Ownership rights in IP addresses? A legal analysis

A new paper on ownership rights in IPv4 addresses has been drafted; it is posted here. Author Ernesto Rubi is a J.D. candidate at Florida International University School of Law. The paper explores the legal framework (or rather, the lack thereof) surrounding IP addresses. The depletion of the IPv4 free pool will, the author contends, intensify arguments about ownership interests in ipv4 address blocks. He concludes that “the concept of IP number ‘ownership’ or claim of right may be inevitable.”

“No known case law exists” clearly establishing either ownership rights or a lack of such rights in IP addresses, according to the paper. Rubi documents all the cases in which U.S. courts, law journals, and ISP contracts consistently refer to IP addresses as being “owned” and refer to ISPs or their subscribers as “owners” of said addresses. According to Rubi, Kremen v. ARIN, a recent decision by a U.S. District Court that is cited by ARIN as support for the claim that property rights do not exist in IP addresses, merely established that ARIN is the only available source for ‘current’ (i.e., non-legacy) address allocations, not that no property rights exist. Moreover, he shows that in the record of Kremen v. ARIN, ARIN’s CEO repeatedly asserted that ARIN had no authority over legacy address blocks. Legacy allocations constitute 44% of the IPv4 address space.

The paper explores various approaches that could be taken to the property status of the address space, including patent rights (not workable), the public trust doctrine, U.S. law regarding telephone numbers, and international law. While declaring IP addresses a “public trust“ seems to be the most feasible way to pre-empt private claims of ownership, this path has a rather undesirable side effect for those favoring a global internet: the only feasible public trustee under this construct would be the U.S. government. Not an outcome the rest of the world is likely to find palatable! And even if one were to take that route, there is, as Rubi points out, “no government contract between ARIN and ICANN or ARIN or any agency or body of the U.S. government.”

By now it should be evident that ARIN is the target of this piece. The section titled “Contract law and ARIN’s contracts – illusory?” provides a frontal assault on the whole contractual regime the North American IR uses to govern its portion of the address space, while a section on “ARIN and antitrust” contends that “ARIN’s exposure to antitrust liability is likely to be a point of contention in future litigation.” The first claim is more original than the latter: the author suggests that ARIN’s Registration Services Agreement (RSA) contracts might be declared “illusory” and thus invalid by courts because of the way ARIN reserves the right to revoke IP address allocations ‘at any time and without prior notice.’ Whether the RSA actually runs this risk of invalidation is debatable; what is clear, however, is that any attempt by ARIN to reclaim scarce address resources on the basis of a “needs assessment” performed years after the initial allocation would indeed generate legal resistance that might invoke this argument. But despite the existence of hotheads within ARIN advocating that it go down that path, ARIN’s management is probably wise enough to avoid that advice.

The author decries the RIRs’ lack of enforcement authority over IP address usage and seems unrealistically worried about the fact that there is “no one technical entity currently in charge of policing the internet routing table.” Here we would caution Rubi to be careful what he asks for. Greater centralization of authority over the internet is as technically risky as it is politically undesirable. Law students who are habituated to well-established, hierarchical forms of national government often view loose, networked forms of transnational governance as problematic or dysfunctional. In fact, decentralization and looseness is a feature, not a bug, in this case. Despite the paper’s attempt to contrast address hijackings (which he thinks are poorly governed) with domain name hijackings (which he thinks are adequately governed because of the centralization of power in ICANN) the truth of the matter is that the former is currently less of a problem than the latter, while the centralization of power in ICANN raises a number of troublesome issues of policy, accountability and geopolitics that would best be avoided.

One very positive aspect of the paper, however, is its emphasis on the need for clearer property rights in IP addresses in order to facilitate market transfers of addresses as we enter an era of IPv4 address scarcity. The paper clearly recognizes the point IGP has been making from 2008 on, namely that scarcity will intensify contention and conflict over legacy address resources and that transferability in response to market forces will alleviate many of those pressures.

One comment

  1. Anonymous

    Hi Milton,
    In addition to your critiques, I would add, inter alia, that the author seems to have not heard of the Specified Transfer Listing Service that was the subject of so much discussion in the previous year.
    This allows for “transferability in response to market forces” while ensuring that Internet numbering resources do not become subject to the same harmful practices we see in the domain name industry.