In a victory for fairness and rule-based Internet governance, an independent review panel (IRP) has decided that ICANN was wrong to deny retailing giant Amazon, Inc. the top level domain AMAZON. Key elements of the decision were unanimous, particularly the conclusion that the Board “cannot accept GAC consensus advice as conclusive.” ICANN’s board, in denying the application, failed to provide sufficient reasons for doing so, the panel ruled. The majority of the panel also concluded that the GAC was at fault for not treating Amazon fairly.

We believe the IRP made the right decision, and we hope the ICANN board and the GAC will come to their senses and simply allow the TLD to be awarded to Amazon and be done with it. The main roadblock to this obvious solution is the perception, which we believe to be misguided, that objections from Latin American countries have some merit and therefore some “win-win” solution, most likely involving payoffs to someone, must be found. This will just muddy the waters and prolong the dispute.

We prefer a cleaner solution with better long-term consequences for Internet governance – a solution that does not encourage governments or other objectors to make claims not based in any law or engage in hold up politics in the future. We need to recognize that certain countries were playing identity politics and making bad arguments, and turn our back on that.

Here’s why the objections to the .AMAZON application are baseless:

  • Good governance requires clear, pre-specified and objective rules – rule of law, not rule of men. Amazon followed ICANN’s application rules. None of the strings applied for by Amazon are listed as ‘geographic names’ in ICANN’s Applicant Guidebook, and thus were not subject to government pre-approval. The new gTLD process also anticipated applications for “brand” top level domains and there was nothing unexpected about it. The objectors to the application were trying to change the rules in the middle of the game.
  • The parties objecting to the application had no application of their own, no alternative ideas about how to use the domain. Indeed, in an attempt to resolve the dispute, Amazon reputedly offered to cover the costs of applying for AMAZONAS or AMAZONIA, but there were no takers. (In the region itself, as well as in many other parts of the world, the region goes by those names) They just wanted to prevent Amazon from having it.
  • Objectors never provided even a minimally plausible case that this appropriation harmed anyone or anything. The .AMAZON application was challenged using ICANN’s “community objection” criteria, but an expert report found that the TLD “would not pose a material detriment to the region or the people who inhabit the geographic region proximate to the Amazon River.”
  • The horse is already out of the barn. If indeed .AMAZON constituted some kind of illegitimate cultural appropriation, then why doesn’t AMAZON.COM? Or all the Amazon trademarks? There is no rational distinction between any of these uses. Can anyone show how the people or environment of the Amazon region have been harmed by the use of these names?

The .AMAZON case is a good example of how, during its new gTLD program, ICANN allowed its processes to be hijacked by politics, usually emanating from the Governmental Advisory Committee (GAC). In numerous cases, pre-established rules and principles were abandoned and application guidelines modified on the fly simply because the GAC wanted to have discretionary power over top level domain applications.

In the AMAZON case this led to a palpably absurd result. A regime that was established to mediate domain name – trademark conflicts denied an application by a company for its own name – a company that had a registered trademark for AMAZON in over 170 nations, and whose use of the name is so well-known globally that it has obtained secondary meaning (in English) that is largely unrelated to the South American region. And it denied the application not because someone else also wanted the name, and not because of any harm to a community, but simply because a few representatives of Latin American governments wanted to flex their muscles.

Regardless of whether Amazon actually gets the TLD, this case has important precedent value. The ruling says that the ICANN board cannot simply defer to consensus GAC advice. If it upholds that advice, it must set forth reasons and explain why the reasons reflect “well-founded and credible public policy interests.” As one of the panelists noted, “For me, the key requirement is that there be a ‘well-founded’ basis for the [Board’s] conclusion.” Once Amazon had rebutted the presumption in favor of following GAC advice, he wrote, “the burden of making that showing became ICANN’s to bear. It failed to do so.”

6 thoughts on “Amazon win sets good precedent

  1. While your argument throughout on behalf of Amazon goes in great detail why Amazon deserves the TLD, I think the IRP case that you may have overlooked or forgotten completely is that of DotConnectAfrica Trust on the .Africa domain and is the one that takes credit for the precedence setting on ICANN’s new gTLD procedural deficiency. This includes the GAC decisions that lacked rationale, a not so independent Geographic evaluation panel and the ICANN board turning a blind eye to credible scrutiny. The Register UK covered in great detail the two IRP cases and comparison, see link The IRP declaration is nearly similar to DCA Trust.

    The two IRP cases, .Africa and .Amazon for DotConnectAfrica Trust and Amazon were represented by the same Lawyers. I think both applications being geo-tlds could be looked at in the same analysis at a greater detail.

    1. You are correct, Martin, the .AFRICA cases are directly relevant to this case. Thank you for making that point. A good summary of that decision by Robin Gross can be found here.
      However, in the .AFRICA case the geographic name evaluation panel actually sided with ICANN, whereas in the .AMAZON case the expert evaluator did not even give ICANN any cover for its decision – the board simply ignored his findings.

  2. You have written in detail about Amazon’s IRP victory over ICANN and in your article, you wittingly and obsequiously tried to portray the IRP as a new precedent-setting effort. Unfortunately, you have not researched properly previous Independent Review Process (IRP) Panel proceedings that have been prosecuted by other parties against ICANN. Therefore, your article which conveys the notion that the Amazon vs. ICANN IRP is precedent-setting is entirely misleading.

    It may interest you to know that the very first IRP process that was instituted against ICANN in 2013 as part of the new gTLD Dispute Resolution Procedures was brought by DotConnnectAfrica Trust (DCA) – an applicant for the .Africa New gTLD string name.

    DCA Trust had prevailed in that IRP following the IRP Panel’s Final Declaration in July 2015 – which actually precedes the Amazon vs. ICANN IRP process by two years. Therefore, the DCA vs. ICANN IRP was the real precedent that was set after that IRP Panel ruled that the actions/in actions of the ICANN Board regarding the treatment of DCA’s .Africa application had violated the ICANN Bylaws and the ICANN Articles of Incorporation.

    It might also be of interest for you to know that before instituting the IRP suit against ICANN at the International Centre for Dispute Resolution (ICDR), DotConnectAfrica Trust had previously exhausted all the prescribed escalation steps in the ICANN Accountability Mechanism – the same dispute resolution path that Amazon had later followed, no doubt, after seeing the results of DCA’s trail-blazing efforts.

  3. Your reply comment to Martin is disingenuous even as you have tried to confound the issue by bringing in something that is unrelated and rather immaterial to your article.

    First, the thematic thrust of your write-up was about ‘precedence’. It has already been shown that your thesis is very wrong. The Amazon IRP was not precedent-setting. Amazon was actually victorious against ICANN; in part, because it relied upon DCA Trust’s winning arguments against ICANN – thus utilizing an earlier precedent that was established in a prior IRP proceeding.
    Second, your analysis on the issue of Geo Name Panel (GNP) agreeing with ICANN is again erroneous and very misleading. At no point during the IRP or in the Final IRP decision did the GNP agree with ICANN – since the GNP works for ICANN and was not part of the IRP process.

    However, the transactions of the DCA Trust IRP proceedings revealed all the wrong-doings of the ICANN GNP – that is, the consultant (InterConnect Communications – ICC) that had been retained by ICANN to perform geographic name evaluations. Such unpardonable wrong-doings (as revealed) had also demonstrated how ICANN staff helped to ‘ghost-write’ an endorsement letter for DCA Trust’s opponent. All these revelations had helped DCA Trust to prove its case against ICANN – who was found guilty by the IRP Panel of discriminatory and inequitable treatment of DCA Trust’s application.

    It may all be confusing, but it helps to analyse all the IRP Final Declarations closely and be better informed before formulating any opinions on the subject. I think it should be part of a journal on the failings of ICANN in its attempt at increasing the domain name space.

    Thank you.

Comments are closed.