Comment Close
Date
Statement
Name 

Status

Assignee(s) and
RALO(s)

Call for
Comments
Call for
Comments
Close 
Vote
Announcement 
Vote Open
Vote 
Reminder
Vote Close
Date of Submission
Staff Contact and Email
Statement Number
07.03.2013"Closed Generic" gTLD Applications

Adopted
13Y, 0N, 1A 

n/an/a27.02.201327.02.201304.03.201305.03.201206.03.2012Karen Lentz
karen.lentz@icann.org 
AL/ALAC/ST/0313/1
Comment/Reply Periods (*)Important Information Links
Comment Open:5 February 2013
Comment Close:7 March 2013
Close Time (UTC):23:59Public Comment Announcement
Reply Open: To Submit Your Comments (Forum)
Reply Close: View Comments Submitted
Close Time (UTC): Report of Public Comments
Brief Overview
Originating Organization:ICANN
Categories/Tags:Top-Level Domains
Purpose (Brief):To receive stakeholder views and suggestions on the topic of "closed generic" gTLD applications.
Current Status:Existing provisions of the New gTLD Program do not provide specific guidance on this issue. Potential new provisions may be considered based on the comment provided and analysis undertaken.
Next Steps:ICANN staff will review comments submitted and will provide a summary and analysis of these comments to the New gTLD Program Committee of the Board of Directors. The Committee will review this feedback as well as the additional research and analysis directed to inform its consideration on this issue.
Staff Contact:Karen LentzEmail:karen.lentz@icann.org
Detailed Information
Section I: Description, Explanation, and Purpose

ICANN is seeking public comment on the subject of "closed generic" gTLD applications and whether specific requirements should be adopted corresponding to this type of application. Stakeholder views are invited to help define and consider this issue. In particular, comments would be helpful in regard to proposed objective criteria for:

  • classifying certain applications as "closed generic" TLDs, i.e., how to determine whether a string is generic, and
  • determining the circumstances under which a particular TLD operator should be permitted to adopt "open" or "closed" registration policies.

The New gTLD Program Committee of the ICANN Board of Directors has discussed this issue and has also directed completion of a set of focused research and analysis items to inform any possible action to be taken. At its 2 February 2013 meeting, the Committee passed the following resolution:

Whereas, the New gTLD Program Committee has received correspondence from the community addressing "closed generic" TLDs and understands that members of the community term a "closed generic" TLD as aTLD string that is a generic term and is proposed to be operated by a participant exclusively for its own benefit.

Whereas, ICANN implemented the Generic Names Supporting Organization (GNSO) policy recommendations on the "Introduction of New Generic Top-Level Domains", and within those policy recommendations there is no specific policy regarding "closed generic" top-level domains (TLDs).

Whereas, members of the community have expressed concerns regarding applications for "closed generic" TLDs.

Whereas, the New gTLD Program Committee considers that it is important to understand all views and potential ramifications relating to 'closed generic' TLDs.

Resolved (2013.02.02.NG01), the New gTLD Program Committee directs the President and CEO to open a 30-day public comment forum on this topic, which should include a call for identification of proposed objective criteria to classify applied-for TLDs as "closed generic" TLDs.

Resolved (2013.02.02.NG02), the New gTLD Program Committee directs the President and CEO to, concurrently with the opening of the public comment forum, request the GNSO to provide guidance on the issue of "closed generic" TLDs if the GNSO wishes to provide such guidance. Guidance on this issue is requested to be provided by the close of the public comment forum.

Resolved (2013.02.02.NG03), the New gTLD Program Committee directs the President and CEO to:

  1. Summarize and analyze all comments submitted in the public comment forum.
  2. Review materials supporting the policy development process resulting in the GNSO policy recommendations on the Introduction of New Generic Top-Level Domains and provide analysis of any discussions relating to the limitations on potential new gTLDs.
  3. Analyze the feasibility of objectively classifying applied for TLDs as "closed generic" TLDs.
  4. Provide an analysis as to whether the public interest and principles of international law are served by adopting a clear approach regarding 'closed generic' gTLDs.
  5. Provide a report to the New gTLD Program Committee informed by the comments received and analysis conducted, including alternatives to addressing this issue.
Section II: Background

Following the publication of the gTLD applications in June 2012, concerns have been brought to ICANN's attention regarding some applications for strings which are labelled as "closed generic." These applications are considered problematic by some due to the proposed use of the TLD by the applicant, e.g., using the TLD in a manner that is seen as inappropriately exclusive, particularly in the sense of creating a competitive advantage. These applications have been the subject of public comments and Early Warnings.

Many of the communications link the issue of registration restrictions for a TLD with the Code of Conduct (Specification 9 to the gTLD Registry Agreement). However, it should be clarified that the Code of Conduct refers to registry-registrar interactions, rather than eligibility for registering names in the TLD. Rather than the Code of Conduct, the true issue of concern being expressed appears to be that in certain applications, the proposed registration policies are deemed inappropriate by some parties.

The New gTLD Program has been built based on policy advice developed in the GNSO's policy development process. The policy advice did not contain guidance on how ICANN should place restrictions on applicants' proposed registration policies, and no such restrictions were included in the Applicant Guidebook.

Defining a "generic" category of strings is a complex undertaking as strings may have many meanings and have implications for several languages. However, there are mechanisms built into the program (e.g., objection processes, GACprocesses) as a means for concerns about specific applications to be considered and resolved as they arise.

Recent correspondence has expressed concerns about the potential impact on competition and consumer choice, as well as phrasing the issue in terms of potential impact on the public interest. The New gTLD Program Committee considers it important to understand all views and potential ramifications relating to "closed generic" TLDs.

Section III: Document and Resource Links
Section IV: Additional Information
None

(*) Comments submitted after the posted Close Date/Time are not guaranteed to be considered in any final summary, analysis, reporting, or decision-making that takes place once this period lapses.


FINAL VERSION TO BE SUBMITTED IF RATIFIED

Please click here to download a copy of the PDF below.

 

FINAL DRAFT VERSION TO BE VOTED UPON BY THE ALAC

Final: 01/03/2013 02:51 UTC   

On the whole, the ALAC does not believe that unrestricted closed generics provide public benefit and would prefer that TLDs -- especially for strings representing categories -- were not allocated in a way that would lock out broad access to sub-domains. Some members of At-Large believe, on principle, that all closed generics are harmful to the public good. Others believe that, while not necessarily being beneficial to end users, closed gTLDs should be allowed as simply being consistent with existing practice for lower-level domains.

However, in developing this response to the Board's request, the ALAC found the issue to be far more nuanced than the above hard positions would suggest. There may be innovative business models that might allow a closed TLD to be in the public interest. An example might be a registry that makes 2nd level names available at no cost to anyone, but retains legal control over them. This is similar to the model used by Facebook and many blog hosting sites. Allowance should be made for applicants interested in widespread sub-domain distribution that do not require domain-name sales as a source of revenue, or for other forms of sub-domain allocation.

Whether a generic-word string is used with its generic meaning or in some other context may also be relevant. The fictitious but famous computer manufacturer, Orange Computers Inc. using the TLD ".orange" might be acceptable, while the same string used as a closed TLD by a California Orange Growers Cooperative (and not allowing access to orange producers from Florida or Mediterranean and South American countries) might well be considered unacceptable.

Allowing this nuanced approach would likely involve a case-by-case review of how a TLD will be used and how its sub-domains will be allocated. Moreover, it would require a contractual commitment to not change that model once the TLD is delegated.

In summary, the ALAC believes that completely uncontrolled use of generic words as TLDs is not something that ICANN should be supporting. However, some instances of generic word TLDs could be both reasonable and have very strong benefits of just the sort that ICANN was seeking when the TLD space was opened. Such uses should not be excluded as long as it can be established that they serve the public interest.


FIRST DRAFT SUBMITTED

This is not so much a first draft as a series of notes and comments I've created while considering the issue. I have a definite bias in my approach to this and these comments will certainly need balance with others who don't quite share my take - Evan



On the At-Large view towards “closed generic” new gTLDs.
 
I am undertaking to try to gel the point(s) of view of the At-Large community on the issue of what has come to be known as “closed generic gTLDs” While gTLDs are by definition generic, this issue specifically refers to the application for TLD strings that may be considered to be dictionary words (and usually, more importantly, potential names of categories or sectors of trade) that the applicant does not want to sell through the usual indirect channel of registrars (and, as applicable, registrar resellers).
 
What follows will start as free form notes and points, that I will hope to gel together into a statement that makes some sense. All input is welcome.
 
(Note: I have personally been very vocal in my lack of opposition to closed generics, but I will attempt to incorporate other views which have been more opposed. I also notice that, as I think about it, that there is a substantial difference between “lack of opposition” and “support”, and that there is actually less divergence than might appear.
 
I will also strive to keep the assembly of comments focused on the point of view of end-users, which may be far different, in this case,  from those of registrants.
 


 
 
I find that the attitudes towards them can be distilled into two general camps:
 
#1: Closed generics are harmful
Closed generics are bad for the public good, and enable monopolies on common words that should not be allowed as a matter of public policy. They will cause end-user confusion by suggesting that a single entity is the definitive authority over a whole category of goods and services (banks, beauty, books). This practise propagates existing existing bad practice by ICANN; here is an opportune time to revert a bad tradition and made the DNS responsive to the needs of the public rather than those of the domain industry.

#2: Closed generics are not great, but not harmful either
Closed generics are at worst benign and at best offer a potential for innovation and allocation of domains beyond the constraints of the traditional ICANN registry-registrar model. From a matter of public policy, ICANN allowing generic words to be privately controlled at the top level is merely consistent with allowing for private ownership of generic words at second and lower level domain names. It is the self-serving constraints on gTLD expansion asserted by existing industry that pose the greatest threat to truly disruptive use of internet domains to advance the public benefit.

Generally, my observation has been that most of At-Large -- and a substantial majority of those within ALAC and At-Large leadership who have expressed a preference -- are in line with the “not harmful” viewpoint.
 
Additional thoughts from Evan:

  • The “harmful” PoV appears to be driven by principle while the “not harmful” PoV seems more driven by evaluating real outcomes

 
Additional thoughts from Carlton
 

  • But as a fundamental fact, the ability to demonetize generic strings at this stage delegitimizes the framework of existence of the entire domain name system to date and undermines ICANN’s entire existence
  • It defines an a priori determination of supportable business models for the domain name system, undeniably intervene in the market by picking winners and losers and is an impermissible selection of business models to favour



Response by Evan to an email from Roberto Gaetano February 21:


It is my take that the public has already been conditioned to understand -- during the decades of the Internet's expansion from technical toy to universal tool -- that the ownership of domain names is not to be trusted.
The public by now EXPECTS that books.com -- or books.ANYTHING -- does not point to a generic open resource, but instead is privately owned. It does not take a huge leap from that understanding to work with/around a reality in which *.book is also privately controlled. For the life of the Internet to date, .com has been considered the "default" by browsers, search engines and the public -- and yet everything in there is held by whoever was first and/or made the highest offer.
The new gTLD program is just an escalation up one level of all the paradigms, philosophies and biases currently affecting second-level domains. The so-called "fair access" rules mean that a new gTLD cannot create innovative allocation rules that circumvent the existing registrars; the existing powers have entrenched themselves well. But, more importantly here, what is also carried over into the TLD program is the concept of names as private property rather than public resource.
> The real question is: in which cases should an operator have exclusive control on a subset of the domain name system, like a TLD.

To the extent that "a subset of the domain name system" also extends to second-level domains, this question was answered decades ago: strings can be private property and are open to the highest bidder. Exceptions have been granted to trademark owners and some (but not all) geographic names, and there is a queue forming for those who want additional restrictions (the Olympics and Red Cross, treaty organizations, GAC whims. and who-knows-what else).
To complain that .books should not be privately owned while books.com should be, is IMO hypocrisy. Same regarding health.com vs .health. There may be semantic differences, and emotional differences, and the domain industry will certainly bemoan lost opportunities to sell, resell and speculate. But there are no differences that functionally matter to end users. Only newcomers to the Internet do not know that the name system is corrupt, and they learn quickly. Realizing that the ownership and use of generic TLD strings is as corrupt as the existing practice in second-level domains will be a surprise to nobody.
Internet consumers are less stupid than consumer groups believe them to be. Nobody goes to ANY domain named "books" expecting it to be a definitive resource. That will not change with the advent of the TLD. People will find their books using search engines or trusted brands such as "amazon", "google" or "kobo", at locations that are actively sought rather than found by accident.
I fully agree that end users would prefer that generic words -- a top AND lower levels -- when used as Internet strings would point to generic areas. But that time has long passed. We have  nonprofits in .COM, a single government monopolizing .GOV, and casinos in .NET. The marketing for .NGO promises what I'd thought what .ORG was supposed to be in the first place. ccTLDs, which are beyond ICANN's control, are (as a whole) a mess. And second level domains have no restrictions on closed versus open, indeed the majority are closed by current definitions.
I understand and sympathize with the sentiment behind the call for an end to "closed generics". But I would not trust whatever mechanism could be concocted to judge that magically .HEALTH is a generic string worthy of special protection from private use but .SPORT is not.
After all, they're just names and pointers, not the content itself.
If consumers are confused by the rollout -- and they will be -- it will be by the sheer diversity of new names and possibly the breaking of application software -- much the same kind that is caused by, say, introducing new area codes in one's phoning area. But it will not be because of anyone pre-supposing that the .BOOKS TLD is being operated as a public service by librarians and then being shocked that it's not. We are WAY past that level of innocence, and new people coming into first time use find out quickly.
- Evan




  • No labels

12 Comments

  1. Thanks to Olivier for the URL another point of view that needs to be integrated in:

    http://atlarge-lists.icann.org/pipermail/gtld-wg/2013q1/001218.html

    Roberto Gaetano 
    Thu Feb 21 17:19:48 UTC 2013

    Sorry to chime in a bit late in the process, but I think that there are a
    couple of points we need to take into consideration before finalizing a
    statement.
    
    To me, there's no point in raising matters of principle like a generic
    public interest vs. the right of businesses to exploit resources to make
    money: we need to give clear examples of the impact of the decisions on new
    gTLDs, which I will try to do below. But first, let me start observing that
    the two questions asked are so much interconnected that it does not make
    sense to answer them separately.
    
    The real question is: in which cases should an operator have exclusive
    control on a subset of the domain name system, like a TLD.
    
    The answer is, IMHO, a new question, which is: what would consumers and
    internet users naturally expect to find under that TLD?
    
    One example is brand TLDs. There is agreement that, for instance, IBM should
    be given exclusive use of .ibm. Why is this? Does IBM have "ownership" on
    the character string "ibm"? No, I would argue that it is to avoid consumer
    confusion, in case users look for some domain under .ibm. I believe that IP
    lawyers have argued this over and over again, in particular to claim
    protection over strings ibm.xyz. As a matter of fact, if we consider the
    computer manufacturer Apple, it is clear that they could be given "control"
    on .apple, even though it is also very clear that they do not own the
    character string "apple", which has also other meanings. So, in the case of
    brands, exclusive use of a brand name is justified in order to avoid
    consumer confusion.
    
    Let's now examine the case of a generic word, let me make the example of
    "health", as this is a specific case that is already under our evaluation
    for a possible objection. What would the average internet user or consumer
    reasonably expect to find under .health? I would expect to find general
    authoritative information about human health, and I would believe that this
    is what generally users would expect to find. So, consumers would be upset
    if they find only partial or biased information, in particular if it is
    described as authoritative, complete and impartial. It might, and I believe
    it will, result in consumer confusion.
    
    This might be an extreme case, because of the importance of the health to
    all of us, but the same reasoning applies to many other generic terms.
    
    What would you expect to find under .book? Personally I would expect to find
    information on any book in the world, maybe in some cases in clearer form,
    in other cases with less skillful presentation, maybe in some cases in
    languages or script I do know, in other cases I don't, but what I do not
    expect is to find books only of a specific editor, or commercialized by a
    specific commercial chain.
    
    So, in an ideal world, the operator of the .book TLD must be open for
    registration under .book to all editors, libraries, distributors, and all
    for profit and not for profit organizations and activities that diffuse
    information about books.
    
    This should be based not on a matter of principle about being against
    private ownership of public resources like common words, but based on the
    need to avoid confusion of the internet users and to protect the consumers.
    If we pursue the former argument, we will end up in a confrontation with the
    people who claim (rightfully so, IMHO) that business shall be allowed to
    make money with the domain name system.
    
    I strongly believe that business as a whole will get a huge benefit if
    internet resources are administered in a way that consumers are not
    confused. To allow control of a TLD with a "closed" model by one entity,
    therefore excluding access to the same TLD to other entities that internet
    users would reasonably expect to find in that TLD, might end up in being a
    huge benefit for one business entity, but having a very bad impact to the
    rest of the business in that area. Ultimately, I believe that it will be
    business as a whole that will also suffer.
    
  2. A response to Roberto's point: would this not be putting ICANN in the position of making decisions based on content?  I.e The content that a user would expect to find.

    1. Anonymous

      Sorry for replying as Anonimous to Avri's question.

      Isn't any closed TLD based on content by definition? Wouldn't any user who accesses any domain under .ibm expect to find information about IBM and not random things?

      Cheers,

      Roberto

      1. Hi Roberto, Hi Roberto, In the arguments I have seen, the referral to "closed" TLDs has less to do with content than with the allocation of subdomains. If Amazon were to obtain the TLD of dot-book, I have no doubt that it and its subdomains will indeed be about books. But that is not what the complaints are about. The argument against Amazon owning dot-book is that it won't allocate subdomains through the current channel of registrars, and would deny access to speculators and/or competitors. People wouldn't be able to privatise, squat and auction subdomains as they could in "open" domains. (This is why registrars are amongst the loudest opponents of "closed generics". ) In the creation of the Applicant Guidebook the domain industry succeeded masterfully in forcing TLD applicants to make a binary choice - go through registrars and enable speculation, or maintain control over the namespace by not reselling anything. ICANN is now paying the price for that awful, innovation-hostile policy.
  3. Anonymous

    Hi, Evan.

    I am sure that if Amazon will operate .books the contents will be about "some" books (most probably, the books you could eventually get from Amazon, i.e. not "all" books - which is my concern). Incidentally, this is orthogonal to the issue about honesty and ethics of registrars.

    My comment was in reply to Avri's observation that ICANN will make a choice on contents if it decided to delegate a TLD that will be "all about books". My reply is that by delegating any brand TLD ICANN would implicitely make a choice about contents: .ibm would reasonably be "all about IBM" (at least all what IBM itself would like to make public). I don't see the difference.

    Cheers, R.

     

  4. Anonymous

    I'd like to know whether at-large is going to adope any uniform policy regarding close generic TLDs OR selectively opposes some close generic strings (like dot Health) but does not care the other closed generic (like dot Book). In the second case, there should be criteria established to select these strings. Also I'd like to know whether the discussion right here is treated as the comments that may impact the new gTLD Review Group's objection decisions stil.

    Hong

      

  5. Hong,

     

    I think as much as anything that is what we are trying to figure out.  So far, I can find much commonality in At-Large postions, so I am not sure what the WG could recommend to the Board.  And trying to define the different postions is a fine exercise, but I am not sure how much it adds to the discussion.

    I don't think this has anything to do with the Review Group, but rather with the community comment call.  the RG is int its own space and the Working Group is only involved in the processes of the RG, not its substance.

    avri

  6. Roberto,

    I guess when I go to a named site, I expect to find whatever it is the designers of the site have put there.  Sometime it is what I am looking for, sometimes it isn't.  

    And while it is true that going to .book, I would expect something about books, I would not be surprised if I found online bookies or discussions about the difficulties of writing the stories for musicals.

    Going to a brick and mortar example,  When I walk into a store called Books, yes I expect to find books and would be surprised to find auto parts. But I expect to find the books the owner wants to sell me.  I do not expect to find books from every publisher or on every subject.  I expect to walk in the door, look around and then figure out if I am intersted in sticking around and maybe even buying something.  Why would I expect my on-line expereince of .book to be any different?

    Then again in the UK, when I walk into a store called Boots, I no longer expect to find boots.

    Avri

    1. Anonymous

      But either (a) any number of bricks and mortar stores could call themselves 'Books' OR (b) none could (instead calling themselves anything from 'Aardvark Books' to 'Zebra Books' or perhaps even 'Amazon Books').

      The point is no single organisation gets to be 'Books' thus forcing everyone else to say "we're 'Aardvark' (we sell books)".

      Alan K

      Scotland

  7. Let the record show this position.  The argument against rests on high principle. Allegedly. My default position is to doubt.  So absent recognition of what is and the willingness to tear it all down and consign it to the rubbish bin of history, what we have here is 'all sound and fury, signifying nothing'.  Unless and until the reasons given for objecting begins:

    1) "The domain name systems valorize words, strings and combinations of characters, known and hitherto unknown, to the property of one person".  Or words to that effect.

    AND 

    2) "closed generic strings illegally expropriates public goods to private profit, refining and extending the business model that currently drives the domain name system." Or words to that effect.

    AND

    3) "Current business models fraudulently extract $ from the global public and explicitly rejects public ownership of words, strings, and various combinations of characters, known and unknown." Or words to that effect.

    AND

    4) "'This conduct is a criminal rejection of global public rights." Or words to that effect. 

    AND

    5) "ICANN is complicit and, as such, is and remains a continuing criminal enterprise". Or words to that effect.

    If constructs of the kind above are not at the top of every recital I see opposed, then I couldn't possibly swallow or successfully navigate the logic that says a privatized string - much less only SOME! - at the top level is inherently bad but at second level, a mild irritation, at best.

    The expectation beggars reason. 

    Carlton

  8. 1st March, 2013

    Memo

    To: ALAC and At Large

    From: Salanieta Tamanikaiwaimaro


    Submissions to the ALAC on Impact of Closed Generics

    Dear All,

    These are some reflections on closed generic TLDs.

    Background

    A “Closed Generic” is a TLD that is a generic term, but domains within that TLD will not be sold to the public.

    Today, there are 22 generic TLDs. These include .COM, .BIZ, .INFO and .NET. Domain names within today’s generic TLDs are available for purchase by the general public. Generic TLDs that are available for purchase by the general public are NOT closed generic TLDs.

    When ICANN held its open application process in June 2012, there were many applicants for Top Level Domains for both branded and generic terms. For example, there were applications filed to create the .BMW Top Level Domain, the .DOT, Top Level Domain, the .SEARCH Top Level Domain, and the .SHOP Top Level Domain. Some of the applicants intend to sell domain names within their proposed new Top Level Domains to the public, while others do not intend to sell domain names within their proposed new TLDs to the public.

    The litmus test in my mind is what is the impact on global public interest? The Affirmation of Commitments (AoC) by the United States Department of Commerce (DOC) and ICANN clearly specify the promotion of competition, consumer trust and consumer choice.[1] There are two ways of examining the situation, one is by looking at the closed generic applications and the other is to look at it from the standpoint of ICANN which is beholden under the AoC. For the purposes of the questions that are being asked of the ALAC, I would like to submit a dual approach to viewing the issues.

    Issues

    The issues that arise are as follows:-

    1. Would the endorsement of “Closed Generic” Applications create a situation or a series of situations whether now or in the future that will restrict competition?
    2.  Would the endorsement of “Closed Generic” Applications create a situation where there is a dominant position within the market?
    3. Would the endorsement of the “Closed Generic” Applications create a restraint in trade of a particular market?
    4. Would ICANN be immune from anti-trust liability?

    Traditionally, the prohibition and control provisions laid out in competition rules basically aims to prevent cartelization and monopolization in markets for goods and services. Such developments in markets inevitably harm consumer welfare which competition rules aim to protect. On the same token, there are instances where some agreement may limit competition to allow for social and economic benefits to pass to the other. In order to ensure that such agreements with a net effect of increasing competition can be made, an exemption regime is regulated in competition law and agreements between undertakings in the same level (horizontal) and different levels (vertical) of the market may be left exempt from the prohibition of the competition rules under an exemption system, provided they are not cartel agreements which are, by nature, out of the scope of exemption.

    The Sherman Antitrust Act also referred to as the Sherman Act prohibits certain business activities that federal government regulators deem to be anticompetitive, and requires the federal government to investigate and pursue trusts, companies, and organizations suspected of being in violation.

    On 4 August 2012, the Honorable Philip S. Gutierrez, United States District Judge ruled in Manwin Licensing International S.A.R.L., et al. v. ICM Registry, LLC, et al.[2] that “anti-trust” claims could be filed over controversial .xxx. See: a. ICANN’s Involvement in Trade or Commerce By its terms, the Sherman Act applies to monopolies or restraints of “trade or commerce.” 15 U.S.C. §§ 1, 2. The identity of a defendant as a nonprofit or charitable organization does not immunize that organization from antitrust liability. NCAA v. Bd. of Regents of Univ. of Okla., 468 U.S. 85, 101 n.22 (1984) (“There is no doubt that the sweeping language of § 1 [of the Sherman Act] applies to nonprofit entities.”). To the contrary, nonprofit organizations that act in trade or commerce may be subject to the Sherman Act. Big Bear Lodging Ass’n v. Snow Summit, Inc., 182 F.3d 1096, 1103 n.5 (9th Cir. 1999) (“A nonprofit organization that engages in commercial activity . . . is subject to federal antitrust laws.”). Rather than focusing on the legal character of an organization, an antitrust inquiry focuses on whether the transactions at issue are commercial in nature. Virginia Vermiculite, Ltd. v. W.R. Grace & Co. – Conn., 156 F.3d 535, 541 (4th Cir. 1998) (“We emphasize that the dispositive inquiry is whether the transaction is commercial, not whether the entity engaging in the transaction is commercial.”). “Courts classify a transaction as commercial or noncommercial based on the nature of the conduct in light of the totality of surrounding circumstances.” United States v. Brown Univ. in Providence in State of R.I., 5 F.3d 658, 666 (3rd Cir. 1993). In any circumstance, “[t]he exchange of money for services . . . is a quintessential commercial transaction.” Id. [My own underlining]

    Each of the generic TLDs presents a market and there are generic brands like .blog which if were closed could pose serious threats to freedom of expression for those who wish to register .blog. Article 19 of the International Covenant on Civil and Political Rights (ICCPR) clearly provides for freedom of expression. The threat of limiting or restricting the ability of persons wishing to acquire .blog poses serious harm to the global blogging community and individuals.

    For the purposes of assessing whether closed generic TLDs should be permitted, it is essential to engage in identifying the market for the TLD and whether there is likelihood that a monopoly or oligopoly would be created that could distort the market and prejudice public interest.

    Under the Sherman Act § 2, 15 U.S.C. § 2[3] monopolizing trade is a felony. Under the circumstances where this trade involves foreign nations such as generic TLD applications that have been made by countries outside the US, then Sherman Act § 7 (Foreign Trade Antitrust Improvements Act of 1982), 15 U.S.C. § 6a will apply in relation to conduct involving trade or commerce with foreign nations.  

    There is the possibility that something which is declared open can be later declared closed, depending on market dynamics and how competition is controlled.At the same time, can what is closed today, later be opened?

    The other issue is who regulates the competition of the gTLD market? Is this supposed to be self regulatory where market forces are left to determine how the pendulum swings or does ICANN or the Applicant of the gTLD given discretionary rights to control its respective gTLD market?

     

    However complex these questions, the litmus test for the ALAC is the impact on global public interest and I would propose that the considerations would be:-

    • Is there a visible threat to the global public interest?
    • What is the nature of the threat/challenge?
    • Is there need to “seal off a market” to preserve competition?
    • Are there generic terms where it is in the public interest to be closed?

     



    [1] Clauses 3 and 9.3 of the Affirmation of Commitment between the United States Department of Commerce and the Internet Corporation for Assigned Names and Numbers

    [2] CV 11-9514 PSG (JCGx), United States District Court, Central District of California, see: http://pdfserver.amlaw.com/tal/icann.pdf

     

  9. The suggested changes is in BLUE.

    Suggested Changes to the Draft Statement

    In considering the matter of closed generic Top Level Domains (gTLDs), ICANN is guided by The Affirmation of Commitments (AoC) between the United States Department of Commerce (DOC) and ICANN clearly specify the promotion of competition, consumer trust and consumer choice.[1] It is also worth noting that there are several issues that surface with closed gTLDs and these include but are not limited to the following:

    1. Would the endorsement of “Closed Generic” Applications create a situation or a series of situations whether now or in the future that will restrict competition?
    2.  Would the endorsement of “Closed Generic” Applications create a situation where there is a dominant position within the market?
    3. Would the endorsement of the “Closed Generic” Applications create a restraint in trade of a particular market?
    4. Would ICANN be immune from anti-trust liability?

    On 4th August 2012, a precedent was established in the Manwin Licensing International S.A.R.L., et al. v. ICM Registry, LLC, et al  where the Honorable Philip S. Gutierrez, United States District Judge ruled that antitrust claims could be filed over .xxx.  Under the Sherman Act § 2, 15 U.S.C. § 2[2] monopolizing trade is a felony. Under the circumstances where this trade involves foreign nations such as generic TLD applications that have been made by countries outside the US, then Sherman Act § 7 (Foreign Trade Antitrust Improvements Act of 1982), 15 U.S.C. § 6a will apply in relation to conduct involving trade or commerce with foreign nations. 

    On the whole, the ALAC does not believe that unrestricted closed generics provide public benefit and would prefer that TLDs -- especially for strings representing categories -- were not allocated in a way that would lock out broad access to sub-domains. Some members of At-Large believe, on principle, that all closed generics are harmful to the public good. Others believe that, while not necessarily being beneficial to end users, closed gTLDs should be allowed as simply being consistent with existing practise for lower-level domains.

    Traditionally, the prohibition and control provisions laid out in competition rules basically aims to prevent cartelization and monopolization in markets for goods and services. Such developments in markets inevitably harm consumer welfare which competition rules aim to protect. On the same token, there are instances where some agreement may limit competition to allow for social and economic benefits to pass to the other. In order to ensure that such agreements with a net effect of increasing competition can be made, an exemption regime is regulated in competition law and agreements between undertakings in the same level (horizontal) and different levels (vertical) of the market may be left exempt from the prohibition of the competition rules under an exemption system, provided they are not cartel agreements which are, by nature, out of the scope of exemption.

    However, in developing this response to the Board's request, the ALAC found the issue to be far more nuanced than the above hard positions would suggest. There may be innovative business models that might allow a closed TLD to be in the public interest. An example might be a registry that makes 2nd level names available at no cost to anyone, but retains legal control over them. This is similar to the model used by Facebook and many blog hosting sites. Allowance should be made for applicants interested in widespread sub-domain distribution that do not require domain-name sales as a source of revenue, or for other forms of sub-domain allocation.

    Whether a generic-word string is used with its generic meaning or in some other context may also be relevant. The fictitious but famous computer manufacturer, Orange Computers Inc. using the TLD ".orange" might be acceptable, while the same string used as a closed TLD by a California Orange Growers Cooperative (and not allowing access to orange producers from Florida or Mediterranean and South American countries) might well be considered unacceptable.

    Allowing this nuanced approach would likely involve a case by case review of how a TLD will be used and how its sub-domains will be allocated. Moreover, it would require a contractual commitment to not change that model once the TLD is delegated.

    In summary, the ALAC believes that completely uncontrolled use of generic words as TLDs is not something that ICANN should be supporting. However, some instances of generic word TLDs could be both reasonable and have very strong benefits of just the sort that ICANN was seeking when the TLD space was opened. Such uses should not be excluded as long as it can be established that they serve the public interest.



    [1] Clauses 3 and 9.3 of the Affirmation of Commitment between the United States Department of Commerce and the Internet Corporation for Assigned Names and Numbers