| Comment/Reply Periods (*) | Important Information Links | |||
| Comment Open: | 16 May 2012 | |||
| Comment Close: | 6 June 2012 | |||
| Close Time (UTC): | 23:59 UTC | Public Comment Announcement | ||
| Reply Open: | 7 June 2012 | To Submit Your Comments (Forum) | ||
| Reply Close: | 28 June 2012 | View Comments Submitted | ||
| Close Time (UTC): | 23:59 UTC | Report of Public Comments | ||
| Brief Overview | ||||
| Originating Organization: | ICANN | |||
| Categories/Tags: | Contracted Party Agreements | |||
| Purpose (Brief): | To receive public comment for Board consideration on the Proposed Revised Process for Handling Requests for Removal of Cross-Ownership Restrictions on Operators of Existing gTLDs. | |||
| Current Status: | Existing registries do not currently have a process to seek removal of cross-ownership restrictions related to names in the TLDs they operate. | |||
| Next Steps: | Following Public Comment, staff will summarize and provide to Board for consideration of revised proposal. The Board will then take action on the Proposal. | |||
| Staff Contact: | Kurt Pritz, Sr. VP, Stakeholder Relations | Email: | kurt.pritz@icann.org | |
| Detailed Information | ||||
| Section I: Description, Explanation, and Purpose | ||||
To seek public comment on a proposed process whereby existing registries may ask to amend their agreements so that they can vertically integrate with respect to selling domain name registrations in the TLDs that they operate. | ||||
| Section II: Background | ||||
On 20 June 2011, ICANN approved a process, by which existing gTLD registries could request relief from cross ownership restrictions on registrars to sell domain names in new gTLDs (see http://www.icann.org/en/resources/registries/removal-cross-ownership). The recommendation was based on extensive independent analysis and community discussion. The registry agreements that are currently in place have cross ownership restrictions. The Board resolution approving the process stated, “consideration of modification of existing agreements to allow cross-ownership with respect to the operation of existing gTLDs is deferred pending further discussions including with competition authorities.” That process did not allow the lifting of restrictions on the existing gTLD registries from owning or controlling a registrar that sells domains in their own existing registries (see http://www.icann.org/en/minutes/resolutions-20jun11-en.htm). In preparation for lifting this last set of restrictions, exchanges have occurred with competition authorities that have expressed interest in this issue. With an understanding and consideration of the concerns and issues raised by those authorities relating to market power, ICANN is now ready to post a revised process that would allow existing registry operators to request removal of cross-ownership restrictions for the gTLDs that they operate. Commencing a public comment period on the proposed revised process at this time is intended to solicit community input for Board consideration. The proposed revised process is posted in redline against the process that was approved on 20 June 2011. | ||||
| Section III: Document and Resource Links | ||||
| http://www.icann.org/en/resources/registries/proposed-removal-cross-ownership-16may12-en.pdf [PDF, 304 KB] | ||||
| 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.
Please click here to download a copy of the PDF below.
First Draft (3 June 2012) by Alan Greenberg
The ALAC and At-Large have multiple opinions on whether the removal
of Cross-Ownership Restrictions for gTLD Operators will be to the
benefit or detriment of users, or in fact, the domain ecosystem.
There is, however, a unified position that whatever the environment
is, with certain constraints, there should be a level playing field
for all gTLD operators.
As such, the ALAC supports the removal of cross-ownership constraints
for existing gTLD operators.
Nevertheless, the ALAC does have one concern with the proposal, and
that is the option for existing gTLD operators to transition to the
new gTLD agreement. That transition would be subject to limits
related to competition issues raised by the removal of the
cross-ownership restrictions. The document is silent on other results
of such a transition, and particularly the removal of price caps on
existing operators.
The ALAC does not believe that there is sufficient proof at this time
to indicate that the new gTLD environment will so significantly
change the gTLD market so that price caps are no longer required for
the dominant gTLDs. As such, no change driven by the removal of
cross-ownership restrictions should at the same time remove the price
caps in the current agreements for dominant gTLDs without substantive
community involvement.
Second (Final) Draft (6 June 2012)
The ALAC and At-Large have multiple opinions on whether the removal
of Cross-Ownership Restrictions for gTLD Operators will be to the
benefit or detriment of users, or in fact, the domain ecosystem.
There is, however, a unified position that whatever the environment
is, with certain constraints, there should be a level playing field
for all gTLD operators.
As such, the ALAC supports the removal of cross-ownership constraints
for existing gTLD operators.
However, the ALAC does have one major concern with the proposal. The
option for existing gTLD operators to transition to the new gTLD
agreement would remove cross-ownership constraints, but it would also
have a very significant other effect that has not been the subject of
virtually any public discussion. Specifically, it would remove the
price caps in the existing agreements, and that is not something that
should quietly be slipped in without careful analysis.
That transition would be subject to limits related to competition
issues raised by the removal of the cross-ownership restrictions. The
document is silent on other results of such a transition, and
particularly the removal of price caps on existing operators.
The ALAC does not believe that there is sufficient proof at this time
to indicate that the new gTLD environment will so significantly
change the gTLD market so that price caps are no longer required for
the dominant gTLDs. As such, no change driven by the removal of
cross-ownership restrictions should at the same time remove the price
caps in the current agreements for dominant gTLDs without substantive
community involvement.
9 Comments
Olivier Crepin-Leblond
Comment on behalf of Evan Leibovitch:
Olivier Crepin-Leblond
Comment from Titi Akinsanmi:
Olivier Crepin-Leblond
Comment from Evan Leibovitch (on Second and Final Draft):
Olivier Crepin-Leblond
Comment from Jean-Jacques Subrenat:
Evan just brought out an aspect I had not noticed. So I too would request the change he is suggesting.Jean-Jacques.
Olivier Crepin-Leblond
Comment from Alan Greenberg:
Evan, perhaps I am missing something, but here is my take.
The comment says that the ALAC is in favour of removing the cross-ownership constraints. This is not conditional (it is for some of us a statement grudgingly made because of a preference for maintaining ownership limits everywhere - but THAT decision was already irrevocably made).
The concern is not with the removal of ownership constraints, but with one of the methodologies to implement this - namely the wholesale transition to the new gTLD contract form.
The comments says we support removing the cross-ownership rules, but NOT by the wholesale move to the new contract which makes a LOT of other changes, most benign to users, some good (such as thick whois) and some bad, such as the lifting of price caps.
Alan
Olivier Crepin-Leblond
Comment by Carlton Samuels:
On Wed, Jun 6, 2012 at 10:47 AM, Alan Greenberg <alan.greenberg@mcgill.ca> wrote:
Olivier Crepin-Leblond
Comment by Evan Leibovitch:
Ah. OK. Thanks for the explanation.
Olivier Crepin-Leblond
Comment by Alan Greenberg:
Staff has their arms full at the moment, so let's let the statement get posted as currently written. Once that is done, I will revise for voting and later re-posting during the reply period) which ends on the Thursday of the Prague meeting, so plenty of time).
Alan
Matt Ashtiani
On behalf of Christopher Wilkinson:
Vertical Integration Between Registries and Registrars
Commentary on the Vertical Integration PDP Working Group
Initial Report, 23 July 2010
Christopher Wilkinson, in his Personal Capacity*.
I. INTRODUCTION
The following comments are based on only reading the VI PDP WG Initial Report.
I was not able to participate in the meetings, lists and conference calls which preceded the preparation of this report. May I express my admiration for the dedication of many members of the ICANN community to this issue during recent months.
My main impression is that the broad drafting of the WG's Charter and the associated GNSO Resolutions has resulted in a wide range of proposals and counter proposals many of which do not address the primary question which is: whether and, if so, when and how, a new registry should be allowed to act as its own registrar, and subject to what thresholds?
I would recommend that GNSO limit its activity at this stage to answering that primary question. Other matters, and particularly reverse integration between registrars and registries, cross-ownership issues, Single Registrant registries etc. could be dealt with separately and with a more reasonable timescale. It would appear from the Initial Report that there is little prospect of comprehensive consensus on the present basis, within the time lines set by the ICANN Board and the new gTLD procedure.
II. SUMMARY
1. The Working Group was initiated by a request to address the topic of “Vertical Integration between Registries and Registrars”. The Initial Report's title is “Vertical Integration between Registrars and Registries”. That is not quite the same thing, and opens up a range of issues that would not have arisen if the initial question had been adhered to. The Initial Report suggests that some of the contributory authors are more interested in the question of cross-ownership and reverse integration between Registrars and Registries, than in answering the initial question.
But the issue that has to be resolved, soon, is whether newly formed TLD Registries, which are likely to be quite small to start with, or which are designed to address specific user communities, should be required ab initio to implement full Registry/Registrar separation, conformant to the current ICANN business model, or not?
That option does not conform to current ICANN policy, for the very good reason that ICANN policy was designed in the mid-1990's to address the issue of a monopoly registry without a distinct registrar function. However, the integrated business model is not exceptional in the ICANN context. Even today, I believe that there still are ccTLDs which operate on the basis of registering domains directly to registrants.
I would submit that the primary question could be answered now in terms of thresholds for initial numbers of registered domains and/or duration of initial periods of exceptions. Not much more.
Most of the other issues could be dealt with subsequently. It would not be productive – and would be a source of further delay - to link the current round of applications for new gTLDs with fundamental changes in the ICANN business model. Rather, we should be talking about limited exceptions to accommodate initially small start-up registries.
2. ICANN is responsible for the conditions of competition and consumer protection in the DNS. In the mid 1990's when ICANN was initially set up, it became clear that the competition authorities in the US and the EU expected ICANN to fulfill that role. As a result, most of the international experience and expertise in this area now resides in the ICANN community. One should not now expect the official competition authorities to take up parts of that responsibility, nor for ICANN to delegate other parts of the responsibility to – yet to be created – external entities.
More generally, the Initial Report, speaks of determining “ ... whether there has been competitive harm in the domain name market”. (page 5, my emphasis.) That is to set the bar far too low. Competition in the Registry market is intrinsically weak. ICANN should continue to be improving the conditions of competition. To-date this has been undertaken through structural separation and price caps. There may well be other ways of improving the conditions of competition in the registration market but most of the alternative proposals set out in the Initial Report would move the DNS market in the other direction.
3. Some of the contributions to the Initial Report make much of the opportunity for innovation and new business models in the DNS market. At one extreme, these may be “unlimited and diverse” (page 58). That would take matters too far. Most business models conform to public company laws and regulations which pre-determine many aspects of business behaviour. The DNS market is not an exception. It is just that, because of the global character of the Internet, the ICANN community enjoys a greater role in determining what the business models should be, which have been generally determined by the ICANN community itself. There may, indeed, be a need for progressive improvements in ICANN's policies, in the first instance because of internationalisation of the DNS. But general transformations of well-understood business models at this time would not be compatible with the efficient launch of the next round of new TLDs.
4. The Initial Report does not address the question of retroactivity of the various policy proposals which are envisioned. How would the proposed changes be applied to the existing Registry/Registrar businesses?
At this stage, in view of the short delays and lack of documented facts and experience, all changes in current policy for the purposes of the next round of gTLD applications should be:
(a) temporary within time lines and thresholds
(b) reversible and
(c) when confirmed, retroactive.
III COMMENTS ON THE PROPOSALS
The following specific comments on the proposals documented in the Initial Report are neither comprehensive nor exhaustive. They just illustrate on a “case-in-point” basis, the general themes outlined above.
1. JN2 Proposal
This proposal does not address the primary question which was initially addressed to the GNSO and resulted in the PDP and the creation of the Working Group.
It is not at all clear why permitting cross-ownership between Registrars and Registries is relevant to the issue. Normally one would create a rule and then - as an exception- allow for temporary deviations for a limited period, e.g. 18 months.
This proposal does the opposite: a rule would be created and applied, but after 18 months the Registrars could petition for the rule to be waived!
2. Free Trade Proposal
This proposal effectively abandons Registry-Registrar separation, not just as an exception, but as a general rule. This would impose a disproportionate burden on ICANN's other regulatory instruments (auditing, compliance, etc.) and is rather optimistic as to the resulting behaviour of the Registration businesses (they are not “Authorities”) in the public interest. That the proponents of this proposal are sensitive to these shortcomings is evinced by their concerns expressed about the potential for “.. abuse and harms of having integrated control of data.” (page 38).
The proposed changes (a) would be permanent and (b) do not address the question of retroactivity (except to suggest that existing ICANN contracts might require “a few adjustments” (sic).
3. RACK+ Proposal
This proposal, which would appear to be close to the Status Quo, maintains the principle of Registry/Registrar separation, but does not address the primary question, either. The question of scale economies for startups and the issue of “orphans” should be addressed.
4. Competition Authority Model (CAM3)
Here, ICANN would (a) maintain structural separation in general but (b) allow 100% cross ownership and full vertical integration as exceptions. ICANN is enjoined to delegate its decisions to a new Panel (the CESP) and - in the event of difficulties - to delegate to national competition authorities.
There are several problems with this approach:
(a) For more than a decade ICANN and the ICANN community has held the experience and expertise in the area of competition policy for the DNS.
(b) There is not uniform international coverage of competition authorities with the appropriate powers and competences. Even the competition authorities in the EU and the US have little experience or precedent in this field precisely because ICANN has been doing that job.
(c) The delays demanded for responses from the public authorities concerned are not very realistic: it is not so much that a competition authority needs a lot of time to treat a specific case, it is rather that those authorities have to prioritise their cases in terms of the scale of abuse and the availability of alternative recourse. (e.g. ICANN).
It is disappointing that relevant national competition authorities were apparently not participating in any stage of the PDP.
5. IPC Proposal
This proposal is likely not quite as “clear and simple” (page 61) as it might seem:
(a) The proposal would afford a very high level of global Trademark protection for a comparatively low level of actual registration and recognition of the brands in question. Thus, in certain respects, global protection on the Internet would be acquired as against trademark registration in only nine jurisdictions, world-wide.
Consequently the proposal would in practice expand “intellectual property rights beyond that granted by the national governments ...” (page 63).
(b) On the other hand, for a startup Internet company, the proposal would require trademark registration and the associated costs in nine jurisdictions in three different ICANN regions, even before starting business. I wonder how many startups would be able to do that.
(c) The putative benefits to the trademark owners of such a system are not self-evident. For these to be realised in practice, quite large corporate websites would have to be rebranded and redirected. Has the resulting consumer confusion and administrative cost been evaluated by the entities proposing this far reaching change in policy?
(d) The Single User. Single Registry (SUSR) option would justify a separate call for proposals with a different time-line. A separate procedure would have to be in place to verify the respective Trademark claims and to collect audited evidence of the numbers of national or regional registrations. There might well be competing claims – equally substantiated – for the same name that had been trademarked in different jurisdictions or sectors. An appropriate arbitration mechanism might be necessary. Auctions are not an appropriate option because they would bias decisions towards the larger entities which would not support a policy of promoting diversity, choice and competition.
* * *
Such issues illustrate the suggestion made above that by unduly broadening the scope of the PDP, ICANN and the GNSO have raised issues that cannot be resolved within the timescale proposed.
____________________
Christopher Wilkinson
55 rue Charles Quint
1000 Brussels, Belgium
13 August 2010