RE: [jig] Getting started

["Gomes, Chuck" <cgomes@xxxxxxxxxxxx>]

Excellent discussion in this string of email messages.  To provide some added 
context, I think it might be helpful to keep ICANN's core value # 5 in mind: 
"Where feasible and appropriate, depending on market mechanisms to promote and 
sustain a competitive environment."  I personally believe this applies to gTLDs 
and ccTLDs.

In policy work there is often a tendency to over standardize by trying to come 
up with one-size-fits-all solutions.  I believe that that is not always in the 
best interests of users. As you all know, the needs of local users vary from 
country to county and region to region.  In my opinion, a good goal is to try 
to find the right balance between global policies and flexibility for local 
markets to be able to customize applications to their needs.

Internet security and stability are needs that fit clearly into the global 
policy arena.  To the extent that variance can be allowed without risking 
security and stability, I think that is healthy.  

The distinction Edmon makes seems useful to me: "resolving the issue for a 
particular TLD may be different from resolving the issue for universal 
acceptance of IDN TLDs".  As long as security and stability is not hampered, 
allowing local entities to resolve issues that best suit their needs, seems 
like a good idea to me.

I also think that Avri raises a very important question about what are ICANN 
issues and what are not.  I encourage the JIG members to keep that in mind as 
the work evolves.

Chuck

> -----Original Message-----
> From: owner-jig@xxxxxxxxx [mailto:owner-jig@xxxxxxxxx] On 
> Behalf Of Edmon Chung
> Sent: Friday, November 27, 2009 6:44 AM
> To: jig@xxxxxxxxx
> Subject: RE: [jig] Getting started
> 
> 
> I think ICANN community has a role to play in promoting IDN 
> TLD acceptance (even if you think ICANN as a policy 
> coordinator doesn't).
> 
> As you mentioned, it is a rather sticky issue... and as a 
> gTLD operator with a gTLD string of more than 3 characters I 
> can tell you from experience that it is useful for us to go 
> about it with some unity... More specifically, resolving the 
> issue for a particular TLD may be different from resolving 
> the issue for universal acceptance of IDN TLDs (e.g. adding 
> one more TLD to a white list vs. utilizing an algorithm to 
> determine the validity of a TLD vs.
> checking a particular list, etc.)
> 
> So I do think this is an issue of common interest between IDN 
> gTLDs and IDN ccTLDs.
> 
> What we can do about it... perhaps is the discussion we can 
> have in this group later.
> 
> Edmon
> 
> 
> > -----Original Message-----
> > From: owner-jig@xxxxxxxxx [mailto:owner-jig@xxxxxxxxx] On Behalf Of 
> > Avri
> Doria
> > Sent: Friday, November 27, 2009 7:25 PM
> > To: jig@xxxxxxxxx
> > Subject: Re: [jig] Getting started
> > 
> > 
> > Hi,
> > 
> > I think it might be better called global interoperability 
> for IDNxTLDS.
> > 
> > Though in order to bother implementing/deploying/replacing an 
> > application
> that
> > supports these new IDNxTLDs, one does need to accept that there is a
> reason to
> > do so.  This is going to get into the sticky issue of 
> market reason to
> bother, just
> > like IPv6, people will need so see an incentive for changing all of 
> > the
> code they
> > run.
> > 
> > I am not sure, though, what part of this is an ICANN issue.  Not 
> > saying it
> isn't, just
> > not sure I understand why it would be.
> > 
> > IETF needs to make sure that the protocol standards worked.
> > ICANN needs to make sure all of the policy issue related to 
> stability 
> > and
> security
> > of the DNS are in place.
> > 
> > Especially with regard to IDNccTLDs, which are the only 
> ones that will
> exist for
> > the first while, the sponsors of those IDNccTLDS will have 
> the market
> imperative
> > of making sure they work in their market area  - which one 
> can assume 
> > is
> primarily
> > in country.
> > 
> > a.
> > 
> > On 27 Nov 2009, at 04:05, Stéphane Van Gelder wrote:
> > 
> > > Thanks Edmon,
> > >
> > > Are you talking about universal acceptance solely at the technical
> level?
> > >
> > > Stéphane
> > >
> > > Le 27 nov. 2009 à 04:23, Edmon Chung a écrit :
> > >
> > >> This sounds like a good conversation and one other clear item of 
> > >> common
> > interest between IDN ccTLDs and IDN gTLDs is the acceptance of IDN 
> > TLDs by systems around the world.  ICANN has been working on the 
> > universal
> acceptance
> > of TLDs: http://www.icann.org/en/topics/TLD-acceptance/
> > >>
> > >> This issue would clearly be heightened when IDN TLDs are 
> introduced.
> > >>
> > >> Perhaps one of the items this group can discuss is the universal
> acceptance
> > of IDN TLDs.
> > >>
> > >> Edmon
> > >>
> > >>
> > >>
> > >>
> > >> From: owner-jig@xxxxxxxxx [mailto:owner-jig@xxxxxxxxx] 
> On Behalf Of
> Terry L
> > Davis, P.E.
> > >> Sent: Friday, November 27, 2009 5:03 AM
> > >> To: 'Andrei Kolesnikov'; jig@xxxxxxxxx
> > >> Subject: RE: [jig] Getting started
> > >>
> > >> Andrei
> > >>
> > >> I think your suggestions are probably critical to success.
> > >>
> > >> Fahd, As the different entities begin to plan on their 
> > >> implementations
> and
> > script standards, I would encourage all to consider the 
> difficulties 
> > of
> managing
> > spelling/script differences to such critical items such as:
> > >> -          Firewalls
> > >> -          Anti-virus
> > >> -          Load balancers
> > >> -          Virtualized services
> > >> -          Replication services
> > >> -          Etc
> > >>
> > >> Getting the standardized and certified tools to manage 
> these will 
> > >> be
> made
> > more challenging by variations in the scripts especially if 
> the tools 
> > must discriminate by the TLD of the domain name as to how 
> they convert 
> > the IDN names.
> > >>
> > >> Remember that the "Puny code" name translations as stored in the 
> > >> DNS
> are
> > unintelligible to anyone, regardless of your native script. 
>  So good 
> > tools
> will be
> > critical to full functionality.
> > >>
> > >> Take care
> > >> Terry
> > >>
> > >> From: owner-jig@xxxxxxxxx [mailto:owner-jig@xxxxxxxxx] 
> On Behalf Of
> Andrei
> > Kolesnikov
> > >> Sent: Tuesday, November 24, 2009 1:09 PM
> > >> To: jig@xxxxxxxxx
> > >> Subject: RE: [jig] Getting started
> > >>
> > >> Fahd,
> > >>
> > >> Just a common sense thought - less mixing of the various 
> scripts, 
> > >> the
> easier
> > way to launch gTLDs. One of the core success factors of 
> Fast Track is 
> > the
> script
> > restriction. If talking about Latin .arab TLD, you may find the
> comfortable
> > situation in case where there is no Latin allowed for second level
> domains, only
> > IDNs, let it be even extended script table, including all scripts 
> > you've
> mentioned.
> > You may find useful the same trick for any Arabic .IDN - prohibit 
> > Latin
> script for
> > the second level.
> > >> The pressure to gNSO and new gTLDs has a name. This is big 
> > >> businesses
> and
> > trademarks from the Latin-script world and there is nothing 
> wrong with it.
> But this
> > pressure can be minimized by restriction for two forms of IDNs:
> > >> -          Only IDN second level domains in new latin gTLDs;
> > >> -          No Latin second level domains in new IDN gTLDs.
> > >>
> > >> So, this is  question of  smart combination of Latin and 
> other scripts.
> Kind of
> > "Fast Barak" (baracca, Italian - fast construction of the 
> building). 
> > This
> can go as a
> > discussion sub-topic for already mentioned.
> > >>
> > >> If to apply for gTLDs in Cyrillic and understanding of 
> the reality 
> > >> with
> TM
> > protection in Russia (and Cyrillic-use countries), I would 
> recommend 
> > any
> applicant
> > to restrict gTLD only to Cyrillic and second/third/etc domains to 
> > Cyrillic
> script only.
> > >> "Big Cyrillic" is larger than Russian one, but it's OK.
> > >>
> > >> --andrei
> > >>
> > >> From: owner-jig@xxxxxxxxx [mailto:owner-jig@xxxxxxxxx] 
> On Behalf Of
> Fahd A.
> > Batayneh
> > >> Sent: Monday, November 23, 2009 3:01 PM
> > >> To: jig@xxxxxxxxx
> > >> Subject: Re: [jig] Getting started
> > >>
> > >> Hi Edmon,
> > >>
> > >> With respect to your first question, I would suggest discussing 
> > >> both
> items
> > since they might be of use to the IDN community.
> > >>
> > >> With regards to the Arabic script in specific and 
> whether policies
> should be
> > applied alike for cc's and g's, my answer is NO! Since the Arabic 
> > script
> consists
> > of Arabic, Urdu, Farsi, Jawi, and Turkish languages, each community 
> > can
> devise
> > their own language table for their cc. However, as for g's, 
> and since
> various
> > countries using the same script might require different variants or
> digits, a
> > different script table would be used. For example, we in Jordan are 
> > going
> to use
> > only the Arabic alphabets we use as an Arabic speaking 
> community with 
> > no diacritics or variants. Also, we will prohibit digit 
> mixing since 
> > digit
> mixing is not
> > required. However, as for the Arabic table devised for 
> registering the
> .arab
> > domain (for the entire Arab region in Arabic), we might require some
> variants and
> > some possible digit mixing!
> > >>
> > >> We could discuss this issue further in our next telephonic 
> > >> conference
> or f-2-f
> > meeting.
> > >>
> > >> Cheers,
> > >>
> > >> -- FAB
> > >> Sent from Amman, Jordan
> > >>
> > >> On Mon, Nov 23, 2009 at 8:41 AM, Edmon Chung 
> <edmon@xxxxxxxxxxxxx>
> > wrote:
> > >> Hi Fahd,
> > >>
> > >> From your explanation, can I understand it to be 
> interest to add an
> item for
> > discussion on one or both of:
> > >> - IDN language table/policy at the root for IDN TLD strings
> > >> - Consistency of language table/policy applied for 2nd/3rd level 
> > >> domain
> > registrations across TLDs
> > >>
> > >> I understand that you are talking specifically about the Arabic 
> > >> script,
> am trying
> > to generalize what you are saying into an item for 
> discussion for this
> group.
> > >>
> > >> With regards to Arabic script specifically, can you share with us
> perhaps
> > whether you think that the policies should be applied across ccTLDs 
> > and
> gTLDs
> > alike when Arabic IDNs are concerned? and why...
> > >>
> > >> The JIG is intended to be a group to discuss issues that 
> are common 
> > >> (or
> > should be viewed as common or leads to an inter-relation) across IDN
> ccTLDs
> > and IDN gTLDs.  Especially policy implementation issues at the root 
> > and
> those
> > which ICANN (or IANA for that matter) would concern itself with in
> coordinating the
> > security and stability of the DNS.  [[Others please correct 
> or add to 
> > this
> :-)]]
> > >>
> > >> Edmon
> > >>
> > >>
> > >>
> > >>
> > >>
> > >> From: Fahd A. Batayneh [mailto:fahd.batayneh@xxxxxxxxx]
> > >> Sent: Monday, November 23, 2009 2:25 PM
> > >> To: Edmon Chung
> > >> Cc: jig@xxxxxxxxx
> > >> Subject: Re: [jig] Getting started
> > >>
> > >> Greetings Edmon and Members of the JIG working group,
> > >>
> > >> First of all, allow me to introduce myself. My name is 
> Fahd, and I
> represent JO
> > (Jordan). I am one of the additions from the ccNSO.
> > >>
> > >> Since I am a member of the Arabic Script community, one 
> challenge 
> > >> we
> are
> > facing is the issue of digit mixing. Other issues that have been 
> > discussed
> and
> > resolvedare the usage of diacritics, as well as zero-width 
> joiners and
> zero-width
> > non-joiners.
> > >>
> > >> Since I am not aware of other scripts or languages other than the
> Arabic script,
> > and since I am not sure if the items I mentioned are of concern to 
> > this
> working
> > group, just thought of bringing them up in case...
> > >>
> > >> Cheers,
> > >>
> > >> -- FAB
> > >> Sent from Amman, Jordan
> > >>
> > >> On Mon, Nov 23, 2009 at 6:37 AM, Edmon Chung 
> <edmon@xxxxxxxxxxxxx>
> > wrote:
> > >>
> > >> Hi Everyone,
> > >>
> > >> I understand that the ccNSO representatives along with some 
> > >> observers
> are
> > added to this list, and we are ready to get started finally :-)
> > >>
> > >> One of the discussions we had in Seoul (during the GNSO/ccNSO 
> > >> launch)
> was
> > that we anticipate the release of a report from the staff IDN team 
> > shortly (regarding IDN TLD length and variant management). 
> It may be 
> > useful for
> this
> > group to observe some of the outcomes from that effort as well.
> > >>
> > >> I wonder if anyone can provide some update on the release of that
> report?
> > >>
> > >> Based on previous discussions the 2 items that we have 
> identified 
> > >> as
> items of
> > common interest were in fact:
> > >> 1. Length of IDN TLD strings (ccTLD vs. gTLD) 2. Variant 
> > >> implementation at the root (for IDN TLD strings)
> > >>
> > >> Does anyone have any suggestion on other potential items 
> to discuss?
> > >>
> > >> Edmon
> > >>
> > >
> 
> 
>