Re: Comments on Karl's draft

[Karl Denninger <karl@Mcs.Net>]

> The following are some comments to Karl's draft.  I make no statement
> whether I agree or disagree with any of the plans or proposals set forth 
> in this document (or any other draft for that matter).  The following are just
> comments (constructive, I hope).
> 
> --=====================_849102749==_
> Content-Type: text/plain; charset="us-ascii"
> Content-Disposition: attachment; filename="KARL.TXT"
> 
> >6)      It has been demonstrated that thousands (ie: 5,000 to 10,000) TLDs
> >        will not break the "root" nameservers.  The currently operational
> >        "root" nameservers handle upwards of 400,000 domains at present as
> >        they also serve the .COM, .ORG and .NET zones.  This is done today
> >        without the servers "falling over".  A load of two orders of
> >        magnitude less in the TLD namespace is unlikely to cause any
> >        negative operational impact on their functionality.
> 
> You do not specify a time frame for implementing 5,000 new TLDs.
> All at once in 1997?  Gradually over a period of time? (to determine
> whether your hypothesis is true (see following statement)).

I don't believe that given the structure put in place (and limitations on
hoarding of TLDs) you'll see 5,000 at once under *any* plausable scenario.

You might see them over the space of two to three years.  That's the
intention.

> >        There is therefore no documentable technical restriction on large
> >        numbers of TLDs operating in parallel.

Again, the restrictions on ownership are enough to insure that this doesn't
happen all at once.

> I think there is another rule that needs to be stated: that the
> ISOC/IANA/IETF, being entrusted with the responsibility for ".",
> shall take all necessary measures that the Internet name space not
> become balkanized and that the Internet continue to function in a
> suitable manner as it does today (probably needs lots of
> embelishment and fixing up but you get the idea).

Is the ISOC/IANA/IETF also going to take up the banner when it comes to
peering agreements and the stupidity that is ALREADY causing balkanization
of the net?

If the I* organizations think THIS is properly their area, then address the
whole problem.   Frankly, DNS is the least of your worries in this arena.

> This might be valid for 1996.  It probably won't be for 1998.  I
> would specify a time frame as well as when such a rule would need to
> be revised.  I would also specify that the total aggregated
> bandwidth by T3 or better.  Having just two T1s doesn't cut it
> anymore.  Your section also leaves the ability for me to connect up
> via T1 to two upstream ISPs - which are only connected via T1 (or
> even less) to a major - who in turn connects to a NAP.  I would go
> for NAP/IX interconnection at at least 2 sites in addition to
> multi-homing.

Uh, not even close.

If you distribute the nameservers then you *automatically* qualify.  So
contract out one nameserver of the two.

DNS isn't a big bandwidth consumer.  Even large registries need less than a
T1 to really operate a nameserver.  This will be true two years, three
years, and five years from now.

> >Each entity holding a TLD shall be required to provide the following
> >administrative services and policies:
> >
> >1)      A means, via the "whois" protocol, to search the database of
> >        second-level domains maintained by this registry and return
> >        common directory information.  This information shall include,
> >        but not necessarily be limited to:
> >
> >        a)      The "owner" of the second-level domain, including
> >                contact name(s), physical address(es), and telephone
> >                number(s) of the persons responsible for the operation
> >                of the second-level domain.
> 
> You realize this contradicts Allisat's position that he will not
> release this information unless via a court order or by the police.

Yep.  Nonetheless, if I'm going to sue someone who holds a domain, I have to
be able to find them.  I don't think its unreasonable AT ALL to require that
if you wish to be listed in a directory (which is what DNS basically is
being used for) that your name, address, and phone number be listed.

That's the point, natch.

And it deflects the whole liability issue from the registry (which otherwise
is going to come crashing down on your own head if you try to play the
secrecy game).

> >4)      An administrative fee of US $500 per annum per TLD to be paid to
> >        the ISOC for its oversight of this process.  The ISOC may delegate
> >        this fee structure and disbursement of funds through open, public
> >        procedure as it sees fit.
> 
> Do you really think $500/yr is enough to cover all the paperwork,
> all the details you listed in section "ISOC/IANA responsibilities",
> dispute resolution, lawyers, deletion, transfer of TLDs, etc.?

1)	Paperwork?  What?  One online or faxed form?
2)	The responsibilities are minimal.  Two hours of someone's time to 
	verify the technical requirements should do it.  You can easily 
	pay that someone out of the $500 and have $300 left.
3)	Dispute resolution?  See the draft; there isn't much that the ISOC
	or anyone else is required to do there.  Disputes are best handled
	by the mechanisms we already have in the countries of relavence to
	the dispute in question.
4)	Lawyers - see (3).
5)	Deletion and transfer - see the draft.  What problem is there again?

--
--
Karl Denninger (karl@MCS.Net)| MCSNet - The Finest Internet Connectivity
http://www.mcs.net/~karl     | T1's from $600 monthly to FULL DS-3 Service
			     | 33 Analog Prefixes, 13 ISDN, Web servers $75/mo
Voice: [+1 312 803-MCS1 x219]| Email to "info@mcs.net" WWW: http://www.mcs.net/
Fax:   [+1 312 248-9865]     | 2 FULL DS-3 Internet links; 400Mbps B/W Internal