Copy Link
Add to Bookmark
Report
Netizens-Digest Volume 1 Number 266
Netizens-Digest Monday, February 1 1999 Volume 01 : Number 266
Netizens Association Discussion List Digest
In this issue:
Re: Re [netz] A Call to Arms?
[netz] Re: A Call to Arms?
[netz] Re: A Call to Arms?
[netz] Re: A Call to Arms?
Re: [netz] Re: A Call to Arms?
[netz] Re: A Call to Arms?
Re: [netz] Re: A Call to Arms?
Re: [netz] Re: A Call to Arms?
----------------------------------------------------------------------
Date: Sun, 31 Jan 1999 10:12:35 -0500
From: Craig Simon <cls@flywheel.com>
Subject: Re: Re [netz] A Call to Arms?
Ronda Hauben wrote:
[much snipped]
> IANA was under DARPA and DARPA was responsible for what went on in IANA.
> And the DoD folks were responsible. There was a line of responsibility
> backed up by penalties for abuse.
Well, of course Postel would have been penalized for crimes like
embezzlement if such things had occurred, but you have been missing the
fact that he had a very free hand. As RFC editor and in his other
functions he saw himself as answerable to the Internet community. USG
oversight provided a certain set of circumstances, yes, but Postel's
agenda had much more to do with building up ISOC and a global Internet.
> This is all being thrown to the wind by the so called
> "self governance" model or "private self regulation model"
> which is essentially setting up a system for abuse as there
> is only reward for abuse.
There is only reward for abuse??
>
> >contractual sanction provided by a few USG-authorized agents working
> >within agencies like the NSF, and now the DOC. In other words, formal
>
> Some of the problem is that NSI was allowed to function in a
> way that is abusive of the obligations of a government contract.
>
> All that they should have been allowed to do was to have an
> administrative fee *not* a way to make profit, from the registrations.
>
> So there was a probelm that had to be solved, and instead of it being
> solved it was used by the U.S. government as a power play to
> make a much worse problem for those who depend on and understand
> the importance of the Internet.
>
> >oversight was once coordinated through agencies of the USG, but the
> >private entities (peers) which constitute the nodes, intersections, and
>
> Not "private" entities. There are all sorts of entities, but in general
> they are autonomous networks that join together to communicate.
> (Many are public or related to public entities like government or
> universities, etc.)
OK, autonomous, but the vast portion of it is privately funded.
> >.highways of the Internet looked to the IANA for guidance regarding
> >standard operatiing procedures, and tended to interact within
> >IETF-inspired venues when they wanted those procedures to be changed.
>
> But the whole process was under obligation to be legitimate that
> the U.S. government DoD oversight and responsibility for IANA
> provided.
I don't understand that sentence, but I think you're repeating your
assertion that everyone was supposed to follow orders of the DoD. That
simply isn't true. NSF and DOD were seeding things, watering them with a
little money, and were happy to see them grow without need for constant
attention.
> >That community, as you know, has been attempting for over three years to
> >terminate NSI's commercial monopoly on .com, .net and, .org
> >registrations. In my view, that community has also been rather noble and
> >forthright in attempting to establish new sorts of DNS oversight
>
> That community has *not* in general been attempting to do anything.
>
> A small group has been trying to (at the instigation of the U.S. govt)
> get themselves a piece of the NSI pie, but not to identify or solve
> the problems.
You are simply missing the point that the different camps in this debate
bring to different interests to the table. There are material reasons
that companies like Cisco, MCI and Sun are promoters of ISOC, and
consequently why the IAHC/IETF positioned itself against NSI, and the
proprietary registry interests. And you miss the fact the the USG policy
is driven by rival material impulses and is therefore beset by
contradictions. It's not some evil conspiracy, just people muddling
through, inventing ideologies to explain their behaviors.
Craig Simon
------------------------------
Date: Sun, 31 Jan 1999 09:57:49 -0800 (PST)
From: Greg Skinner <gds@best.com>
Subject: [netz] Re: A Call to Arms?
Ronda Hauben <ronda@panix.com> wrote:
>Do you know who the "powers that be" in this situation are Greg?
>Do you have an idea of who wants this? Is it Clinton? Or Gore? Or
>both? Or others?
The NTIA is part of the executive branch. The objectives that have
been put forth by Magaziner seem consistent with what Clinton and
Gore's objectives are for stimulating the US (and world) economy by
opening up markets and creating competition. So yes, it seems that
Clinton and Gore want this.
Why they want it is because the people who voted for them want
economic growth, low unemployment, etc.
Whether or not those who voted for them wanted the Internet to be the
vehicle for this is debatable.
>Greg, are there law suits that you know of about what ICANN is doing?
>Or about what the NTIA is doing?
The law suits I have referred to are the numerous domain name
disputes, some of which have named past net.luminaries (e.g. Postel).
Refer to Ellen Rony's site for more info.
>Also they asked if I knew who was behind what is happening with
>creating ICANN and why this is being done.
I don't know why this is such a mystery. Postel approached several
people. Mike Roberts did as well. This has been reported ad
nauseam. One can trace Mike Roberts' involvement in similar matters
back to the NSF backbone privatization.
>I'm surprised. I just described what is happening to a relative.
>She knows very little about the Internet or about computers.
>But I told her how the U.S. government is claiming it will give
>control over central points of control of the Internet to a private
>entity it is creating. [...]
>Her comment was that this is the greatest giveaway ever.
>It wasn't hard for her to understand what is happening at all.
Naturally, if you make it sound like the USG is giving away resources,
it will sound like a gross injustice. On the other hand, if you point
people to the online documentation, and have them ask direct questions
of the individuals involved, they will be able to make more informed
judgments.
I challenge you to have your relative join the online discussions, and
interact with the people involved. Let's see what her opinions are
after about a year's worth of involvement with the issues.
As I wrote before, some people I know who have been involved with the
Internet from its earliest days agree with you. Others do not, and
still others are either undecided, apathetic, or too busy to give the
issue much consideration.
- --gregbo
------------------------------
Date: Sun, 31 Jan 1999 12:47:50 -0400
From: kerryo@ns.sympatico.ca (Kerry Miller)
Subject: [netz] Re: A Call to Arms?
Craig Simon wrote:
{ > That $50 a year charge for only those allocated would lead to $100
{ > billion alone. That the control over the root server system etc give
it
{ > additional power. That NSI had an income of $900 million (almost a
{ > billion dollars) just from selling domain names. (gross income)etc..
{
{
{ I'm surprised as well. According to the current numbers at
{ domainstats.com there are about 3 million registrations in .com, about
{ 300,00 in .net, and about 250,000 in .org.
{
{ 3.55 million at $35 per year works out to around $125 million. This is
not
{ small change, to be sure, but nevertheless not large compared to the
other
{ sorts of things going on the computer and telecom industries. And who
ever
{ proposed a $50 annual charge for an IP number? No one that I know of.
Every provider/ resellers page Ive seen that mentions cost says
basically,
"InterNIC
registration fees are separate from ours. They
currently charge $100 for a two year registration, and
$50 per year thereafter."
Has that changed recently, or is there a difference between IP
registration and name registration?
{ >The question is how to establish a new body to provide formal
{ >public oversight of these activities, recognizing that they are
{ >essential for the continued existence of a unified and scalable
{ >Internet.
Greg and I went through this last year, but perhaps you can help us out:
are names (not IP numbers) essential, rather than a convenience to
humunks; and what does 'scalability' mean in this context (i.e. what
alternative do you find undesirable)?
kerry
------------------------------
Date: Sun, 31 Jan 1999 23:00:28 -0800 (PST)
From: Greg Skinner <gds@best.com>
Subject: [netz] Re: A Call to Arms?
Ronda Hauben <ronda@panix.com> wrote:
>The problem is that the U.S. government set up a power play for
>an illegitimate purpose.
>And lots of folks like the 5 IANA advisory council folks, Roberts,
>Farber, Cerf, Bradner, Landweber, etc. seem to have gone along
>with it and helped carry it out, rather than exercising the kind
>of ethical obligation they have as computer scientists to say that
>this shouldn't be happening.
>But there have been a number of folks who have also said that
>the whole idea of privatizing is a problem. And it is.
>These essential functions of the Internet are too important to be
>put in commercial hands.
>And they need protection by something that has means for asserting
>resonsibility and punishing abuse.
>They need government and scientific support. That was the basis
>for my proposal which was built on the way Usenet and the Internet
>were built. And no one in power even looked at it. That
>meant they weren't trying to solve any problem. If you are
>trying to solve a real problem you don't cut off any avenue
>as that is the avenue where the solution may come from.
I think one thing you may need to accept is that even the most
powerful people are not always able to prevent change. It may be the
case that some of the ICANN backers prefer a noncommercial, research
Internet to what we have now. But they may not have the power to
change the minds of the many individuals and organizations who want to
pay lots of money to use the Internet for commercial purposes. So
rather than seeing the Internet split up into a bunch of autonomous
networks, they have tried to set up a structure that will allow them
to continue to communicate.
- --gregbo
------------------------------
Date: Mon, 01 Feb 1999 08:31:34 -0500
From: Craig Simon <cls@flywheel.com>
Subject: Re: [netz] Re: A Call to Arms?
> Craig Simon wrote:
> { >The question is how to establish a new body to provide formal
> { >public oversight of these activities, recognizing that they are
> { >essential for the continued existence of a unified and scalable
> { >Internet.
>
> Greg and I went through this last year, but perhaps you can help us out:
> are names (not IP numbers) essential, rather than a convenience to
> humunks; and what does 'scalability' mean in this context (i.e. what
> alternative do you find undesirable)?
The system can work without names. You can reach www.flywheel.com at
206.136.140.17, but the use of names generally makes Internet locations
easier to remember, guess, and communicate. It also makes it easier to
move content from one physical location to another (changing IP
addresses) without having to tell the rest of the world what the new
address is. Since the vast majority of today's users have no idea what a
URL's underlying IP address is, if the DNS "went dark," the Internet
community would be thrown into a crisis.
Scalabilty here refers to creating something that can grow and expand
without disrupting connectivity and interoperability within the existing
user base. It's not as though the system is fetterless, but it is
apparent to me that IETF members take pride in creating non-proprietary
standards that can endure for the longest possible time (as opposed to
processes of innovation in commercial environments which are often
premised on quickening obsolescence).
As I see things, the DNS/ICANN debates are roooted in questions of
whether a body of specialists should be enabled to provide continuing
oversight for certain essential functions, or whether the development of
the system should be left up to the less constrained play of commercial
actors. My bias is slipping into this characterization, of course. It's
probably evident that I believe the "old guard" has done a fairly good
job vetting innovations before they become standards.
I'm not a technical guru on the inner workings of these things, but I've
picked up a few things by listening to people who are. Hope this helps.
Craig Simon
------------------------------
Date: Mon, 1 Feb 1999 00:41:50 -0400
From: kerryo@ns.sympatico.ca (Kerry Miller)
Subject: [netz] Re: A Call to Arms?
Craig,
{ the use of names generally makes Internet locations
{ easier to remember, guess, and communicate. It also makes it easier to
{ move content from one physical location to another (changing IP
{ addresses)
Ah, I see a light! If IP are geographically determined, then the vaunted
'dimensionlessness' of cyberspace rests entirely on having *names (or
some other set of non-IP characters) in a DNS database.
But this seems to indicate that the real 'scalability' problem is in IP
addresses, because its limited to [0-9] and only 12 of em at that. Now,
since (after the alphabet) its 'easier to remember' numbers, I have a
suggestion: why not shift the IP to some other part of the 8-bit
address-space, and give digits back to namespace? (It may not seem like
it now, but I foresee the day when a nice, compact, almost elegant name
like http://206d.136e.140f.17g will be a highly valuable property...)
{ ... non-proprietary standards that can endure for the longest possible
{ time
Are you saying scalability = durability?
{ As I see things, the DNS/ICANN debates are roooted in questions of
{ whether a body of specialists should be enabled to provide continuing
{ oversight for certain [*]essential[*] functions,
For sure, if the ICANN edifice rests on building a *convenient* system,
the squabbling is never gonna end! So where is the discussion going on
that focusses on this distinction in particular?
=========
Greg,
{ But they may not have the power to change the minds
{ of the many individuals and organizations who want to pay lots of money to
{ use the Internet for commercial purposes. So rather than seeing the
{ Internet split up into a bunch of autonomous networks, they have tried to
{ set up a structure that will allow them to continue to communicate.
{
Is there any reason the comm interests would want to see the net split
up? Is there any reason any other large administrative entity would like
to see everybodys traffic flowing through a _single network?
Do you suppose its time to call for a little diversity, again?
kerry
------------------------------
Date: Mon, 1 Feb 1999 13:17:22 -0500 (EST)
From: Jay Hauben <jay@dorsai.org>
Subject: Re: [netz] Re: A Call to Arms?
Greg Skinner <gds@best.com> wrote:
> Ronda Hauben <ronda@panix.com> wrote:
>> The problem is that the U.S. government set up a power play for
>> an illegitimate purpose.
>> And lots of folks like the 5 IANA advisory council folks, Roberts,
>> Farber, Cerf, Bradner, Landweber, etc. seem to have gone along
>> with it and helped carry it out, rather than exercising the kind
>> of ethical obligation they have as computer scientists to say that
>> this shouldn't be happening.
>> But there have been a number of folks who have also said that
>> the whole idea of privatizing is a problem. And it is.
> I think one thing you may need to accept is that even the most
> powerful people are not always able to prevent change. It may be the
> case that some of the ICANN backers prefer a noncommercial, research
> Internet to what we have now. But they may not have the power to
> change the minds of the many individuals and organizations who want to
> pay lots of money to use the Internet for commercial purposes. So
> rather than seeing the Internet split up into a bunch of autonomous
> networks, they have tried to set up a structure that will allow them
> to continue to communicate.
The technical principle upon which the Internet has been built is that
the interconnection will be among autonomous networks. Robert Kahn called
this the principle of "open architecture". Louis Pouzin anticipated
this principle in his work in France. The Internet IS a bunch of
autonomous networks, must of which are public like milnet and the
networks at all the public universities in the US and throughout the
world. There are private networks as well that connect as part of
the Internet but they all up until now act according to the principle
of network autonomy. If new players want onto the Internet they too
should set up an autonomous network, adopt IP as their network layer
protocol and play by the Internet rules. ICANN's back room
backers whoever they maybe are not content with the Internet rules.
They want to control the crucial functions of the Internet so
they can change the rules to allow QoS, privacy, whatever over the
whole of other people's networks.
What I don't understand is how people who know the technical principles
upon which the Internet is built so easily disregard those principles
when backing the US Gov't-ICANN MoU. I do see there are at least two
different views of the Internet. One that the Internet is a logical
connection of some 50,000,000 computers and 150,000,000 people who
basically use the Internet to communicate among each other. And two
those who see the Internet as 120,000 routers and various physical
media links. But on either view, the computers are acknowledged to be
grouped on various differnet autonomous networks as are the routers.
Many of the compters and many of the routers are in libraries,
universities, government offices, etc as are many of the routers so
much of the Internet is publicly owned. Yet in order to create the
ideological ground work for the imposition of commercial dominance,
knowledgable people join in trying to create the myth that the
Internet is private and that if commerce is not allowed to dominate
the Internet will "split up into a bunch of autonomous networks".
> --gregbo
Jay
------------------------------
Date: Mon, 1 Feb 1999 10:51:16 -0800 (PST)
From: Greg Skinner <gds@best.com>
Subject: Re: [netz] Re: A Call to Arms?
kerryo@ns.sympatico.ca (Kerry Miller) wrote:
>Ah, I see a light! If IP are geographically determined, then the
>vaunted 'dimensionlessness' of cyberspace rests entirely on having
>*names (or some other set of non-IP characters) in a DNS database.
I don't wish to speak for Craig here, but there are a few issues that
I feel need clarification.
IP addresses were not, at their inception, geographically determined
(although to a certain extent, they were organizationally
determined).
It was also the case that (convenient) names were assigned to hosts,
and used thereafter (rather than requiring IP addresses, or NCP
addresses in ARPAnet times), before DNS existed.
With the coming of CIDR and provider-based addressing, IP addresses
began to take on more geographic attributes, because they were
assigned regionally. However, if you look at the heritage IP
addresses (the /8s I have referred to the in the past), you'll see
that they do not necessarily imply a geographical or regional
boundary. (For example, my company has one of the legacy /8s, 16/8,
but those addresses are assigned worldwide.)
>But this seems to indicate that the real 'scalability' problem is in
>IP addresses, because its limited to [0-9] and only 12 of em at
>that. Now, since (after the alphabet) its 'easier to remember'
>numbers, I have a suggestion: why not shift the IP to some other part
>of the 8-bit address-space, and give digits back to namespace? (It
>may not seem like it now, but I foresee the day when a nice, compact,
>almost elegant name like http://206d.136e.140f.17g will be a highly
>valuable property...)
I don't understand everything you've written here, but it is true that IP
addresses are becoming highly valuable properties. At this point, IP
blocks of /19 or larger (someone correct me if I'm in error) are
considered valuable because they are not subject to route filtering.
(This subject is treated in depth at ARIN's site, www.arin.net.) ARIN
and other regional registries are responsible for giving out these
addresses, which are in turn assigned by providers, resellers, etc.
Some providers charge customers (above and beyond the cost of
providing connectivity) for sub-blocks of their assigned space. Also,
some organizations have rented, or sold, their IP address block.
These and other related practices are controversial, in light of
ARIN's stated policies that IP addresses are not to be traded or
sold. (I would like to point out here that according to some things I
have read on one of ARIN's mailing lists, they are aware of the fact
that IP addresses are being sold, but they have not been able to come
up with concrete proof of sales. Also, if someone were actually found
guilty of selling an IP address block, it's not clear that a court of
law would uphold the ARIN policies.)
There seems to be a growing movement of people who declare their net
existence via the IP address. So it is not only possible, but
acceptable, for people to specify addresses such as http://18.62.0.6
when there is no name associated with that address. But this method
of address specification has not proved over time to be desirable, by
and large, by the Internet community.
>Are you saying scalability = durability?
To a certain extent, scalability = durability, in that the system can
still be used effectively even if the number of entities that use the
system becomes very large. The old way of maintaining host tables was
not scalable in this regard, because the number of hosts became too
large to effectively maintain in one table, and changes became too
frequent. Additional services were also desired that were easier to
implement in DNS, such as canonical names and address classes. (For
example, the DNS can be used to map names to other types of
addresses.)
>For sure, if the ICANN edifice rests on building a *convenient*
>system, the squabbling is never gonna end! So where is the
>discussion going on that focusses on this distinction in particular?
I don't believe it is the "ICANN edifice." It is as I wrote above --
something that has evolved over time and has been found to be more
desirable than the alternatives.
If you wish to propose alternatives, or even raise consciousness, I
suggest you bring your ideas to a forum like the IETF mailing list, or
perhaps one of the IETF DNS working groups. People like Bob Allisat,
for example, have debated on IETF that DNS should be administered on
the desktop.
>Is there any reason the comm interests would want to see the net
>split up? Is there any reason any other large administrative entity
>would like to see everybodys traffic flowing through a _single
>network?
There is already some Internet fragmentation caused by the use of
firewalls, proxies, NAT devices, etc. Most of this is not visible to
end users. There is a possibility that some of the comm interests
might leave the Internet and take their services to something like AOL
or @Home, where you would have to pay the subscriber fee and purchase
proprietary hardware and software to access those services. I think
this is the sort of thing the old guard are trying to keep from
occurring. As to your second question, I'm not sure what you mean by
"a single network;" no one has proposed (as far as I am aware) that
anything like that should happen.
>Do you suppose its time to call for a little diversity, again?
You need to convince the people who are actively involved in
developing and managing Internet resources that they should follow
your ideas. Those people are on IFWP, IETF, inet-access, NANOG, etc.
- --gregbo
------------------------------
End of Netizens-Digest V1 #266
******************************
