|
Jeffrey A. LeVee (SBN 125863)
UNITED STATES DISTRICT COURT CENTRAL DISTRICT OF CALIFORNIA
Despite plaintiffs' strenuous effort to persuade this Court to the contrary, what is before the Court is a very simple contract dispute. Plaintiffs argue that Defendant Internet Corporation for Assigned Names and Numbers ("ICANN") has breached the Registrar Accreditation Agreements ("RAA") that ICANN has entered into with each of the plaintiffs by proposing to amend ICANN's contract with a third party. A plain reading of the RAA, however, shows nothing that would prevent or restrict ICANN from entering into new or amended agreements with any other party. What plaintiffs really seek here is a significant reformation of the RAA to provide them with a veto over any action by ICANN that, in their view, adversely affects their commercial interests. The RAA provides absolutely no basis for any such relief and it would be completely contrary to ICANN's public interest mission. The RAA is a fairly simple document, setting forth in separate sections the obligations of both parties. Plaintiffs' primary argument is that ICANN cannot take any action that affects plaintiffs' commercial interests without following the "Consensus Policy" requirements of the RAA. But the "Consensus Policy" provisions of the RAA clearly apply only to those instances in which ICANN seeks to impose new obligations on registrars. What plaintiffs are arguing is that the RAA "Consensus Policy" provisions should be interpreted to impose obligations on ICANN generally, whenever ICANN takes actions that might have some impact on registrars, or even more broadly on domain name policy generally. This is not what the RAA says. The only place in which any "Consensus Policy" obligations are set forth in the RAA is subsection 4.1, which speaks to the circumstances in which ICANN may impose additional obligations on registrars (such as the three plaintiffs). It is revealing that plaintiffs never discuss that subsection of the RAA, seeking instead to focus on subsection 4.2, which does nothing more than delineate an illustrative list of topics that could be the subject of new registrar obligations, and subsection 4.3, which sets forth the procedures that must be followed in those circumstances where a Consensus Policy is to be established. Plaintiffs argue that these subsidiary subsections 4.2 and 4.3 somehow "obligate Defendant to ensure that any new policies or specifications identified in the Agreements and imposed on Registrars are approved by a consensus of Internet stakeholders" (Motion at 5:13-15), notwithstanding the plain language to the contrary in subsection 4.1. While that is not accurate -- those subsections merely define the obligation expressed in subsection 4.1 -- plaintiffs have, perhaps inadvertently, with this articulation exposed the basic flaw in their argument: the ICANN action challenged here does not impose any obligation on registrars. It is important to understand what plaintiffs are really complaining about. Some registrars, including plaintiffs, have created products that they sell to consumers. Those products are not the result of any ICANN process or consensus; they represent market decisions made by these economic actors without any ICANN input, involvement or oversight. The development of these products was not governed by the RAA; they required no approval by ICANN or anyone else; they were, in fact, created wholly outside the ICANN process. This fact alone illustrates why plaintiffs' argument overreaches: plaintiffs' own actions, which clearly had "an impact on registrars" (and, indeed, involved a new product offered by some registrars), were certainly not the result of any consensus process. Now, VeriSign has sought to create a product that will compete with those offered by plaintiffs. Plaintiffs obviously fear that this new product may be preferred by consumers once that option is available to them. But VeriSign, unlike plaintiffs, was not able to simply create and offer this product without any ICANN involvement, because VeriSign is a registry, not a registrar, and in that role it operates not under the RAA but a different agreement. VeriSign's registry agreement with ICANN requires ICANN approval for any new registry product that will be offered for sale, a provision that is intended to prevent evasion of the price ceiling set in the registry agreement so as to prevent monopoly pricing by the registry, for which some customers have no practical substitute1. Plaintiffs are attempting here to take commercial advantage of this wholly independent agreement and its requirement that ICANN approve this new registry product so as to prevent the introduction of new competition -- competition they fear will reduce the revenues that they have been able to generate from their products that did not require ICANN approval. Plaintiffs seek to wrap this rather unattractive position -- that this Court should prevent VeriSign from offering consumers an additional and potentially more attractive option to the products plaintiffs offer -- in a bunch of public policy/competition/contract mumbo-jumbo, because it is difficult to explain otherwise why plaintiffs should be protected from this new competition and why consumers should be deprived of this new (and less costly and more manageable) option. In doing so, plaintiffs ignore the plain language of the RAA and seek to invent obligations that do not exist. To be very clear on what is happening here: this is an attempt to use this Court to exclude new competition on the basis of imaginary contractual obligations that are flatly inconsistent with the actual language of the RAA. This effort should be summarily rejected. The fact, as opposed to the fiction, is that ICANN, in its ongoing effort to promote consensus and broad participation wherever possible, voluntarily sought community input on the pros and cons of the proposal by VeriSign to modify its registry agreement. This was not required by any agreement, including the RAA, but was thought to be a desirable approach to what was anticipated would be a controversial subject. Plaintiffs argue that by seeking public input ICANN invoked the "Consensus Policy" requirement of the RAA, but in fact it was a voluntary initiative by ICANN, as was obvious from all the contemporaneous discussion. There were a wide range of views expressed by various parts of the ICANN community, including registrars such as the plaintiffs,2 and in the end, ICANN’s Board took those views into consideration and came to a decision. Plaintiffs now propose to redefine the RAA so that ICANN is required to develop a "Consensus Policy" anytime there are "new principles established for the allocation of domain names." (Motion at 10:19-20). They argue, in a way that is flatly inconsistent with contemporaneous statements of ICANN during the process that it initiated,3 that ICANN agreed with this interpretation. But because the record is clear that ICANN did not agree with this interpretation, plaintiffs' argument boils down to a contention that, because ICANN voluntarily sought public input, without a contractual obligation to do so, this creates a basis for reforming the RAA to require such an obligation. This is a remarkable and unsupportable notion of contract interpretation, and if adopted by the Court would have dramatic negative implications on the ability of ICANN to carry out its mission. The structure and purpose of the "Consensus Policy" term in the RAA was not intended to protect ICANN-accredited registrars from competition but only from additional obligations imposed by ICANN. An amendment to VeriSign's registry agreement allowing it to offer the WLS would impose no obligations on plaintiffs or any other registrars but would simply present them with an additional competitive challenge. In short, plaintiffs ask this Court to rewrite the RAA and to protect them from having to respond to this new competition. This is not consistent with the language of the RAA, nor with the public interest. The motion for preliminary injunction should be denied. ICANN is a not-for-profit corporation that was organized under California law in 1998. Pursuant to a series of agreements with the United States Department of Commerce ("DOC"), ICANN is responsible for administering certain aspects of the Internet's domain name system. (Halloran Decl., ¶ 13.) Among its various activities, ICANN accredits companies known as "registrars" that make Internet "domain names," such as "cnn.com" or "pbs.org," available to consumers. (Id., ¶ 15.) Each registrar enters into an RAA with ICANN that permits it to sell the right to use domain names in a particular domain (such as ".com," ".net," ".biz" and so forth). (Id.) Registrars, in turn, contract with consumers and businesses that wish to register Internet domain names. (Id.) Typically, those contracts last one or two years, and at the end of that term, the consumer is given the option to renew the contract so as to retain that particular domain name. Separately, ICANN also contracts with Internet "registries." Each "top level domain name" -- such as .com, .net, .biz and so forth -- is operated by a single registry that functions similar to a phone book, making sure that each name registered in that domain is unique. Registries offer a variety of services that, for example, permit consumers to check to see if a particular domain name has already been registered and when the name is set to expire. (Id., ¶ 8.) A registry that wishes to offer new services for a fee must obtain ICANN's permission via an amendment to its registry agreement. VeriSign's Wait Listing Service Beginning in late 2001, VeriSign proposed to offer the WLS, which would operate by permitting ICANN-accredited registrars, acting on behalf of customers, to place reservations for currently registered domain names in the .com and .net top-level domains. (Id. ¶ 16.) Only one reservation would be accepted for each registered domain name. Each reservation would be for a one-year period. Registrations for names would be accepted on a first-come/first-served basis, with the opportunity for renewal. (Id.) VeriSign would charge the registrar a fee, which would be no higher than $24.00 for a one-year reservation and would be the same for all registrars. (Id.) The registrar's fee to the customer would be established by the registrar, not by VeriSign. In the event that a registered domain name is not renewed and is thus to be deleted from the registry, VeriSign would check to determine whether a reservation for the name is in effect, and if so would automatically register the name to the customer. If there is no reservation, VeriSign would simply delete the name from the registry, so that the name is returned to the pool of names equally available for registration through all registrars, also on a first-come/first-served basis. (Id.) VeriSign proposed to implement the WLS for a twelve-month trial. At the end of the trial, ICANN and VeriSign would evaluate whether the WLS should be continued. (Id., ¶ 17.) In order to provide this service and charge a fee, VeriSign is required by its registry agreement with ICANN to obtain ICANN’s approval to modify that agreement, which requires a modification to the registry agreement for the offering of any new registry service for which a fee will be charged. (Id., ¶ 18.) In addition, the Memorandum of Understanding between ICANN and the U.S. Department of Commerce, which requires ICANN to submit for DOC approval any material change to the .com and .net registry agreements between ICANN and VeriSign. (Id. at ¶ 18.) Presently, several registrars, including the three plaintiffs, provide their own form of "wait list services." (Id., ¶ 24.) (As reflected in plaintiffs' declarations, some of these services commenced after VeriSign proposed the WLS.) Under these services, a consumer who wants to register a particular name that is already registered by someone else may sign up, and in many cases pay in advance, for the opportunity to try to obtain that name when, and if, it is deleted at some point in the future. (Id.) Upon receiving such a request from a consumer, the registrar would then watch for the particular name to be deleted and, if and when that happened, immediately attempt to register it. (Id.) However, none of these services can provide a customer with any certainty that a particular domain name will be registered to it (if and when the name is deleted from the registry) because there may be numerous registrars that have sold to different customers the chance to obtain the right to use the very same deleted name, and only one of those registrars will be successful in registering that name for its customer. In contrast to the various “wait list services” offered by Plaintiffs and other registrars, the WLS would permit a consumer to sign up with any participating registrar to be placed on the waiting list for a particular name if there was not already a WLS registration for that name, and such a registration would guarantee that consumer the right to register that particular name should it subsequently be deleted. (Id., ¶ 25.) This description illustrates why plaintiffs fear the introduction of this new product: the WLS will offer the certainty that none of the plaintiffs can offer today. The WLS will not affect current domain name registrations. (Id. at ¶ 19.) An existing registrant will continue to be the registrant of its domain name for so long as it continues to renew the domain name in a timely fashion and to meet the requirements of its chosen registrar. (Id.) A WLS subscription matures into an actual domain name registration only when a domain name is finally deleted by the registry. (Id.) Likewise, the WLS will not change the manner in which a deleted domain name is processed when there is no WLS subscription for the domain name. (Id. at ¶ 20.) If the domain name has not been redeemed or renewed, the deletion of the domain name is effectuated by the registry and the domain name ceases to exist in the registry database until and if registered again at some time in the future. (Id.) In the absence of a WLS subscription, the deleted domain name becomes available for creation and registration through any ICANN-accredited registrar on a first-come/first-served basis, just as it was before WLS. (Id.) All ICANN-accredited registrars will have an equal opportunity, at the same wholesale price, to participate in the WLS. (Id. at 22.) Registrars also have the option of not participating, since the WLS is an entirely optional service. (Id.) If they elect not to participate in the WLS, registrars may still register, delete, transfer or otherwise make registered domain names available in the secondary market (e.g., auctions, person-to-person transactions, etc.). The WLS services at the registrar level might be differentiated through customer service, marketing, registrar value-added services, or other creative actions, and through competitive retail pricing. (Id.) As this explanation demonstrates, plaintiffs’ argument (Motion at 12:20-25) that the creation of the WLS will be anticompetitive is flatly wrong: the only competitive effect that it will have is the introduction of a new product to the marketplace. Registrars can continue to vigorously compete with each other in the sale of domain name registrations, in the sale of WLS subscriptions, and (if they choose) in the sale of other deleted domain name services such as those currently offered by plaintiffs. Plaintiffs obviously fear that their existing deleted name services will not fare well in this new competition with WLS subscriptions, and that they will generate smaller profits in this more competitive environment. But this is hardly a concern that should trouble this Court, and certainly is not a basis for invoking or revising the provisions of the RAA. Under subsection 4.1 of the RAA (Halloran Decl. at ¶ 15, Ex. 2), all ICANN-accredited registrars agree to comply with new or revised "policies" that apply to all registrars and are developed during the term of the agreement, provided they are established according to a consensus process described in subsection 4.3 and in the circumstances prescribed in subsection 4.1.2 (a.k.a. "Consensus Policies"). (Id. at ¶ 27.) Registrars thus contractually agree that, through this process, they may be compelled to take action in compliance with a duly-established Consensus Policy without an amendment to their RAAs. (Id.) In essence, subsection 4.1 permits ICANN to impose new policies on all of its registrars, if adopted as set forth in the RAA, without requiring each of its 170 or so registrars to sign new agreements. This is the only place in which “consensus policy” development is even discussed in the RAA, and its only impact is to require ICANN to follow certain procedures when it seeks to impose additional obligations on registrars generally. This provision insures that registrars are not obligated to comply with any ICANN action or policy that was not developed pursuant to the RAA's specified procedures or is not a result of a negotiated amendment to the RAA. Decision to Proceed with the WLS Contrary to the inference that plaintiffs seek to leave with the Court, ICANN has been clear from the beginning of the WLS process that it did not consider the issue of a possible amendment of the VeriSign registry contract to be subject to any consensus policy requirement. (See footnote 3, supra.) It did, however, believe that it was appropriate to seek public input as an aid to its decisional process, and therefore it invited input from various ICANN constituencies, including the Registrar Constituency of which plaintiffs are a part. On March 10, 2002, ICANN’s “Registrar Constituency” issued a position paper opposing the WLS and urging ICANN to withhold permission for its implementation. The registrars supporting the paper, to nobody's surprise, were those who already had their own version of a wait-list service in place, including the plaintiffs in this action. Several registrars that did not offer such wait-listing services dissented from the paper. (Halloran Decl. at ¶ 40.) On August 23, 2002, the ICANN Board determined that the WLS “promotes consumer choice” and that the “option of subscribing to a guaranteed ‘wait list’ service is a beneficial option for consumers.” For these reasons, the Board approved a resolution (Resolution 02.100) authorizing (with certain conditions, imposed largely to address the stated concerns of registrars) ICANN's President and General Counsel to negotiate appropriate revisions to VeriSign’s registry agreements to allow for the offering of the WLS. (See id. at ¶ 41, Ex. 15.) On September 9, 2002, after the Board had approved the WLS, counsel for Dotster, Inc. (“Dotster”) submitted a letter to ICANN and then filed a formal request for reconsideration of the Board’s decision regarding the WLS. As is its usual practice, ICANN posted a copy of Dotster's letter on its website. (Id. at 42, Ex. 16) On May 20, 2003, ICANN’s Reconsideration Committee determined that Dotster’s request lacked merit and recommended that the Board take no action on it. (Id.) On July 16, 2003, plaintiffs initiated this litigation and filed a request for a temporary restraining order, which the Court denied via its order of July 18, 2003.4 Plaintiffs took no further action in the case until they filed their motion for preliminary injunction. It is a "'fundamental principle that an injunction is an equitable remedy that does not issue as of course.'" Miller For And On Behalf Of N.L.R.B. v. Cal. Pac. Med. Ctr., 991 F.2d 536, 539 (9th Cir. 1993) (quoting Amoco Prod. Co. v. Village of Gambell, 480 U.S. 531, 542 (1987)). To obtain a preliminary injunction, the moving party must establish: 1) a strong likelihood of success on the merits; 2) that the balance of irreparable harm favors the moving party; and 3) that the public interest favors the issuance of an injunction. Regents of the Univ. of Cal. v. Am. Broad. Co., Inc., 747 F.2d 511, 515 (9th Cir. 1984). Where the public interest may be affected, the Court must examine each of these three elements in turn. Sammartano v. First Jud. Dist. Ct., 303 F.3d 959, 965 (9th Cir. 2002); see also Caribbean Marine Serv. Co., 844 F.2d at 674 (9th Cir. 1988) (this "traditional test" is typically used in cases involving the public interest). I. ICANN HAS NOT BREACHED ITS REGISTRAR ACCREDITATION AGREEMENTS WITH PLAINTIFFS. A preliminary injunction should not issue where plaintiffs are unlikely to prevail on the merits of the alleged claim. See Associated Gen. Contractors, Inc. v. Coalition for Econ. Equity, 950 F.2d 1401, 1405 (9th Cir. 1991) (affirming denial of preliminary injunction because plaintiff had little chance of succeeding on the merits); Goldie's Bookstore, Inc. v. Super. Ct., 739 F.2d 466, 470-72 (9th Cir. 1984) (reversing grant of preliminary injunction given weakness of plaintiffs' section 1983 claim and interests implicated). Plaintiffs cannot demonstrate any probability of success on the merits, let alone a "strong" probability, because the record clearly shows that ICANN has not breached the RAA. Section 4 of the RAA does not require ICANN to initiate a consensus-driven process before amending VeriSign's Registry Agreement to allow for the WLS. Plaintiffs' interpretation of section 4 of the RAA is contrary to the RAA's plain meaning and contrary to ICANN's mission and statements throughout the discussion of the WLS. Plaintiffs' interpretation of subsection 2.3 of the RAA is equally flawed.
Plaintiffs assert that ICANN breached its RAA with plaintiffs by failing "to obtain a consensus among Internet stakeholders . . . before the establishment of any policy affecting the allocation of registered domain names, in this case the implementation of the [WLS]." (Motion at 1:6-11.) However, the RAA contains no such requirement. Instead, the RAA requires the consensus development process only when ICANN seeks to impose new obligations on the registrars that are the signatories of the RAA. In all other circumstances, and a fortiori here, where ICANN's negotiations with VeriSign over a possible amendment to a wholly different agreement will impose no obligations on plaintiffs or any other registrars, the RAA's consensus policy provisions create no restrictions of any kind on ICANN's conduct or decisions. Specifically, subsection 4.1 of the RAA, which is the only subsection that sets forth any Consensus Policy requirement, applies only if and when ICANN seeks to compel registrar action without amending the RAA:
Nothing in the above-quoted language creates any obligation upon ICANN to act only by consensus where no registrar action is compelled. Indeed, where no registrar action is compelled, subsection 4.1 is irrelevant.5 Because subsection 4.1 applies only where ICANN is seeking to compel registrars to comply with some policy without amending the RAA, subsection 4.1 does not govern the process or extent to which ICANN chooses to involve the Internet community, including accredited registrars such as plaintiffs, in any of its decision-making activities that do not seek to compel registrar action. The WLS plainly is one of those activities. The WLS will be effected by an amendment to ICANN's Registry Agreement with third party VeriSign, not an amendment to the RAA. Registrars may choose to offer the WLS service once it is available, but no registrar will be obligated to do so -- just as no registrar is obligated to participate in any current form of "wait-listing" service such as those presently offered by the plaintiffs (and, indeed, most registrars do not offer any form of "wait-listing" service). Because the WLS does not impose any obligations on any registrars (much less purport to amend the RAA), the Consensus Policy provisions of section 4 of the RAA are simply not applicable, and thus the fact that they were not followed cannot, as a matter of law, constitute a breach of the RAA by ICANN. Although subsection 4.1 is the only subsection of the RAA that sets forth when ICANN must adopt a Consensus Policy (i.e., when it seeks to compel registrar action without amending its contract with each registrar), plaintiffs argue that subsection 4.2.4 somehow imposes an independent obligation on ICANN to develop a Consensus Policy anytime there are "new principles" established for the allocation of domain names. (Motion at 10:19-20.) Plaintiffs' argument ignores the plain language of subsection 4.2, which merely enumerates some topics for which ICANN may compel registrar action through new or revised specifications or policies contained in a Consensus Policy:
The plain language of subsection 4.2 does not create an independent obligation and require, as plaintiffs urge, that a Consensus Policy process must be implemented any time domain name allocation is affected (and particularly when no registrar action is compelled). See Cal. Civ. Code § 1641 ("the whole of the contract is to be taken together, so as to give effect to every part, if reasonably practicable, each clause helping to interpret the other.") (Emphasis added.) See also Fidelity & Deposit Co. v. Curtis Day and Co., 1993 WL 128073, 1 (N.D. Cal. 1993) (defendants' argument misconstrued contract term; language in preceding sentences made term entirely inapplicable until condition precedent occurred). Because no registrar compulsion was or is contemplated by any amendment to its registry agreement with VeriSign, ICANN had no obligation to follow any particular procedure. Although ICANN sought community input in a variety of ways -- because it believed this was useful under the circumstances -- ICANN was not acting "pursuant" to any RAA provision, as plaintiffs contend. (Motion at 6:8 11.) Plaintiffs argue that ICANN's recognition that some portions of the Internet community might oppose the WLS, and the fact that it sought input from the Internet community, are evidence that ICANN was contractually required to invoke the "Consensus Policy" procedure set forth in subsection 4.3 of the RAA. (Motion at 5:16-6:15.) But this is flatly inconsistent with the plain language of the RAA, and with ICANN's position throughout the WLS discussions.6 ICANN's website is replete with postings, including analyses by its General Counsel, Committee reports, and Board resolutions that show that ICANN's position has always been that the amendment of its contract with VeriSign to allow for the WLS is not a consensus policy issue under the RAA. (See Halloran Decl., 35-36, Exs. 3 10.) Moreover, the Names Council, which includes plaintiffs and the rest of the Registrar Constituency, and whose review and recommendations plaintiffs rely on so heavily in their papers, obviously understood this point, since its reports make no mention of the Consensus Policy provisions of the RAA.
Plaintiffs’ half-hearted claim that ICANN separately breached the RAA by acting inconsistently with subsection 2.3 (Motion at 12:1-12) also is without merit. Subsection 2.3 imposes a general obligation on ICANN to ensure that registrars have adequate appeal procedures through reconsideration and independent review policies with respect to a registrar's rights, obligations and role. Plaintiffs’ argument assumes that registrars have some inherent right or obligation to operate a wait-listing service, which clearly they do not, as reflected by the fact that plaintiffs never submitted their services for ICANN’s approval, and the RAA does not address “wait-listing" services. For this same reason, plaintiffs’ argument that they were entitled to an independent review board to evaluate their “protest” to the WLS (Motion at 12:26-28) again assumes the correctness of plaintiffs’ argument that the WLS implicates rights and obligations under the RAA in the first instance. As explained above, since the registrars do not have the right to require ICANN to conduct a “Consensus Policy” process every time ICANN wants to amend a registry agreement, the registrars obviously were not entitled to the “independent review board” process with respect to the WLS. (See Halloran Decl., ¶ 35, Ex. 9.) II. THE HARDSHIP PLAINTIFFS CLAIM FROM POTENTIAL WLS COMPETITION IS NOT IRREPARABLE AND IS OUTWEIGHED BY THE POTENTIAL HARM TO ICANN.
Plaintiffs argue that a preliminary injunction is necessary because "[i]mplementation of the WLS policy would have an immediate, discernible but unquantifiable adverse impact on plaintiffs' goodwill, reputation, earnings, and market share as well as their ability to maintain their existing customers." (Motion at 13:18-20.) The Ninth Circuit has held, however, that such "[s]ubjective apprehensions and unsupported predictions of revenue loss are not sufficient to satisfy a plaintiff's burden of demonstrating an immediate threat of irreparable harm." Caribbean Marine Serv. Co. v. Baldrige, 844 F.2d 668, 675-76 (9th Cir. 1988); see also L.A. Coliseum, 634 F. 2d at 1201. While it may or may not be true that consumers will prefer the WLS to plaintiffs' services, the fact that plaintiffs may make less money after the introduction of this new competition does not mean they have a claim against ICANN, much less a claim that supports injunctive relief. See L.A. Coliseum, 634 F.2d at 1202 (reversing district court's grant of preliminary injunction because loss of revenue rarely constitutes irreparable injury). "Mere injuries, however substantial, in terms of money, time and energy necessarily expended . . . are not enough. The possibility that adequate compensatory or other corrective relief will be available at a later date, in the ordinary course of litigation, weighs heavily against a claim of irreparable harm." Id. (citing Sampson v. Murray, 415 U.S. 61, 90 (1974); see also Stanley v. Univ. of So. Cal., 13 F.3d 1313, 1320 (9th Cir. 1994) (quoting Beacon Theaters, Inc. v. Westover, 359 U.S. 500, 506-07 (1959) (the inadequacy of legal remedies is a prerequisite to the issuance of an injunction). Plaintiffs' argument that they will suffer harm to the goodwill and reputations of their "well-established" wait-listing businesses is unsupported by credible evidence (as opposed to plaintiffs' conclusory statements), and in any event easily outweighed by the harm to the public interest described in Part III below. (Motion at 14:6-24.) See Oakland Tribune, Inc. v. Chronicle Publ’g Co., 762 F.2d 1374, 1377 (9th Cir. 1985) (holding that conclusory statements by interested parties that plaintiffs would suffer the loss of reputation, competitiveness, and goodwill did not support a finding of irreparable loss). Plaintiff Go Daddy's wait-list service is hardly "well-established"; it was not launched until April 2003, only a few months ago.7 (See Parsons Decl., ¶ 5.) And although the service apparently now accounts for $100,000 in revenues per month (id. at ¶ 5), it remains difficult to comprehend Go Daddy's argument that, with revenues of $10 million in 2001 and over 300 employees, it will be unable to handle customer inquiries regarding the WLS and will suffer harm to its reputation. (Id. at ¶ 3; Halloran Decl. at ¶ 48, Ex. 17.) Plaintiffs have likewise made no factual showing to support their contentions that introduction of the WLS would somehow cause their wait-listing services to be "unavailable," as opposed to unattractive, to interested customers. The alleged harm to plaintiffs' goodwill and reputation is unsupported and a mere recasting of their claimed potential monetary injury in different terms; it does not support injunctive relief. See L.A. Coliseum, 634 F. 2d at 1202 (plaintiffs’ claimed loss of substantial goodwill and market value were “but monetary injuries which could be remedied by a damage award.”). The cases plaintiffs cite are easily distinguished. In Gilder v. PGA Tour, Inc., 936 F.2d 417, 425 (9th Cir. 1991), the Ninth Circuit upheld an injunction to prevent the PGA from banning an existing product. Unlike the situation in Gilder, an injunction here would prevent the introduction of a new product and competitor simply to protect the products offered by plaintiffs and others like them. Moreover, the product the PGA would have banned was used by several golf professionals, whose individual livelihoods depended on the continued availability of the product. In contrast, plaintiffs will continue to be able to offer their wait-list services after the introduction of the WLS, and plaintiffs do not show that their companies (as opposed to their wait-list services) would not survive the introduction of the WLS.8 Because any conceivable harms are fully calculable and compensable by money damages, the motion for a preliminary injunction should be denied.
Plaintiffs seek an injunction that would interpret the RAA to require ICANN to "obtain a consensus among Internet stakeholders . . . before the establishment of any policy affecting the allocation of registered domain names, in this case the implementation of the [WLS]." (Motion at 1:6-11.) This interpretation of the agreement has no connection to reality and would be contrary to ICANN's public interest mission. Under plaintiffs' interpretation, registrar consensus would be required before ICANN could enter into any agreement with a third party that might affect domain name allocation in any respect. This would literally turn upside-down the entire relationship between ICANN and its accredited registrars, giving registrars (whose very existence is predicated upon an ICANN decision to permit them to offer domain name registrations to the public under certain clearly-defined conditions) a veto over any ICANN action that they believed was inconsistent with their private economic interests (as opposed to the public interest that ICANN is established to advance). Moreover, all of ICANN's existing registry agreements -- for example, with respect to the new top level domains in .biz, .name, and .info -- would immediately be legally suspect; none was the product of consensus policymaking, and no one has ever suggested that they had to be. This would be a dramatic reformation of the RAA and an equally dramatic blow to the effectiveness of ICANN. The contrast between the harm to plaintiffs, which is entirely economic and compensable in damages if and when they were ever to prevail on the merits, and the potential harm to ICANN of the entry of an injunction on the basis of the arguments presented here, is stark. Since similar consensus policy provisions also appear in many other ICANN agreements -- all also limited to those situations where ICANN seeks to compel action or conduct by the signatory to the agreement without having to negotiate an amendment to that agreement -- presumably all those other actors would also have a similar veto over ICANN actions that might adversely affect their economic interests. (See Halloran Decl., ¶ 49.) The result, quite logically, would be institutional paralysis, and the potential end of ICANN as an effective organization for coordinating certain technical and related policy issues for the Domain Name System ("DNS"). (Id.) ICANN is a body that seeks to develop consensus wherever possible. (Id.) Indeed, that is its principal reason for existence. (Id.) ICANN maintains open and transparent processes; it regularly posts on the Internet its minutes, transcripts of its meetings, and other important information. Indeed, its website contains virtually a day-to-day description of ICANN’s activities. 9 Because the Internet is a global resource, it is extremely difficult as a practical matter, and highly undesirable as a conceptual matter, for the nations of the world to seek individually to set policy for important technical elements of the Internet such as the DNS. (Id.) Thus, the realistic options for appropriate coordination of technical aspects of the Internet are a multinational treaty organization or a global private sector organization like ICANN, where governments and private actors come together to attempt where possible to create consensus policies that will allow the Internet to continue to grow as an engine of global commerce and communication. (Id.) For now, the world has chosen the private sector route, on the theory that if that can succeed, it will be more efficient and effective than a treaty organization. (Id.) If plaintiffs' view was to prevail, and ICANN were prevented from taking any significant actions unless it was able to achieve a consensus from all of the many constituencies that participate in ICANN, it seems inevitable that ICANN would fail. It would be particularly ironic if ICANN's failure would be precipitated by a judicial rewriting of its contractual relationships -- contracts that ICANN did in fact draft in the first instance and compel each registrar to sign as a condition of receiving accreditation -- generated by a fear of new competition that exists within a small subset of the ICANN community and an even smaller subset of the overall Internet community. Nothing in the RAA requires or justifies this result. In short, plaintiffs are not likely to succeed on the merits of their action. For this reason alone, their motion should be denied. III. THE PUBLIC INTEREST, WHICH ICANN SEEKS TO ADVANCE, WOULD NOT BE SERVED BY THE ISSUANCE OF AN INJUNCTION. Because the injunction plaintiffs seek would affect the public, this Court must examine whether the public interest would be advanced or impaired by the issuance of the requested injunction. See Caribbean Marine Serv. Co., 844 F.2d at 674 (reversing injunction based in part on district court's failure to identify and weigh the public interests at stake); L.A. Coliseum, 634 F.2d at 1200. There is no doubt that the public interest will be impaired if a preliminary injunction issues in this case. Plaintiffs argue that the public interest falls in their favor because "the WLS preempts the competitive process that currently exists and allows only VeriSign to control when, and now if, domain names expire." 10 (Motion at 12:19-25.) Plaintiffs argue that WLS would be a monopoly and, thus, anticompetitive. This is truly disingenuous. What is really at stake here is the continued ability of multiple economic actors to individually sell the same product over and over again, as opposed to the circumstance where a higher quality product (because it is guaranteed) is sold only once. These plaintiffs will have the opportunity, if they choose, to sell WLS subscriptions. But because VeriSign, the registry, will accept only one subscription per name, it will be more difficult for plaintiffs to continue to sell multiple "reservations" for the same name. The antitrust laws are designed to "protect competition, not competitors." U.S. v. Syufy Enterprises, 903 F.2d 659, 668 (9th Cir. 1989). The only members of the "public" who might be injured are plaintiffs. Plaintiffs' waitlisting business involves selling an opportunity for a potential registrant to get in line, along with however many others to whom other registrars have sold the same opportunity to get in line for the very same domain name. It is no doubt in plaintiffs' economic interest to try to prevent this from changing, as evidenced by the effort they are putting into this litigation, but this is hardly evidence of harm to the public interest, in this case properly represented by ultimate consumers of the products in question. The introduction of the WLS will not require or compel the elimination of the current services offered by plaintiffs, and they are free to continue to seek to sell those non-guaranteed chances to anyone who will buy them. What plaintiffs fear is that consumers will recognize that those products have become relatively less attractive to the new WLS product, and that they will instead prefer a WLS subscription. But the fact that consumers may find the WLS -- with its guarantee of access if and when the desired name becomes available -- to be more attractive than current services is obviously a pro-competitive, not an anti-competitive, result. In fact, consumers will be better off, not worse off, with the introduction of a new competitive alternative, which offers some features, such as the guarantee of access to a newly available name, that are simply not available today. Plaintiffs' position is strikingly similar to that of the defendants in Regents of the Univ. of Cal. v. Am. Broad. Cos., 747 F.2d 511, 512 (9th Cir. 1984). In that case, the defendants argued unsuccessfully that a contract's exclusivity provision should be enforced to prevent competition in the broadcast of college football games. In weighing the public interest, the Ninth Circuit, citing the Supreme Court's decision in NCAA v. Regents, 468 U.S. 85, 104 (1984), found that "the public interest is served by preserving the competitive influence of consumer preference in the college broadcast market." The defendants should not be permitted, the Court held, to "unilaterally determine that the public would not have the choice of viewing an admittedly popular college football game." See Regents, 747 F.2d at 521. As in Regents, the public interest here clearly weighs against, not for, the injunctive relief requested. The introduction of the WLS will not force any of the plaintiffs out of business; it will not eliminate the existing products from the competitive marketplace; and it will give consumers a new alternative that will dramatically simplify the system of acquiring deleted domain names, offering an option that is in the aggregate less expensive, more understandable and more certain. Because it is clear that the public interest would not be served by the issuance of an injunction, plaintiffs' motion should be denied. IV. IF A PRELIMINARY INJUNCTION ISSUES, PLAINTIFFS SHOULD BE REQUIRED TO POST A SIGNIFICANT SECURITY BOND. "No restraining order or preliminary injunction shall issue except upon the giving of security by the applicant, in such sum as the court deems proper." Fed. R. Civ. P. 65(c). While plaintiffs contend no bond is needed (and cite to a case in which no bond was requested), a bond is both necessary and appropriate here because the issuance of a preliminary injunction would cause a very significant disruption in ICANN's operations and contractual relations, as described above and in the accompanying Halloran Declaration. Indeed, the consequences may be catastrophic. As a result, a very significant bond would be required to protect the interests of the global Internet community, and to compensate for the potentially significant harm that might result from even a temporary imposition of injunctive relief in this matter. Under these circumstances, a bond in the amount of $25 million or more would not be inappropriate. Plaintiffs' motion for injunctive relief should be denied because plaintiffs' interpretation of the parties' contract is wrong on its face, and there is no threat of irreparable harm. Instead, the issuance of an injunction would harm consumers, prevent new competition, and potentially strike a fatal blow to the ICANN experiment in private-sector coordination of the Internet, a global public resource.
Notes: 1. In addition to ICANN approval, any material change in the VeriSign registry agreement also requires approval by the U.S. Department of Commerce. (See declaration of Daniel E. Halloran ("Halloran Decl.”), ¶ 18.) 2. It is perhaps instructive that the comments from registrars tracked fairly directly with whether the commenting registrar offered a product that would be threatened by the new product (the Wait Listing Service, or "WLS") proposed by VeriSign. Those registrars who offered wait-listing types of services opposed WLS; other registrars that did not offer wait-listing services supported WLS. (See Stahura Ex. 3, pp. 17-18.) 3. Numerous documents posted on ICANN's website demonstrate without a doubt that neither ICANN nor the Names Council, which includes the Registrar Constituency (of which plaintiffs are a part), has ever taken the position that an amendment to VeriSign's Registry Agreement to allow the WLS requires a Consensus Policy under the terms of the RAA. (Halloran Decl., ¶ 35.) For example, ICANN's General Counsel recommended, in his first analysis of VeriSign's request for an amendment to its Registry Agreement, "that the Board establish the following procedure for obtaining public comment to illuminate its consideration of [the WLS]" -- because no such procedure existed at the time. (Id.. Ex. 3 (emphasis added).) Subsequently, the Transfers Task Force issued a Final Report, which was adopted by the Names Council (which includes the Registrar Constituency), that assumed that the Consensus Policy procedure of the RAA did not apply. (Id., Ex. 4.) The General Counsel's second analysis of the WLS, which detailed the various steps ICANN was voluntarily taking to develop consensus on the WLS (if a consensus was possible), makes no reference to an RAA requirement. (Id., Ex. 5.) The Board's final resolution approving the WLS, which sets forth a thorough discussion of why the Board reached its decision on the WLS makes no mention of a contractually required consensus policy process. (Id., Ex. 6.) ICANN's Reconsideration Committee later, in response to a request for reconsideration by one of the plaintiffs that for the first time raised this issue, explicitly stated that the WLS was not a consensus policy issue, a decision that ICANN's Board subsequently adopted. (Id., Exs. 7, 10.) Furthermore, plaintiffs did not adopt this position when, in March 2001, ICANN sought public comment on a different set of modifications to VeriSign’s registry agreements. See id., 37, Exs. 11-13. 4. ICANN noted in its opposition to the motion for temporary restraining order that it had not yet reached a definitive agreement with VeriSign and that the Department of Commerce had not, therefore, approved the amendment to VeriSign's registry agreement. This status remains true today, as Mr. Halloran explains in his declaration, although ICANN remains hopeful that these matters will be resolved by October 27, 2003, which is the date by which VeriSign hopes to begin the WLS service. If not, VeriSign will not be able to offer that service at that time, consistent with its registry agreement with ICANN. Plaintiffs' suggestions that ICANN has somehow "manufactured" evidence for the purposes of this litigation are obviously false and unwarranted, and ICANN will not indulge them further. See Halloran Decl. at ¶ 44. 5. Likewise, where no registrar action is compelled, ICANN is not required, contrary to plaintiffs' argument, to divulge its contract negotiations with third parties or to submit such decisions to an Independent Review Panel. (Motion at 12:11-13:14.) Any such obligations only arise in the context of a Consensus Policy, which the present conduct at issue is demonstrably not. 6. Although ICANN's General Counsel recommended that ICANN seek community input on this issue, his advice to the Board did not even reference the terms of the RAA, because that agreement was not relevant to the evaluation of the WLS. (Stahura Decl. at ¶ 17, Ex. 4.) 7. Thus, this service was established 8 months after the Board decision to proceed with negotiations to amend VeriSign's Registry Agreement to allow for the WLS. 8. Regents of the Univ. of Cal. v. Am. Broad. Cos., 747 F.2d 511 (9th Cir. 1984) does not support plaintiffs' position either. In Regents, the Ninth Circuit granted a preliminary injunction to avoid restricting consumer choice in the television viewing of football games. Here, plaintiffs seek an injunction to prevent consumer access to a new service option. And whereas the plaintiff in Rent-A-Center, Inc. v. Canyon Television and Appliance Rental, Inc., 944 F.2d 597 (9th Cir. 1991), sought an injunction to enforce a covenant not to compete, plaintiffs here seek an injunction that would, in essence, create a covenant not to compete. 9. Plaintiffs criticize ICANN for not publishing its negotiations with VeriSign concerning the WLS, but it obviously would handicap ICANN in its contract negotiations if each aspect of those negotiations was conducted in public. Nonetheless, openness and transparency are part of ICANN's core values (see Halloran Decl., Ex. 14 (ICANN Bylaws)), and most of ICANN’s other activities are, in fact, published on ICANN’s website, as a perusal of www.icann.org shows. 10. As plaintiffs well know, the WLS has nothing to do with when or if domain names expire.
Comments concerning the layout, construction and functionality of this site should be sent to webmaster@icann.org.
(c) 2001 The Internet Corporation for Assigned Names and Numbers All rights reserved. |
