[governance] Is really Bulgarian Cyrillic .бг (.bg)? - Recommendation

[Daniel Kalchev]

On 04.11.11 07:38, Imran Ahmed Shah wrote:
>
> Dear Friends,
>
> *Context:*
>
> Actually, the visual resemblance of TLD string had not been studied 
> thoroughly and precisely in early stages, when the ccTLDs were being 
> allocated with ASCI codes (Latin Characters). That is why the 18 ccTLD 
> script has confusing similarity in visual text with each other.
>

Not exactly...
The reason these strings have "similarity" problems is because, these 
strings are listed in an internationally accepted list of country codes. 
Those who accepted the list and the "possible confusion" are way, way 
larger and more important in this world than ICANN. In this sense, 
ICANNs stance on this matter is simply arrogant.

The use of existing two letter country codes as ccTLD strings was a 
decision made long before ICANN existed and has never been disqualified. 
Apparently, such existing similarity, within the same (ASCII) script has 
not led to any instability for Internet nor impacted it's security.

> However, this kind of study become the focus  of the DNS stability 
> teams when the IDN ccTLDs were being introduced and they had to 
> exercise their authority for the refusal, easily say, No. According 
> the Fast Track they do not have liability and responsible to justify 
> their decision or recommendations to the public.
>

This is correct. The expert panel says: "We were given specific 
instructions by ICANN and for this given input, we are supposed to 
produce this output". ICANN says: "we did not make this decision. We 
gave the application to the stability panel and they said no".
The specific "instructions" are kept secret. I could speculate that 
today's ICANN staff may not even know what they were.

> That is why, confusions are being developed among public and technical 
> community trust on the transparent decision making process of the 
> ICANN. Internet recognized by ASCI codes (Latin Characters) and simply 
> Cyrillic.бг has resemblance with Latin .6r (digit six + r). I also 
> underact that Bulgarians would not like leave the abbreviated string 
> by selecting alternatively full name of the Country in Cyrillic 
> Language script.
>

It is interesting to note, that digits are not permitted as TLD names.

It is also interesting to note, that until recently, everybody was 
talking about .бг and .br, now more and more talk about .бг and .6r.
Perhaps because we already pointed out that the UNICODE table of 
confusable characters do not list any match for the Latin 'r' and a 
Cyrillic character.


> *Review on Requirement Analysis:*
>
> Now, my question is with the Bulgarians Internet Community that do 
> they really want to go for the cyrillic language script and feel much 
> benefit out of it by having Cyrillic.бг in parallel to the English .bg 
> then what is the solution? (because in Korea, the Korean Internet 
> community do not feel much benefit and advantage to have domain names 
> with IDN ccTLD with Korean Language). And I believe that the public 
> and technical communities and ICANN are unaware with the future 
> (upcoming) prospects and potential of the failures of IDN TLDs framework.
>

During my talks with ICANN staff on the subject, I raised the point that 
one should look in perspective. In the past, it was viewed that IDN TLDs 
should not be similar to ASCII TLDs. However, reality is that the Latin 
alphabet on which ASCII is based is actually an minority alphabet. The 
use of other alphabets is more significant and as Internet becomes more 
and more "internationalized" (which is rather poor choice of 
terminology, by the way), ASCII labels will have less and less 
importance. What is more, ICANNs own great new IDN gTLD initiative will 
spring a lot more IDN new TLDs than ASCII TLDs.
So at some point things will be actually reversed and ASCII TLD 
applications will have to make sure they are not similar with some other 
script's characters. Why not Cyrillic... ;-)

Some background:
I am obviously with the BG ccTLD Registry. As such, in theory I should, 
according to some people, have no interest in supporting the development 
of an IDN ccTLD for Bulgaria. That is of course speculation. Cyrillic is 
deeply involved in the Bulgarian culture -- some even go that far to 
claim that Bulgaria is the originator of the Cyrillic script.
Further, the abbreviation "БГ" is how Bulgarians identify the country in 
our own language. There was no other Cyrillic abbreviation in use for 
that purpose, ever.

When the IDN Fast Track process started, we initiated a number of pools 
in Bulgaria to inquire what the community opinion on the best IDN ccTLD 
would be. All previous pools were more or less informal and it was 
always 'БГ". These pools were formal this time. I am aware of three big 
pools: one done by the BG Registry (you can see the results on 
https://www.register.bg), another done by the Bulgarian Government and 
yet another done by the Uninet Association (who were experimenting at 
that time with a .бг in alternate root environment). The pool made by 
Register.BG was to all BG TLD contact persons -- therefore anyone who 
has had anything to do with a BG domain name. Therefore, we consider it 
most representative of the "community opinion" and it is somewhat 
related to your question. The pool by the Government wasn't very popular 
in participation, perhaps due to the short time frame an d lack of 
enough publicity. All pools suggested the prevalent choice being '.бг' 
and therefore this is why such application was made.

Later, when the applicant was informed of the expert panel opinion, 
there was a second pool made by the Government, with the special 
question "what OTHER string you prefer". The response was overwhelming, 
this time with much greater participation. The prevailing majority of 
answers were:

- we want .бг.
- if we are not going to get .бг, we do not want any IDN ccTLD.

With this public opinion, it is understandable that our Government are 
not looking for any other option anymore.

More specific on your question:

It was always assumed (at least by Register.BG) that all delegations 
under a Cyrillic TLD will be in Cyrillic. This is the whole point to 
have a Cyrillic TLD: to be able to type the entire domain name in 
Cyrillic. You already can register IDN domain names under the ASCII BG, 
such as президент.bg (president.bg), but these are not extremely 
popular, primarily because everyone now knows that you can have IDN.IDN 
names and.. sort of wait for this to happen (ICANN to sour our their 
internal confusion).

In respect of the pending application and eventual assignment of .бг, 
Register.BG has made a proposal to the Government, that should both TLDs 
be handled by the same registry, a form of 'bundling' may be 
appropriate, for example if one registers президент.бг they get 
президент.bg, subject to the applicable restrictions under BG of course: 
that is, you (at least under current rules) cannot have око.bg (око 
being in Cyrillic).

It is also interesting to note, that for many years, registries that 
have implemented IDN registrations, restrict the possible labels so that 
no string confusable labels are possible with (say) Cyrillic and ASCII 
characters. But note: this only applies to exact character match, not to 
"possibly similarly looking in some font". It is sad this is not the 
policy at the root level as well.


> *Recommended Solution:*
>
> In order to resolve this conflict I do not understand that appealing 
> to ombudsman could resolve the conflict, because they provisioning 
> allows to refuse. I would recommend that Internet Community and 
> Internet Governance supporting organizations should write to the ICANN 
> Board and requesting them to authorize ccNSO and GNSO with a mandate 
> to develop a Working Group, and that working groups would address 
> these issues and to sort out the way to compensate IDN TLDs 
> applicant’s to promote Internationalized Domain Name System. And the 
> allocation of the Cyrillic.бг string may be allowed with the 
> recommendation of that specific working group to the board of ICANN.
>
>

This has already happened. In a very ICANN-ish way...

On the San Francisco ICANN meeting, when it became clear that not only 
the Bulgarian case is halted, but also the Greek application and also 
the EU IDN application in Bulgarian and Greek (Bulgarian and Greek as 
well as the Cyrillic and Greek alphabets are official within the EU and 
according to the "EU law" are to be fully supported). Then, a 
sub-working group was created under IDN ccPDP WG1 to produce suggestions 
to ICANN how to handle these cases.
That working group eventually reported at the Dakar ICANN meeting... on 
the EU case only, stating in effect (you will excuse my imprecise 
citation) "well... we found out that those strings are confusingly 
similar, but it appears that they are confusingly similar with one 
another (?). We also discussed the matter with EURid (the EU registry) 
and agree that as long as they register only Cyrillic names under the 
Cyrillic TLD and Greek names under the Greek TLD, everything is fine. 
Therefore we recommend that ICANN approves this application.."

Double standard? Why the Bulgarian and Greek cases were not considered?

Indeed, it may happen that the full IDN ccPDP process will permit 
Bulgaria to have .бг at some point in the future. This is somehow not 
dependent on ICANN or the 'expert panel' opinions.

But then the question will remain: How come the .бг TLD was, let me cite 
from the 'expert panel' report:



.. and, in the future it will be less confusable?

Why was then Bulgaria delayed with it's Cyrillic TLD implementation and 
why was all this several years long attempt to undermine ICANNs 
credibility to the community?


Finally, I would like to comment thus:

If someone (ICANN in this case) is tasked with the heavy responsibility 
to make a decision in situation like this, effectively permitting or not 
a country's own language/script to be used and they are to consider the 
opinion of an expert panel, with whom they have a contract.. and it is 
apparent, that this opinion is accepted by practically no one, then a 
prudent one (ICANN) will simply seek the opinion of another expert 
panel, or two (to make easier choice). That would resolve any and all 
doubt. What is more, ICANN staff is required to do so according to the 
Fast Track Implementation Plan.

There are all kinds of theories and first hand knowledge why all this 
has happened.. but let's give ICANN chance to fix this stupid situation 
and clear their image, before these things go public.

Daniel
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.igcaucus.org/pipermail/governance/attachments/20111104/02fcfaca/attachment.htm>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: gffihbcb.png
Type: image/png
Size: 78810 bytes
Desc: not available
URL: <http://lists.igcaucus.org/pipermail/governance/attachments/20111104/02fcfaca/attachment.png>
-------------- next part --------------
____________________________________________________________
You received this message as a subscriber on the list:
     governance at lists.cpsr.org
To be removed from the list, visit:
     http://www.igcaucus.org/unsubscribing

For all other list information and functions, see:
     http://lists.cpsr.org/lists/info/governance
To edit your profile and to find the IGC's charter, see:
     http://www.igcaucus.org/

Translate this email: http://translate.google.com/translate_t