<HTML>
<HEAD>
<TITLE>Re: [governance] India's communications minister - root server misunderstanding (still...)</TITLE>
</HEAD>
<BODY>
<FONT FACE="Calibri, Verdana, Helvetica, Arial"><SPAN STYLE='font-size:11pt'>As someone who has consistently expressed opposition to the US unilateral position as regards root zone file authorisation, let me say also that I do not believe Parminder’s scenario 2 is workable.<BR>
<BR>
There is a high risk involved in looking for a better geopolitical location of root zone servers. The risk is that state partners – who would surely want to be involved in selection of sites – would also want a say in changes to the root zone. The risk of a split root increases significantly. I don’t think enough is gained by the change to justify the significant risks involving split roots or political interference.<BR>
<BR>
I come back to my original position – and perhaps the only one where we might get some agreement and also even the possibility of some action. The authorisation role is completely unnecessary, whether carried out by USA or UN or whatever. Please do not transfer it to another body – just remove it. The authorisation is based on recommendations involving a set of very consultative and exhaustive procedures. Once the ICANN processes recommend a change after these consultations, let that be the final authorisation. <BR>
<BR>
<BR>
I can perceive a situation where USA might actually accept that proposition, consistent with increasing independence of ICANN. I cant see a situation where they transfer their authorisation function to any other body.<BR>
<BR>
Ian Peter <BR>
<BR>
<BR>
<HR ALIGN=CENTER SIZE="3" WIDTH="95%"><B>From: </B>parminder <<a href="parminder@itforchange.net">parminder@itforchange.net</a>><BR>
<B>Reply-To: </B><<a href="governance@lists.igcaucus.org">governance@lists.igcaucus.org</a>>, parminder <<a href="parminder@itforchange.net">parminder@itforchange.net</a>><BR>
<B>Date: </B>Wed, 08 Aug 2012 16:46:35 +0530<BR>
<B>To: </B><<a href="governance@lists.igcaucus.org">governance@lists.igcaucus.org</a>><BR>
<B>Subject: </B>Re: [governance] India's communications minister - root server misunderstanding (still...)<BR>
<BR>
<BR>
<BR>
On Wednesday 08 August 2012 12:35 PM, Dr. Alejandro Pisanty Baruch wrote:<BR>
<BR>
<BR>
</SPAN></FONT><BLOCKQUOTE><FONT FACE="Calibri, Verdana, Helvetica, Arial"><SPAN STYLE='font-size:11pt'> <BR>
</SPAN></FONT><FONT SIZE="2"><FONT FACE="Courier New"><SPAN STYLE='font-size:10pt'> Parminder, <BR>
<BR>
<BR>
<BR>
it may be useful to separate your problem into two parts:<BR>
<BR>
<BR>
<BR>
<BR>
1. authorization for changes in the root;<BR>
<BR>
</SPAN></FONT></FONT><FONT FACE="Calibri, Verdana, Helvetica, Arial"><SPAN STYLE='font-size:11pt'> <BR>
</SPAN></FONT></BLOCKQUOTE><FONT FACE="Calibri, Verdana, Helvetica, Arial"><SPAN STYLE='font-size:11pt'> <BR>
Thanks for the kind advice, Alex. In fact, I have insisted repeatedly that I am only dealing with the second part as below. The first part was dealt in an earlier discussion in June with the subject line 'oversight'.<BR>
<BR>
</SPAN></FONT><BLOCKQUOTE><FONT FACE="Calibri, Verdana, Helvetica, Arial"><SPAN STYLE='font-size:11pt'> <BR>
</SPAN></FONT><FONT SIZE="2"><FONT FACE="Courier New"><SPAN STYLE='font-size:10pt'> <BR>
<BR>
<BR>
<BR>
2. operation of the independent root servers, including their submission or not to an outrageously arbitrary and deletereous change in the root.<BR>
<BR>
</SPAN></FONT></FONT><FONT FACE="Calibri, Verdana, Helvetica, Arial"><SPAN STYLE='font-size:11pt'> <BR>
</SPAN></FONT></BLOCKQUOTE><FONT FACE="Calibri, Verdana, Helvetica, Arial"><SPAN STYLE='font-size:11pt'> Yes, this alone is the issue under consideration here.<BR>
<BR>
<BR>
</SPAN></FONT><BLOCKQUOTE><FONT FACE="Calibri, Verdana, Helvetica, Arial"><SPAN STYLE='font-size:11pt'> <BR>
</SPAN></FONT><FONT SIZE="2"><FONT FACE="Courier New"><SPAN STYLE='font-size:10pt'> <BR>
<BR>
<BR>
<BR>
That, I think, will help you parse the apparent contradictions. We all have a problem with the first's asymmetric-power situation; the second is a fail-safe mechanism for the potential excesses of the first. <BR>
<BR>
<BR>
</SPAN></FONT></FONT><FONT FACE="Calibri, Verdana, Helvetica, Arial"><SPAN STYLE='font-size:11pt'> <BR>
</SPAN></FONT></BLOCKQUOTE><FONT FACE="Calibri, Verdana, Helvetica, Arial"><SPAN STYLE='font-size:11pt'> <BR>
So you agree that independence of root server operators indeed serves as a 'fail-safe mechanism for the potential excesses' of the unilateral root changing power with the US. <BR>
<BR>
In that case, you may agree that making MORE sure that the root operators are MORE independent of US gov will make the system MORE fail-safe or capture-resistant. In practical terms I mean what if instead of the present distribution of root server operators, 9 in the US and 3 in US friendly countries, we have these servers distributed in a more geopolitically equitous manner - as I suggested, for a start RIRs of Africa, LA and Asia Pacific get one each, and perhaps one more in each of these continents at a reputed public technical institute. What do you say?<BR>
<BR>
Lets first agree on the need and desirability of such re-allocation, before we go to the question of how to do it. <BR>
<BR>
(apologies for some repeat language from my email to Roland)<BR>
<BR>
parminder <BR>
<BR>
<BR>
</SPAN></FONT><BLOCKQUOTE><FONT FACE="Calibri, Verdana, Helvetica, Arial"><SPAN STYLE='font-size:11pt'> <BR>
</SPAN></FONT><FONT SIZE="2"><FONT FACE="Courier New"><SPAN STYLE='font-size:10pt'> <BR>
<BR>
<BR>
<BR>
(Fail-safe does not mean "it cannot fail"; it means "if it fails it devolves to a safe state", sort of when well designed elevators go out of electrical power they don't sink to the bottom and crash, nor just get stuck; they fall to the next floor down and open the doors)<BR>
<BR>
<BR>
<BR>
<BR>
(As a side: it may be valuable for IT4Change to recruit the assistance of some Internet engineers, for example by forming an all-volunteer Technical Advisory Board, if you don't find this too meddlesome. I've seen such an Rx work wonders in other, similar organizations elsewhere and it's a win-win. If too meddlesome please ignore. Again, happy to be corrected by those more knowledgeable.)<BR>
<BR>
<BR>
<BR>
<BR>
Yours,<BR>
<BR>
<BR>
<BR>
<BR>
Alejandro Pisanty<BR>
<BR>
</SPAN></FONT><SPAN STYLE='font-size:10pt'><FONT FACE="Tahoma, Verdana, Helvetica, Arial"> <BR>
</FONT></SPAN></FONT><FONT FACE="Courier New"><SPAN STYLE='font-size:11pt'> <BR>
</SPAN></FONT><FONT SIZE="2"><FONT FACE="Tahoma, Verdana, Helvetica, Arial"><SPAN STYLE='font-size:10pt'> <BR>
</SPAN></FONT></FONT><FONT FACE="Courier New"><SPAN STYLE='font-size:11pt'>! !! !!! !!!!<BR>
</SPAN></FONT><FONT SIZE="2"><FONT FACE="Tahoma, Verdana, Helvetica, Arial"><SPAN STYLE='font-size:10pt'> <BR>
</SPAN></FONT></FONT><FONT FACE="Courier New"><SPAN STYLE='font-size:11pt'>NEW PHONE NUMBER - NUEVO NÚMERO DE TELÉFONO<BR>
</SPAN></FONT><FONT SIZE="2"><FONT FACE="Tahoma, Verdana, Helvetica, Arial"><SPAN STYLE='font-size:10pt'> <BR>
<BR>
<BR>
<BR>
<BR>
</SPAN></FONT></FONT><FONT FACE="Courier New"><SPAN STYLE='font-size:11pt'>+52-1-5541444475 FROM ABROAD <BR>
</SPAN></FONT><FONT SIZE="2"><FONT FACE="Tahoma, Verdana, Helvetica, Arial"><SPAN STYLE='font-size:10pt'> <BR>
<BR>
</SPAN></FONT></FONT><FONT FACE="Courier New"><SPAN STYLE='font-size:11pt'>+525541444475 DESDE MÉXICO <BR>
</SPAN></FONT><FONT SIZE="2"><FONT FACE="Tahoma, Verdana, Helvetica, Arial"><SPAN STYLE='font-size:10pt'> <BR>
<BR>
</SPAN></FONT></FONT><FONT FACE="Courier New"><SPAN STYLE='font-size:11pt'>SMS +525541444475 <BR>
Dr. Alejandro Pisanty<BR>
UNAM, Av. Universidad 3000, 04510 Mexico DF Mexico<BR>
<BR>
Blog: <a href="http://pisanty.blogspot.com">http://pisanty.blogspot.com</a><BR>
</SPAN><FONT SIZE="2"><SPAN STYLE='font-size:10pt'> LinkedIn: <a href="http://www.linkedin.com/in/pisanty">http://www.linkedin.com/in/pisanty</a><BR>
Unete al grupo UNAM en LinkedIn, <a href="http://www.linkedin.com/e/gis/22285/4A106C0C8614">http://www.linkedin.com/e/gis/22285/4A106C0C8614</a><BR>
Twitter: <a href="http://twitter.com/apisanty">http://twitter.com/apisanty</a><BR>
---->> Unete a ISOC Mexico, <a href="http://www.isoc.org">http://www.isoc.org</a><BR>
. . . . . . . . . . . . . . . . <BR>
</SPAN></FONT></FONT><FONT SIZE="2"><SPAN STYLE='font-size:10pt'><FONT FACE="Tahoma, Verdana, Helvetica, Arial"> <BR>
</FONT><FONT FACE="Courier New"> <BR>
<BR>
</FONT></SPAN></FONT><FONT FACE="Times New Roman"><SPAN STYLE='font-size:12pt'> <BR>
<HR ALIGN=CENTER SIZE="3" WIDTH="100%"> <BR>
</SPAN></FONT><FONT FACE="Tahoma, Verdana, Helvetica, Arial"><SPAN STYLE='font-size:11pt'><B>Desde:</B></SPAN><SPAN STYLE='font-size:12pt'> <a href="governance-request@lists.igcaucus.org">governance-request@lists.igcaucus.org</a> [<a href="governance-request@lists.igcaucus.org">governance-request@lists.igcaucus.org</a>] en nombre de parminder [<a href="parminder@itforchange.net">parminder@itforchange.net</a>]<BR>
<B>Enviado el:</B> miércoles, 08 de agosto de 2012 01:38<BR>
<B>Hasta:</B> <a href="governance@lists.igcaucus.org">governance@lists.igcaucus.org</a>; Norbert Klein<BR>
<B>Asunto:</B> Re: [governance] India's communications minister - root server misunderstanding (still...)<BR>
<BR>
</SPAN></FONT><SPAN STYLE='font-size:12pt'><FONT FACE="Times New Roman"> <BR>
<BR>
Norbert, <BR>
<BR>
<BR>
On Wednesday 08 August 2012 07:08 AM, Norbert Klein wrote:<BR>
<BR>
<BR>
</FONT></SPAN><BLOCKQUOTE><SPAN STYLE='font-size:12pt'><FONT FACE="Times New Roman">I do NOT understand what the debate here is about - discussing the location of the 12, or of the many mirrors - when it is a debate over possible changes in the political control of this system.<BR>
<BR>
Only what happens or does not happen on the Alpha Server makes any difference (and it is replicated down the lines throughout all the sub-systems) I understand. Wrong?<BR>
<BR>
So any question about control of the 12 and the mirrors is only about technical details. If the "control" question is pointing at anything else but the Alpha Server it is not changing anything fundamentally. Correct or wrong?<BR>
<BR>
</FONT></SPAN></BLOCKQUOTE><SPAN STYLE='font-size:12pt'><FONT FACE="Times New Roman"> <BR>
You have asked a good question - what is the debate here :)<BR>
<BR>
You seem to agree with Carlos that the political issue is ONLY vis a vis the control over the alpha server, which we now know is in fact not the alpha server that a new 'stealth server'. All other root servers, including their anycast extensions, simply and ONLY reflect the root zone file, and so it does not matter who controls them. As for location, there has not been any known difficulty to locate new anycasts anywhere. Fair enough. <BR>
<BR>
Now, I will have to take you, and others who may still be with us, to a long discussion on 'US's oversight' over CIRs - chiefly the IANA function, that took place in June on this list. David was greatly involved in it. When I and others argued why US cannot be relied on to have the unilateral authority to change the root file at its will - the MAIN argument by David and others was; the 13, or at least 9, root zone operators will very likely simply refuse to publish a file so changed by the US. This 'system feature' was listed as the MAIN defence that things are not as problematic as some of us are making them to be. McTim, Lee and others made the same argument of the 'independent decision making' by root server operators, to minimise what was seen as the 'scare' over US's fiddling with the root in its own interest. At the end of this email I provide a few quotes from among several on how this single argument was repeated employed. <BR>
<BR>
Whereby, when we argue about the problem with US's unilateral control over the root, the argument of 'independence of root operators' is invoked. Such independence means that the '13 root operators' systems is seen, if required, as being able to go beyond simply reflecting the root zone file. Well, it has to be one of the two;<BR>
<BR>
(1) Either, root operators can and will ONLY reflect the root zone file in the 'stealth server', whatever happens - in which case, we should not use the argument of their deemed independence in discussions on problems vis a vis US's unilateral IANA oversight powers <BR>
<BR>
(2) Or, indeed, at least potentially, root operators can refuse to publish what is considered as an improperly changed file by the US, and support the internet system continuing to work on the basis of the original 'proper' file - whereby, it is useful to redistribute root server operator-ship among agencies that together are more likely to resist US unilateralism. <BR>
<BR>
One of the above two must be true, and both cant be true, because they are logically exclusive arguments. It cant be that (2) is true in a discussion over IANA authority, but it becomes untrue when we discuss distribution of root server operators in a geo-political even and just manner. This alone is my case.<BR>
<BR>
I can accept either (1) to be true, in which case the argument of independence of root server operators to publish what they want should NOT be used in an IANA related argument (David, McTim, Lee et all, are you there :) )<BR>
<BR>
Or I can accept (2) to be true, in which case, I will appeal to Carlos for sympathy to the argument that redistribution of root server operation authority may be useful to be considered, while agreeing that IANA authority is a much more important question. <BR>
<BR>
(To be fair to David, he has said even in the present thread of discussion that 'The diversity of architecture ( of root server operators) and lack of centralized control is seen as a feature as it reduces the opportunities for "capture". If I surmise right, Carlos, and perhaps you, Norbert, do not think this of being of any real significance.)<BR>
<BR>
So, indeed there are real difference of views between, for instance David and Carlos, on the political significance of root server operator's independence (or absence of it) - and thus of political significance of who the 13 root server operators are.<BR>
<BR>
Such independence (or absence of it) of root operators, especially in the face of an eventuality of US's rogue behaviour, thus remains a key political issue, and in good part is the point of debate here. The answer to this question would determine whether it is worth the effort to consider reallocating root server operation authority in a more equitous manner. <BR>
<BR>
parminder <BR>
<BR>
<BR>
<BR>
<BR>
<BR>
</FONT></SPAN><BLOCKQUOTE><SPAN STYLE='font-size:12pt'><FONT FACE="Times New Roman"><BR>
<BR>
Norbert Klein<BR>
<BR>
<BR>
-- <BR>
<FONT COLOR="#151B8D">Norbert Klein<BR>
<a href="nhklein@gmx.net">nhklein@gmx.net</a><BR>
<a href="http://www.thinking21.org">http://www.thinking21.org</a><BR>
</FONT> <BR>
<BR>
<BR>
<BR>
</FONT></SPAN><BLOCKQUOTE><SPAN STYLE='font-size:12pt'><FONT FACE="Times New Roman"> <BR>
This is the only place in which there is NTIA-authorized/controlled<BR>
change in the root (the so-called "IANA function"), and all the other 12<BR>
and the hundreds of Anycast servers just replicate - the Anycast servers<BR>
being replicators of replicators in nearly all cases (except for six<BR>
replicating directly from a.root-servers.net).<BR>
<BR>
A new gTLD/ccTLD will never become alive if NTIA does not give the<BR>
"nihil obstat" to insert it in this file in this "mother of all<BR>
servers", which interestingly (or coincidentally, depending on your<BR>
level of paranoia :)) sits very close to CIA headquarters in Virginia.<BR>
NTIA also must become aware of *any* modification intended in existing<BR>
ccTLD or gTLD records in the root zone file, whatever the Affirmation of<BR>
Commitments says.<BR>
<BR>
If a saboteur explodes this server installation (each one of the 13 is<BR>
actually a cluster for resilience and security), does the Internet stop?<BR>
No, of course, the net of replicators will make sure the Internet<BR>
continues to operate fine. But no more changes in the root, Virginia,<BR>
until the "mother server" is rebuilt in Virginia :)<BR>
<BR>
If there is a worldwide revolt agains the USA regarding the DNS, can the<BR>
Anycast net operate and be modified without resorting to one of the 13<BR>
servers (an Anycast server is by agreement used tied to one of the 12<BR>
"master replicators", the F, I, J and L being the most popular for this)?<BR>
<BR>
Technically, yes, of course, but...hmmm... I think it is better to keep<BR>
a dialogue with the USA instead. :) Aside from the root servers, 16 of<BR>
the largest 20 DNS servers in the planet are in the USA, hosting many<BR>
millions of domain pointers to Web services *worldwide* -- millions of<BR>
websites in Latin America, for example, depend on these servers and<BR>
corresponding hosting services.<BR>
<BR>
Is this talk necessary at all? I think this is abundantly common<BR>
knowledge since the root system's 13 servers started to operate...<BR>
<BR>
frt rgds<BR>
<BR>
--c.a.<BR>
<BR>
On 08/07/2012 02:17 AM, parminder wrote:<BR>
<BR>
</FONT></SPAN><BLOCKQUOTE><SPAN STYLE='font-size:12pt'><FONT FACE="Times New Roman"> <BR>
David,<BR>
<BR>
On Sunday 05 August 2012 10:40 PM, David Conrad wrote:<BR>
<BR>
</FONT></SPAN><BLOCKQUOTE><SPAN STYLE='font-size:12pt'><FONT FACE="Times New Roman"> <BR>
Parminder,<BR>
<BR>
On Aug 5, 2012, at 5:40 AM, parminder <<a href="parminder@itforchange.net">parminder@itforchange.net</a><BR>
<<a href="mailto:parminder@itforchange.net>">mailto:parminder@itforchange.net></a>> wrote:<BR>
<BR>
</FONT></SPAN><BLOCKQUOTE><SPAN STYLE='font-size:12pt'><FONT FACE="Times New Roman"> <BR>
Now, we know that there are three kinds of root servers, the<BR>
authoritative root server (in which changes are made to the root file<BR>
vide the IANA process), 13 root servers and then the any number of<BR>
mirrors that can allegedly be created by making an investment of 3k<BR>
usd .<BR>
<BR>
</FONT></SPAN></BLOCKQUOTE><SPAN STYLE='font-size:12pt'><FONT FACE="Times New Roman"> <BR>
No.<BR>
<BR>
There is a "distribution master".<BR>
<BR>
</FONT></SPAN></BLOCKQUOTE><SPAN STYLE='font-size:12pt'><FONT FACE="Times New Roman"> <BR>
So, well, apologies for referring to the root zone file as the highest<BR>
level of root zone server; I should perhaps simply have said 'the<BR>
highest level of Internet's root architecture'. However, your chastising<BR>
may be biased. Someone, quite unlike me, with deep technical training<BR>
like Daniel said is a recent email;<BR>
<BR>
"As already mentioned, there are hundreds of root server instances.<BR>
Each of these is an actual root server."<BR>
<BR>
Isnt this statement as or more untrue, in a discussion where we are<BR>
mainly speaking about actual 'control' over the root file. The hundreds<BR>
of root servers mentioned above are NOT 'actual root servers'. An actual<BR>
root server is a shorthand for an actual root server operator, who<BR>
exercises control (at least potentially) over the root zone file that he<BR>
publishes. (I learnt this from my earlier discussions with you on the<BR>
IANA authority and the US.) The 'ill-informed' Indian minister seems<BR>
rather better informed than 'technical experts' here on this particular<BR>
issue. He seems to know better which is a true or actual root server and<BR>
which is not. Quote from the same interview where he quite wrongly said<BR>
that Internet traffic flows through 13 root servers (he should have<BR>
said, internet traffic, in a way, gets directed by 13 root servers).<BR>
<BR>
<BR>
"Currently, India's mirror servers reflect the data but without<BR>
mechanisms of control and intervention."<BR>
<BR>
Clearly what some 'technical experts' stress and what they suppress (or<BR>
forget to mention) depends on their techno-political proclivities. Isnt<BR>
it obvious!<BR>
<BR>
In response to my another email, you have asked me to "provide examples<BR>
of supposed 'statements of technical facts' that are ''thoroughly<BR>
wrapped in a certain techno-political viewpoint". Apart from the above<BR>
example, I will try and find others in your email below :)<BR>
<BR>
<BR>
</FONT></SPAN><BLOCKQUOTE><SPAN STYLE='font-size:12pt'><FONT FACE="Times New Roman"> <BR>
(snip)<BR>
<BR>
That's all. There are no special "13" machines that are the "true<BR>
root servers" from which other lesser machines mirror the root zone.<BR>
<BR>
</FONT></SPAN></BLOCKQUOTE><SPAN STYLE='font-size:12pt'><FONT FACE="Times New Roman"> <BR>
Well, you did understand early in this discussion that the argument is<BR>
not about 'true root servers' but about 'true root server operators', so<BR>
why dont we stick to the real point of contestation rather than create<BR>
strawmen and defend against them. From your email of a few days ago<BR>
<BR>
"The concern (as I understand it) is that the administration of<BR>
those root servers is in the hands of 12 organizations, of which 9<BR>
are US-based. " (David)<BR>
<BR>
Yes, true. It is this what we are discussing here, not the network<BR>
latency problem. In that email, you understood the concern right. It is<BR>
about root server operators, and the term '13 root servers' is loosely<BR>
used to mean '13 root server operators'. That is the real issue, and it<BR>
was the issue that bothered the Indian and the African ministers the<BR>
latter being wrongly, if not mischievously, retorted to in terms to<BR>
availability of root server mirrors - a very different issue. Similarly,<BR>
this current discussion is continuously pulled towards the convenient<BR>
description of geographic extensions through mirrors of root servers,<BR>
away from the real issue of 'concentration' (against distribution) of<BR>
power to change root file or resist changes to root file that is with<BR>
the root server operators and none at all with anycast mirror operators.<BR>
<BR>
It is very interesting that when I did that long discussion with you,<BR>
David, on the US's unilateral IANA authority, your almost entire case<BR>
was based on how the root server operators are really independent (which<BR>
is the same thing as saying they have 'power') and this is the insurance<BR>
against any US mischief with the root zone file. However, now when we<BR>
are discussing the power of root server operators, which is<BR>
geo-politically very unevenly distributed, the 'power' with the root<BR>
server operators is sought to be so minimized as to be completely<BR>
evaporated. The focus is repeatedly sought to shifted to how anyone can<BR>
set up a root server and that those who speak about 13 root servers<BR>
(meaning, root server operators) being not distributed well enough are<BR>
merely stupid!<BR>
<BR>
How does what appears to be the 'same fact' take such very different<BR>
manifestations in two different political arguments? This is what I mean<BR>
by 'technical advice' being warped by strong techno-political<BR>
viewpoints. I am not making any personal accusation. I am stating a<BR>
sociological 'fact'.<BR>
<BR>
<BR>
</FONT></SPAN><BLOCKQUOTE><SPAN STYLE='font-size:12pt'><FONT FACE="Times New Roman"> <BR>
(snip)<BR>
<BR>
</FONT></SPAN><BLOCKQUOTE><SPAN STYLE='font-size:12pt'><FONT FACE="Times New Roman"> <BR>
What I see is that, while there are of course clearly very<BR>
significant differences between these three layers or kinds of root<BR>
servers, much of the 'technical input' on this list that I have come<BR>
across seem to focus on the non-difference and greatly underplay the<BR>
difference.<BR>
<BR>
</FONT></SPAN></BLOCKQUOTE><SPAN STYLE='font-size:12pt'><FONT FACE="Times New Roman"> <BR>
As discussed above, the distinction you are making doesn't exist.<BR>
<BR>
</FONT></SPAN></BLOCKQUOTE><SPAN STYLE='font-size:12pt'><FONT FACE="Times New Roman"> <BR>
Well!! See above for the distinction. A clear distinction that you did<BR>
understand and articulate in your earlier email in terms of<BR>
concentration of ability for "administration of those root servers is in<BR>
the hands of 12 organizations, of which 9 are US-based. " There is<BR>
obvious and very important distinction between the 'power' of root zone<BR>
operator and someone operating a mirror. This distinction is the very<BR>
basis of the whole discussion in this thread. But you have easily and<BR>
conveniently dismissed, or minimised, distinctions between the root file<BR>
layer, root zone layer and anycast mirror layer, esp between these two<BR>
latter layers . This is done through a unilateral decision to speak<BR>
about one thing when the other party is speaking about quite another, or<BR>
at least another aspect of the issue - which here is the issue of<BR>
'control' rather than availability of root file for resolving queries.<BR>
<BR>
<BR>
</FONT></SPAN><BLOCKQUOTE><SPAN STYLE='font-size:12pt'><FONT FACE="Times New Roman"> <BR>
</FONT></SPAN><BLOCKQUOTE><SPAN STYLE='font-size:12pt'><FONT FACE="Times New Roman"> <BR>
This I think is politically motivated, though disguised as factual<BR>
neutral/ technical information.<BR>
<BR>
</FONT></SPAN></BLOCKQUOTE><SPAN STYLE='font-size:12pt'><FONT FACE="Times New Roman"> <BR>
Conspiracy theories are tricky things as it makes it difficult to<BR>
communicate.<BR>
<BR>
</FONT></SPAN></BLOCKQUOTE><SPAN STYLE='font-size:12pt'><FONT FACE="Times New Roman"> <BR>
:). I made it clear at the onset that I am trying to argue that when a<BR>
group has strong political inclinations - as the so called technical<BR>
community has - its technical advice gets accordingly wrapped... Call<BR>
it my conspiracy theory, but at least I am upfront. But also (try to )<BR>
see how the technical community sees deep conspiracies in every single<BR>
political utterance from the South. Worse its conspiracy theory is<BR>
further compounded by a 'stupidity theory'. Double insult!<BR>
<BR>
</FONT></SPAN><BLOCKQUOTE><SPAN STYLE='font-size:12pt'><FONT FACE="Times New Roman"> <BR>
(snip)<BR>
<BR>
You misread. The 13 IP(v4) address limitation due to the default<BR>
maximum DNS message size still exists. While there are now ways<BR>
around this limitation (specifically, the EDNS0 extension to the DNS<BR>
specification), these ways are not universally supported and as such,<BR>
cannot be relied upon, particularly for root service.<BR>
<BR>
</FONT></SPAN></BLOCKQUOTE><SPAN STYLE='font-size:12pt'><FONT FACE="Times New Roman"> <BR>
No, I dont think I misread. Just that the fact remains that the number<BR>
13 can be expanded without much difficulty, but you are not too<BR>
interested to explore that direction while I am (again, political<BR>
proclivities intervene). Wasnt introducing multilingual gtlds also<BR>
considered a bit 'difficult to rely upon' just a few years back.<BR>
Finally, political considerations helped get over that unnecessary and<BR>
exaggerated fear. It depended who were taking the decisions, the US<BR>
centric ICANN establishment earlier, but the same establishment with<BR>
some WSIS related fears and cautions in the second instance.<BR>
<BR>
<BR>
</FONT></SPAN><BLOCKQUOTE><SPAN STYLE='font-size:12pt'><FONT FACE="Times New Roman"> <BR>
</FONT></SPAN><BLOCKQUOTE><SPAN STYLE='font-size:12pt'><FONT FACE="Times New Roman"> <BR>
So if indeed it is not, why not breach it and make people of the<BR>
world happy.<BR>
<BR>
</FONT></SPAN></BLOCKQUOTE><SPAN STYLE='font-size:12pt'><FONT FACE="Times New Roman"> <BR>
Even if it were possible, I sincerely doubt everyone having their own<BR>
root server would make the people of the world happy.<BR>
<BR>
</FONT></SPAN></BLOCKQUOTE><SPAN STYLE='font-size:12pt'><FONT FACE="Times New Roman"> <BR>
This is 'the' most important point - whether there is any justification<BR>
at all to increase the number or root servers and/or to reallocate /<BR>
redistribute them in a manner that is politically more justifiable and<BR>
thus sustainable. I will take it up in a separate email.<BR>
<BR>
regards<BR>
parminder<BR>
<BR>
<BR>
</FONT></SPAN><BLOCKQUOTE><SPAN STYLE='font-size:12pt'><FONT FACE="Times New Roman"> <BR>
</FONT></SPAN><BLOCKQUOTE><SPAN STYLE='font-size:12pt'><FONT FACE="Times New Roman"> <BR>
Even within the limit of 13, why not allocate root servers in a<BR>
geo-graphically equitable manner, as Sivasubramanian has suggested,<BR>
especially when it seems to make no difference at all to anyone. Why<BR>
not make all these ill-informed ministers happy.<BR>
<BR>
</FONT></SPAN></BLOCKQUOTE><SPAN STYLE='font-size:12pt'><FONT FACE="Times New Roman"> <BR>
As mentioned in a previous note, the operators of the root servers are<BR>
independent (modulo "A" and "J" (through the Verisign contract with<BR>
the USG) and "E", "G", and "H" (operated by USG Departments), albeit<BR>
each of these operators deal with their root servers differently). How<BR>
root server operators distribute their instances is entirely their<BR>
decision. To date, there has apparently been insufficient<BR>
justification for those root server operators to decide to distribute<BR>
their machines in a "geo-graphically equitable manner".<BR>
<BR>
With that said, there are at least two root server operators ("L"<BR>
(ICANN) and "F" (ISC)) who have publicly stated they are willing to<BR>
give a root server instance to anyone that asks. Perhaps the<BR>
ill-informed ministers could be informed of this so they could be happy?<BR>
<BR>
<BR>
</FONT></SPAN><BLOCKQUOTE><SPAN STYLE='font-size:12pt'><FONT FACE="Times New Roman"> <BR>
I read that there is no central control over the 13 or at least 9 of<BR>
these root servers. Is it really true?<BR>
<BR>
</FONT></SPAN></BLOCKQUOTE><SPAN STYLE='font-size:12pt'><FONT FACE="Times New Roman"> <BR>
Yes. The diversity of architecture and lack of centralized control is<BR>
seen as a feature as it reduces the opportunities for "capture".<BR>
<BR>
<BR>
</FONT></SPAN><BLOCKQUOTE><SPAN STYLE='font-size:12pt'><FONT FACE="Times New Roman"> <BR>
Is the 13 root server architecture not something that is aligned to<BR>
what goes in and from the authoritative root server.<BR>
<BR>
</FONT></SPAN></BLOCKQUOTE><SPAN STYLE='font-size:12pt'><FONT FACE="Times New Roman"> <BR>
Root server architecture is independent of how the root zone is<BR>
distributed.<BR>
<BR>
<BR>
</FONT></SPAN><BLOCKQUOTE><SPAN STYLE='font-size:12pt'><FONT FACE="Times New Roman"> <BR>
If it is, why can these root servers not be reallocated in the way<BR>
tlds have been reallocated. Can they be reallocated or cant they?<BR>
<BR>
</FONT></SPAN></BLOCKQUOTE><SPAN STYLE='font-size:12pt'><FONT FACE="Times New Roman"> <BR>
In practical terms, the "reallocation of a root server" boils down to<BR>
transferring the root server's IP address and telling the new owner<BR>
the zone transfer password.<BR>
<BR>
Before the DNS became a political battleground, root server<BR>
"reallocation" occurred (extremely infrequently) when (a) the person<BR>
to whom Jon Postel "gave" the root server changed employers or (b) the<BR>
assets of the organization running the root server were acquired by<BR>
another company. Today, "reallocation" of a root server would either<BR>
require the existing root server operator voluntarily giving the root<BR>
server IP address to a different organization or that IP address would<BR>
have to be "taken" by eminent domain or somesuch.<BR>
<BR>
<BR>
</FONT></SPAN><BLOCKQUOTE><SPAN STYLE='font-size:12pt'><FONT FACE="Times New Roman"> <BR>
I also read that the it is not about 13 physical root servers, but 13<BR>
root server operators,<BR>
<BR>
</FONT></SPAN></BLOCKQUOTE><SPAN STYLE='font-size:12pt'><FONT FACE="Times New Roman"> <BR>
Well, 12 operators (since Verisign operates two root servers).<BR>
<BR>
<BR>
</FONT></SPAN><BLOCKQUOTE><SPAN STYLE='font-size:12pt'><FONT FACE="Times New Roman"> <BR>
so the number 13 is about the root server ownership points, and not<BR>
physical location points.<BR>
<BR>
</FONT></SPAN></BLOCKQUOTE><SPAN STYLE='font-size:12pt'><FONT FACE="Times New Roman"> <BR>
In the sense that there are 13 IP(v4) addresses that are "owned" by 12<BR>
organizations. Geography is largely irrelevant.<BR>
<BR>
<BR>
</FONT></SPAN><BLOCKQUOTE><SPAN STYLE='font-size:12pt'><FONT FACE="Times New Roman"> <BR>
Therefore what is needed is to reallocate the ownership points in a<BR>
geo-politically equitious manner. As Siva suggests, probably one to<BR>
an Indian Institute of Technology.<BR>
<BR>
</FONT></SPAN></BLOCKQUOTE><SPAN STYLE='font-size:12pt'><FONT FACE="Times New Roman"> <BR>
Somewhat as an aside, my understanding is that efforts to provide<BR>
infrastructure (not root server infrastructure specifically albeit the<BR>
same folks do provide anycast instances for a root server operator) in<BR>
India were blocked by demands for bribes greater than the value of<BR>
hardware being shipped into the country (see<BR>
<a href="http://permalink.gmane.org/gmane.org.operators.nanog/100786">http://permalink.gmane.org/gmane.org.operators.nanog/100786</a>).<BR>
<BR>
<BR>
</FONT></SPAN><BLOCKQUOTE><SPAN STYLE='font-size:12pt'><FONT FACE="Times New Roman"> <BR>
Why this is not done, or cant be done are the real questions in the<BR>
present debate. Any answers?<BR>
<BR>
</FONT></SPAN></BLOCKQUOTE><SPAN STYLE='font-size:12pt'><FONT FACE="Times New Roman"> <BR>
Sure. You are assuming a top-down model that does not exist. There is<BR>
no single entity that can dictate to the root server operators "you<BR>
will give your root server to IIT". You and others that care about<BR>
this are free to make the case to (say) Verisign that it would be in<BR>
their corporate best interests for them to relocate administrative<BR>
control of one of their root servers to India, but it would be up to<BR>
Verisign (or perhaps more accurately, its shareholders) to make that<BR>
decision.<BR>
<BR>
<BR>
</FONT></SPAN><BLOCKQUOTE><SPAN STYLE='font-size:12pt'><FONT FACE="Times New Roman"> <BR>
Is the real problem here that if root server allocation issue is<BR>
opened up, countries would like to go country-wise on root servers<BR>
(as the recent China's proposal for 'Autonomous Internet') which will<BR>
skew the present non-nation wise Internet topology (other than its US<BR>
centricity), which is an important feature of the Internet.<BR>
<BR>
</FONT></SPAN></BLOCKQUOTE><SPAN STYLE='font-size:12pt'><FONT FACE="Times New Roman"> <BR>
No. Placement of root servers has no impact on Internet topology.<BR>
Really. Distributing root server instances can be helpful in reducing<BR>
root query latency and improving resiliency in the event of network<BR>
disruption. That's pretty much it. Opening up the "root server<BR>
allocation issue" is a red herring, particularly given pretty much<BR>
anyone can get a root server instance if they care and are willing to<BR>
abide by the restrictions inherent in operating a root server.<BR>
<BR>
Merging a subsequent note:<BR>
<BR>
On Sunday 05 August 2012 06:10 PM, parminder wrote:<BR>
<BR>
</FONT></SPAN><BLOCKQUOTE><SPAN STYLE='font-size:12pt'><FONT FACE="Times New Roman"> <BR>
' administrative access will not be available' to the anycast<BR>
operator to his own anycast server. <BR>
<BR>
</FONT></SPAN></BLOCKQUOTE><SPAN STYLE='font-size:12pt'><FONT FACE="Times New Roman"> <BR>
Yes. However, if you ask anyone familiar with computer systems, you<BR>
will be told that if you have physical access to a machine, you can<BR>
gain control of that machine. Obtaining such control would violate<BR>
the terms by which the machine was granted, but that's irrelevant.<BR>
<BR>
<BR>
</FONT></SPAN><BLOCKQUOTE><SPAN STYLE='font-size:12pt'><FONT FACE="Times New Roman"> <BR>
This is a pretty centralised control, not at all the picture one got<BR>
from all the technically well informed insiders who seem to suggest<BR>
on this list that everything is open, uncontrolled and hunky-dory and<BR>
kind of anyone can set up and operate root servers.<BR>
<BR>
</FONT></SPAN></BLOCKQUOTE><SPAN STYLE='font-size:12pt'><FONT FACE="Times New Roman"> <BR>
I'm getting the impression that you read what you prefer to read, not<BR>
what is actually written. No one (to my knowledge) has suggested<BR>
"everything is open, uncontrolled and hunky-dory". Root service is<BR>
considered critical infrastructure and is treated as such, so anyone<BR>
asserting it is "open and uncontrolled" would be confused at best.<BR>
Can you provide a reference to anyone making this suggestion?<BR>
<BR>
As for "hunky-dory", I suppose some folks would say the way the root<BR>
servers are operated is "hunky-dory". I am not among them.<BR>
<BR>
<BR>
</FONT></SPAN><BLOCKQUOTE><SPAN STYLE='font-size:12pt'><FONT FACE="Times New Roman"> <BR>
Was the African minister really so wrong, or even the Indian minister? <BR>
<BR>
</FONT></SPAN></BLOCKQUOTE><SPAN STYLE='font-size:12pt'><FONT FACE="Times New Roman"> <BR>
Yes. Really.<BR>
<BR>
Regards,<BR>
-drc<BR>
<BR>
<BR>
</FONT></SPAN></BLOCKQUOTE><SPAN STYLE='font-size:12pt'><FONT FACE="Times New Roman"> <BR>
</FONT></SPAN></BLOCKQUOTE><SPAN STYLE='font-size:12pt'><FONT FACE="Times New Roman"> <BR>
</FONT></SPAN></BLOCKQUOTE><SPAN STYLE='font-size:12pt'><FONT FACE="Times New Roman"> <BR>
</FONT></SPAN></BLOCKQUOTE><SPAN STYLE='font-size:12pt'><FONT FACE="Times New Roman"> <BR>
<BR>
<BR>
</FONT></SPAN><FONT SIZE="2"><FONT FACE="Courier New"><SPAN STYLE='font-size:10pt'> <BR>
<BR>
</SPAN></FONT></FONT><FONT FACE="Calibri, Verdana, Helvetica, Arial"><SPAN STYLE='font-size:11pt'> <BR>
</SPAN></FONT></BLOCKQUOTE><FONT FACE="Calibri, Verdana, Helvetica, Arial"><SPAN STYLE='font-size:11pt'> <BR>
<BR>
<HR ALIGN=CENTER SIZE="3" WIDTH="95%"></SPAN></FONT><FONT SIZE="2"><FONT FACE="Consolas, Courier New, Courier"><SPAN STYLE='font-size:10pt'>____________________________________________________________<BR>
You received this message as a subscriber on the list:<BR>
<a href="governance@lists.igcaucus.org">governance@lists.igcaucus.org</a><BR>
To be removed from the list, visit:<BR>
<a href="http://www.igcaucus.org/unsubscribing">http://www.igcaucus.org/unsubscribing</a><BR>
<BR>
For all other list information and functions, see:<BR>
<a href="http://lists.igcaucus.org/info/governance">http://lists.igcaucus.org/info/governance</a><BR>
To edit your profile and to find the IGC's charter, see:<BR>
<a href="http://www.igcaucus.org/">http://www.igcaucus.org/</a><BR>
<BR>
Translate this email: <a href="http://translate.google.com/translate_t">http://translate.google.com/translate_t</a><BR>
</SPAN></FONT></FONT>
</BODY>
</HTML>