MEMORANDUM AND ORDER
GERTNER, District Judge.
I.
INTRODUCTION
This case involves a dispute over the use of nearly identical tráde and service-marks.
The Plaintiff, Alta Vista Corporation, Ltd. (“AVC”), is the senior user of the servicemark “Alta Vista” in New York City and Los Angeles, where it started to conduct business in 1993. The Defendant, Digital Equipment Corporation (“Digital”), introduced a nearly identical mark, “Alta-Vista,” across the country in 1995. Digital is therefore the junior user of this closely related family of marks in New York, City and Los Angeles. To prevent confusion to its clients and business contacts, AVC seeks a preliminary injunction to bar Digital from using the name “AltaVista,” in the two areas where AVC is the senior user. For the reasons stated below, AVC’s motion is DENIED.
II.
BACKGROUND
The facts in this case are largely uncontested. AVC is a small (three full-time employees), London-based company that “manages) markets, sells and exploits intellectual property assets,” such as books, films, and television shows. In effect, it is a specialized literary services agency, providing labor-intensive representation and management to its clients, mostly writers seeking to publish books, or produce films or television shows.
AVC began using the servicemark “ALTA VISTA” in New York City and Los Angeles in 1993, but did not seek to have a more general presence in the United States until 1997. In January of that year it took out its first advertisement in an American publication; in April it created a Web page,
giving Americans an easy way to find out about their services; and in September of that year its trademark on “Alta Vista” was issued.
Digital is a large corporation, involved in a variety of “high tech” and media enterprises. In December 1995, it launched its AltaVista Internet search engine.
Through it, Digital provides one of the leading Internet search services.
Digital was not aware of AVC or its use of the name “Alta Vista” when naming its search engine. It chose the name in a brainstorming session because it means “high view” in Spanish, which seemed apt for a Web site that would offer a view of the entire Web. Nonetheless, it has been using a name almost identical to AVC’s name for over two and a half years.
After launching the search engine, Digital introduced an array of Internet-related computer software, starting in May 1996. It markets the software under numerous “AltaVista” names, including AltaVista Directory, AltaVista Mail, AltaVista Forum, and AltaVista Firewall.
More recently, Digital announced an agreement with Amazon.com, one of the leading booksellers on the Web. According to the agreement, Amazon.com is prominently advertised on the AltaVista site, and- a search done on the AltaVista search engine can also be done with the click of a mouse button in Amazon.com’s database. Furthermore, Digital gets a commission of 2.5% of the purchase price of every book bought on Amazon.com prompted by an initial search on the AltaVista search engine.
In January 1998, Digital announced an agreement with'- OneZeró Media, Inc. (“OZM”) to create a partnership between the AltaVista search engine and OZM’s national TV series, Wild Wild Web.’ According to Digital’s press release, “the companies will co-produce an entertainment area on AltaVista, based on programming from the Wild Wild Web.’ AltaVista users will gain access to a wide variety of content about pop culture, music, film, sports, entertainment news and events, TV, comedy, radio, games and literature.”
AVC contends that by making these agreements with OZM, Amazon.com, and
others, Digital has “expanded its AltaVista business into a full-fledged media company, which develops and promotes many forms of intellectual property.” In so doing, it brings its trademark into competition with AVC’s senior mark in the areas of New York City and Los Angeles. The result, AVC claims, is that it is suffering “reverse confusion,” a condition in which a less well-known senior user’s goodwill and identity are overwhelmed by a more well-known junior user’s use of a confusingly similar mark. Claiming that this reverse confusion results from a violation of Section 43(a) of the Lanham Act, 15 Ú.S.C. § 1125(a), AVC seeks a preliminary injunction enjoining Digital from using the disputed mark in New York City and Los Angeles.
Digital responds that there is no substantial likelihood of confusion between the marks of the two parties. It acknowledges that the marks are similar, but claims that they make different impressions in their respective commercial contexts. More importantly, it claims that it is not “a full fledged media company,” and that it has not expanded into “media development, publicity and promotion servicés.” In addition, it claims that any confusion AVC has to deal with is
de minimus,
while the cost of changing the name of Digital’s Al-taVista search engine would be very high, significantly higher than even the proportional cost to the much smaller AVC of clarifying any confusion that may arise from the similar marks.
III.
ANALYSIS
A.
Preliminary Injunction Standard
Having recently addressed this topic in
Digital Equipment Corp. v. AltaVista Technology, Inc.,
960 F.Supp. 456, 472 (D.Mass.1997), I quote myself at length (internal square brackets omitted):
The granting of a preliminary injunction in actions for trademark infringement under § 43(a) of the Lanham Act requires a showing that: (1) the plaintiff may suffer irreparable injury absent an injunction; (2) equitable balancing weighs in favor of plaintiff when measuring the harm the injunction would do to defendants; (3) plaintiff is likely to succeed on the merits; and (4) public interest favors granting the preliminary injunction.
See, e.g., Camel Hair and Cashmere Institute of America, Inc. v. Associated Dry Goods Corp.,
799 F.2d 6, 12 (1st Cir.1986).
In the context of trademark infringement, if [AVC] can show it is likely to prevail on the merits of its infringement claims ..., two consequences follow. First, there is a presumption that [AVC] will suffer irreparable injury absent an injunction.
See id.
at 14-15. Second, the preliminary injunction will be held to be beneficial, and in the public interest, “given the societal value of full disclosure and fair competition, together with the policy of the law to provide at least minimal protection to established trade names.... ”
Hypertherm, Inc. v. Precision Products, Inc.,
832 F.2d 697, 700 (1st Cir.1987);
see also Calamari Fisheries, Inc. v. The Village Catch, Inc.,
698 F.Supp. 994, 1015 (D.Mass.1988) (“Preventing consumer confusion is clearly in the public interest.”).
Thus, in the trademark context, the “heart of this test for preliminary injunctions is the second and third steps, which present the question whether the harm caused plaintiff without the injunction,
in light of
the plaintiffs likelihood of eventual success on the merits, outweighs the harm the injunction will cause defendants.”
Calamari Fisheries,
698 F.Supp. at 1005 (noting that this test applies “particularly in actions arising out of the Lanham Act”).
960 F.Supp. at 472.
As I explain below, AVC is right to claim that special attention must be paid to the strength of the junior user’s mark in cases involving reverse discrimination. But this affects only the analysis of one factor in determining plaintiffs likelihood
of success; it does not change the overall legal standard for a preliminary injunction in a trademark infringement action. Thus, with this one very limited exception, I follow the First Circuit’s treatment of reverse confusion as just another type of confusion. “The principal trademark statute does not speak of ‘ordinary,’ or ‘reverse,’ confusion. It refers simply to copying that is ‘likely to cause confusion,’ without dividing confusion into types.”
DeCosta v. Viacom Intern., Inc.,
981 F.2d 602, 608 (1st Cir.1992).
See also Star Financial Services, Inc. v. Aastar Mortgage Corp.,
89 F.3d 5, 10 (1st Cir.1996) (citing
DeCosta
and referring to “so called ‘reverse confusion’ ”).
B.
Likelihood Of Success
I
will first address AVC’s likelihood of success on the merits. To succeed under the Lanham act, AVC needs to show: “1) that [it] uses, and thereby ‘owns,’ a mark; 2) that the defendant is using that same or a similar mark; and, 3) that the defendant’s use is likely to confuse the public, thereby harming the plaintiff.”
Star Financial Services, Inc. v. Aastar Mortgage Corp.,
89 F.3d 5, 9 (1st Cir.1996). Digital does not contest that AVC owns, uses, and is the senior user of the “Alta Vista” mark in New York City and Los Angeles. In addition, there is no doubt that the two marks have a strong resemblance, at least out of context. The more difficult question is whether the resemblance is such as to lead to confusion. As such it is part of the third and central question: Is Digital’s use likely to confuse the public?
As I noted in
Digital Equipment,
960 F.Supp., at 476, the First Circuit, in
Pignons S.A. de Mecanique de Precision v. Polaroid Corp.,
657 F.2d 482 (1st Cir.1981), established the following non-exhaustive list of factors used in assessing the “likelihood of confusion” between marks:
. the similarity of the marks; the similarity of the goods; the relationship between the parties’ channels of trade; the relationship between the parties’ advertising; the classes of prospective purchasers; evidence of actual confusion; the defendants’ intent in adopting its mark; and the strength of the plaintiffs mark.
Id.
at 487;
see also Astra Pharmaceutical Products, Inc. v. Beckman Instruments, Inc.,
718 F.2d 1201, 1205 (1st Cir.1983) (enumerating the
Pignons
test, noting these factors are “to be used as guides in assessing the likelihood of confusion”); and
Aktiebolaget Electrolux v. Armatron Int’l, Inc.,
999 F.2d 1, 4 (1st Cir.1993) (“No one factor is conclusive as to likelihood of confusion, and the district court must consider each.”). I turn then, to an application of
Pignons
’ eight factors to the “Alta Vista” mark.
1.
Similarity of the Marks
As AVC points out, the marks are highly similar. Textually, the only difference is that Digital’s mark does not have a space between the words “Alta” and “Vista.” In addition, both marks superimpose the text over an image of a mountain range within a rectangular frame. But, as Digital points out, the marks should not be compared in the abstract, out of context. “A realistic evaluation of consumer confusion must attempt to recreate the conditions in which buying decisions are made, and the court should try to determine not what it would do, but what a reasonable purchaser in market conditions would do.”
Calvin Klein Cosmetics Corporation, v. Lenox Laboratories,
815 F.2d 500, 504 (8th Cir.1987).
I agree with Digital that when viewed in their respective contexts,. the m.arks would be hard to confuse. Take, for example, their appearance on the Web. A Web user who arrives at Digital’s Alta-Vista search site can easily tell that the service offered is an Internet search service. Likewise, a Web user who arrives at AVC’s site can easily tell that the service offered is that of a literary agency.
Therefore I conclude that in context they are not particularly similar marks.
2.
Similarity of the Goods and Services
AVC argues that by forming a special connection with Amazon.com and OZM, Digital has entered the publishing and television markets, areas where AVC has been actively working in the United States since 1993. AVC also points to Digital’s promotion and review of current movies, another area of commercial activity for AVC.
The question is whether being involved in the same media (books, television, and movies) constitutes provision of similar goods and services. The District Court for the Southern District of New York explains well why more is needed:
To escape the seemingly obvious conclusion that the bulk of plaintiffs services are not competitive with defendant’s, [plaintiff] contends that both are in the same general class of “educational services.” However, this type of classification is so broad as to be meaningless for purposes of determining that the products are proximate.... By increasing the level of generality, any products can be made to appear to fall in the same class. Aspirin and easy chairs could be characterized as “comfort products.” Jet planes and roller blades could be characterized as transportation products. Such semantic exercises simply are not helpful in assessing likelihood of confusion.
Mejia And Associates, Incorporated, v. International Business Machines Corp.,
920 F.Supp. 540, 548 (S.D.N.Y.1996) (internal citations omitted).
See also Lang v. Retirement Living Publishing Co., Inc.,
949 F.2d 576, 582 (2d Cir.1991) (defendant’s magazine that catered to older adults and plaintiffs publishing house did not produce confusingly similar products because they were both “in the field of publishing”).
This leaves the question of how tough the standard should be. Consider that the
Pignons
court distinguished the goods offered by two different single lens reflex camera manufacturers because one sold low priced “instant” cameras, and the other sold high priced traditional cameras with a very different look. 657 F.2d at 487-88.
See also Lang,
949 F.2d at 582 (defendant’s magazine that catered to older adults and plaintiffs publishing house are different goods and services). Contrast
Digital Equipment,
where I found the companies provided similar goods because both “provide search engine services ..., sell computer software under the “AltaVista” mark... [and] utilize their Websites to advertise.” 960 F.Supp. at 477.
When examined under this exacting standard, the similarity between Digital’s and AVC’s services evaporates. AVC offers literary services, and Digital does not. The fact that Digital has a special arrangement with Amazon.com does not fundamentally change the picture. The parties contest whether the right metaphor for the relationship between Amazon.com and Digital is that of an advertiser to a magazine. It does not matter. Even if I were to treat Amazon.com as another subsidiary of Digital, the services are still quite distinct. Amazon.com sells books to the public, and provides a bulletin-board like space on which people can post reviews of the books Amazon.com sells. AVC promotes and develops intellectual property not for the public at large, but for publishers and producers. They come at the same material from completely different ends, and thus their services are fundamentally dissimilar and noncompetitive. The same holds,
mutatis mutandis,
for OZM.
AVC makes one more argument in this category. It points out that Digital recently published an article on its “Mar-ketspace I-School” Web page (www.mar-ketspace.altavista.digital.com) which encouraged authors to publish on the Web. This advice is directly counter to the advice given by AVC, and AVC contends
that it makes the two companies competitors in the ■ area of advising authors on publication- strategies. This would be a compelling point if such advice on publishing were a major theme of the Mar-ketspace I-School Web page, but it is not. It was just one of thousands of articles that have appeared on that site. No reasonable reader would infer from the fact that Digital runs a site that carried that one article, that it had acquired an interest in running a literary agency of the sort that would compete with or could be confused with AVC’s.
3.
Channels Of Trade
4.
Forms Of Advertising
5.
Classes Of Prospective Purchasers
These factors are often analyzed together.
See Digital Equipment,
960 F.Supp. at 477; and
Astra Pharmaceutical,
718 F.2d at 1206. The differences in the channels of trade are substantial. AVC conducts a labor-intensive, personal form of business. Digital is a mass marketer of products and impersonal services such as Internet searches.
There is a similarity in the form of advertising insofar as both obtain clients from their, sites on the Web. But AVC also solicits business by “cold calling” potential customers; something which Digital does not do.
And while both parties do business with sophisticated clients — those who use the Internet and those who seek to create and promote intellectual property — this does not tilt in favor of finding confusion. For not only is AVC’s clientele a small subset of the people who are likely to use the Web, but the sophistication of those clients is sufficiently high that they can be .expected to know the difference between an Internet search service and a literary agency.
See Astra Pharmaceutical,
718 F.2d at 1206 (citing the sophistication of hospital chemistry lab staff and pharmacy staff as the primary reason confusion was unlikely to exist);
see also Pignons,
657.F.2d at 489 (“Those most likely to buy an expensive, sophisticated camera in a specialty camera store are also least likely to be confused by any similarities in Polaroid’s and Pignons’ marks.”).
6.Evidence Of Actual Confusion
AVC cites two instances of actual confusion experienced by its clients. First, when Mr. Peter Cox (“Cox”) of AVC contacted Mr. Isadore Rosmarin (“Rosmarin”) and handed him an AVC business card, Rosmarin — a television producer, writer and director who considers himself “fairly computer literate” — wondered if the name “Alta Vista” did not refer to the search engine. When told that AVC was not the search engine, Rosmarin asked if it was owned or sponsored by the AltaVista search engine, and wondered “who the legitimate owner of the name ‘Alta Vista’ was.” This concern seems to have haunted Rosmarin, as Cox had to spend “some time trying to explain the duplicative use the name ‘Alta Vista’ ” to him.
Second, Mr. Gregor Crosby (“Crosby”) — the husband and “formal manager” of one of AVC’s clients — after hearing about the AltaVista search engine, wondered whether AVC “might have been associated with Digital’s search engine.” He claims that if AVC “had the backing of a very large company, we may have sought greater remuneration in the contract.”
There are three reasons why these examples do not help AVC’s cause. First, they are the only two examples of customer confusion to which AVC can point.
Given that the parties have been using a nearly identical name for two and a half
years, that is a rather insubstantial showing of confusion. Of course, the small number must be put in context. AVC does not have millions of customers as some other businesses have had where the low number of confused persons was telling.
See, e.g., Scott Paper Co. v. Scott’s Liquid Gold, Inc.,
589 F.2d 1225, 1231 (3d Cir.1978) (nineteen misdirected letters over four years “extremely minimal evidence” and insufficient to support a finding of confusion in the marketplace, given the millions of customers of both companies). Nevertheless, it is a relevant factor.
More telling is the fact that the confusion did not run deep. Rosmarin went on to become one of AVC’s clients, and Crosby’s wife remained one of AVC’s clients. As the court in
Lang
held, “the relevant confusion is that which affects the purchasing and selling of the goods or services in question.” 949 F.2d at 582 (internal quotation marks omitted). While there may have been some potential for the confusion to affect the decisions of Rosmarin and Crosby, it seems in the end that it did not.
The third reason this confusion is not very telling is that it is not really the sort of confusion which counts under trademark law. Rosmarin’s confusion, in particular, seems to have been a concern that either AVC or Digital was “usurping the rightful name of the other.” The claim that a concern with usurpation or piracy should be encompassed by trademark law was explicitly rejected, however, in
DeCosta:
[T]o adopt [such a] theory would undermine an important limitation central to the law of trademarks, the limitation of trademark protection to the protection of marks
as used
on particular goods to identify their source or sponsor.... If falsely being thought a pirate were an actionable harm, no one could safely use a mark ever previously used by another, no matter how different the product, place of sale, or class of buyer.
981 F.2d at 609 (internal quotation marks omitted).
The only other evidence of confusion that AVC cites is an article in the
Guardian
newspaper.
Referring to the name “Alta Vista,” the article reads: “You thought it was a web browser, but Mr. Cox has adopted the name for a new literary agency....” The author of this article is clearly confused about who is the senior user of-the mark, but as with Rosmarin, the confusion reflects a concern with piracy, not with the underlying goods or services sold. As such it is not relevant to trademark protection. Indeed, no one reading the article would think that services provided by AVC were really traceable to or underwritten by Digital.
7.
Digital’s’ Intent in Adopting its Mark
This factor is not relevant here as AVC does not claim that Digital showed any bad faith in adopting the “AltaVista” mark.
8.
The Strength of the Marks
“Under the Lanham Act strong marks enjoy the greatest protection against infringement.”
International
Association
of Machinists and Aerospace Workers, AFL-CIO v. Winship Green Nursing Center,
103 F.3d 196, 206 (1996). Surprisingly, there is no consistent measure of what constitutes the strength of a mark, but five factors can be gleaned from the recent case law: (1) being registered; (2) being in use a long time; (3) being widely promoted; (4) being renown in the relevant field of business; and, (5) being distinctive or having strong “secondary meaning.”
The distinctiveness aspect of strength falls on a spectrum, which describes marks from least deserving to most deserving of protection as follows: “(1) generic, (2) descriptive, (3) suggestive, and (4) arbitrary and fanciful.”
Digital Equipment,
960 F.Supp. at 478. (citing
S.S. Kres
ge Co. v. United Factory Outlet, Inc.,
598 F.2d 694, 696 (1st Cir.1979).
AVC’s mark is registered; it has been in use longer than Digital’s use of the nearly identical mark; being on the Internet it is now reasonably widely promoted; it is not clear just how well known'it is in its field; and it is at least a suggestive mark, if not an arbitrary or fanciful one for a literary agency. This makes it a fairly strong mark deserving of protection.
In addition, AVC argues that what is more relevant in a case of reverse confusion is the strength of Digital’s junior mark. In making this argument they focus on the factors most relevant to the possibility of confusion, factors (3) and (4), being widely promoted and widely renowned. Though I noted when discussing the preliminary injunction standard that the First Circuit has not recognized any distinct features of a reverse confusion case,
I agree with AVC that there should be a limited exception for strength of a junior user’s mark. The more widely renowned a junior user’s mark is, the more likely it is to do whatever harm it might do to the senior user.
See Dreamwerks Production Group, Inc. v. SKG Studio,
142 F.3d 1127, 1130 n. 5 (9th Cir.1998). And as I noted in
Digital Equipment,
Digital’s AltaVista mark is “a strong and well known mark into which significant amounts of money and energy have been invested.” 960 F.Supp. at 478.
The problem for AVC is that this factor only helps them if some of the other factors have indicated that they are suffering wrongful confusion. This publicity component of the strength factor is, in effect, a multiplier. If there is some cause for confusion, then Digital’s having a well known mark would exacerbate that confusion. But there is no such cause of confusion. See
Equine Technologies, Inc., v. Equitechnology, Inc.
68 F.3d 542, 546 (1995) (district court did not err in being guided by six of seven factors that pulled one way, rather than strength, which pulled the other).
Ci
Relative Harm to AVC and Digital
I turn now to the relative harm that would befall AVC if I deny it a preliminary injunction, or Digital if I were to grant AVC a preliminary injunction.
1.
De Minimus Cost to AVC of Explaining that Alta Vista is not Affiliated with Digital
It should be relatively easy
for
AVC to explain to its clients, potential clients, and business contacts that it is not affiliated in any way with Digital’s AltaVista search engine. It could include a brief statement to that effect on its brochures and on its Web page. Indeed, AVC already takes a few minutes to explain to new clients and potential clients that there is no connection between itself and Defendant’s AltaVista search service. There is no reason to think that a brief explanation cannot clear up any confusion that may exist, given that the causes of confusion are so attenuated.
AVC complains that by forming a relationship with Web21 for the operation of the “lOOHot” website (www2.web21.com/av), Digital has associated itself with pornography on the web, and that this will tarnish the mark.
If AVC can establish that “AltaVista” is generally associated with pornography, it may have an important point. But even so, it would only be important if AVC could also show that people regularly assume that it is somehow connected with the tainted version of that mark. If there,is no confused belief in a connection, then there is no actionable harm. Having found little reason to think there is general confusion between the parties’ different uses of nearly identical
marks, and finding this concern to be still speculative, I cannot agree that AVC will suffer a serious harm if I deny it the preliminary injunction it seeks. The harm they will suffer seems most likely to be the minimal cost of setting the record straight for those people who are curious to know if there is any connection between AVC and Digital.
2.
Large Cost to Digital of Changing its Name
By contrast, the cost to Defendant of changing the name of its search engine subsidiary would be quite substantial. First, the change would have to be global. Because the Internet is a global network of computers, there is no way to change the name of the search engine only in New York and Los Angeles. Second, it would cost millions of dollars to replace “AltaVis-ta” on the Web sites where it now appears, and on all the printed materials and packaging containing the name. Third, it would also cost millions of dollars to advertise a new name to give it the kind of name recognition that “AltaVista” already has for Digital. It is worth noting that it would be imperative to spend such large quantities of money to advertise the name change because Digital’s income depends on traffic on its search engine site, which of course depends on name recognition. Fourth, it would be very difficult to change the AltaVista name on all the Web sites that now contain it because the links are run independently, and because it would be difficult for Digital to even figure out who runs all the different sites. Fifth, it would undermine the good will Digital has built up in the name to change it because Internet users would find an error message if they tried to use their old bookmarked sites,
and this could cause them to wonder if Digital’s search engine had simply been shut down.
IV.
CONCLUSION
The question has been whether the harm AVC would suffer without the injunction,
in light of
the likelihood of eventual success on the merits, outweighs the harm the injunction would cause Digital. Because I think it is unlikely that AVC would win on the merits, and because I think the harm it would suffer clarifying that it is not affiliated with Digital is less than the harm an injunction would cause Digital, I conclude that the answer is no.
Therefore, AVC’s motion for a preliminary injunction [docket # 1] is DENIED. SO ORDERED.