October 2017 - Gnso-newgtld-wg-wt2

Notes and Action Items: New gTLD Subsequent Procedures PDP WG - Track 2 - 05 October 2017
by Julie Hedlund Oct. 5, 2017

Oct. 5, 2017

Dear Work Track members, Please find below the action items and discussion notes from today’s call. These high-level notes are designed to help Work Track members navigate through the content of the call and are not a substitute for the chat transcript or the recording. The referenced document is available at: https://docs.google.com/spreadsheets/d/1kua4x0sLOXy5ZStMkzqG3oYnbkzbxCNMMIG…. Because of the large volume of comments in the chat room staff did not endeavor to capture excerpts. Please reference to the chat room transcript distributed by the GNSO Secretariat and on the wiki at: https://community.icann.org/display/NGSPP/2017-09-14+New+gTLD+Subsequent+Pr…. Best, Julie Julie Hedlund, Policy Director -------------------------------------------------------------------------- Notes and Action Items: New gTLD Subsequent Procedures PDP WG - Track 2 - 05 October 2017 1. SOIs: Kiran Malancharuvil: I have a new SOI. 2. Closed Generic Discussions: -- Last week we discussed If we allowed closed generics how would we allow objections to be filed against them. -- Difficult to find the standing or where those had a good basis to follow through with. What the leadership did was look at the case the Kathy Kleinman brought up as well as others. -- Looked at whether the objection to the closed generic had standing. If we can look at community objectioins and see the common themes maybe we can find a way to allow for objections. See Matrix: https://docs.google.com/spreadsheets/d/1kua4x0sLOXy5ZStMkzqG3oYnbkzbxCNMMIG… -- If you go down the list all except for one were community based objections. -- We can start to prepare a test, that could be similar to what was used in the last round. .BOOK Case Against Amazon: -- Ignore the standing requirement. Even if they found no standing they looked at the merits of the claim. -- Objector said that if you gave .book to Amazon they will engage in monopolistic practices. Panel says you can't judge an application for a closed generic for monopolistic actions until you have some activity. Can't assume that the closed generics in general will necessarily restrain trade or cause material detriment. Whether they are good or not is a matter of policy, not for the panel. -- Contrast with .mobile decision: In a case such as this where a party has shown it is a community and there is substantial opposition in that community to a particular party having a closed registry on that gTLD, there is a strong likelihood that there is a material detriment. Discussion: -- What will that get us if the GAC just rejects it again? How would we deal with that? -- The GAC has provided its advice. If you look at it the advice basically said -- it didn't outright ban closed generic. It said there should be a finding that the TLD is in the public interest. This is looking at the other side of the coin. Unless there is a likelihood of material detriment it can't be said that it is not in the public interest. Beijing GAC Advice: https://www.icann.org/en/system/files/correspondence/gac-to-board-18apr13-e… -- On .books: Sounds like the decision was exactly right, but our policies don't have to mirror the law. Given the lack of a clear direction in the policies it is not surprising that the panel would turn to the law. If you are talking about monopolistic behaviour then you are using legal terms. We need a clear statement of policy against which a prospective TLD can be measured. Absent that we will get the GAC's idea of what the policy should be and objectors saying what they think the law or policy should be, and panels that will go back to the law. -- Applications that are for the public interest -- the GAC has stated that dictionary terms with proposed exclusive access where it can be determined that there is harm to the public interest. If we can develop clear policy an applicant for a dictionary if the applicant can show it could help the public interest in some way that could help. -- The panel used the AGB for the test. On monopoly they were quoting the objector. That wasn't introduced by the panel. Everyone should read the limited public interest objector -- #8 on the matrix -- .broker. Using the criteria it found that it was not contrary to generally accepted norms -- couldn't be sustained by international law. -- We should look to the law for guidance anyway for policy as we do for other types of names and for the legal rights objections. These decisions tackle the pros and cons we are talking about. -- The community objection wasn't designed for closed generics. Monopolistic isn't a standard. But CTIA did show a material detriment before the fact. Go to the definition of generic word. -- Don't think we are looking at the outcomes from the objections as the path forward. We aren't looking at the community objection as the path forward, we are looking at the evidence presented. -- Having to prove a community up front takes away from the entrepreneurs and innovators. -- Are we looking at domain names and trademarks as the same thing? -- There are ways to take trademark law and apply it to domains when you aren't talking about a trademark. There is the question of whether we are making a policy preference for the use of top-level domains to sell second-level domains to the public. Or do we not show a preference for the second-level domain sales of top-level domains. It seems like there is a definite leaning that we want top-level domains to be used to give different people the right to second-level domains. If there is no principle how do you decide what is a closed generic? Are we only talking about nouns? How about adverbs? I think we need to figure out what we are trying to accomplish or prefer. I think we do need to avoid using the term monopoly. -- People are saying since only one party can use the space -- is it "use" or "register"? In the .music decision from AAIM against Amazon they state it is not possible to conclude that this is a case of likely or concrete damage. Different opinions on the same type of issue. Try to avoid etting a blanket rule without allowing any kind of showing something is in the public interest. -- We've looked at public interest objection, it doesn't quite fit; we've looked at community objections and it doesn't quite fit; we've looked at trademark law it doesn't quite fit. ICANN needs to decide if it won't choose the business model it likes or didn't consistently, if not then it needs to be clear what business model it likes and how it is applied. -- Interesting point: There is nothing in policy that deters the innovative business models or an exclusive operation of a registry. ICANN's core value is the registration of second-level domain names. That doesn't say how the registry has to operate. -- There are considerable misunderstandings of what was in the AGB. Why people said they wanted closed generics was to lock out competition. -- We are discussing having some form of visibility into the business plans within the boundaries being for the public interest. -- Don't think we were favoring one business model over another. We have to consider the public interest. -- We need to be careful about guessing the motivation of applicants. Closed generics provided the opportunity for the registry to innovate with new services. -- The trademark spec 13 discussion is considered outside of this discussion. -- Most people who were preparing applications, and working on the AGB, everyone thought of the business model. One of the decisions cites Avri while she was chairing the committee saying that this was a model that was expected. -- What matters is do we keep the general rule opposing all closed generics, or do we look at the public interest and see if there is a detriment. We should avoid a blanket policy. -- Do want to ask the same question again: Could we as a WG consider that we do agree with the concept of innovative TLDs, and not all should be for the registration by third parties. We need to consider all gTLD types as well. If we were to propose this within the public interest -- showing that the applicant can say how they plan to use this in an exclusive way in the public interest, could we see that as a path forward. -- First, we should not favor as a policy matter models that involve the sale of second-level domains to third parties. As to what is in the public interest I would not say that that model is more favorable to the public interest than other forms of innovation. If we foster the current model we won't end up with innovation. -- We (Australia) considered having .bank as closed and they only people who could use it were registered with our regulator. That was an example of the public interest in giving certainty that yes it was a generic but you could be sure that it was registered and regulated. -- Our challenge is to find the way in between. Evaluating whether something is in the public interest is difficult, but don't see how we can get around that. -- We should take the next step and figure out what the criteria would be for a closed generic that adds some benefit. Brainstorm some criteria that might indicate the level of innovation. -- ICANN operates on an "ask permission" model. prefer that they should be allowed unless they are showing harm. Should have criteria that we don't want -- to protect from. -- A permissionless Internet used to be what we said we believed in. Good to deveope criteria in the negative. -- We are arguing up front, not after the fact -- the threshold should be made up front that you will be operating in the public interest. -- The public interest shouldn't only be community interest. Don't think we should have a not-for-profit idea of what a closed generic might be. -- Not suggesting that we adopt the community-based criteria. -- If we went with the permissionless model could see an analogy with vertical integration. You can vertically integrate but you had to live by a code of conduct to make sure you are not taking too much advantage. Could set up criteria that you have to meet for a closed generic.

1 0

Agenda for WT2 Call at 20:30 UTC 5 October 2017
by Michael Flemming Oct. 5, 2017

Oct. 5, 2017

Dear All, Please note that the agenda for this week's call is as below. 1. Welcome 2. SOI Updates 3. Closed Generic Discussion 4. AOB Please note that this week will be a continuation of our discussion from last week on Closed Generics. In addition to the NGPC decision, there were several objections filed based against closed generics from the 2012 round. One of these, .mobile, was mentioned on the last couple of calls. Turns out there was more than one decision on this issue, but with opposite results. The assignment is to see if we can extract common themes in the findings to create a path forward that may allow certain closed generics to proceed while addressing the potential harms. I look forward to discussing with everyone. Kind regards, Michael Flemming

4 7

For SubPro Sub team 2: Continuing discussion on Closed Generics
by Kathy Kleiman Oct. 3, 2017

Oct. 3, 2017

*** This is intended for SubPro Subteam 2. If it has not reached the right list, could you kindly send me the email address. *** Dear Subteam 2, I’ve been reading the discussion of the last meeting, and want to share some thoughts before the call this week. I hope this helps to inform and share information in this important discussion. 1. “Actual harm” is a purposely impossible standard. Very little requires proof of “actual harm” in the real world – it’s proof of “likelihood of harm” that is the basis of most standards, particularly in trademark law. For example, the test for trademark infringement is a “/likelihood of confusion/.” The test for “imminent danger” is where the danger exists which could “/reasonably be expended to cause death or serious physical harm.” /If I build a swimming pool in my backyard, I don’t have to wait until a child drowns to build the fence – the danger is foreseeable and avoidable – and I must act now. 2. When it comes to a “Likelihood of Harm,” we have tons of evidence of the likelihood of harms of Closed Generics, and a great deal of precedent. A. The protections against monopolization of generic words used in their generic sense date back almost three quarters of a century (1946) in US trademark law and can be recited by every trademark attorney. /"Generic word are the weakest types of 'marks' (and cannot even qualify as 'marks' in the legal sense) and are never registrable or enforceable against third parties. //*Because generic words are the common, everyday name for goods and services and everyone has the right to use such terms to refer to their goods and services, they are not protectable. *//Be aware that if you adopt a generic term to identify your goods or services, you will not be able to prevent others from using it to identify potentially competing products or services."/Basic Facts about Trademarks, USPTO, /https://www.uspto.gov/sites/default/files/documents/BasicFacts.pdf [emphasis added]/ *B. *Thus, the US Trademark Office would never allow one milk producer, milk pasteurizer, or milk distributor to monopolize or trademark the word “milk” for its own purposes. Milk for the limited purpose of a line of young children’s clothing, sure, but never to allow Borden Milk Products to monopolize the word “milk” as opposed to Horizon Organic Milk*. * 2)The high standard of the “likelihood of material detriment” WAS met when CTIA – The Wireless Association ®, representing the Mobile Industry, brought it’s Objection against Amazon for monopolize the gTLD planning to own all .MOBILE domain names to solely support its mobile business, and not that of the industry as a whole. /CTIA - The Wireless Association (R) (USA) vs/ Amazon EU S.A.R.L. (Luxembourg), Decision 1-1316-6133/. CTIA did meet the very high and difficult standard of showing that the Closed Generic application of Amazon for .MOBILE and the Panelist found that Amazon’s ==> “/would create a likelihood of material detriment to the rights or legitimate interests of a significant portion of the community to which the string may be explicitly or implicitly targeted."/ 3)Hundreds of commenters from around the world – a larger and wider spectrum than in any ICANN comment proceeding before – responded to the Board’s call for clarification and understanding of the issues of Closed Generics by stating clearly and precisely the “likelihood of harm” they would suffer if Amazon owned the Internet space of “.BOOK” or Google owned all the domain names of .SEARCH. In the comment period, some of the largest and best known companies in the world explained why Closed Generics were a “restriction of competition to the detriment of consumers.” One was the attorney Martin Broden, an attorney for Inter IKEA Systems B.V. presenting IKEA’s concerns with ICANN: “Preventing others from registering under descriptive gTLDs means restriction of competition to the detriment of consumers. The current proposal is contrary to ICANN’s intention with the launch of new gTLDs, namely to broad the domain name space and to spur competition for the benefit of all.” https://forum.icann.org/lists/comments-closed-generic-05feb13/msg00049.html *Others included Microsoft, the McCarthy Institute for Intellectual Property and Technology Law, Spanish Booksellers Federation, Asia Cloud Computing Association, Yves Rocher, Prudential Insurance Company, even the US Postal Service. The eloquent and informative Microsoft letter is attached. * 4) As was noted during the Subteam discussions, dozens of GAC Early Warnings were filed specifically on Closed Generics. The Governments shared with us the basic principles of competition and trademark outlined above. _Overall:_ We made a great compromise in the original round – we allowed .BRANDS to be Closed – defined as the Registry being the sole Registrant and owning all of the domain names. Thus we created the space for experimentation of the brand, and .AWS, among others, is well underway with their models – all of which tie their New gTLD to their products and services. We clarified the policy of the First Round that Generic gTLDs would be open – and that the Registry would sell domain names and service all within its industry or product/service group. You’ll see attorneys and companies long in the ICANN space tell us that this clarification by the ICANN Board was what they thought the rule was all along. Not new policy, but clarified one. They view the ICANN Board's work barring Closed Genrics as clarifying and restating the policy as originally intended – if a New gTLD Registry is not a .BRAND, then the Registry /cannot own all of the second level domains. /No single business can control all of the domain names of its industry. I hope this bit of history and analysis has been helpful to the discussion. I think the world would be happy to revisit these issues, if needed, but that would delay the next round of New gTLDs which none of us want to do. Best, Kathy

4 3