RE: [wg-c] Re: MHSC Position Paper (unavailable), and your comments

["Roeland M.J. Meyer" <rmeyer@mhsc.com>]

> Behalf Of Eric Brunner
> Sent: Wednesday, October 20, 1999 6:48 AM
>
> Netscape Lite 4.6/Export, 04-May-99 abnormally exits when
> attempting to
> display the contents of this html reference. If you want to fax it, my
> number is 781 359 5196.

Alternatively, you could pick it up directly from the web-site, using http
transfer, at <http://download.dnso.net/dox/politicks/> as well as the
<http://www.dnso.net/library/> in a form that is convenient for you. I have
Word2K, PDF and PS, please ignore the other format references. I consider
PDF to be the common form and I use Acrobat Exchange v3.0 to generate it.
The Postscript file is printable on any postscript printer. I don't
understand your problem with Netscape, I have tested it with both Netscape
Communicator 4.5 and 4.7 (domestic, 128-bit encryption, both cases), as well
as MS-IE v4.0. I suspect that you have a local problem.

> Could you (in plain text form, please) expand on how Kent's
> proposal is avoidably complex and over specified?

As far as being overly complex, do a staffing analysis. The process is quite
convoluted. How many people would be employed by that process? Also, it
makes the presumption that the TLDs are institutionalized. This is a
presumption that I vehemently disagree with. This is not a presumption
either Johnathan nor I make.

> If secondary nameservers are a solution to any technical
> issue other than service interruption, but to the policy issue of which of
two or more
> conflicting claimants to authoritative status to select, then
> could you explain why cache corruption is a non-trivial issue in the
> bind development and dns operational communities?

Cache corruption is a configuration issue these days and not germane to the
point. In fact, most of this is a configuration issue. Being in the DNS
operational community myself, I have had only a single case of root-cache
corruption, three years ago, on a crufty old version of BIND. I am using
BIND8.2 now with no problems. I have three root-servers behind four tiers of
telco-style firewall, one-each to a tier.

> Do you expect Paul Vixie to endorse the claim that mutual secondaries
> solves the problem presented by two (or more) simultanious assertions
> to authoritative nameserver status? I don't, but I could be wrong.

I don't expect Paul to endorse anything but strict hierarchy, as he always
has, religiously. I believe that, as he gets jerked around with
f.root-servers.net, he will eventually come around to a different view, but
I do not presume to predict what that view might become. I watch those
developments with interest.

>      "... and allowing the customers themselves to take over operation
>       of the TLD registry ..."
>
> We usually check out our ideas by their ability to scale, if that is
> appropriate to the idea. At which indicies of scale (number
> of customers)
> 10x1, 10x2, 10x3, 10x4, 10x5, 10x6, 10x7, 10x8, ... does the mechanism
> for cooperative registrant election of the registry take less
> time than the process of selecting a successor to Jon Postel?

If there are that many customers to a TLD registry then the registry has no
reason to fail, does it? The presumed failure-mode is one of insufficient
customer demand (lack of $cash$). What will, in all likelyhood, happen is
that the customers would rather move to another registry. If cash isn't the
problem, then the customers could form a new registry, for their TLD,
facilitated by the root-registry. The probability of this happening with
very populous TLDs approaches NIL rapidly. Even the negative-revenue based
approach (name.zero) might work, given a sufficiently large population.

Also, given that, one year later, we still don't have a successor to Jon
Postel and we aren't even looking for one, I would state that it would
certainly take less time.

> There are three statements in your comments on Dave's
> comments on Jon's
> draft that I simply don't understand.
>
> You characterize the discussion of an argued weakness of a for-profit
> motivation as a corner case. The issue is whether the assumption that
> for-profit business rules of necessity lead to specific
> outcomes, e.g.,
> "innovation".
>
> If the assertion Jon makes isn't idealistic as Dave observes, then it
> is empirical, we can find it somewhere in the existing history of this
> industry. The transition from hosttables occurred during the
> SRI tenure of the NIC contract. What innovation could you offer to
> illustrate your
> comments on the error in Dave's claim that Jon's paragraphs on the
> topic of for-profit registries and their probable business
> trajectories is "an eyes-closed, idealistic assertion that is demonstrably
false:?

I would simply point at the history of technological development, in this
century, and those private, for-profit, companies that lead that growth. In
contrast, observe similar growth within the USSR, under
centralist-bureaucratic rule. Also view the post-revolutionary bureaucracy
in France. It wasn't until the advent of Napoleon that they focused on
technology again.

> Finally there is this statement ...
>
> 	NSI was started by academics and scientists...
>
> In September 1991, the Data Defense Network Network Information Center
> (DDN NIC) relocated from SRI International in Menlo Park,
> California to
> Government Systems Incorporated (GSI), Chantilly, Virginia. It (name
> service) transitioned from academics and scientists (SRI) to
> marginally
> competent, but "REAL capitalists" (a sub-2nd-tier phone company is how
> I remember them then).

Having worked for telco's, in management, for a number of years, I liken
them more as a communal organization than a capitalistic one. You really
should read some of Alex G. Bell's papers on corporate organization. He was
quite revolutionary for his day. In other words, I dispute your
characterization, of US-based telcos, as capitalistic. It also explains the
marginal competence.

> The characterization you offer of the origins of NSI is demonstrably
> false. Given the strength of your subsequent arguement, as prose at
> any rate, you should have done your homework. DDN Management Bulletins
> are available if you need them, as are the filings of GSI and its
> successors in interest.

Looking at the qualifications of those that started NSI, I would argue that
there isn't a decent marketer in the bunch. I would argue that, aside from
Jim Rutt, there still isn't one. However, one good marketer is all that it
takes. Jim is good. Gabe was too risk-averse, mainly because he didn't
really know what he was doing, IMHO. It is the main reason we are here
today. ARIN has similar problems. Kim Hubbard is decent, but she is no
marketer. She isn't helped that ARIN is set up in a non-capitalistic manner,
as their current cash problems continue to prove. That style of organization
can only continue long-term with continuous external subsidies. The same can
be said of ICANN.

It is not marketing's job to sell you a ticket to a "river styx cruise", it
is their job to make you want one, whether it is available for sale or not.
No decent marketer will work without suitable payment, it is how they keep
score.

> Since this issue (the legal history of NSI) is revisited in
> every set of pleadings drafted by competent counsel for plaintifs in
> actions against NSI, another place you can go to test the truth of your
claim (do your
> homework) is to pleadings, NSI's generic recitation or those
> of plaintif's
> counsel.

What, pray tell, do these pleadings have to do with the business acumen of
those that started NSI?