In addition to those arguments made during the hearing on Plaintiff OCLC, Inc's Motion for Temporary Restraining Order on June 21, 2022 and continued on June 24, 2022, Defendants Clarivate, Plc, Clarivate Analytics (US) LLC, ProQuest LLC, and Ex Libris (USA), Inc. (“Defendants”) offer the following additional arguments and support in opposition to OCLC's Motion. Defendants maintain that no TRO should be entered under the circumstances, but in the event the Court disagrees with that position, the revised TRO that has been proposed by OCLC remains vastly overbroad.
I. OCLC'S MOTION FOR TEMPORARY RESTRAINING ORDER SHOULD BE DENIED IN ITS ENTIRETY.
A temporary restraining order, like other forms of preliminary injunctive relief, is an “extraordinary remedy which should be granted only if the movant carries his or her burden of proving that the circumstances clearly demand it.” Overstreet v. Lexington–Fayette Urban Cty. Gov't, 305 F.3d 566, 573 (6th Cir.2002). The party seeking preliminary relief “bears the burden of justifying such relief, including showing irreparable harm and likelihood of success.” McNeilly v. Land, 684 F.3d 611, 615 (6th Cir. 2012). “Although no one factor is controlling, a finding that there is simply no likelihood of success on the merits is usually fatal.” Gonzales v. Nat'l Bd. of Med. Examiners, 225 F.3d 620, 625 (6th Cir. 2000); accord Jolivette v. Husted, 694 F.3d 760, 765 (6th Cir. 2012).
In this case, the circumstances do not warrant the issuance of any form of TRO, including the narrower proposed order that OCLC has submitted at the Court's urging. OCLC has asked this Court to grant it a TRO to either halt, or at least, drastically curtail Defendants' development of their innovative MetaDoor search tool that will—when it is launched no sooner than February 2023—facilitate peer-to-peer searching across the electronic “MARC” records used in libraries' own catalogs in order to permit those institutions to make their cataloguing processes more efficient. OCLC seeks this drastic step not on the basis of any intellectual property or trade secret claim but, rather, on the basis of a purported tortious interference with contract. OCLC asserts that its members' use of OCLC's WorldCat database—a repository for largely member-contributed metadata records—forever ties those institutions to OCLC's platform and prevents any other players in this market from offering improved technologies for those institutions' resource-sharing endeavors.
Critically, OCLC does not assert a copyright or any other claim of ownership to the “WorldCat records” over which it is now claiming a monopoly. OCLC in fact expressly disclaims any such right in its agreements and policies with its members. (See Section 2.C to the WorldCat Rights and Responsibilities for the OCLC Cooperative, attached as Exhibit B to the Rozek Declaration.) These same policies also expressly permit OCLC's members to share freely— with members and non-member institutions alike—any and all metadata catalog records corresponding to works in their own library holdings, whether those records are derived from or copied verbatim from records as they appear in WorldCat. (See id. at section 3.A.) Defendants' MetaDoor product is being designed as a tool to facilitate precisely the sort of resource sharing and crowd-sourcing activities that these institutions are already doing and are expressly authorized to do. It is not a database or repository of records. It does not interface with WorldCat or the records in WorldCat. It is simply a search tool that will facilitate searching by participating institutions across other such institutions' own existing catalog records. No records are provided to or retained by MetaDoor either during its development or, eventually, after it is launched. (See Compl. ¶ 95, n.8.) That Defendants are not themselves libraries or educational, cultural, or scholarly institutions within the ambit of section 3(A) of the WorldCat Rights and Responsibilities is irrelevant. MetaDoor simply facilitates transfers of records between those covered institutions, and such transfers are unquestionably permitted under OCLC's agreements. As a result, Defendants are not tortuously interfering with the libraries' contracts with OCLC simply by making those types of transfers easier and more efficient.
Without any intellectual property right or indeed any contractual right to do so, OCLC essentially asks this Court to grant it a monopoly over any metadata that it has touched over the last 55 years of its existence. It seeks to control not only the records compiled in its own database—i.e., exact copies of WorldCat records—but all the records in its members' libraries that have ever been shared with WorldCat, even where some or all of the metadata contained in those records (with the exception of OCLC's control number (“OCN”) designation) originated from the member library either initially or through further enhancements by the library. OCLC has no basis to claim ownership and control over that underlying metadata to prevent the peer-to-peer institutional sharing that the MetaDoor tool facilitates.
In addition to there being no legal basis to curtail this alleged interference with any OCLC contracts, there is no exigency that would demand the drastic action OCLC asks this Court to take. MetaDoor does not launch for at least another seven months. Until then, no peer-to-peer sharing via MetaDoor of any OCLC member's catalog records will occur, and the alleged harm that OCLC has identified (i.e., the diminished viability of WorldCat in the face of a free alternative) cannot and will not materialize. Moreover, there is no—nor will there ever be—any mass downloading or copying of records out of WorldCat in order to hand them over to Defendants or non-member institutions, as OCLC suggested in its filings. That is not how MetaDoor works. Independent of OCLC's inability to prove a likelihood of success on the merits, there is no basis to curtail Defendants' pre-launch development efforts or its work to obtain consents from potential customers—consents that expressly state that the customer has 100% control over what it shares via the MetaDoor tool based on its own understanding of its authority to do so.
While disrupting the development and marketing of MetaDoor will not prevent any irreparable or immediate harm to OCLC, it will cause significant harm to Defendants. By drastically curtailing which records Defendants' development partners can make available from their own catalogs for use in MetaDoor development, Defendants' development efforts will be set back substantially and may even require development to restart from scratch (if that is even possible) if current development partners are lost. One element of MetaDoor development consists of building out a Global Title Index of the holdings of each participating institution, which will facilitate searching across participating institutions for multiple versions of catalog records corresponding to the same title or work. That process will be drastically impacted if Defendants are prohibited from indexing any title for which the institution's catalog record was ever made available to WorldCat. Yet OCLC undoubtedly can claim no proprietary interest in the titles of the works held by a member library and, thus, can hardly justify an injunction that would prohibit Defendants from creating an index of those titles. MetaDoor development also includes efforts to improve the search functionality across participating partners' catalogs. There too, development will be greatly curtailed if the data set across which the search tool can operate represents only a fraction of the development partners' records. For these reasons, declaring any catalog record bearing an OCN number as off limits to the development of MetaDoor will significantly impair if not stop MetaDoor development for the duration of the TRO.
For these reasons, no TRO should issue, even in the narrowed form OCLC has now proposed.
II. NO TRO SHOULD BE ENTERED BECAUSE OCLC FAILED TO ADEQUATELY PLEAD PERSONAL JURISDICTION OVER DEFENDANTS.
As an additional, independent ground for denying OCLC's Motion, OCLC has failed to plead facts that would support a finding of personal jurisdiction over any of the Defendants. None of the Defendants is incorporated or headquartered in Ohio (see Compl. ¶¶ 14-17), and there is no basis alleged in the Complaint to conclude that the affiliations of any Defendant (let alone every Defendant) with Ohio are sufficiently “continuous and systematic” to support the exercise of general jurisdiction. Daimler AG v. Bauman, 571 U.S. 117, 127 (2014) (quoting Goodyear v. Dunlop Tires Ops., S.A. v. Brown, 564 U.S. 915, 919 (2011)).
Moreover, the Complaint fails to allege conduct that would suffice to establish specific jurisdiction under, at minimum, the due process clause of the Constitution. See, e.g., Burger King Corp. v. Rudzewicz, 471 U.S. 462, 471- 72 (1985); Marrik Dish Co., LLC v. Wilkinson CGR Cahaba Lakes, LLC, 835 F. Supp. 2d 449, 456 (N.D. Ohio 2011) (outlining the three-part test for specific jurisdiction). OCLC alleges that a single project manager employed by one of the Defendants who gave a presentation regarding MetaDoor was located in Cleveland (see Compl. ¶ 76) and otherwise relies on the conclusory allegation that “Defendants' representatives have reached out to OCLC customers in Ohio” (id. ¶ 120). Those allegations do not establish a sufficiently substantial connection to Ohio. See, e.g., NTCHWest Tenn, Inc. v. ZTE Corp., 761 F. App'x 485, 490 (6th Cir. 2019) (finding that “general allegation[s]” that only establish an “attenuated affiliation” are insufficient to meet the standard of minimum contacts). Indeed, given that OCLC alleges that Defendants have contacted OCLC customers “all over the world” (Compl. ¶ 120 (emphasis added)), that allegation would render Defendants subject to suit in virtually every single state. The allegation that OCLC was injured in Ohio is also an impermissible basis to exercise specific jurisdiction in this context. See Burger King, 471 U.S. at 474 (“Although it has been argued that foreseeability of causing injury in another State should be sufficient to establish such contacts there when policy considerations so require, the Court has consistently held that this kind of foreseeability is not a ‘sufficient benchmark' for exercising personal jurisdiction.” (quoting World-Wide Volkswagen Corp. v. Woodson, 444 U.S. 286, 295 (1980))). While Defendants intend to fully brief this issue at the appropriate time, OCLC's failure to adequately plead personal jurisdiction is another reason that the Court should deny OCLC's motion for a temporary restraining order.
III. THE PROPOSED TRO, EVEN AS NARROWED BY OCLC, IS NOT TAILORED TO PREVENT IMMEDIATE, IRREPARABLE HARM TO OCLC.
A TRO is not needed currently to prevent OCLC members from sharing so-called “WorldCat records”— however defined—with non-members, because there is no peer-to-peer sharing of records outside of the MetaDoor development partners prior to the platform going live. MetaDoor is not expected to launch until February 2023 at the earliest, which is well beyond the time frame where extraordinary temporary measures such as a TRO or, indeed, even potentially a preliminary injunction are generally used. Thus, the only appropriate consideration for this Court is whether the circumstances warrant curtailing of Defendants' pre-launch activities
During the development phase, various development partners' catalog records are only made available to Defendants through Ex Libris's Alma ILS platform, to which each partners already subscribes, and only for purposes of developing and improving the software, including building out its global title index function and improving the searching and automation functions to align with individual users' defined preferences. These are records that already reside within the development partners' catalogs. Indeed, the way MetaDoor is being designed to function is to allow access to other institutions into a sharing institution's own catalog system. No partner is being asked to engage in mass downloading of additional records directly from the WorldCat database in order to populate some sort of database akin to WorldCat. MetaDoor is not a database or a repository of records; it is merely a peer-to-peer sharing tool. When MetaDoor is operational, the records shared between libraries will never be in the custody of Defendants.
OCLC's proposal to limit MetaDoor development by prohibiting Defendants' access to and/or use of any socalled “WorldCat records” or any “metadata or records and metadata derived from the same” is not well-founded as a means to preserve a status quo. As discussed above, OCLC member institutions already engage in and are permitted to engage in sharing of their own catalog records and even “enhanced” records within the WorldCat database that correspond to works within their own holdings in order to further their own institutional goals, which would include efforts to facilitate their cataloguing processes. But even setting aside the merits issues, OCLC's proposal is impractical. Even if “WorldCat records” is defined to reach only exact copies of records contained within the WorldCat database, there is no reasonable mechanism to police compliance. There is no reasonable means to identify whether a record within an institution's catalog is identical to a record contained within the WorldCat database without a manual, record-by-record inspection of every record to be shared and a comparison of those records to the record for the same item contained in the WorldCat database—to which Defendants (as non-subscribers) do not themselves have access.
As for catalog records that are not exact copies but that may include metadata derived from records in the WorldCat database, a tracing of metadata sources would be even harder (and OCLC's claims to control over derivative works even more tenuous). OCLC's proposal to use the presence of an OCN number is not a suitable proxy for a “WorldCat record or metadata.” The OCN number is not an indicator of authorship or of ownership of any or all the metadata present in a record (again, OCLC disclaims ownership of any individual records in its database). It is merely a tracking number to associate catalog metadata records that may exist in the world in numerous versions with the version of the record contained within the WorldCat database that corresponds to the same work. The OCN number is used primarily as a means for OCLC to deduplicate records as they are shared with OCLC by its member institutions. Indeed, for these reasons, OCLC has stated publicly that OCN numbers should be treated as in the public domain,1 and OCLC has expressly disavowed the OCN number as an indication that a record originated with OCLC.2 OCLC's suggestion that an OCN number's presence on a record itself provides proof of OCLC's right to control the use of all the metadata contained within that record, simply because, at some point, that record may have been made available by a member to the WorldCat system, is inconsistent with OCLC's acknowledged lack of an proprietary right in the individual records and metadata within WorldCat and with OCLC's own policies and member agreements. If the proposed TRO is granted and interpreted in the manner OCLC suggests, it may give OCLC a constructive proprietary right over records for which the OCN number represents the only contribution authored by OCLC itself.
1 https://help.oclc.org/Library_Management/WorldShare_Reports/Report_objects/ Report_objects_A_to_Z/Report_objects_M-P#O:
The OCLC Control Number is a unique, sequentially assigned number associated with a record in WorldCat. The number is included in a WorldCat record when the record is created. The OCLC Control Number enables successful implementation and use of many OCLC products and services, including WorldCat Discovery and WorldCat Navigator. OCLC encourages the use of the OCLC Control Number in any appropriate library application, where it can be treated as if it is in the public domain.
OCLC's proposed TRO, as revised, would further restrict obtaining consents from potential customers who are OCLC members or even from communicating with them concerning the sharing of any records that would fit within the ill-defined category of “WorldCat records or metadata.” It would further prohibit any institution from partnering with or assisting Defendants to the extent that these arrangement “induce” WorldCat customers to share such “WorldCat records or metadata,” thereby removing any notion of intentional interference and making Defenants the ultimate guarantors of their customers' compliance with third-party contracts. To be clear, nothing about Defendants' communications with or obtaining consents from prospective customers in any way induces any breach of those customers' contracts with OCLC. Any such communications and consents are, at this point, purely prospective with respect to any peer-to-peer sharing. Moreover, Defendants' MetaDoor consent form already indicates that customers will share only what they wish, on the terms they wish to, and what they determine and represent they are authorized to share. Thus, these forms do not have any conceivable role in inducing any breach of contract, and OCLC has not pointed to any evidence of immediate, irreparable harm caused from obtaining these prospective consents. They are not designed to give otherwise potential WorldCat customers a way of accessing WorldCat. Customers are not being encouraged to share WorldCat records or download records from WorldCat to include in what they share. Such an order would only result in the chilling of legitimate commercial activity.
IV. TO THE EXTENT THE COURT DETERMINES THAT A TRO IS WARRANTED, IT SHOULD BE TAILORED TO ADDRESS IMMEDIATE HARMS ARISING FROM AN ALLEGED TORTIOUS INTERFERENCE WITH OCLC'S CONTRACTS.
To the extent any TRO is warranted under the circumstances, which Defendants strongly dispute, it should be tailored to address only imminent threats of irreparable harm to OCLC, and only such harm that might arise from conduct that would amount to tortious interference with OCLC's contracts. For these reasons, without conceding that any TRO is appropriate, Defendants have revised the proposed TRO (see attached as Exhibit A) to illustrate an order that would more narrowly tailor the relief to any potential immediate or irreparable harm. In particular, the revision limits the scope of enjoined communications to communications concerning records or metadata “directly from WorldCat” (see ¶ 1), in recognition of the ways in which institutions have intermingled records in their own catalogs that may contain OCN numbers and which OCLC members are free to share among themselves; and we have proposed a similar limitation to paragraph 2. We have proposed revisions to paragraph 3 to limit the scope of the enjoined activity to actions that could plausibly result in any immediate harm—i.e., Defendants' retention of records over which OCLC asserts certain contractual rights, or those records being made public through MetaDoor. Finally, we have added a paragraph clarifying that Defendants may communicate with OCLC customers about activities that are indisputably permitted under the WorldCat Rights and Responsibilities—specifically, the sharing of records among institutions.
V. SHOULD ANY TRO BE ENTERED, OCLC SHOULD BE ORDERED TO POST A BOND COMMENSURATE WITH THE DISRUPTION CAUSED TO DEFENDANTS' DEVELOPMENT EFFORTS.
As noted above, the TRO OCLC has proposed would drastically curtail Defendants' ongoing development efforts. Should it extend for any significant length of time, or should it require partnering institutions to exclude from the development process a substantial amount of their own library catalog records, then Defendants would lose the benefit of a considerable amount of the development work it has already conduct and may lose its current development partners entirely. This would result in substantial losses of development resources already expended.
Defendants have incurred several millions of dollars in development costs for MetaDoor to-date. Even estimating conservatively the extent of those development efforts that will be lost were a TRO to be entered, Defendants propose that a suitable bond amount would be at least $500,000.
For these reasons, and as further elaborated upon by Defendants during the hearing on OCLC's Motion, OCLC's Motion for a TRO should be denied. Should the Court decide to proceed with consideration of OCLC's request for a preliminary injunction, Defendants respectfully request that any hearing or decision on that Motion occur only after a full opportunity for the parties to brief that issue.