Thank you very much for your prompt reply and comments.

First of all, I am glad that ICANN has kept seperate accounting for the new gTLD program, which is clearly a prudent way to handle it. However, if we look at all the income brought in by the program, it will also include the auction revenue, same as the USD $175K a piece for applicants. The only difference is, when there is high demand for a certain TLD, the price went higher due to the law of supply and demand. Thus, not considering this auction revenue was either a mistake at the first place, or not seeing the very nature of it, or both.

Secondly, regarding domain parking, CCT-RT identified seven cases/reasons as the following (CCT-RT's Draft Report, page 33, see attached):

• The domain name does not resolve.
• The domain name resolves, but attempts to connect via HTTP return an error message.
• HTTP connections are successful, but the result is a page that displays advertisements, offers the domain for sale or both. In a small number of cases, these pages may also be used as a vector to distribute malware.
• The page that is returned is empty or otherwise indicates that the registrant is not providing any content.
• The page that is returned is a template provided by the registry with no customization offered by the registrant.
• The domain was registered by an affiliate of the registry operator and uses a standard template with no unique content.
• The domain redirects to another domain in a different TLD.

Furthermore, this Draft Report also states that "nTLDStats reports that, by one measure, about 63% of the domains in new gTLDs are currently parked" (same page as above). By any measure, most of these do not serve ICANN's original purpose of the new gTLD program. (The last case could be either a trademark holder's defensive registration, or merely to broaden its reach into new gTLDs.)

Thus, this report's Recommendation #5 (with High priority) is to "Collect Parking Data". That is, "ICANN should regularly track the proportion of TLDs that are parked with sufficient granularity to identify trends on a regional and global basis" (page 50, same report). Thus, CCT-RT is recommending ICANN itself being the arbiter for "real usage".

Thirdly, I fully agree with you that certain sections of ICANN, primarily from the registry/registrar side, may resist discouraging domain parking out of their own financial interests. However, ICANN, according to its mandate as defined by the Bylaws, is to protect the "public interest". As I understand it, this means foremost to protect the end-users' interest. Otherwise, ICANN will become a DNS industry association and advocator, no longer to be trusted by the public. Or even worse, to become a for-profit organization if ICANN first considers its own financial interests.

Especially for us at At-Large, we are supposed to specifically represent the end-users' interests. After being with ALAC for about a year and half, especially from the experience at CCT-RT, my understanding is, the essence of ICANN's multi-stakeholder structure is to bring all parties representing different interests together and to negotiate in order to reach an agreement that can be acceptible by all. Thus, as end-user/consumers' interest often do not coincide with the industry's interest, our responsibility is to stand out and speak out. My personal view is, this case related to domain parking and speculation is exactly one of such cases that calls upon us for duty.

Regarding ICANN's staff, both at ALAC and CCT-RT, my experience is that they all provide excellent service and assistance, while their personal opinions are fairly neutral when there happens to be disagreements. As to the "fairly paranoid legal staff" of ICANN, I do not have much experience so far. But I expect them to face the truth, including recognizing "significant negative impacts" of the new gTLD program, as well as recognition of harm and liability to trademark holders on defensive registrations.

Having said all the above, I also agree with you that, using some of the auction revenue to provide refunds may encounter resistance from some parts within ICANN. However, I believe about all suggestions for allocating this fund will have a similar situation. So be it. This is why ICANN has us at At-Large and ALAC. Although there is no guarantee that we will win every fight in protecting end-users/consumers' interest, at least we have stood out and spoken out.

----- Original Message -----

From: Evan Leibovitch

Sent: Monday, May 15, 2017 12:08 AM

Subject: Re: [At-Large] Auction Proceeds - where we are and what you can help

On 14 May 2017 at 07:48, Kan Kaili <kankaili@gmail.com> wrote:

First of all, I do not want to exclude other ways of spending the auction revenue, but only to use part of the funds for refund to applicants.

We're mixing things in a way that isn't helpful.

ICANN accounting has separated the gTLD program out, so there is already an understanding of what the fees in excess of expenses are. It is reasonable to advocate for a refund from that that amount.

This email thread regards the disboursement of funds from auction proceeds, which is a completely separate source of funds and not at all related to the delivery of the gTLD expansion program. By its very nature it is above and beyond the official monies collected for administration of the expansion.

However, for those applicants who applied for new gTLDs for REAL usages, ICANN clearly over-charged them. This is not ICANN's fault, but ICANN's original calculation of costs was too high without considering the auction revenue which happened beyond expectation. Thus, we OWE them a refund.

OK. Who is the arbiter of "real" usages?

Therefore, as ICANN's original purpose of the new gTLD program was to facilitate usage of new domain names, refunding those who paid the hefty $175K for real usage of new gTLDs would only be natural. In addition, if ICANN's refund is proportional to the real usage of domain names but excluds those being parked, it could motivate registries/registrars to discourage domain parking in the future.

An interesting notion. But the industry, which comprises more than half of ICANN's policy-making body and financially benefits from speculation, would aggressively resist any such attempt. Also keep in mind that since ICANN's source of income is domain sales, a reduction in speculative domains financially damages ICANN itself. So ICANN as an institution will also oppose any move to reduce domain speculation.

Furthermore, it is already recognized that the new gTLD program also has its down-sides.

It is recognized by some stakeholders. As I suggested above, the domain industry and ICANN itself will vigorously resist any acknowledgement that there is significant negative impact to gTLD expansion. Just look at the pushback you receive in the CCT-RT from ICANN staff.

One of those is trademark holders are often forced to spend money to "defensively register" domain names in new gTLDs, with some of the costs substantial. Thus, using some of the auction revenue to subsidize those who suffered from this new gTLD program would also be reasonable.

This is an interesting concept, but in practice absolutely impossible to implement. Such a subsidy might be interpreted as not only recognition of harm but also a recognition of liability. As such, ICANN's fairly paranoid legal staff would never let this idea get any traction.

Cheers,

- Evan