ICANN|ARCHIVES

Internet Corporation for Assigned Names and Numbers

At the ICANN Public Forum to be held 12 March 2001 in Melbourne, Victoria, Australia there will be discussion of a proposal to revise the agreements between ICANN and VeriSign, Inc. (formerly Network Solutions, Inc.). The proposal is described in more detail in a topic paper that was posted on 1 March 2001.

Under the agreements between ICANN and VeriSign currently in effect, VeriSign operates the .com, .net, and .org registries until 10 November 2003, but has the contractual right to obtain a four-year extension of that term (until 10 November 2007) in the event it completes the legal separation of ownership of its Registry Services business from its registrar business by divesting all the assets and operations of one of those businesses by 10 May 2001. VeriSign has indicated that to obtain the extension it is prepared to sell its registrar business by 10 May 2001.

The registrar-sale condition was negotiated as part of a set of agreements between ICANN, VeriSign, and the United States Department of Commerce signed in November 1999. Those agreements contain various requirements for structural and operational separation of VeriSign's registry and registrar businesses. The requirements were established to protect registrar-level competition by preventing VeriSign from using its operation of the .com, .net, and .org registries to competitively advantage its registrar business. Because of doubts about the practical effectiveness of those separations, the agreements sought to encourage separation of the ownership of these two businesses by giving VeriSign the right to extend its term until November 2007 by selling either the registry or its registrar.

One significant aspect of the proposal for revision of the contracts now being considered is to replace the registrar-sale condition VeriSign must meet to obtain the term extension with a series of new provisions intended to promote registry-level competition. Under the proposal, the contracts under which the three registries are operated would be separated (with VeriSign giving up operation of .org in December 2002, the agreement covering .net expiring in January 2006, and the .com agreement lasting until November 2007) but VeriSign would not be required to sell its registrar business. (Under the proposal, the structural and operational separation requirements between VeriSign's registry and registrar businesses would remain in place.)

The attractiveness of the present proposal to the Internet community therefore depends in part on whether the increased opportunities for registry-level compeition outweigh any benefits of continuing the registrar-sale condition in promoting registrar-level competition. Thus, the current state of competition in the .com, .net, .org registrar business is an important factor in assessing whether the proposal would benefit the Internet community. To provide a better factual basis for community discussion, the following market-share data has been gathered.

NOTE: The information that appears below is believed to be accurate but is not guaranteed.

A. Percentage, by Registrar, of Total Registrations Under Sponsorship in the .com/.net/.org Registry

As of the end of each quarter, ICANN receives data showing the total number of registrations sponsored by each ICANN-accredited registrar in the .com/.net/.org registry. The percentage of total registrations in the .com/.net/.org regisry sponsored by some of the larger registrars (those typically having more than a 1% share) at the end of the last five quarters was as follows:

Registrar	31 December 1999	31 March 2000	30 June 2000	30 September 2000	31 December 2000
Network Solutions, Inc.	91.71%	73.26%	62.54%	58.06%	52.90%
Register.com, Inc.	4.63%	9.30%	10.29%	11.47%	11.91%
Tucows.com, Inc.	0.00%	1.23%	4.16%	5.89%	7.08%
BulkRegister.com, Inc.	0.36%	4.26%	6.22%	6.48%	6.65%
Melbourne IT, Ltd.	0.72%	2.71%	3.84%	3.24%
CORE Internet Council of Registrars	0.92%	2.66%	3.43%	3.49%	3.46%
Internet Domain Registrars	0.73%	2.12%	2.25%	2.07%	2.11%
Dotster, Inc.	0.00%	0.88%	1.50%	1.64%	1.63%
eNom, Inc.	0.05%	0.25%	0.35%	0.87%	1.30%

B. New Registrations

This data has been supplied by VeriSign. In general, it shows the portion of add-domain commands executed by the registry that are attributable to VeriSign's registrar business. It does not reflect deletions of expired names, transfers to other registrars, etc. The data is shown for the five quarters ending on the dates below:

	Quarter Ending
Registrar	31 December 1999	31 March 2000	30 June 2000	30 September 2000	31 December 2000
Network Solutions, Inc.	70%	37%	35%	39%	35%

C. Net New Registrations

This data is calculated from the registry-supplied registration-counts described under item A above. It shows the proportion of the total registry growth that is attributable to the growth in the number of registrations sponsored by each registrar. The percentages for registrars typically accounting for more than 3% of net new registrations are shown below. Please note that the percentages for VeriSign/Network Solutions may be somewhat depressed due to (a) its deletion of large numbers of unpaid names that it registered in 1999 and early 2000 (VeriSign has been obligated to require pre-payment of registrations only since 10 March 2000); (b) customer transfers of names from VeriSign to other registrars; and (c) the fact that most registrations coming up for renewal (and thus those subject to non-renewal) are legacy registrations sponsored by VeriSign.

	Quarter Ending
Registrar	31 December 1999	31 March 2000	30 June 2000	30 September 2000	31 December 2000
Tucows.com, Inc.	0.00%	4.35%	12.54%	12.62%	15.97%
Register.com, Inc.	24.69%	21.17%	13.14%	16.02%	15.26%
Network Solutions, Inc.	46.61%	26.39%	31.89%	40.66%	14.26%
BulkRegister.com, Inc.	3.12%	14.17%	11.81%	7.49%	7.92%
Melbourne IT, Ltd.	5.67%	7.76%	7.08%	0.88%
eNom, Inc.	0.40%	0.78%	0.64%	2.88%	4.48%
CORE Internet Council of Registrars	6.32%	7.08%	5.63%	3.71%	3.31%

One-time events or other artifacts make it inapproproriate to rely on any single number in the above data. Readers should consider the data as a whole.