[kictanet] Day 2 0f 10:-IG Discussions- DNS, Root Servers, IP Addresses

[Michuki Mwangi]

Hi Walu, et al,

John Walubengo wrote:
> 
> I hope Michuki has not gone too technical to make most of us silent.

Yes me too :)


> Anyhow, I wish to take this opportunity to thank Mich for his input
> on this discussion.  His technical responses are quite true but they
> remain 'work-arounds'. Work-arounds are similar to someone asking to
> learn HOW to fish but someone responds by PROVIDING the fish - and
> hoping you will forever shut up thereafter because you have had your
> fill anyway.
> 

Using the word workaround(s) is probably IMHO giving the wrong 
impression and so does the analogy :). Lets go back to the history of 
the Internet and start there. For all intentions, the folks who designed 
it did not intend for it to be what it is today. It was a US military 
project supported by scientists. It was later adopted for the academic 
community and now its the bread and butter for the whole world.


> Take for example the point that root-servers are NOT really US
> Government managed but are actually in private sector management.
> Very true, day to day management of these root-servers is largely
> under private sector but these private sector act according to ICANN
> Policies which are subject to US Dept of Commerce (their very own
> Ministry of Trade).
> 

So going back to the history - this is purely a legacy issue. When the 
first contract for ICANN/IANA was being signed with the DOC am not sure 
if i have read anywhere any significant concerns over the contract. In 
any case before ICANN/IANA came into being - everything was run by one 
by name of John Postel!. So why is it that then the ITU being far and 
large did not raise its concerns again at that point in time? - well 
maybe the internet was not mature then as it is now. Now that the 
internet is mature - is it a high time we got involved?. Well maybe it 
is, however, how do we want to get involved? is it by contributing to 
its growth or just its management?. What do i mean, for instance, here 
in the developing world there are few (if any at all) folks involved in 
the Internet Engineering Task Force IETF where Internet protocols are 
developed for the betterment of the Internet. However, we have not been 
left behind when it comes to the discussion of how the Internet should 
be run!. If i was to take the US position, i would also wonder why 
everyone seems to have taken such a huge interest in the running and 
little or no interest in its technical development! actually i would be 
very skeptical to say the least in letting go!. But thats just me :)

> 100 root-servers using anycast technology? Very true. Infact in Kenya
> we have our very own instance of this wonderful technology at KENIC.
> It allows our DNS queries to be locally serviced rather than
> traversing expensive links to the US.  But again, that is still the
> 'provided fish' I mentioned; we do have a solution but as a country
> still lack the oversight to decide on the which, when, where and how
> such solutions shall be deployed, now and in the future.
>

So having said the above - inherently, the DNS has a technical 
limitation of a max of 13 name-servers. In this case, anycast is a 
workaround - but its also a technology that addresses the inefficiencies 
  of the internet protocols.

> International Domains? Thanks to extensive pressure, this is
> happening as Mich reports for the Japs and possibly others to follow
> (i was not aware).  Previously the position was that the internet can
> break up the moment non-latin characters get on board the DNS system.
> Imagine a more representative and proportionate oversight role for
> internet resources:- such solutions would have definitely been on
> board much earlier.
> 

Maybe this could have happened, but only if we thought that the Internet 
was going to be useful tool then. But how many regions do actually take 
the resource seriously as others do?. Those that do i.e Japan, Sweden 
and the US are way ahead in many respects both policy and technical 
development. Thats probably why they even host the Root Servers that are 
outside the US - they were there from the start.

The only way we can get involved, because most of the work starts at the 
technical level before it goes to the policy development and not 
vise-versa. Only the engineers see the technical limitations and propose 
  solutions/workarounds the the policy development and implementation 
process kicks in. For many years IDNs have been under discussion at the 
technical level both at the IETF and IANA. Now that its clear how it 
will be technically implemented, the policy discussions are ongoing who 
will manage the new implementation in a way that preserves the stability 
and security of the Internet as it is today. Because if anything was to 
happen that breaks the Internet as it is today as a result of policy 
implementations, well ....


> IP address management beyond ICANN? i.e. under two or more
> Organisations? Oh yes, I quite agree that this looks like an
> impossible if not outrageous proposal. But that's because we are so
> used to the 'if it ain't broken, do not fix it' phylosophy. But
> introducing competition at IP address and even DNS management level
> could revolutionalise the way ICANN manages this space or resources.
> ITU in particular has been pushing to provide internet citizens with
> optional service at IP and DNS level. What's wrong with splitting the
> IP address range or DNS name space into two and charging ICANN and
> ITU to provide oversight for the management of both? Collaborative
> competition so to speak.
> 

Well for a start, that if two organizations were involved - they would 
both have different policies. The IANA and the RIR's policy was based 
more on aggregation and conservation of the address space. Now with IPv6 
the policy is more on aggregation than conservation. Would they share 
the same policy especially if their member driven policy development 
processes?.


> I hope this is not too technical - techies can hide behind jargon to
> confuse policy makers. 

I would want to believe that the policy makers have clue in the space 
they are developing policy's for or at least its in their area of 
expertise. If thats the case, then there shouldnt be much confusion IMHO :)

Good morning!

Michuki.



-- 
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.