In a letter sent by FairSearch.org (FairSearch) to ICANN they urge ICANN to reject the new gTLD application for .Search filed by Google even after Google changes their application to make it an opened extension.
“I write to apprise ICANN that Google’s recent attempt to amend its .search Generic Top-Level Domain (gTLD) application is an effort to evade Google’s responsibilities to address objections through the dispute resolution process at the ICC International Centre for Expertise.”
“Instead of responding directly to the areas of concern raised in objections to Google’s .search application that are on record, Google now seeks to alter its application. ”
“The proposed amendments do not remedy the anticompetitive effects that would result from a dominant company controlling a crucial entry point for Internet search. ”
“Instead, they (Google) attempt to shift the focus away from legitimate objections. At the same time, the proposed amendments proffer new procedures that are laden with their own problems.
Google should not be permitted to continue to put off the substantive discussion about the competitive risks associated with Google’s ownership of a search gTLD.
FairSearch respectfully submits that by proposing amendments to its original application and thereby increasing the surface area of competition concerns, Google is engaging in a counterproductive exercise at this stage of the gTLD application process. For these reasons, ICANN and the ICC should consider Google’s .search application as it stands.
FairSearch serves as a reputable voice for the Internet search community to preserve transparency, fairness, consumer choice, and innovation in search.
“On March 13, 2013, FairSearch filed objections to applications by Google subsidiary Charleston Road Registry (hereinafter referred to as Google) for the .search, .map, and .fly gTLDs. FairSearch was concerned particularly by Google’s request to obtain and to operate the .search gTLD as a “closed registry with Google as the sole registrar and registrant.”
Google proposed to make .search proprietary by “allow[ing] Google to manage the domain space for its Google Internet search offerings.”
“If Google controls .search, Google would have both the incentive and the ability to become even more of a bottleneck in search than it already is.”
“Google would unilaterally control the domain name that describes its core function while being able to exclude other competing search engines from such beneficial association”
“In a recent communiqué, ICANN’s Governmental Advisory Committee (GAC) states that restricted registration policies should require safeguards, and, “[f]or strings representing generic terms, exclusive registry access should serve a public interest goal.”
The GAC identifies .search as one such generic term requiring the Registry Operator to apply for a code of conduct exemption in order to operate the gTLD as a closed domain.
Thus, under its original application, the burden is on Google to demonstrate that administering the .search gTLD as a closed registry would somehow serve the public interest—a difficult showing to make given that Google’s proposal would seriously impair competition from search rivals.
On April 6, 2013, Google requested the opportunity to amend its .search application. It now proposes to operate the search top-level domain instead as a “redirect service on the ‘dotless’ search domain (http://search/)”
This about-face is an acknowledgment that operating the .search gTLD as a closed registry would pose too obvious a harm to competition—even for Google. Nevertheless, Google’s proposed amendments continue to describe a system for managing the search gTLD in a way that would impair search competition.
This letter lays out initial reactions to Google’s proposed amendments and why those amendments should not be accepted. If ICANN agrees to approve Google’s revisions to the .search application, FairSearch intends to file additional comments within thirty (30) days of such revision.
I. Procedural Concerns: No Business Should Have to Go Through Its Biggest Competitor to Gain Access to a Common Resource that is Critical to Competition.
If ICANN grants Google’s search gTLD application, ICANN would confer upon Google even greater incentive and ability to exclude competitors from the search markets. Such a result would not further ICANN’s goal of “foster[ing] diversity, encourag[ing] competition, and enhanc[ing] the utility of the [Domain Name System].”6
Google proposes to “limit registration to only application developers. [Google] plans to require registrars to confirm that a domain applicant is an application developer via an established process.”7
Yet Google does not specify how or why it would limit registration.
As online search evolves, neither ICANN nor the community of vertical, general, or to-be-developed search experiences can be assured that they will come within the Google-defined parameters for applicants.
Moreover, Google alone would determine applicants’ rights and privileges to access the search gTLD.
Applicants would have no bargaining power to change unfair or onerous terms. Google also reserves the right to monitor registrants and to enforce suspensions.
Consequently, Google may limit competitor registrations and suspend competitors’ access to the search gTLD.
Additionally, Google retains the right “to charge different prices for unique second-level domains within the gTLD .
Google does not describe on what basis it would charge different prices, much less any criteria for how it would set prices.
Google merely assures that it will not “price discriminate among ICANN-accredited registrars.”
Without stating more, Google asks ICANN to trust that Google will set prices reasonably.
“Trust” is not a sufficient strategy to address competition concerns.
In short, under its amended proposal Google reserves too much authority to decide how it alone will manage the search gTLD.
ICANN’s GAC advises that registries for generic terms operated on a restricted access basis should be administered “in a transparent way that does not give an undue preference to any registrars or registrants, including itself, and shall not subject registrars or registrants to an undue advantage, But Google’s amended application provides no such guarantee of transparency. <
Google, the dominant global search provider with a history of anticompetitive conduct, cannot be trusted as an honest broker. As the European Commission recently found, Google indeed promotes its vertical search services in way that unduly diverts traffic away from Google’s competitors in specialized search towards Google’s own specialized search service
FairSearch respectfully suggests that it is too tall an order to ask nascent competitors and search entrants to entrust their most significant competitor with provision of fair access to what likely will be a critical resource to online search competition—the search gTLD.
II. Privileged Access to Competitively Significant Data: Google Would Be Positioned to Collect User Queries and Other Sensitive Information from the Search gTLD.
Google offers assurances in its proposed amended application that it will “ensure the appropriate level of privacy and security will be met for its users.”
Google plans to encrypt data “to ensure third parties cannot access personally identifying information or other sensitive data as it crosses the Internet.”
Notably absent from Google’s assurances, however, is any acknowledgment that Google may use such access for its own search and related services. <
Many competitors in search seek to obtain sources of data so that they can serve more relevant results. Nowhere in Google’s application does Google promise that it will not access queries, search results, or user data that pass through the search gTLD. That loophole is gaping considering that Google may be able to capture data from competing services and thereby piggyback off of the value created by competitors without compensating them.
In addition to Google’s access to competitively sensitive data, Google promises to run the search gTLD as a “redirect service” in which it provides users the option to elect a default search engine.<
What Google’s application does not specify is how Google will consider which competitors to list as one of those “default options.” Nor does it say whether users will have the ability to select default vertical search services as well, or whether the election will be for general search engines only. Finally, Google makes no assurances that it will not seek to override a user’s choice even after the searcher makes an initial search engine selection.
III. Stifles Innovation: Google’s Proposal to Standardize Search Architecture would not Promote Innovation.
Google’s application proposes to “encourage websites with search functionality to adopt common query frameworks. . . .”
Google asserts that this change “may allow the emergence of new services and make it easier for users to switch between search functionality.” Google< describes such a Google-managed search gTLD as a “simple technical standard describing how users and other software can interact with search functionality within the TLD.”
Despite Google’s claim that this standardization will promote innovation, it likely will do the opposite. To the extent Google attempts to standardize search interfaces, that will merely prevent other search engines from developing new and innovative interfaces and ranking algorithms. And unlike most standards that develop collaboratively to create the best technology that poses the fewest frictions for existing competitors, Google has every reason to refine a “standard” that hews closest to its own search architecture. As a result, Google will have the ability to ensure that the transition to the search gTLD is painless for itself and costly for everyone else.
IV. Security Concerns: Google’s Proposal to Run the gTLD as a “Dotless” Domain Name Would Pose a Serious Threat to Security.
Lastly, Google does not address concerns posed by a number of security experts—including ICANN—that have evaluated “dotless” domains.19 According to the February 2012 report from ICANN’s Security and Stability Advisory Committee (SSAC), “[d]otless domains will not be universally reachable and the SSAC recommends strongly against their use.”20 Adopting any proposal to the contrary not only would circumvent the SSAC’s recommendation but also would undermine ICANN’s dedication to “preserving the operational security and stability of the Internet “
Programmers have designed web browsers and enterprise Intranet sites for a world in which dotless domains are not prevalent on the Internet but instead are reserved for enterprise Intranets. As such, dotless domains would present serious challenges to enterprise security architectures. Introducing dotless domains also would require reconfiguration of web browsers in order to address the ambiguous destinations that dotless domains pose. While they may eventually innovate to be able to utilize dotless domains in a way that ameliorates these technical concerns, neither a critical mass of Intranets nor of web browsers will be designed to work with dotless domains in the short- to mid-term future. Even implementing a dotless search domain in the interim could create massive and widespread risks to security.
It is noteworthy that one of Google’s engineers, Ian Fette, considers dotless domains insecure.
Mr. Fette provided feedback during the SSAC’s efforts to develop its report on dotless domains.
In a subsequent public comment he submitted on October 3, 2012, Google’s Mr. Fette notes that “the migration to using the new TLDs in a ‘dotless’ fashion may simply not be possible in a secure manner . . .” and predicts that “at least for the next three years for sure (as I have concerete [sic] evidence for this time period), such an evolution [to dotless domains] would be actively harmful.“
Having opposed dotless domains with good reason, Google should not now be permitted to foist a dotless search gTLD on the Internet with ICANN’s endorsement.
Google’s proposed amended application for the search gTLD simply cannot address how Google, as the dominant search provider, is capable of controlling a search entry point in a way that allows search competition to flourish—Google cannot be trusted to take on such a task. Further, there is no process in place that could ensure that Google adheres to any of the assurances made in its proposed amendment. Finally, a dotless domain proposal is fraught with security and stability concerns.
Consequently, even if ICANN considers accepting Google’s amendment, FairSearch urges ICANN ultimately to deny Google’s application for the search gTLD.
Ben Hammer FairSearch.org