Latest news of the domain name industry

Recent Posts

.hotel losers gang up to threaten ICANN with legal bills

Kevin Murphy, August 30, 2016, 08:26:49 (UTC), Domain Registries

The six losing applicants for the .hotel new gTLD are collectively threatening ICANN with a second Independent Review Process action.
Together, they this week filed a Request for Reconsideration with ICANN, challenging its decision earlier this month to allow the Afilias-owned Hotel Top Level Domain Sarl application to go ahead to contracting.
HTLD won a controversial Community Priority Evaluation in 2014, effectively eliminating all rival applicants, but that decision was challenged in an IRP that ICANN ultimately won.
The other applicants think HTLD basically cobbled together a bogus “community” in order to “game” the CPE process and avoid an expensive auction.
Since the IRP decision, the six other applicants — Travel Reservations, Famous Four Media, Radix, Minds + Machines, Donuts and Fegistry — have been arguing that the HTLD application should be thrown out due to the actions of Dirk Krischenowski, a former key executive.
Krischenowski was found by ICANN to have exploited a misconfiguration in its own applicants’ portal to download documents belonging to its competitors that should have been confidential.
But at its August 9 meeting, the ICANN board noted that the timing of the downloads showed that HTLD could not have benefited from the data exposure, and that in any event Krischenowski is no longer involved in the company, and allowed the bid to proceed.
That meant the six other applicants lost the chance to win .hotel at auction and/or make a bunch of cash by losing the auction. They’re not happy about that.
It doesn’t matter that the data breach could not have aided HTLD’s application or its CPE case, they argue, the information revealed could prove a competitive advantage once .hotel goes on sale:

What matters is that the information was accessed with the obvious intent to obtain an unfair advantage over direct competitors. The future registry operator of the .hotel gTLD will compete with other registry operators. In the unlikely event that HTLD were allowed to operate the .hotel gTLD, HTLD would have an unfair advantage over competing registry operators, because of its access to sensitive business information

They also think that HTLD being given .hotel despite having been found “cheating” goes against the spirit of application rules and ICANN’s bylaws.
The RfR (pdf) also draws heavily on the findings of the IRP panel in the unrelated Dot Registry (.llc, .inc, etc) case, which were accepted by the ICANN board also on August 9.
In that case, the panel suggested that the board should conduct more thorough, meaningful reviews of CPE decisions.
It also found that ICANN staff had been “intimately involved” in the preparation of the Dot Registry CPE decision (though not, it should be noted, in the actual scoring) as drafted by the Economist Intelligence Unit.
The .hotel applicants argue that this decision is incompatible with their own IRP, which they lost in February, where the judges found a greater degree of separation between ICANN and the EIU.
Their own IRP panel was given “incomplete and misleading information” about how closely ICANN and the EIU work together, they argue, bringing the decision into doubt.
The RfR strongly hints that another IRP could be in the offing if ICANN fails to cancel HTLD application.
The applicants also want a hearing so they can argue their case in person, and a “substantive review” of the .hotel CPE.
The HTLD application for .hotel is currently “On Hold” while ICANN sorts through the mess.

Tagged: , , , , , , , , , , , , , , ,

Comments (16)

  1. Richard Funden says:

    So much salt…

  2. With respect to the RfR, it is important clarify that an alliance qualifies per the AGB: With respect to ‘Delineation’ and ’Extension,’ it should be noted that a community can consist of […] a logical alliance of communities (for example, an international federation of national communities of a similar nature). (AGB, 4-12). Per the .HOTEL’s community applicant’s definition: “a logical alliance of members, with the associations and the marketing organizations maintaining membership lists, directories and registers that can be used, among other public lists, directories and registers, to verify eligibility against the .hotel Eligibility requirements” (.HOTEL community application, 20A). It is clear that the .HOTEL applicant did not mention exclusively “industry” in its community definition, which is what this RfR incorrectly states. The .HOTEL community applicant clearly met the CPE requirements based on the language. As a side note, our community definition for the music community was: “Strictly delineated and organized, logical alliance of communities of similar nature that relate to music” (DotMusic music community application 20A). It is exactly – word-to-word – per the AGB language and what the AGB allows to qualify as a community. However, in a miraculous manner the EIU/ICANN Staff omitted our definition from the CPE Report. In other words, we got graded without our definition considered, analyzed or mentioned in the CPE despite thousands of letters “explicitly” defining our definition by supporters and tens of thousands of pages of Application Materials explicitly outlining our definition. A recent .KIDS BGC RfR decision stated that the “explicit” definition from 20A must be considered and a “logical alliance” qualifies if there is awareness. A reasonable person would conclude (just like the AGB does) that an “alliance” has awareness, especially when they have submitted letters supporting and being members of that particular “alliance” which the general public will commonly associate the string with (i.e. a match). Given the Dot Registry decision, it appeared ICANN Staff was “intimately involved” with the CPE process. This is not what ICANN Staff/BGC stated in their determinations to our DIDP request. It appears something is inconsistent here.

    • Kevin Murphy says:

      Is the AGB not referring to alliances that pre-date the new gTLD program, rather than alliances that were created purely in order to game the system?

      • Great question Kevin.
        To fully answer the “construed” community and gaming issue then one should consider how each community functions and what the eligibility requirements are. Restricting eligibility to members of an alliance of specific community organizations, while excluding all other legitimate community members and organizations is what the AGB tried to prevent and what the “construed” definition is all about. This is how the other music community applicant got in trouble by defining a static group of organizations as the community definition and restricting eligibility to only members of those specific 42 organizations. In contrast, the DotMusic “alliance” definition does not explicitly mention specific “organizations” purposely because that would be construing a community by discriminating against legitimate members of the community.
        Once you start defining the community and eligibility in relation to specific organizations and their members then it would exclude and discriminate against all other entities that have a legitimate interest in registering because of their association to the string. Such a definition could not possibly constitute the entire community as commonly known by others and would fail to meet the nexus because the community would not match the string. As such, it would be “construed.”
        In antithesis, the community defined in DotMusic’ application “relates to ALL music-related constituents using an all-inclusive, multi-stakeholder model” i.e. does not relate to specific organizations or add restrictions to prevent legitimate entities from registration. Further, the alliance of music communities (delineated by constituent types) predated 2007. As per the DotMusic application, “only eligible members of the Music Community can participate” (20E). All must cohere and are united under general principles of international copyright (i.e. of similar nature), international treaties and conventions, such as the Berne Convention that predated 2007. Copyright is what the Music Community has to abide to cohesively, regardless whether you are a musician, label, publisher or any other music constituent type. In other words, copyright ties the community members together because music is a regulated sector (due to copyright). Further, the community defined cohesively uses “commonly used classification systems such as ISMN, ISRC, ISWC, ISNI,” which all predated 2007. Per the DotMusic application indicates, “the Music Community encompasses global reaching commercial and non-commercial stakeholders, and amateur stakeholders.” These stakeholders pre-dated 2007.
        There seems to be a misunderstanding within the ICANN community between “definition” of the community and “delineation.” There are critical differences depending on how an application was written in relation to the AGB rules. According to the AGB instructions, if it is a “community consisting of an alliance of groups (e.g. individuals, organizations or business), details about the constituent parts are required” (See AGB, 4-12, Notes, 20A). For example, DotMusic’s definition consisted of an “alliance” so the community as a whole (in the form of constituent groups/categories) was described and delineated. In other words, to fulfill the mandatory AGB and “nexus” requirements, all category types that match the community (i.e. are essential for the community to function or complement each other to form a whole) must be delineated in the application (and they were). For DotMusic, the music constituent types detailed in the application pre-dated 2007.
        It is also noted that DotMusic received a lot of support after submitting the application in 2012. The DotMusic application’s definition accommodated this because DotMusic’s mutually inclusive community definition and eligibility requirement was associated to only members belonging to the alliance of communities “related to music.” In other words, DotMusic did not construe a static community in 2012. The definition is dynamic and all-inclusive of all constituent category types related to music. This also guarantees transcience into the future. If DotMusic’s definition was “construed” then how is it that any music-related constituent is able to qualify as part of the music community and be eligible for registration going forward? Further, if it were “construed” then how could “music communities” be able to identify themselves as part of the community defined and support our application years after 2012 (i.e. and by supporting it means having the requisite awareness of community defined)?
        As indicated, the devil is in the details. The .HOTEL community applicant defined their community as a “logical alliance” but eligibility was not restricted to specific to hotel organizations or associations nor did the community applicant explicitly define the hotel “industry” as the community. As such, the .HOTEL community application qualifies because it did not discriminate, kept eligibility related to hotels and did not overreach because it restricted eligibility to hotel-related constituents and only those that were essential for the hotel community to function.
        The intended goals of CPE were to pass community applications that were genuine, did not discriminate against legitimate members of the community associated with the string, restricted registration to community members, had appropriate community-based registration policies and had demonstrable support i.e. the application was based on “trust” as envisaged by the GNSO’s 2007 Principles for launching the New gTLD Program (“where an applicant lays any claim that the TLD is intended to support a particular community, that claim will be taken on trust.” See, p.94).

      • I looked into what the BGC determined in prior RfR cases that mention “logical alliance.” As you know, per the Bylaws, BGC determinations have precedential value.
        According to the DotKids BGC declaration 16-6, the BGC determined that: “The CPE Panel is only asked to inquire when the community DEFINED BY the applicant was formed.” (, p.11).
        Footnote 40 of the BGC Determination 16-6 specifies: “the first CPE criterion ‘relates to the community as explicitly identified and defined according to the statements in the application.'” (p.11)
        It is easy to see in DotMusic’s application that the formation of the community defined pre-existed 2007 as DotMusic EXPLICITLY mentioned. In DotMusic’s words: “The Community has bought, sold, and bartered music for as long (“LONGEVITY”) as it has been made…The foundation for the structured and strictly delineated Community only resulted from the interplay between the growing music publishing business and an emerging public music concert culture in the 18th century (“PRE-EXISTING”).” (DotMusic Application, 20A).
        According to the AGB, an option that fulfills the criteria for a community definition under “Community Establishment” is a logical alliance of communities: “a community can consist of … a logical alliance of communities. All are viable as such, provided the requisite awareness and recognition of the community is at hand among the members. Otherwise the application would be seen as not relating to a real community.”
        The BGC confirms this in its determination that the option of a “logical alliance” qualifies per the AGB: “the Guidebook provides that ‘a community can consist of logical alliance of communities provided the requisite awareness and recognition of the community is at hand among the members.’” (BGC DotKids Determination 16-6, p.12)
        In other words, as long as members have the awareness and recognition of the community (by for example (i) self-identifying as being part of the community or (ii) by filing a letter to support and verify that they are a community member organization associated with the community) then the “alliance” community definition qualifies.
        The CPE Report for DotMusic states: DotMusic’s Application “bounds community membership by way of well-defined categories. Therefore the Panel has determined that the applicant provides a clear and straightforward membership definition.”
        So if, according to the DotMusic CPE Report, the Music Community’s (i.e. the alliance’s) membership definition is “well-defined categories” related to music then how can one conclude that the “alliance” of constituent type categories were construed or that they did not pre-exist 2007? You cannot have it both ways.
        In sum, the EIU purposely omitted the “alliance” definition in its DotMusic CPE Report and “construed” its own definition (contravening DotMusic’s EXPLICIT community definition) to find a way not to pass DotMusic. The CPE Report stated: “In addition to the member categories, the applicant also includes in its application a more general definition of its community (20D),” (CPE Report, p.3). So where is the “explicit” definition from 20A? The EIU using “general” terms from 20D violates the AGB’s “explicitly identified and defined” criterion for defining the community from 20A. Where exactly is the explicit “alliance” community definition mentioned in the CPE Report? It is mentioned nowhere because then DotMusic would have qualified to pass CPE. So much for transparency and a CPE Report that is “defensible” and “compelling.”
        With respect to .HOTEL community application, it is clear to me that they too qualify to pass CPE based on the AGB and its definitions.

  3. Rubens Kuhl says:

    Analyzing .hotel, .spa, .music, .gay and .llc/.llp, as a group, it’s clear that there are both false positives and false negatives among this group. I saw different opinions on which ones should have got community priority and which ones should have not, and I won’t get into there since it would become an endless discussion; but the commonality is that no one is able to agree with all evaluations or disagree with all evaluations. For me, it’s a clear indicator that all of them should be scraped, but that’s not what’s going to happen.

    • @RubensKuhl You should include .radio and .eco in that mix

    • “For me, it’s a clear indicator that all of them should be scraped, but that’s not what’s going to happen.”
      There is zero chance they will be scraped because some that have passed CPE have already signed registry agreements with ICANN (.ECO, .RADIO, .OSAKA). On the flip-side the majority that have not passed CPE and have been resolved privately or via an ICANN auction (e.g. .SHOP) have already signed registry agreements. Most have in fact launched and are in the root.

  4. Desiree says:

    @KevinMurphy, I’d like to understand your source of information in making the statement that ICANN was not involved in the actual scoring of Dot Registry’s CPEs?
    “It also found that ICANN staff had been “intimately involved” in the preparation of the Dot Registry CPE decision (though not, it should be noted, in the actual scoring) as drafted by the Economist Intelligence Unit.”
    The IRP Panel indicated in 93. of the IRP Declaration that “Moreover, EIU did not act on its own in performing the CPEs that are the subject of this proceeding. ICANN staff was intimately involved in the process…”

    • I agree with Desiree’s assessment as well.
      It is clear that based on the evidence that ICANN Staff contributed in the grading.
      For example, the CPE draft final reports that showcase ICANN was more than intimately involved are Exhibits C-42 to C-44 from the Dot Registry IRP case (See
      Can you imagine the content of documents that ICANN did not publicly release because they were either covered under NDA during the IRP or “protected” by the attorney-privilege loophole?

      • Kevin Murphy says:

        I would not leap to the conclusion that ICANN staff were involved in the scoring based purely on my imagination. I’d need to see some evidence first.

    • Kevin Murphy says:

      I read the exhibits disclosed in the IRP. The suggested edits ICANN made were to the wording of the report, not the scoring. If you have seen any evidence that ICANN was involved in the scoring, please share the links.

      • Desiree says:

        @KevinMurphy. There is an NDA in place in this IRP which prohibits disclosure. So, the limited exhibits to the IRP Declaration do not paint the whole picture. The IRP Panels stated, “intimately involved” and did not indicate anything regarding icann’s involvement in scoring one way or another. You should retract “(though not, it should be noted, in the actual scoring)” .

      • Kevin,
        If ICANN did not influence the scoring then why did ICANN pass everything through the attorney-client privilege loophole?
        If ICANN did not influence the scoring then why did ICANN mandate NDAs?
        If ICANN did not not influence the scoring then why did ICANN remove Quality Control when ICANN knew this was the most controversial and subjective process in the New gTLD Program.
        If ICANN did not influence the scoring then why did ICANN state in the Despegar case and in communication with many CPE applicants that they had no Quality Control (when as you pointed they were making serious edits to CPE reports).
        I think the collective evidence is pretty clear. If ICANN did not influence the scoring then they should disclose all the documents many applicants requested in DIDP and the documents the RIP panel requested. ICANN did not.
        Not sure how you can conclude whether or not ICANN was involved in the scoring. Based on the evidence, it is clear ICANN guided the decisions. Unless ICANN transparently discloses the documents requested then the “intimate” involvement points to the scoring as well. What is ICANN hiding that is so important that they cannot disclose? Your readers can reach their own conclusions.

        • Kevin Murphy says:

          If evidence exists that ICANN influenced the scoring, why did Dot Registry’s lawyer fail to mention it during the IRP hearing?
          Why did the IRP panel, in its declaration, choose to focus on some nonsense about where the word “research” came from, instead of talking about evidence that ICANN was influencing the actual scoring?

Add Your Comment