ICANN ICANN Email List Archives

[gnso-idn-wg]


<<< Chronological Index >>>    <<< Thread Index >>>

Re: [gnso-idn-wg] MP3 recording link for the GNSO IDN -wg teleconference 27 February 2007

  • To: "Bruce Tonkin" <Bruce.Tonkin@xxxxxxxxxxxxxxxxxx>, gnso-idn-wg@xxxxxxxxx
  • Subject: Re: [gnso-idn-wg] MP3 recording link for the GNSO IDN -wg teleconference 27 February 2007
  • From: Tan Tin Wee <tinwee@xxxxxxxxxxxxxx>
  • Date: Wed, 07 Mar 2007 09:13:44 +0800

Hi Bruce,

Thanks for the email. No need to be so formal, calling me
by my family name Tan, just call me, Tin Wee, that's the
given name. Just one of those convention things.
See http://www.apng.org/old/namecard/name-convention1.8.html

> I tend to think that the latter scenario is easier - at least in the
> short term - as the fastest way to set up a new gTLD is probably to use
> an existing TLD service provider (either gTLD or ccTLD) .  If the TLD is
> successful then the funds earnt could be used to build infrastructure
> within the poorer country.

So even though what you say is not being offensive, or what you say is
not even basically wrong, it does open up all the questions about expection
and issues I want to raise below. So I am not responding to you directly so much
as to being reminded use the very good points you made
(which is pretty widespread thinking amongst many of us)
to explain my concerns so that we won't end up with groupthink.

I have discussed this with Adam Peake who has advised me as such:
Policy must be designed to allow for local choice that meets the needs and
wishes of the language community concerned. So a flexible policy, e.g.
- minimum technical standards (a new IDN TLD that doesn't work won't
  help anyone),
- submission fee reasonable for a fighting chance amongst "poorer country"
  operators, e.g. USD less than 10,000 up front pegged to GDP or PPP for the
  country of registration or incorporation or origin (so it won't be prohibitive
  for the poorer nations,
- fees that are either weighted to GDP (etc, many measures for this) etc.etc
are critical to the success of the IDN TLD roll-out.

So I believe that the success of an IDN implementation by ICANN will be
measured by both process and outcome. And the jury will not just be
Westernised people looking at it, but judged also through the eyes of the
man on the bus in Urumqi, or the haunting green eyes of the Afghan girl
in the Nasir Bagh refugee camp in Pakistan. Will IDNs work for them?
And who is it that made it work for them?
http://news.nationalgeographic.com/news/2002/03/0311_020312_sharbat.html

Expectation 1. Overdue IDN and escalated expectations
-----------------------------------------
IDNs used to be, in 1998, an urgent concern that domain names
in local languages be made available to local communities
as soon as possible. A decade later, the stakes and expectations have risen.

The expectation is no longer just that of getting IDNs out
as fast as we can, or as efficiently as we can. Those who
needed IDNs have already carried out work-arounds, studied
English in the past ten years and figured out how to work
the Internet in ASCII. It is therefore a given, that IDNs must
be rolled out expeditiously, and it has been so for the past
10 years. Now, more is expected.

Expectation 2: If IDNs can be launched in my language/script
for my people to use, by my people, it will be a successful launch.
---------
Let's say, few people outside of Myanmar (Burma) speak Burmese
http://www.omniglot.com/writing/burmese.htm or use its
scripts, perhaps, emigrants or exiles in USA, or in neighbouring
countries such as Bangladesh and Thailand or southern China.

Suppose if the Myanmar govt wished to set up their .BURMESE
TLD, almost all of the infrastructure might be operated
out of a "poorer" country, run for their people by their
people (or their military junta in this case). And if trade
and the Internet is limited, relatively speaking, compared
with the rest of the ASEAN countries, demand for .BURMESE
IDN domain names may not be that high.
See http://www.omniglot.com/writing/burmese.htm
http://www.ethnologue.com/show_language.asp?code=mya
Therefore, if the Myanmar government can run a .BURMESE TLD, IDN
implementation would have been a success. It may help open up
the Internet into a country which is now run by a military junta
despite democratic elections. (whether the Internet will be
censored or not is another question which we won't go into).

Expectation 3: that new operators appear and the TLD
operations are not monopolised by a select few and
that there is geographical diversity in where operators
come from (as per GAC)
----------------------
If .TAMIL TLDs are requested by the Tamil diaspora,
one can conceive of the estimated 35,000 of Tamilians in
USA having the means of operating a .TAMIL TLDs, many
of them working for IT companies based in California.
But if despite this, a company based in Tamil Nadu state
in India were to win the bid to run the TLD and run it
with technical proficiency, then the folks doing the IDN
implementation and  policies must have been doing
something seriously right.

But if all of them end up being run by Afilias or Verisign
or whoever perceived to be from the West, then, one might have
some doubts about the processes, policies and preferences of the
system (just citing illustrative examples and not intended
to sound offensive to any of the cited parties), and
arising from taking the easiest way out of a tricky situation.

But having said this, I do accept that it does make a lot
of sense for the fastest deployments to involve experienced
and established existing TLD operators to run the IDN TLDs;
However for them to be favoured in this sensitive day and age,
it might invite the outrage of those who feel the humiliation
at the condescension of such policies. I remember the days in 1998
of helping APNIC CEO Paul Wilson to explain to unhappy China officials that
the fact that they are not issued enough IP addresses despite their
huge population was not because of the competence issue but that
procedures of APNIC have to be followed, and if there is any
competence gap,  APNIC will run courses and workshops to educate
IP address applicants on how to get themselves address blocks that
they need. What we need to see is for the largesse of ICANN to
be spent on educating other people how to get themselves a successful
bid of a registry or registrar licence. This was what APNIC did
and is still doing.

An authority that holds supreme power
over resources that people need, has to engage and be seen
to engage in outreach and assistance and affirmative action
that will help the same people access these resources,
or it will erode that very legitimacy that maintains its authority.


Expectation 4. Of outreach, assistance, support...

 an existing TLD service provider (either gTLD or ccTLD) .  If the TLD is
 successful then the funds earnt could be used to build infrastructure
 within the poorer country

Indeed, and funds already in the budget of ICANN should also be put to good use in educating people on how to build Internet infrastructure especially in the area of how to maintain a best practice TLD operation with stability and security etc, and is not predicate on a successful TLD run by a foreign party on local soil. We're pretty much long past colonialistic or paternalistic approaches in a globalised world today but I do understand that at some point of time in the history of mankind all of major civilisations latest on the top of the totem pole, may have been guilty of such "us-superior-them-inferior" thinking. So outreach, assistance, support etc has to be carried out with sensitivity and humility, in recognising that a particular "poorer" country may even still have visible remnants of a far glorious, technologically advanced and prosperous past worthy of our respect.

 I note that the infrastructure for many ccTLDs are not currently
 operated within the Country associated with the ccTLD.

Indeed, and so, no doubt, while even Singapore might still find it more logical and reasonable with cold hard calculative numbers to outsource technical help for our ccTLD to an external operator (although they still run it in Singapore as Ram would probably clarify), as you have mentioned above, it is still the decision that the country has to take on its own, and not imposed upon by anyone else, however obvious the efficacy and however sensible the idea might be. For that I have no objections to this outsourcing.

Can you clarify what you intend by the first sentence above.
 For example, are you envisaging success menas that all the

Finally, success for ICANN means meeting all the above expectations in concert, choreographed with finesse and diplomatic savvy that a global organisation deserves to display in good measure.

These first baby steps, always the most difficult, would probably
be the most crucial ones in the first two decades of the global
information revolution as we strive to build
an equitable global information infrastructure that is
unprecedented in the history of mankind.

bestrgds
tin wee

Bruce Tonkin wrote:
Hello Tan,

To be specific, if new gTLD awardees do not come from poorer
countries, there must be something wrong with our processes
or procedures. While this may be unfair to the hard work that
we put in, they do have a point.

Can you clarify what you intend by the first sentence above.

For example, are you envisaging success menas that all the
infrastructure might be operated out of a poorer country, or simply that
the organisation awarded the TLD be from a poorer country?

I tend to think that the latter scenario is easier - at least in the
short term - as the fastest way to set up a new gTLD is probably to use
an existing TLD service provider (either gTLD or ccTLD) .  If the TLD is
successful then the funds earnt could be used to build infrastructure
within the poorer country.

I note that the infrastructure for many ccTLDs are not currently
operated within the Country associated with the ccTLD.

Regards,
Bruce Tonkin








<<< Chronological Index >>>    <<< Thread Index >>>

Privacy Policy | Terms of Service | Cookies Policy