Opinion
07-cv-665-bbc.
November 17, 2008
OPINION and ORDER
In this civil suit, plaintiff Board of Regents of the University of Wisconsin System challenges a decision of the United States Patent and Trademark Office's Trademark Trial and Appeal Board granting defendant Phoenix Software International, Inc.'s petition for cancellation of plaintiff's trademark registration. Plaintiff relies on § 21 of the Lanham Act, 15 U.S.C. § 1071(b), which authorizes parties to seek a remedy by civil action when they are dissatisfied with a decision of the Trademark Trial and Appeal Board. Both parties use the trademark "CONDOR" for a computer software goods. Plaintiff uses the mark for computer operating software that is downloaded from the net to utilize idle computers within a network. Defendant uses its mark for online programming development for mainframe systems. On the parties cross-motions for summary judgment, the only question is whether the identical marks used in the general field of computing create a likelihood of confusion for consumers, as defendant contends, or whether the differences in the computer products for which the software is sold, in the trade channels, in the conditions under which sales of the products are made and other factors eliminate the possibility of confusion, as plaintiff argues.
I conclude that plaintiff's position is correct and that the board erred when it considered the actual nature of the parties' goods and misapplied the burden of proof to its determination of a likelihood of confusion. Moreover, defendant has failed to proffer any new evidence in support of a finding of likelihood of confusion. Therefore, defendant's motion for summary judgment will be denied, plaintiff's motion for summary judgment will be granted and the board's decision will be reversed.
As an initial matter, plaintiff filed a motion to strike evidence and other documents filed by defendant with its reply brief. Dkt. #69. Plaintiff contends that defendant violated this court's rules when it waited until it filed its reply brief to submit a second declaration of Fred Hoschett, dkt. #67. I agree. Defendant was instructed to "include ALL factual propositions [it] consider[ed] necessary for judgment in [its] favor" in its initial proposed findings of facts. The court's procedure does not allow the movant a second round for submitting additional facts, only an opportunity to dispute any facts proposed by the nonmovant when the movant files its reply. Procedure to be Followed on Motions for Summary Judgment, I.B.3, III, attached to Preliminary Pretrial Conference Order, dkt. #19. "A movant's response to the nonmovant's additional proposed findings of fact is not an opportunity to include facts previously overlooked."Taurus IP, LLC v. DaimlerChrysler Corp., 534 F. Supp. 2d 849, 854 (W.D. Wis. 2008).
From the parties' proposed findings of fact, I find the following facts to be material and undisputed.
UNDISPUTED FACTS A. Parties
Plaintiff Board of Regents of the University of Wisconsin System is an institution of higher education and an agency of the state of Wisconsin with its principal place of business in Madison, Wisconsin.
Defendant Phoenix Software International, Inc. is a California corporation with its principal place of business in Los Angeles, California.
B. Defendant's Registration and use of its Mark
Defendant's first attempt to obtain a federal registration for its CONDOR mark was in 1979, and was followed by attempts in `81, `85, and `95. At the time defendant first attempted to obtain registration, the CONDOR mark had already been registered by the Condor Computer Corporation in association with "computers."
Defendant's Fred Hoschett explained that the Phoenix Software was "quite different" from the products sold under Condor Computer Corporation's mark because "they were work-station based." In spite of Condor Computing Corporation's mark, Phoenix continued to use the CONDOR mark with its software.
On January 7, 1997, defendant received a federal registration, Registration No. 2,028,364, for the mark CONDOR for "COMPUTER SOFTWARE FOR ON-LINE PROGRAMMING DEVELOPMENT, LIBRARY MANAGEMENT AND SYSTEMS UTILITIES FUNCTIONING ON MAINFRAME SYSTEMS." Defendant claims a date of first use of June 1, 1978 and a date of first use in commerce of October 18, 1979 for its CONDOR mark. Phoenix has not used the mark on any product except the Phoenix Software.
C. Plaintiff's Registration and Use of its Mark
On March 13, 2001, UW received a federal registration, Registration No. 2,434,630, for the mark CONDOR for:
COMPUTER NETWORK OPERATING SYSTEM SOFTWARE, DOWNLOADED FROM A GLOBAL COMPUTER NETWORK, THAT DELIVERS LARGE AMOUNTS OF COMPUTATIONAL POWER BY UTILIZING IDLE COMPUTING RESOURCES IN A NETWORK OF INDIVIDUAL COMPUTER WORKSTATIONS
Phoenix did not oppose the registration of UW's CONDOR mark at the time it was published for opposition. UW claims a date of first use of February 1, 1988 for its CONDOR mark.
In 2000, the year before it obtained its federal registration in Condor, the University of Wisconsin provided 3,738 downloads of its software. The number of downloads increased to 15,155 in 2004, an increase of over 400 percent from 2000. The first Condor week offered by the University of Wisconsin in 2000 attracted about 20 participants and lasted one day. Since then, the Condor week has grown into a four day event attracting more than 150 participants. The University of Wisconsin experienced a continuation and expansion of grant funding and a corresponding investment in the Condor project between 2000 and 2004 and continued to develop new versions and releases of the UW Software between 2000 and 2004.
D. Cancellation of Plaintiff's Mark
On January 6, 2004, Phoenix filed a petition to cancel UW's registration for the mark CONDOR, in a cancellation proceeding identified as Cancellation No. 92042881. On September 26, 2007, the Trademark Trial and Appeal Board granted Phoenix's petition to cancel UW's Registration No. 2,434,630.
In determining whether to cancel plaintiff's registration, the board determined whether there was a likelihood of confusion between the parties' goods in light of a 13-factor test set forth in In re E.I. DuPont Nemours Co., 476 F.2d 1357, 1361 (C.C.P.A. 1973) (factors relevant to test for likelihood of confusion include: (1) similarity of marks; (2) nature of goods or services as described in application or registration; (3) similarity of trade channels; (4) conditions under which sales are made, including whether sales made are careful and sophisticated or "impulse"; (5) fame of prior mark; (6) number and nature of similar marks; (7) nature and extent of actual confusion; (8) length of time of concurrent use without actual confusion; (9) variety of goods on which mark is used; (10) market interface between applicant and owner of mark, including laches and estoppel; (11) extent of applicant's right to exclude others from use of its mark; (12) whether potential confusion is de minimis or substantial; (13) any other fact probative to effect of use).
1. Similarity of marks
2. Nature of parties' goods and similarity of trade channels
Octocom Sys., Inc. v. Houston Computers Services, Inc. 918 F.2d 937942
The board quoted the identification of the goods in the parties' respective registrations and turned to consider the differences between the goods described. It stated that"[o]ne of the differences between [the parties' respective] software is the fact that [defendant's] software functions on `mainframe systems' while [plaintiff's] goods are used in connection with a `network of individual computer systems.'" The board then considered the parties' competing evidence regarding that difference. First, it cited to the deposition of plaintiff's mainframe coordinator, who stated that, as far as he could "tell," an organization having no mainframe and not involved in developing software applications for mainframe computers would have no use for defendant's Condor program.
The board considered defendant's evidence that the products were sufficiently related, which all came from defendant's president, Fred Hoschett. The board cited Hoschett's testimony that: the Condor software could run "on someone's desktop, as if it were a mainframe" and could operate "in a network of workstations"; defendant has "many customers that use defendant's Condor software that do not have mainframes; defendant's software was "a toolbox of functionality to be used essentially by anyone who uses a computer to assist them in doing their jobs, whether it is programming software, submitted [sic] batch jobs and queuing batch jobs, or managing the environment or managing the resources"; and when Hoschett first read a description of plaintiff's Condor software, it "made an emotional impact" on him and he initially believed that the description was about defendant's product. Finally, the board noted testimony by plaintiff's then Mainframe Coordinator, Clark Jones, who conceded that "there might be [some] incentive" for software manufacturers to seek to operate in both mainframe and network environments.
The board acknowledged that there was a "clear" difference between the products and stated that "the main difference appears to be that petitioner's goods are used in a mainframe environment while respondent's goods are used in a network of individual computer workstations." In addition, the board considered plaintiff's argument that defendant's evidence that its product may be used in either environment was insufficient to show likelihood of confusion because "whether [defendant's] software is running on a PC via a mainframe emulator, or is running on a mainframe computer, is a distinction without a difference. In either case, the software is being used by mainframe programmers for mainframe programming purposes."
Nonetheless, the board concluded that the products were sufficiently related to weigh in favor of likelihood of confusion, stating that "[t]here is at least some evidence in the record that the parties' respective software performs the same general functions and the evidence does not demonstrate the goods are used in distinctly different fields" and "there is no clear division between the parties' software that would cause us to conclude that these products are not related." In reaching its conclusion, the board reasoned that "to the extent that [the parties'] software can both be used on workstations (even if not the same workstations), it is more likely that the same consumers would encounter these products." The board considered also the fact that plaintiff's goods were identified as "downloadable from a global computer network," but concluded that the products could not be distinguished on that basis.
3. Condition under which sales are made
Turning to the next factor, the board determined that purchasers of either product "would have some level of skill and sophistication to the extent that they are programming mainframe computers or networking computer workstations to increase computational resources. It also is apparent that these purchases would be made with some care." Nonetheless, the board concluded that "even sophisticated purchasers who would be exposed to advertising for [defendant's] and [plaintiff's] software sold under the identical marks that could perform similar functions are likely to believe that there is some association or relationship between the sources of the software."
4. Evidence of actual confusion
The board determined that there was no evidence of actual confusion. Defendant attempted to submit some evidence of confusion in the form of internet searches for "Condor Software" and a Wikipedia entry for plaintiff's mark, the board concluded that it was not sufficient to establish actual confusion. At the same time, the board disagreed with plaintiff's contention that the absence of actual confusion weighed heavily in its favor. It concluded that the parties' current marketing practices were relatively limited and neither party had "conducted a large scale, public marketing campaign," citing evidence that plaintiff's did not advertise its product at all and defendant spent only $65,000 on marketing. The board refused to "assume that both parties will continue to operate in such a limited manner" and noted that plaintiff had reported that it intended to expand its operations. It concluded that the expanding operations made the chance that confusion would occur more likely, undermining the importance that actual confusion had not occurred.
The board then balanced the factors to determine whether defendant had met its burden to prove likelihood of confusion by a preponderance of the evidence. It summarized its findings related to the factors, stating that: the parties' marks are identical; "the parties' software performs similar functions and, therefore, we cannot find that they are used in unrelated fields"; and "[e]ven sophisticated purchasers would likely believe that there is some relationship or association between the sources of the goods under these circumstances."
E. New Evidence Related to Likelihood of Confusion
1. Defendant's Product
The Phoenix Software functions only in a computing environment connected in some manner to a mainframe computer system and cannot operate in a network of individual workstations that are not connected to a mainframe computer. The software is designed to be used on a mainframe, although it may be used on non-mainframe computers through the use of emulation software. An entity may use the Phoenix Software for on-line development of programs designed to run on mainframe systems, management of libraries of mainframe programs and mainframe systems utilities functioning. As a component of its software for on-line development of mainframe programs, plaintiff's software allows its users to submit batch jobs to local and remote computers through a network of computers to more effectively utilize and balance the available computing cycles.
Typically, an organization owns a mainframe computer in order to support its business applications such as accounting, payroll and budget. True mainframe computer systems are generally expensive computing systems that are extremely reliable and secure and capable of enormous throughput, although small computers such as microcomputers or personal computers may use special software to "emulate" a mainframe. Mainframes are centrally managed and maintained, remotely accessed and optimized for the movement of data. The consequences of installing the wrong software on a mainframe could be severe, so deciding which software to install on a mainframe computer requires careful consideration.
When an organization needs a specific application or program in order to exploit the functionality of its mainframe computer, it can either employ programmers in-house or contract the programming work to a third party expert. The end-users of the Phoenix software include mainframe systems administrators and mainframe systems application developers operating on IBM or compatible mainframe operating systems. Mainframe system administrators learn about mainframe software products and new developments in the field by talking to other mainframe system administrators, through mainframe publications, going to mainframe trade shows and conferences and by employing consulting firms. People who work with mainframe computers actually self-identify as a community because it is a niche of the IT community dedicated to supporting mainframes and not other platforms or operating systems.
Software for mainframe computers is not typically available under an open source license. Phoenix charges a one-time, up-front license fee of $15,000 to $50,000 for its software in addition to an annual maintenance charge of 15% of the then current license price. In addition, a purchaser of Phoenix's Software would need a license to run the IBM software necessary to run the mainframe system tied to the Phoenix Software. Such a license runs anywhere from $30,000 to $300,000 a year, depending upon the specifics of the licensee's organization. As of April 2006, Phoenix had 100 or less active licensees of its software. The average time to consummate a sale of the Phoenix Software is three months.
Phoenix advertises the Phoenix Software on the Internet, at trade shows and through product brochures. Phoenix's sales brochure states that it serves "the mainframe software community." The trade shows at which Phoenix advertises the Phoenix Software are AFCOM, OS/390, SHARE, GUIDE, WAVV and various IBM trade shows.
Phoenix spent $64,480.00 in total marketing expenditures, including sales training, personnel expenses, market research, trade shows and travel expenses for the Phoenix Software in 2000, and spent $66,786.00 on the same items in 2003. Phoenix's only competitors are IBM and Computer Associates.
2. Plaintiff's Product
The University of Wisconsin began offering its Condor software in June 1988 and has offered the UW Software continuously since 1988. Professor Livny chose the name "Condor" for the software; at the time he was not aware of any other computer or software products offered under the Condor mark. In spite of his vocation as a computer science professor and the extent and thoroughness of his scholarly activities, Professor Livny never encountered the Phoenix Software prior to this case.
The University of Wisconsin distributes its software under an open source license free of any cost to the consumer. The UW Software enables "grid" computing to bring the principles of distributed computing to its users. "Grid" computing is a metaphorical description based on the power grid, where computing power is seamlessly available to users via a distributed computing infrastructure. The term "high throughput computing" was coined by Professor Livny to articulate and differentiate the concept behind the UW Software from high performance computing. The goal of the "Condor Project" is to develop, implement, deploy, and evaluate mechanisms and policies that support high throughput computing on large collections of distributively owned computing resources.
The UW Software operates in an environment of loosely-coupled computers, which are individual workstations and PCs that maintain hardware, software and ownership autonomy while serving as part of a pool of computational resources. The UW Software links together a network of individually-owned computer resources to create a "greater computational system" from the individual computers in the network while allowing each computer to maintain its autonomy. The purpose of the UW Software is to enable individual users to exploit under-utilized computer resources on idle workstations within the network/pool of individual computer workstations.
The UW Software does not run under mainframe operating environments and could not be installed on a mainframe computer. (The parties dispute whether a mainframe could become part of the "power grid" upon which the UW Software operates.) Moreover, the UW Software is not used to build applications, does not provide a programming environment for building applications and has no components that are useful for providing a programming environment. (Defendant attempts to dispute this fact by noting that the UW Software provides explicit support for certain types of programming environments, allowing users to write and execute applications to these interfaces on machines managed by the UW Software. However, this fact does not show that the software "provides a programming environment" for building the applications or has components that are useful for providing such an environment. Instead, it simply shows that the UW Software does not conflict or interfere with the identified programming environments. Plaintiff's proposed finding of fact remains undisputed.)
The primary consumers of the UW Software have academic, scientific or engineering backgrounds. They include individuals responsible for the computing infrastructure of scientific research groups, like high energy physics researchers, engineers and computer scientists that have preexisting, certified applications. The people downloading and installing the UW Software generally require a systems-level understanding in order to configure the network and make computing resources available to users within the network. However, the UW Software is free and available for anonymous download, so others may use it as well.
The user communities for the UW Software include such groups as the high energy physics community, the DOE National Labs, biology and computer science departments, and industrial groups like Micron Technology, which has a Condor pool of 12,000 Windows machines. A research group or laboratory uses the UW Software to create a pool of computing resources by interconnecting a collection of individual workstations. When a user in a Condor pool has a job that requires processing resources in excess of his individual workstation's capacity, the UW Software identifies and schedules computing resources on idle computers in the Condor pool to complete the job in a time that is much faster than his individual workstation could do alone.
The University of Wisconsin relies on word-of-mouth communication among scientists or groups of scientists and presentation in various academic and scientific forums to promote its software. In addition, it holds an annual meeting, called Condor Week, to unite the community of University of Wisconsin's Condor researchers and users. The University of Wisconsin also hosts a "Condor" website and provides an email address on its web site designed to respond to questions about the product. Since the University of Wisconsin started tracking inquiries submitted to this email address in 1998, it has received just short of 14,000 discrete contacts. The University of Wisconsin has never experienced or received reports of any instances of actual confusion regarding the source of the Phoenix Software and the UW Software.
Since the University of Wisconsin began tracking numbers in 1998, it has seen an upward growth in licensed downloads of its software, e.g. from 3,001 in 1999 to 17,193 in 2005.
3. Overlap between the parties' products
4. Actual Confusion
OPINION A. Standard of Review of Board's Decision
The case comes before the court pursuant to 15 U.S.C. § 1071(b) for review of the decision of the Trademark Trial and Appeal Board to cancel plaintiff's mark. For cases brought to the district court under § 1071(b), the court "sits in a dual capacity. It is an appellate reviewer of facts found by the TTAB and is also a fact-finder based on new evidence introduced to the court." CAE, Inc. v. Clean Air Engineering, Inc., 267 F.3d 660, 674 (7th Cir. 2001). In this setting, a court affords deference to the findings of fact made by the board but considers the board's decision de novo to the extent the parties present new evidence. Id. The board's findings of fact are properly reviewed under the standard set forth in 5 U.S.C. § 706, which requires the court to set aside findings and conclusions "unsupported by substantial evidence." Id. at 675-76 (citing Dickinson v. Zurko, 527 U.S. 150 (1999)). Under this standard, the board's findings of fact must be upheld so long as they are supported by "such relevant evidence as a reasonable mind might accept as adequate" to support that finding. Id. (citing Consolidated Edison Co. v. NLRB, 305 U.S. 197, 229 (1938)).
The parties dispute exactly how to apply the substantial evidence standard to the board's finding that there is a likelihood of confusion between the parties' marks. According to plaintiff, the board's ultimate conclusion that there exists a likelihood of confusion is not a finding of fact that receives deference, but instead is a legal conclusion that should be reviewed de novo. Plaintiff cites M2 Software, Inc. v. M2 Communications, Inc., 450 F.3d 1378, 1381-82 (Fed. Cir. 2006) andPalm Bay Imports, Inc. v. Veuve Clicquot Ponsardin, 396 F.3d 1369, 1371 (Fed. Cir. 2005) in support of its position. I disagree. In this circuit, as in most circuits, the question of likelihood of confusion is considered a question of fact.AutoZone, Inc. v. Strick, 543 F.3d 923, 929 (7th Cir. 2008) (citing Seventh Circuit cases); 4 J. Thomas McCarthy, McCarthy on Trademarks and Unfair Competition, at §§ 23:71 and 23:73 (majority of circuit courts hold that likelihood of confusion is issue of fact). Because likelihood of confusion is a question of fact, the board's finding on the matter will be given deference, not to be set aside if supported by substantial evidence. However, to the extent the parties have submitted new facts relevant to the likelihood of confusion analysis, those facts will be considered de novo and the board's determination will be considered in the context of the newly-considered evidence. CAE, 267 F.3d at 676 (concluding that district court applied proper standard of review when it acted as fact finder with respect to new evidence while also affording deference to board's findings).
Because the question whether the board's cancellation was proper is now before the court in the context of the parties' cross motions for summary judgment, I may reverse or affirm the board's cancellation at this stage only if "the evidence [related to the question of likelihood of confusion] is so one-sided that there can be no doubt about how the question should be answered."Id. at 677 (internal quotations omitted).
B. Likelihood of Confusion
The Court of Appeals for the Seventh Circuit applies a seven-factor test to determine likelihood of confusion, similar to the test set forth in Dupont:
(1) the similarity between the marks in appearance and suggestion;
(2) the similarity of the products;
(3) the area and manner of concurrent use;
(4) the degree and care likely to be exercised by consumers;
(5) the strength of the plaintiff's mark;
(6) any actual confusion; and
(7) the intent of the defendant to "palm off" his product as that of another.Autozone, 543 F.3d at 929.
In finding a likelihood of confusion, the board considered the following factors: similarity between the parties' marks, the similarity of the products, the area and manner of concurrent use, the degree and care likely to be exercised by consumers and the extent of actual confusion. The board's conclusion that there was a likelihood of confusion must come out of its findings that the marks are identical and the products and their area and manner of use overlap because it found that no other factor weighed in favor of likelihood of confusion. The other factors weighed in favor of no confusion, if only slightly: the purchasers of either product were found to be sophisticated consumers and there was no evidence of actual confusion.
Plaintiff challenges the board's findings related to the similarity of the products and the area and manner of concurrent use. First, plaintiff contends that the board erred when it found the parties' products were sufficiently similar to make confusion likely because the board strayed away from the description of goods described in the parties' marks into the actual nature of the parties' goods. As plaintiff points out, the Court of Appeals for the Federal Circuit has long held that, for the purpose of determining whether a mark may be registered must be determined "on the basis of the identification of goods set forth in the application regardless of what the record may reveal as to the particular nature of applicant's goods." Octocom, 918 F.2d at 942 (citing cases). Although the Court of Appeals for the Seventh Circuit has not addressed this issue, I see no reason why it would decline to follow the Federal Circuit on this matter.
I agree with plaintiff that the board erred when it found the goods were similar because it based its decision on the actual nature of the parties' goods. The board suggested that it was considering the nature of the parties' goods only "[t]o the extent that these facts provide some information about the market and purchasers of these goods"; however, its reasoning shows otherwise. The board noted that the identifications of the parties' respective goods were distinct because defendant's product functions on mainframe systems and plaintiff's is used in a network of individual computer systems. However, it found the products sufficiently similar to be confusing because "[t]here is at least some evidence in the record that the parties' respective software performs the same general functions." Because there was no evidence that mainframe systems in general overlap with systems of networks of individual computer, the board's finding that the products were similar lacked substantial evidence.
Nor does the new evidence submitted by the parties support a finding that the products are sufficiently related to make confusion likely. The language in the parties' registrations limits the goods in ways that distinguish them: plaintiff's product is limited to "network[s] of individual computer workstations" and defendant's product is limited to "mainframe systems." M2 Software, 450 F.3d at 1382 (court must consider scope limiting language in description of goods when determining nature of goods in cancellation proceedings). Although defendant attempts to show that the distinction between mainframe systems and non-mainframe systems is disappearing, the evidence falls far short of establishing this point. It is undisputed that mainframes are specialized systems with distinct features such as high reliability, speed and a specialized operating system; the growing "overlap" between the two systems is nothing more than the presence of an interface that allows non-mainframe systems to emulate mainframe systems and may allow certain non-mainframe programs to operate on mainframe systems. No reasonable person could find that "mainframe systems" now overlapped with "networks of individual computer workstations" such that the parties' product descriptions could be considered confusingly similar.
The board's error on this point alone does not mean its ultimate finding was in error. The board also relied on the fact that there was some evidence that the parties' identical marks were used in the same field. The board noted that plaintiff's witness had acknowledged that "there might be [some] incentive" for software manufacturers to seek to have software that operates in both mainframe and non-mainframe environments and that "to the extent that [the parties' respective] software can both be used on workstations (even if not the same workstations), it is more likely that the same consumers would encounter these products." However, the board did not point to any evidence of actual overlap in trade channels or the field of use of the products. The board appeared to recognize that it had no evidence of overlap before it when it found only that "the evidence does not demonstrate the goods are used in distinctly different fields." To the extent the board relied on this "finding" to reach its conclusion of likelihood of confusion, it was error. Plaintiff's mark is presumptively valid and may only be cancelled by a preponderance of evidence. 15 U.S.C. § 1057; West Florida Seafood, Inc. v. Jet Restaurants, Inc., 31 F.3d 1122, 1125 (Fed. Cir. 1994). Therefore, it was not plaintiff's burden to show the board that the "goods are used in distinctly different fields"; instead, it was defendant's burden to show that they are not. Defendant's attempt to submit new evidence of overlapping trade channels does not improve this deficiency. Defendant attempts to submit new evidence by nestling it in a response to plaintiff's proposed findings of fact. Dft.'s Resp. to Plt.'s PFOF, dkt. #51, ¶ 79. However, the evidence defendant submits of overlapping trade channels was not responsive to the fact proposed and defendant did not offer the evidence in its own proposed finding of fact as required by this court's procedure and will therefore be disregarded. Procedure to Be Followed on Motions for Summary Judgment, I.B.1., II.B-D and Helpful Tips for Filing a Summary Judgment Motion in Cases Assigned to Judge Barbara Crabb, attached to Preliminary Pretrial Conference Order (February 21, 2008), dkt. #19. Even if I were to consider the additional evidence, it would do little to help defendant's cause. It shows only that both plaintiff's and defendant's products are used by universities and other companies. There is no evidence that the same universities and companies purchase the product and, more important, that the same customers or purchasers within any of the institutions and companies purchase both products. Electronic Design Sales, Inc. v. Electronic Data Systems Corp., 954 F.2d 713, 717 (Fed. Cir. 1992) (citing Astra Pharmacy Products, Inc. v. Beckman Industries, Inc., 718 F.2d 1201, 1206 (1st Cir. 1983)) (mere purchase by same institution insufficient to establish similarity of trade channels or overlap of customers because "likelihood of confusion must be shown to exist not in a purchasing institution, but in a `customer or purchaser'").
After setting aside these two findings by the board, there is nothing left to its finding that a likelihood of confusion exists between the parties' marks. The board found that the purchasers of the parties' products would be sophisticated buyers and that there was no evidence of actual confusion; without evidence that the products and trade channels are similar or overlapping, no reasonable person could find a likelihood of confusion exists under these circumstances. See, e.g., M2 Software, 450 F.3d at 1383 (upholding board's finding of no likelihood of confusion between marks after finding that parties' multimedia CD-Roms were not similar and had different channels of trade; one party's product was used strictly in the film and music industries and the other's was used in the fields of pharmacy and medicine.) The defendant submits no other new evidence that would support a finding that consumers are likely to be confused by the parties' identical marks. Because the board's finding is not supported by substantial evidence, and because the new evidence before the court shows that defendant cannot establish a likelihood of confusion between the parties' respective marks, plaintiff's motion for summary judgment will be granted and the board's decision cancelling plaintiff's mark must be reversed.
ORDER
IT IS ORDERED that
1. Plaintiff Board of Regents of the University of Wisconsin System's motion to strike, dkt. #69, is GRANTED.
2. Defendant Phoenix Software International, Inc. motion for summary judgment, dkt. #37, is DENIED.
3. Plaintiff's motion for summary judgment, dkt. #41, is GRANTED. The judgment of the Trademark Trial and Appeal Board is REVERSED with directions to the board to reinstate Registration No. 2,434,630.