ICANN approves .XXX red-light district for the Internet

Agreed. And ICM made damn well sure that they had the ways and the means to wage a considerable and sustained amount of legal pressure by selling over a quarter million pre-registrations at $75 each, generating over $20M in revenue...

Stefan Fouant

See http://www.rfc-editor.org/rfc/rfc3675.txt.

Thanks,
Donald

Doesn't look to me like anyone is trying to mandate the use of .xxx. I would
be opposed if they were.

Mandate bad. Options good.

-Bill

Two comments from two commenters:

I can't seem to find anyone that would benefit from this, with the exception
of Stuart and ICM's shareholders.

                        ... I expect the board and staff really
really would not want to have to answer questions under oath like "who
did you talk to at the US Department of Commerce about the .XXX
application and what did you say?" and "why did you vote against .XXX
when they followed the same rules as the TLDs you voted for?"

The first assumes that a beneficiary should exist that is distinct from the applicant-sponsor.

In the case of .aero, SITA itself ceased to exist in the form it existed at the time of application. At that time it was a non-profit cooperative "open to anyone operating aircraft for the transport of passengers, mail or cargo and to other organisations whose primary business is in the air transport industry", having 728 members in 2003, 581 of which were airlines. It is now an IT shop.

The beneficiaries of the existence of .aero may be limited to Afilias and the entity known by the initials "SITA", independent of how competent the mission of .aero is executed.

In the case of .travel, a 2004 round sTLD applicant ICANN approved, a post-delegation reorganization took place resulting in significant bulk sales of no observable connection to the travel industry. This situation has not drawn formal attention from ICANN for contractual compliance reasons.

In the case of .jobs, also a 2004 round sTLD applicant ICANN approved, a similar situation has drawn formal attention from ICANN for contractual compliance reasons.

In sum, the absence of beneficiaries other than the applicant or its successor in interest, and the registry services platform operator, for sponsored registries approved in the 2000 and 2004 new gTLD rounds is not an exceptional condition. Neither is it a universal condition, as .cat, .coop, .museum, clearly serve the beneficiaries claimed in their respective applications, and are non-profits operating in the public interest.

The second assumes the principle liability that exists is specific to a single application.

While possible, this fails to place a controversy in its complete context, and assumes an implied pattern of conduct by an agency of government at a point in time reflects a continuous primary issue of that agency. The Bush-Cheney Administration's lack of commitment to accountability and transparency is a matter of record, or gaps in the record, to make the obvious pun. Yet accountability and transparency have been required to implement in the transition from a MoU to a subsequent relationship between a private corporation and the Department of Commerce. The current Administration's public comments began with Deputy Assistant Secretary Anna Gomez' observation that there is "no statutory authority", and continues to Secretary Larry Strickland's observation at Silicon Flatirons that accountability and transparency must be acted upon by June of this year.[1]

The liability, only in theory, untested as yet, whether the specific liability cited above, or a general liability, may include whether ICANN is exercising delegated rule making and is therefore subject to the Administrative Procedures Act of 1946, as are other 501(c)(3)s to which an agency of government has delegated rule making.

Well, that's enough for a Sunday morning sermon on the beneficiaries of sTLDs, whether pew safe or not, and the cloud of liabilities that surround a corporation that manages a contract originally between the Department of Defense and SRI International.

Eric

[1] Keynote Remarks of Assistant Secretary Strickling at the Univeristy of Colorado's Silicon Flatirons Conference | National Telecommunications and Information Administration

                        ... I expect the board and staff really
really would not want to have to answer questions under oath like "who
did you talk to at the US Department of Commerce about the .XXX
application and what did you say?" and "why did you vote against .XXX
when they followed the same rules as the TLDs you voted for?"

The first assumes that a beneficiary should exist that is distinct
from the applicant-sponsor.

On the contrary. Since it is clear that all of the other sTLDs have
failed to attract the predicted support from their nominal
communities, why should a similar lack of support for .XXX make any
difference?

The second assumes the principle liability that exists is specific to
a single application.

While possible, this fails to place a controversy in its complete
context, and assumes an implied pattern of conduct by an agency of
government at a point in time reflects a continuous primary issue of
that agency.

Heck no. I expect that were a case to bring documents to light, they
would show that what ICANN said to the US government was at odds with
what they were saying in public. I know none of us would find that at
all surprising, but we're not a judge looking at the contracts.

R's,
John

                         ... I expect the board and staff really
really would not want to have to answer questions under oath like "who
did you talk to at the US Department of Commerce about the .XXX
application and what did you say?" and "why did you vote against .XXX
when they followed the same rules as the TLDs you voted for?"

The first assumes that a beneficiary should exist that is distinct
from the applicant-sponsor.

On the contrary. Since it is clear that all of the other sTLDs have
failed to attract the predicted support from their nominal
communities, why should a similar lack of support for .XXX make any
difference?

First, you (John Levine) are free to interpret the comments of Stephen Fouant (the author of the first comment) any way you please, his statement (below) suggested to me that he expected to find a beneficiary other than the applicant (consistent with rfc1591 aka ICP-1), and he did not also assert that beneficiaries other than the applicant do not exist for all other sTLD applicants, or perhaps all other gTLD operators.

I can't seem to find anyone that would benefit from this, with the exception
of Stuart and ICM's shareholders.

However, you're free to assert the contrary, though to what is unclear to me.

Next, on what basis do you make the claim that .coop and .cat have failed to attract the predicted support from their nominal communities?

The second assumes the principle liability that exists is specific to
a single application.

While possible, this fails to place a controversy in its complete
context, and assumes an implied pattern of conduct by an agency of
government at a point in time reflects a continuous primary issue of
that agency.

Heck no. I expect that were a case to bring documents to light, they
would show that what ICANN said to the US government was at odds with
what they were saying in public. I know none of us would find that at
all surprising, but we're not a judge looking at the contracts.

I don't think you caught the sense of my point that the transparency and accountability issue may transcend any specific case or controversy, however, as I pointed out, all theories of ICANN liability wait for a first test, and so are all equally hypothetical.

Eric

Next, on what basis do you make the claim that .coop and .cat have
failed to attract the predicted support from their nominal communities?

Arithmetic, mostly. There are 40,000 co-ops in the United States,
160,000 in Europe, and apparently several million world-wide, yet
there are only 6700 domains in .COOP. I would find it hard to say
that under 3% takeup was significant support.

The population of Catalonia is about the same as that of Switzerland
or Hong Kong. There are 47,000 domains in .CAT, over 200,000 in .HK,
and about two million in .CH. Of those 47,000, about 7,000 have DNS
on Nominalia's servers, and spot checking suggests most of those are
parked.

I suppose one could argue in both cases that the existence of any
registrations at all shows "support", in which case .MUSEUM is a
rousing success, too.

R's,
John

PS:

I don't think you caught the sense of my point that the transparency
and accountability issue may transcend any specific case or
controversy, however, as I pointed out, all theories of ICANN
liability wait for a first test, and so are all equally hypothetical.

You're certainly right that it's hypothetical, since as far as I can
recall, no case against ICANN since Karl Auerbach's has gone to trial,
but I don't see how this disagrees at all with my theory that ICANN
fears discovery because it would be embarassing. It would show how
opaque and unaccountable ICANN is.

Next, on what basis do you make the claim that .coop and .cat have
failed to attract the predicted support from their nominal communities?

Arithmetic, mostly. There are 40,000 co-ops in the United States,
160,000 in Europe, and apparently several million world-wide, yet
there are only 6700 domains in .COOP. I would find it hard to say
that under 3% takeup was significant support.

Do you attach any significance to the restriction that the .coop operator has to use non-cooperatives as sales channels and the primary means of relations with cooperatives as registrants?

I do, I thought we had an initial exemption from the registrar requirement at the Montevideo meeting, for .museum and for .coop.

Additionally, do you attach any significance to the absence of a restriction on the incumbent monopoly operator from accepting or retaining registrations from cooperatives?

Note, that cooperatives with registrations in the legacy monopoly name spaces could be, but are not, accounted for revenue purposes, as .coop registrants.

The population of Catalonia is about the same as that of Switzerland
or Hong Kong. There are 47,000 domains in .CAT, over 200,000 in .HK,
and about two million in .CH. Of those 47,000, about 7,000 have DNS
on Nominalia's servers, and spot checking suggests most of those are
parked.

The Nominalia issue is one registrar. The .cat name space has been available for only 5 years, the .hk and .ch name spaces since 1986. The rate of growth for .cat has been 10k/yr for each of five years, and assuming no changes, will reach the relative densities of western European national name spaces.

Given our difference in perspectives, and values, I'll stop now.

Eric

Arithmetic, mostly. There are 40,000 co-ops in the United States,
160,000 in Europe, and apparently several million world-wide, yet
there are only 6700 domains in .COOP. I would find it hard to say
that under 3% takeup was significant support.

Do you attach any significance to the restriction that the .coop operator has to use non-cooperatives as sales channels and the primary means of relations with cooperatives as registrants?

No. They knew about that when they applied.

The application for .COOP is archived on the ICANN web site. They predicted with "90% confidence" that they'd have over 100,000 registrations within four years and with "50% confidence" that they'd have 300,000 registrations. They failed.

Note, that cooperatives with registrations in the legacy monopoly name spaces could be, but are not, accounted for revenue purposes, as .coop registrants.

Hmmn, counting people who've decided not to use .COOP as indications of support for .COOP. That's very creative. You sure you don't work for ICANN?

The Nominalia issue is one registrar. The .cat name space has been available for only 5 years, the .hk and .ch name spaces since 1986. The rate of growth for .cat has been 10k/yr for each of five years, and assuming no changes, will reach the relative densities of western European national name spaces.

Actually, if you look at the registry reports, there was a burst of about 18,000 domains in .CAT the first year, the annual growth rate has been considerably less than 10K/yr and it is if anything slowing down. From the Nov 10 report, the most recent one ICANN has published, to today, the growth is about 1000, which extrapolates to under 3500/yr, so it'll catch up with the nearby ccTLDs several centuries from now, if ever. I can't find the business plan of the .CAT application on ICANN's web site, but I'd be pretty surprised if it predicted numbers anywhere near that low.

Regards,
John Levine, johnl@iecc.com, Primary Perpetrator of "The Internet for Dummies",
Please consider the environment before reading this e-mail. http://jl.ly

This is purely speculative.

The correct answer to the question vis a vis 2007-2010 is that a
process specified by the ICANN by-laws was followed. This included,
significantly, review by an independent non-binding arbitration panel
agreed to by both parties. This took some time.

Shortly after that 3 member panel issued their recommendation, 2 felt
ICANN was violating their own by-laws in rejecting .XXX, one felt (all
this is of course overly brief) the by-laws allowed them to reject or
accept the application.

Then, in December 2010 (Cartagena) the issue was referred to the GAC,
ICANN's Govt Advisory Committee who had some shared power in the final
decision. Or let's say what power the GAC had was then also an issue,
besides their substantive recommendations. And at the San Francisco
ICANN meeting earlier this month the ICANN Board of Directors approved
a motion to move .XXX along (you are free to read the motion.)

That's much of what happened 2007-2010.

AS AN ASIDE I don't know that the mere threat of an "expensive" or
"embarrassing" lawsuit would be very compelling. Most everything you
allude to is practically public knowledge anyhow.

For a $65M/year organization like ICANN lawsuits are perhaps not their
favorite way to spend money but they're not terribly expensive, it's
not a hugely complicated case and would likely only revolve around
whether ICANN adhered to their own by-laws or similar (contracts and
representations with ICM, prevailing law, undue influence, etc.)

Maybe you meant there was a possibility of an expensive judgment?

Well, a lot of things happen between the filing of a lawsuit and a
judgment, including the possibility of out of court settlements of
various sorts which could include simply yielding, i.e., saying ok you
(ICM) can have .XXX.

My guess is if ICM won such a lawsuit, they get .XXX and perhaps ICANN
would have to remunerate them some expenses, and if ICM lost then they
lost, they'd get nothing. But this is all counter-factual.

Arithmetic, mostly. There are 40,000 co-ops in the United States,
160,000 in Europe, and apparently several million world-wide, yet
there are only 6700 domains in .COOP. I would find it hard to say
that under 3% takeup was significant support.

Do you attach any significance to the restriction that the .coop
operator has to use non-cooperatives as sales channels and the
primary means of relations with cooperatives as registrants?

No. They knew about that when they applied.

You are mistaken. This was a lively subject of negotiation involving Louis Touton and the parties. I was involved as well. There was real shock when Louis came back from the Registrar Constituency with the message that rather than the initial registrar-free budget of initial registrations, the working number was _0_.

The application for .COOP is archived on the ICANN web site. They
predicted with "90% confidence" that they'd have over 100,000
registrations within four years and with "50% confidence" that they'd
have 300,000 registrations. They failed.

See above.

Note, that cooperatives with registrations in the legacy monopoly
name spaces could be, but are not, accounted for revenue purposes,
as .coop registrants.

Hmmn, counting people who've decided not to use .COOP as indications
of support for .COOP. That's very creative. You sure you don't work
for ICANN?

In 2007 I consulted for the IANA function, writing some perl code to process the RT queues and generate reports for the IETF, but otherwise, no.

Again, communicating is elusive. Verisign, Afilias, NeuStar and CORE all operate more than a single registry. The original SRS proposal by Kent Crispen, Dave Crocker, Roberto Gateano, and Sylvan Langenbach placed the locus of competition in the registry function. The choice to place the locus of competition in the registrar function does not prevent ICANN from revisiting that choice. The distinction between a registry as a contractual entity, and one or more back end operators, allows a registry to have a registrant as a revenue source, and a party back end operator, not necessarily the same corporate entity or an affiliate of the registry to have the same registrant as a revenue source. Just as Verisign was required to participate in the redelegation of .org, Verisign could be required to revenue share for registries its market power harms, in this case, a registry created for cooperatives.

The Nominalia issue is one registrar. The .cat name space has been
available for only 5 years, the .hk and .ch name spaces since 1986.
The rate of growth for .cat has been 10k/yr for each of five years,
and assuming no changes, will reach the relative densities of
western European national name spaces.

Actually, if you look at the registry reports, there was a burst of
about 18,000 domains in .CAT the first year, the annual growth rate
has been considerably less than 10K/yr and it is if anything slowing
down. From the Nov 10 report, the most recent one ICANN has published,
to today, the growth is about 1000, which extrapolates to under
3500/yr, so it'll catch up with the nearby ccTLDs several centuries
from now, if ever. I can't find the business plan of the .CAT
application on ICANN's web site, but I'd be pretty surprised if it
predicted numbers anywhere near that low.

I'll ask Nacho or Jordi tomorrow morning to comment. You could be right.

Eric

No. They knew about that when they applied.

You are mistaken. This was a lively subject of negotiation involving Louis Touton and the parties. I was involved as well. There was real shock when Louis came back from the Registrar Constituency with the message that rather than the initial registrar-free budget of initial registrations, the working number was _0_.

If their application was predicated on ICANN changing the rules, I can't feel very sorry for them. And in any event, it's a bit much to claim that the difference between 300,000 registrations and 6400 registrations is that people have to find a registrar. If there were really 293,600 people eager to register if they could only find a coopful registrar, I'd expect we'd have a few,

Actually, if you look at the registry reports, there was a burst of
about 18,000 domains in .CAT the first year, the annual growth rate
has been considerably less than 10K/yr and it is if anything slowing
down. From the Nov 10 report, the most recent one ICANN has published,
to today, the growth is about 1000, which extrapolates to under
3500/yr, so it'll catch up with the nearby ccTLDs several centuries
from now, if ever. I can't find the business plan of the .CAT
application on ICANN's web site, but I'd be pretty surprised if it
predicted numbers anywhere near that low.

I'll ask Nacho or Jordi tomorrow morning to comment. You could be right.

It's all in the reports on the ICANN web site, except for the current count which I got by grepping the zone file. No secrets there.

Regards,
John Levine, johnl@iecc.com, Primary Perpetrator of "The Internet for Dummies",
Please consider the environment before reading this e-mail. http://jl.ly

Who is forcing them?

Tony.

My favourite sTLD from the 2004 round is .post

Tony.

Do a 'whois ibm.biz', 'ibm.info', 'ibm.org', or 'ibm.us' ('ibm.net' appears to
be slightly different) and ask yourself why those registrations exist.

The only reason they exist is so that IBM can stick this in the DNS:

ibm.info. 86400 IN TXT "Visit www.ibm.com for information about IBM products and services"
ibm.info. 86400 IN TXT "v=spf1 mx/24 -all"
ibm.info. 86400 IN MX 100 ns.watson.ibm.com.
ibm.info. 86400 IN SOA ns.watson.ibm.com. dnsadm.us.ibm.com. 2010073000 21600 3600 1209600 86400
ibm.info. 86400 IN A 129.42.38.1
ibm.info. 86400 IN NS ns.watson.ibm.com.
ibm.info. 86400 IN NS ns.almaden.ibm.com.

1.38.42.129.in-addr.arpa. 28800 IN PTR redirect.www.ibm.com.

So some miscreant can't register 'ibm.biz' for themselves for lulz and profit.

You don't think miscreants would do that? Might want to look into the checkered
past of 'whitehouse.com' sometime.

What bothers me is that most companies are now going to be forced to
purchase .xxx domains simply to keep someone else from buying it and
sullying the company's image.

Who is forcing them?

Their lawyers.

Regards
Marshall

What bothers me is that most companies are now going to be forced to
purchase .xxx domains simply to keep someone else from buying it and
sullying the company's image.

Who is forcing them?

Their lawyers.

given that domain registration is cheaper than billable hours I don't
see the problem.

Well, you don't need to wait for .xxx you have things like
http://www.radio.co.ck/

Or http://www.budget.co.ck/ ..