[governance] Re: Alternative DNS systems and net neutrality

[Dan Krimm]

I do understand that ICANN's relationships with the RSOs are informal --
there is no direct authority.  That suggests that if ICANN oversteps its
"consensus" that RSOs could decide to walk away from ICANN.  An interesting
idea, though that seems unlikely given the transaction costs of
coordinating the root outside of ICANN.

As for "some things" not using DNS, well, fine, but you did not contradict
the other part of my statement that "everyone" uses DNS.  And I would guess
that the great majority of Internet applications indeed do use DNS, even if
there are a few exceptions along the way.  Are there any numbers as to how
many Internet uses (not just *types* of uses, but *instances* of uses) use
DNS as opposed to others that do not?

You suggest that "it might go away in a decade" -- but what do we do in the
meantime for the decade upon us now?  In the meantime, it remains
important, here and now.

Nope, I still don't agree that DNS is a molehill.  Just because certain
things are possible in principle doesn't mean that they are likely in
practice.  Even in a "free market"...  ;-)

If you don't incorporate transaction costs into your equations, you get the
wrong answers when you calculate the results.

Bottom line:  DNS remains ineliminably a broadly political issue, that
deserves political representation and systematic public accountability in
addressing that public policy.

Dan



At 7:08 PM +0300 11/26/07, McTim wrote:
>Hello all,
>
>I have been away for the last 10 days, so just catching up now.
>
>On Nov 17, 2007 12:13 AM, Dan Krimm <dan at musicunbound.com> wrote:
>> Whatever, Veni.  This is just rhetorical jousting.
>>
>
>Or a firmly held opinion that's opposed to yours.
>
><snip>
>
>>
>> And excuse me, how is ICANN "responsible for" only a "small segment of the
>> Internet"??
>
>Because, via IANA it only has a very narrow role to play.
>
> Isn't ICANN responsible for the *entire DNS* over the *entire
>> Internet*?
>
>Not at all.  DNS is a distributed hierarchy.  Most DNS administration
>is done near the edge of the network. ICANN is responsible for a very
>small fraction of it!
>
>Let's say I want to visit your website, and I have never been there
>before (and neither has any other customers of my ISP).  Let's also
>say for the purposes of this discussion that no other customers of my
>ISP have been to a .com site either (or the cache of my DNS server has
>just been flushed.)
>
>I put http://www.musicunbound.com/ in my url window, and my cranky old
>box (called sentry.bushnet.net) makes a query for the Address record
>for the above URL.
>
>Well, since the cache has just been flushed it doesn't "know" how to
>reach this site (even if it ever has had it in cache).  But sentry
>"knows" how to reach the rootservers, so it asks one of these.   It
>gets a response saying basically "dunno that, but I DO know how to
>find the NS for .com, heres the address of that one, go ask it", so
>sentry, armed with the address of .com's NS, asks the same question to
>.com's NS (say d.gtld-servers.net). at	192.31.80.30, which says
>"dunno, but go see ns2.dreamhost.com. at 66.33.206.206, he can tell
>you", so sentry troops down the pipe again to 66.33.206.206 asking
>ns2.dreamhost.com what the adress of www.musicunbound.com/ is, and
>because it's an authoritative server for the zone, ns2.dreamhost.com
>says, you need the webserver at 208.113.195.100, and my browser can
>fetch it.
>
>The point of that long story is just to illustrate that ICANN via IANA
>has helped to populate the rootzone, and signed various deals with
>Verisign to admin the .com zone (this example), and a rootserver (or
>2).  However, it is certainly possible that my query went thru a
>non-ICANN/non Versiign administered rootserver(there are 10 of them).
>My local recursive caching, forwarding NS did much of the heavy
>lifting, and the bulk of the work is in setting up and maintaining the
>NSs at the edges, (your edge and mine).
>
>ICANN has nowt to do with the various routers, switches and servers
>between my browser and your website, those are owned by ISPs large and
>small.  Of course, these run on IP, but ICANN really has very little
>to do with IP address distribution either, since that is done by the
>RIRs and LIRs.  Of course IANA distributes to the RIRs, but that's
>purely an administrative function.
>
>
>> Everyone and everything on the Internet uses the DNS,
>
>Who told you this?   Some things things don't.
>
>because
>> DNS is in effect the main gatekeeper to any content or applications on the
>> net
>
>It's not.  DNS is a layer of misdirection allowing humans to type in
>things they can recall, a handy "phone book" if you will.  It might go
>away in a decade, and we would have something better in place.
>
><snip>
>
>>
>> But if we are to resolve this, we can't avoid dealing with the political
>> elephant.  Better to look straight at it and deal with it on its own terms,
>> however inconvenient it may be to those in power for everyone else to
>> actually see it.  But be realistic: elephants are too big to sweep under
>> the rug.
>
>My opinion, oft-stated in this forum is that it's a molehill, that
>some see as a mountain.
>
>--
>Cheers,
>
>McTim
>$ whois -h whois.afrinic.net mctim
>____________________________________________________________
>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