[council] Output of brainstorming session on lessons learnt from the previous introduction of new gTLDs since 1999

["Bruce Tonkin" <Bruce.Tonkin@xxxxxxxxxxxxxxxxxx>]

Hello All,

Below is a copy of the points made by those present in the meeting in
Washington.  It does not represent any consensus but simply is a list of
points made by participants.   This list was used to help with the
discussion on whether to continue with the introduction of new gTLDs.

Regards,
Bruce Tonkin


Negligible impact on security and stability.

New strings and strings with more than 3 characters that were not
interoperable with End-user application software caused reliability
problems.

Whole system needs to absorb a new TLD across all software before fully
interoperable.

No institutional mechanism to inform technical, software development
community and potential users.

Not enough education that new TLDs have been introduced.   Little
knowledge amongst Internet users of the new TLDs.

Selection and implementation process time consuming, expensive and
unpredictable

Registry-registrar protocol was standardised (EPP)

Sunrise program difficult to design

Limitation on the number added caused problems for other applicants that
met selection criteria

Independent evaluators an improvement after first round

Some Selection criteria not objective, clearly defined, and measurable
enough to allow independent evaluation to be effective

Contracts too constraining to allow a registry operator to evolve their
business model in response to market needs

No guarantee of financial gains from operating a new TLD

Long established TLDs have a powerful legacy advantage over new TLDs.


The switching costs for an existing registrant of a domain name from one
TLD to another is significant.

The legacy TLDs are still continuing to grow strongly in registrations
and at higher rate than the new TLDs.


Selection process was not a good judge of what succeeded in the market.

Selection process doesn't scale.

Individual negotiations of registry agreements after Board approves new
TLD also time consuming.

Discretionary processes can be hijacked politically.

Registry operator business models may be limited by the distribution
channel of all ICANN accredited registrars.

Small TLD is OK if meets the needs of the community that has put forward
and doesn't exclude others that are within that Community.

The new gtlds introduced so far do not yet cater for parts of the
international community that use character sets other than the limited
set from the ASCII character range.   This has also led to a growth in
alternative root implementations and applications work arounds (e.g
browser plug-ins).

A policy is required for the introduction of IDNs at the top level, and
need to consider the political and cultural environments as demand for
these IDNs is increasing.

Core of the Internet adapts faster than the edges of the Internet.

Participation of registries, registrars and resellers, end users
required in testing, and identifying clearly the objectives of the test,
policy implications, and measuring the outcomes of the test.

Need to consider whether to set a price and if so, how price is set in
the registry agreement and how it impacts end-users.

Describe reasoning/objectives behind "proof-of-concept" rounds and
whether objectives of new TLD introductions have been met.

Concern about whether open TLDs have resulted in new registrants
compared to existing registrants in a legacy TLD simply registering to
protect the brand.

For existing registrants in a legacy TLD that register in a new TLD, how
many of these use the new registration to create a separate website, or
a separate user of email, rather than simply use email or URL forwarding
to the existing registration in the legacy TLD, or change their
advertising/marketing materials to explicitly reference the new TLD in
an email or website address.

Registry operators have learnt more about the market for new TLDs which
may assist a new operator when launching a new TLD.