maps are therefore "original" and deserving of copyright protection. Nielsen urges that the boundaries are "the central protectable element" of the maps. (Pl.'s Br. at 13.) Truck Ads has not addressed the issue independently, but has suggested that the DMA maps simply display "unprotectible `lot' boundaries." (Def.'s Resp. at 12.)
In urging that the boundaries of its DMA maps are protectable, Nielsen points to City of New York v. Geodata Plus, LLC, 537 F. Supp. 2d 443 (E.D.N.Y. 2007), in which the City of New York sued Geodata for infringing on the City's copyright in a database of real property information. In determining that maps depicting tax lots could be subject to copyright, the Geodata court observed that the individuals who programmed the database upon which the maps were based had scanned paper maps, manually traced them, and in doing so, "made numerous independent decisions concerning the number and location of coordinates used to depict various shapes in the maps . . . [and] concerning which features to incorporate and which to exclude." Id. at 451. As a result, the court observed, the maps were "not identical" to the publicly available maps upon which they were based. Id. The court emphasized that the process of creating the maps involved more than "rote copying," and instead "involved a distinctly human element in which many independent decisions had to be made." Id. at 452. In sum, "although none of the factual representations in the [database copied by defendant] . . ., such as the tax block and lot boundaries themselves, are protectible, there is sufficient originality in the [database] as a whole to justify copyright protection under the Feist standard." Id. at 451.
Nielsen also cites Rockford Map Publishers, Inc. v. Directory Service Company of Colorado, 768 F.2d 145 (7th Cir. 1985), another case examined in this court's ruling on the motion to dismiss. In Rockford Map Publishers, the Seventh Circuit found that a company that copied plaintiff's plat maps had violated plaintiff's copyright. Rockford Map Publishers had created the maps by starting with aerial photographs from the Department of Agriculture, tracing the topography, then tracking down land titles and using the descriptions to draw boundary lines. Id. at 147. The Seventh Circuit anticipated Feist in recognizing that "[t]he copyright laws protect the work, not the amount of effort expended." Id. at 148. The test was whether "the compiler made a contribution — a new arrangement or presentation of facts," and concluded that plaintiff met that test because "[t]easing pictures from the debris left by conveyancers is a substantial change in the form of the information." Id. at 149. The court held that the defendant "may make all the plat maps it wants; everyone is free to repeat the facts contained in the deeds. But it may not use Rockford Map's work as a template." Id. at 149.
Rockford Map and Geodata support Nielsen's argument that its contribution to the mapmaking process supports copyright protection. The DMA regions themselves, as explained previously, are not "facts" in the same way as plat lines or tax lots. If anything, this case is a stronger one because they are, instead, factual estimations; that is, DMA regions are products of estimation and extrapolation — they cannot be looked up at a county assessor's office. In one case more on point, Nester's Map Guide Corp. v. Hagstrom Map Co., 796 F. Supp. 729 (E.D.N.Y. 1992), the court considered whether a taxi guide that listed cross streets and address numbers could be copyrighted. Id. at 733. The court noted that "because the guide has limited space for listing cross streets, [the plaintiff] selected only the important and most helpful cross streets based upon his knowledge of New York. Moreover, many of the numbers assigned to each cross street do not correspond to the actual numbers of the buildings at that intersection. He approximated the vast majority of building numbers in order both to detect copying and to make them easier to remember." Id. at 732-33. The court concluded that the creator of the guide had exercised sufficient creativity to support a copyright. Id. at 734.
That conclusion is consistent with authoritative commentary. Nimmer's treatise explains the distinction between the types of facts that a listing of addresses would constitute, and those factual estimations that merit copyright protection:
A continuum of creative effort and activity exists in reference to factual material and estimations. Estimates and judgments that round off, make more readable, or provide suggested solutions to factual issues involve creative work of an important and often high level. These cases differ from the argument that copyright should apply because of the effort involved in collecting material in exactly the correct way to allow copyright protection: They argue that the derivation of the estimate involved creative work, expressed in the estimates themselves.
Raymond T. Nimmer, 1 INFORMATION LAW § 3:24 (May 2011). Nielsen's DMA boundaries closely resemble the factual estimations that Nimmer identifies, as the creation of the boundaries involves the very type of creative derivation of estimates that he describes. As discussed previously, no "listing" of addresses within a DMA, similar to the white pages, exists — the DMA itself is necessarily a creatively derived estimate.
The court thus concludes that the boundaries depicted in Nielsen's DMA maps are protectable, and that Nielsen is entitled to judgment in its favor if it shows that Truck Ads copied those boundaries.
2. Infringement
Whether that decision can be made at this stage requires careful consideration. As a general rule, "summary judgment is disfavored on the substantial similarity issue in copyright cases, [and is only] . . . clearly appropriate . . . if, after viewing the evidence and drawing every inference in the light most favorable to the nonmoving party, the court concludes that no reasonable jury could find substantial similarity of both ideas and expression between the works at issue." Worth v. Selchow Righter Co., 827 F.2d 569, 571 (9th Cir. 1987) (citation and quotation omitted); "[B]ecause the question of substantial similarity typically presents an extremely close question of fact, . . . questions of non-infringement have traditionally been reserved for the trier of fact." Peter F. Gaito Architecture, LLC v. Simone Development Corp., 602 F.3d 57, 63 (2d Cir. 2010); Schiller and Schmidt, Inc. v. Wallace Computer Services, Inc., No. 85 C 4415, 1989 WL 152860, at *3 (N.D. Ill. Nov. 20, 1989) (Leinenweber, J.).
In the court's view, the boundaries that appear in Truck Ads' maps (Pl.'s Ex. F) are strikingly similar to those in Nielsen's DMA maps (Pl.'s Ex. C, D, E). There are, to be sure, differences in terms of the color schemes used, in labeling, and in other aspects of the layout. Truck Ads points out that among the differences between its map and Nielsen's map are "its elimination of detail; its size, shape, and density of informative legends; and its establishment of conventions relating to color or design to represent geographical or other features." (Def.'s Resp. at 15.) Notably, while these differences are genuine, they do not affect the central protectable element at issue — the DMA boundaries. Other differences between Truck Ads' maps and those of Nielsen also appear to be insignificant. (Def.'s Br. at 12.) That Truck Ads' map is smaller than Nielsen's is immaterial — if all that were needed to escape copyright infringement is a change in the size of an image, the laws would have little force. That Truck Ads also has DMA maps for territories outside the United States is likewise immaterial; Nielsen alleges that Truck Ads infringed its maps of the United States, not Europe, Africa, or elsewhere.
Although Nielsen has made a compelling showing, the court is nevertheless reluctant to award judgment in Nielsen's favor on this issue. The issue of substantial similarity will require the factfinder to examine the detailed maps submitted by Truck Ads and Nielsen. These maps each contain approximately 210 regions, and there are at least some differences: of the 210 DMA regions on the Truck Ads 2006 maps and Nielsen's 2004-05 map, 108 have identical boundaries; 51 differ by the assignment of a single county; and 28 differ by the assignment of just two counties. (Pl.'s 56.1(a) ¶ 32(a)-(d).) The court is not prepared to say that a jury would necessarily conclude that these differences are so insignificant that Nielsen is entitled to prevail or, alternatively, that the differences are significant enough to defeat a finding of infringement. Genuine disputes of material fact preclude summary judgment for either party on this issue.
C. Independent Creation
Finally, Truck Ads could rebut a finding of copying based on access and substantial similarity if it can show that the maps were independently created. Wakeen Doll Co., 272 F.3d at 450. Independent creation is shown if the defendant "created its own work without copying anything or if it copied something other than the plaintiff's copyrighted work." Id. (citing 3 Melville B. Nimmer David Nimmer, NIMMER ON COPYRIGHT § 12.11[D], at 12-175 (2001)).
There is arguably a basis in the record for such a defense in this case. Harris offered an explanation of the process he used in creating Truck Ads' maps, and Defendant has suggested that the TollFree500 maps are either not Nielsen's own or not the ones at issue in this lawsuit. Truck Ads nevertheless has not formally made such an argument, either in support of its own motion for summary judgment or in opposition to Nielsen's. The court declines to address it further, other than to note that Truck Ads remains free to argue independent creation at trial.
CONCLUSION
Nielsen's motion for partial summary judgment [120] is granted in part and denied in part. Truck Ads' motion for summary judgment [122] is denied.
Q. So how did you obtain or create those maps that were displayed on your website? A. The Web hosting company that we had hired at the time, you know, we instructed them that, you know, we determined that the DMA footprint was the way to go, and they said, all right, we'll talk to the folks that have the DMA maps. I subsequently found out it was Nielsen. I said, fine, let's talk to them and, you know, get a prelaunch license or whatever you want to call it. Q. What was the name of the Web hosting company you were using? A. Oh, jeez, there were so many. I honestly can't remember. We had a half a dozen because they were crashing and burning. I honestly can't remember. Q. So it's your understanding, then, once you instructed them that you wanted to use DMAs, they went out and obtained DMA maps from Nielsen which ultimately ended up on the 500Plus Website? A. Right. . . . Q. Okay. I'm going to hand you what's been marked as deposition exhibit 2. It says, Nielsen media research, at the top. Do you recognize this document? A. No. Q. Do you have any knowledge whether the Web hosting company that obtained the DMA maps on your behalf would have clicked through this agreement when they obtained the DMA Maps? A. If it was an online download, yeah, I would assume they would have had like a user agreement, click here for terms of use, and download their — yes. (Eucare Dep. at 24-26.)