[governance] RE: GeoTLD

[Karl Auerbach]

Avri Doria wrote:

You ask so many good questions.

As regard your question about root operators following some authority 
for choice of TLDs:

Imagine if everybody who operates a root system - which ultimately could 
be every user[*] - did not look to any authority other than his/her/its 
own perception of self interest.  For the vast majority of providers 
this means going with the herd - which, as we are learning from swarm 
behavior theories - means that pretty much every provider will offer the 
TLDs that we all know and love.  Those who chose not to provide these 
will probably fail or, like some divergent religious groups, be quite 
happy in their splendid isolation.

But, and this is the important part, each provider would be free to add 
anything else - new and uncommon TLDs - that strikes their fancy.

Why might something strike their fancy?  Most likely because the new TLD 
operator is holding out a handful of cash.

Where is a new TLD going to obtain that cash?  Who knows.  Maybe it's 
from some city, like Berlin, that is willing to buy its way in to enough 
roots that it becomes something that every root operator feels it must 
carry.  Or perhaps it is a part of revenues derived from registry fees 
or data mining.  Who knows.  And who cares? (as long as it is not illegal.)

This kind of independent, self interested, decision makings means that 
we end up with a very nice way for new TLDs to be born, try to grow, and 
maybe even grow up to be one of the big ones that every root operator 
feels that it must include in its inventory.  The flops will flop and 
fade, as they should, into obscurity and oblivion.

Even in networking protocols we have that kind of thing - once Novell 
and IAX/Netware and IBM's SNA ruled the networking landscape.  But this 
idea of TCP, particularly with some boosts from UC Berkeley and 
companies like FTP Software, Proteon, TGV, Intercon, and Sun came out of 
the back rooms and knocked the socks off the old champions.

All I am suggesting is two things: 1) that we stop auto-bespattering 
those who want to try to do things differently (but still within the 
scope of internet standards) than has been the routine and 2) that we 
take care to build institutions of internet governance only when there 
really and truely is a thing that needs the heavy hand of "governance" 
rather than the less coercive hands of user and provider choice 
(something we call "free competition") practiced within the scope of 
broadly accepted civil and legal rules.

You suggested that ICANN might be viewed as a "clearinghouse".  I'm 
having trouble reconciling (pun intended) my sense of that word - a 
place where transactions are presented and accounts are settled - to the 
role that ICANN fills: that of making choices among competing 
allocations.  Clearinghouses don't say 'no' to transactions, but ICANN, 
on the other hand, does seem highly preoccupied with denying people the 
opportunity to make transactions and investments.  So, I don't see the 
analogy.

You asked a good question about why, if there is nothing that stops this 
from happening, why it has not.

That's got several answers.

First is that there have been several people and groups who have done 
such a horribly bad job of it that it has created a broad impression 
that any and every group that might try it anew will do an equally bad 
job.  (By-the-way, I measure the quality of the job done by looking at 
the quality with which name query packets are turned into name response 
packets - to my mind the front-office task of registering and 
transferring names not where we should be measuring quality, although 
the practice has been to use that as the sole metric.)

And second is that ICANN has engaged in a kind of warfare upon those who 
suggest the idea of competing root systems.  Indeed, some of ICANN's 
statements over the years, particularly with regard to New.net, could 
possibly have been construed as business libel.

Remember the Hush-A-Phone case - 
http://en.wikipedia.org/wiki/Hush-a-Phone_v._FCC - This was a situation 
in which some folks invented a plastic hand to focus a speaker's voice 
into those old carbon microphones found on phones through the 1960's. 
That hand also helped prevent others from overhearing.

Anyway, AT&T, along with the FCC swore up and down on a stack of bibles 
that this entirely passive device - a plastic hand - would cause 
operators to go deaf, cause telephone repairman to be electrocuted, and 
otherwise cause the internet of that era, the telephone system, to 
collapse into a pile of sparking and smoking wires (like computers do in 
movies.)

That was, of course, patently false - which is what the courts 
ultimately found (after 26 years).  But it does indicate the energy with 
which entities, including technical ones, like the FCC, will go to 
protect their status quo.

I may not be a conscious thing, but every generation of technical folks 
are proud of their creations and often lose perspective when evaluating 
changes to those creations.  So it is not surprising that one of the 
grumbles that comes out of the IETF rather frequently is that newcomers 
lack "clue".

And third, we have the presence of the US Government standing behind all 
of this.  That looming presence suggests to any investor that before 
they can open their doors for business they may have to go to the mat 
with the largest, richest, and most powerful government on earth.

As a consequence, those who have money to invest have been into a 
position in which they have to ask whether they want to A) invest into 
something that will face condemnation, righteous sounding condemnation 
whether it is right or not, and perhaps the wrath of the US government 
or B) invest in what they hope will be the next Google.

In other words, ICANN has scared away the investors and investment 
dollars.  Just look at what ICANN did to those 40 groups that paid 
$50,000 each just to apply for a new TLD and have been strong along now 
for 7 years.  And then, those few that did get through the gauntlet had 
to endure what amounted to an ICANN run colonoscopy of their business 
and financial matters (but not their ability to run DNS servers.)

Consequently, even though it is entirely feasible to establish competing 
root systems, it is not an attractive investment opportunity.  The 
reason that it is shunned is not that there isn't money to be made - 
there is, lots of it from data mining for commercial and 
governmental/intelligence purposes, but because those who propose the 
idea are painted as pariahs and nutjobs.

[*] There was once a system called "Grass Roots" which was a website in 
which the user could select among the (then) thousands of TLDs, and in 
cases of conflict, select which among the conflicted offerings.  Then a 
zone file was produced, and a named.conf file, that could be plopped 
into Bind so that any *nix machine could become a highly personalized root.

		--karl--






> 
> On 27 Dec 2007, at 14:16, Karl Auerbach wrote:
> 
>> Avri Doria wrote:
>>
>>> With the authority  St Aquinas brought in as a bolstering argument.
>>
>> I've always found it amusing to cite a Saint,an authority, as an 
>> authority to make an argument, based on an assertion of authority, why 
>> arguments that are assertions of authority are weak arguments.
> 
> Yeah i thought that was clever especially since, at least periodically, 
> he was one of the foremost authorities for an authoritarian religion.  
> especially since 1914 with the 24 propositions.  if i remember correctly 
> they left the on about arguing from authority out of the list.
> 
>>
>>
>>> Note: even the use of 'consistent root' as opposed to' single 
>>> root'... are logically the same thing.
>>
>> Not exactly.
> 
> No, not exactly.  but perhaps logically.
> 
>>
>>
>> The idea is that multiple providers of the same thing is rather 
>> different than exactly one provider of that thing.
> 
> But doesn't each on those multiple providers accept the single reference 
> root voluntarily?    At least to the extent that they accept it as the 
> authoritative root.  In a sense don't we already have a market of 
> sorts?  If the root operators wished to accept other authorities, they 
> could.  Couldn't they?  I.e as it stands it is a voluntary arrangement 
> where those who could do otherwise, choose not to.
> 
> In what way is that domination from the center?  To me it looks rather 
> like a bottom up clearinghouse model.  while i don't argue that ICANN is 
> a rose garden (well maybe it is, but i digress), i don't think that 
> comparing it to the politburo is quite accurate.
> 
>>
>>
>> The reason that that difference is important is that it distributes 
>> the way in which decisions are made - each provider makes its own 
>> choices regarding which inventory it will carry.
> 
> In this case the individual root operators are making the decisions.  
> Also as far as I understand there are few TLD registries outside of the 
> ICANN approved list that these root operators could support if they 
> wished.    Why don't they?
> 
> Also, who forces the rest of us to accept this particular model.  sure 
> it might be challenge to find a new mapping model but nothing makes any 
> of us accept the current model.  with IP addresses, sure, we have little 
> choice.  but with domain names - where is the outside compulsion?
> 
>>
>>
>> It is indeed the kind of difference between a planned central economy 
>> - the ICANN method - and a free market.  The difference is choice made 
>> in one place versus a choice made by each person.  The difference is 
>> that between "Top down" - the ICANN way - and true "bottom up" choices 
>> made through the aggregated individual choice of each user of the net.
> 
> As I indicated above this looks more like a clearinghouse model to me 
> and not a planned central economic model.  We use the ICANN names 
> because they are easy and they suit us.  if they didn't we would find 
> something else.
> 
> 
>>
>> The reason that I use the word "consistency" is that it reflects the 
>> core need: that users don't want to be surprised.
> 
> and they don't want to have work too hard.  or be different from their 
> neighbors (which i think is a transitive property especially given 
> cultural hegemony, but i digress again)
> 
>>
>>
>> Even singular DNS is not perfectly consistent - apart from its 
>> built-in inconsistency that occurs as a side effect of its built-in 
>> asynchronous update of information, DNS names are not consistent even 
>> over short periods of time - we are all familiar with names going away 
>> and changing.
> 
> So are you arguing that the need for consistency is not absolute?  
> Probably not.  but we only would tolerate so much insecurity before we 
> found a new tool.
> 
>>
>>
>> But the larger question is the this: Does consistency require a 
>> Procrustean exact identity so that every root server group offers 
>> precisely the same suite of TLDs or can there be some variation around 
>> the edges?
>>
>> There are some who believe in the former, the "mirror" form of 
>> consistency.
>>
>> Of course, internet technology and the end-to-end principle make it 
>> impossible to require and enforce that there be exactly and precisely 
>> one provider of name mapping services.
> 
> Exactly.  and yet, there is still is just one dominant name mapping 
> service, even though there have been various attempts to start others.  
> Is this perhaps a market 'decision'?  did we vote with our feet?
> 
>>
>>
>> I have been arguing for the less strict form of consistency - in which 
>> experimentation and growth can occur in at the edges and, if such 
>> experimentation proves popular, it can spread to the core.
> 
> What stops that from happening now?  Just because ICANN/NTIA won' bless 
> it, doesn't mean people could not do it - if they found a way to 
> convince the users it was something they wanted.  Now, in a planned 
> economy model, we might be able to force people to try the new stuff, or 
> in the supermarket model we could put the stuff in a prominent rack by 
> the cash registers, but in this model, people have to want it to buy 
> it.  and it better be easy to use.
> 
> (e.g. just i l use almost all of the IM services, if it was easy i would 
> probably use multiple name mapping services.  but even though i am 
> mildly willing, i never found it easy enough to register for any of the 
> other name services, so to date, i still use just the ICANN name mapping 
> service.)
> 
>>
>>
>> By this I mean that different root providers would offer new TLDs, in 
>> conjunction with the core TLDs offered by everybody.  That's why I 
>> call these new, experimental things "boutique" TLDs - because they are 
>> not found everywhere and those who use them know (or should know) that 
>> they are not using mainstream products.
> 
> Incidentally I have been using the term boutique TLDs also - but i think 
> in a somewhat different way, I use it to refer to those that would only 
> attract a small group of registrants, even if this was a ICANN blessed 
> registry.   i assume your definition precludes the notion of ICANN base  
> boutique registries.
> 
>>
>>
>> As for the word "propaganda": your word, not mine.  If I am somehow 
>> guilt of using strongly colored words and analogies I am not alone.
> 
> yes, but you do it so well.
> 
>>
>>
>> (When I read your note I imagined you grinning when you used that word 
>> - and knowing you as a person with a constructive intent, I know that 
>> it was used with an intent to tickle our discussion in a positive, 
>> even amusing, way.)
> 
> Well true.  As I said i think propaganda is an art form.  and I have 
> been known to employ it on occasion, but i don't think i have ever done 
> it as well as you did.  and hence my note of praise.
> 
>>
>>
>> Analogies are powerful analytical tools.  Should we avoid them?  If 
>> so, given that humans tend to think via analogies, how could we do so?
> 
> no i don't think we should avoid them.  but sometimes they might be so 
> strong that they could trip the message on its face in a Godwinesque 
> prat fall.
> 
>> Procrustean
>> Soviet Union
>> Standard Oil Trust
>> Orwellian "NewSpeak"
> 
> and of course
> 
>>> "propaganda"
> 
> cheers,
> 
> 
> a.
> ____________________________________________________________
> You received this message as a subscriber on the list:
>     governance at lists.cpsr.org
> To be removed from the list, send any message to:
>     governance-unsubscribe at lists.cpsr.org
> 
> For all list information and functions, see:
>     http://lists.cpsr.org/lists/info/governance

____________________________________________________________
You received this message as a subscriber on the list:
     governance at lists.cpsr.org
To be removed from the list, send any message to:
     governance-unsubscribe at lists.cpsr.org

For all list information and functions, see:
     http://lists.cpsr.org/lists/info/governance