Do any ISPs or web hosting companies have publically available statements on their web sites stating that they will not support the new new.net domains and why they won't? I am getting more requests from users to change our DNS root servers to support this and wanted to see what others tell their users. Any IETF/ICANN statement available?
Thanks,
Hank
Do any ISPs or web hosting companies have publically available statements
on their web sites stating that they will not support the new new.net
domains and why they won't? I am getting more requests from users to
change our DNS root servers to support this and wanted to see what others
tell their users. Any IETF/ICANN statement available?
RFC 2826 IAB Technical Comment on the Unique DNS Root
randy
It would better be termed "IAB Political Comment on the Unique DNS Root."
It is stupid and irresponsible to setup a new DNS root. End of story, read
2826.
And.. if you let new.net do it then every other capitalist in the world
will start doing it and then the Internet will become disfunctional, and
what will that achieve?
Stick to the single root, stand firm on this policy and new.net wont get
far!
Steve
It is stupid and irresponsible to setup a new DNS root. End of story,
read
2826.
And.. if you let new.net do it then every other capitalist in the world
will start doing it and then the Internet will become disfunctional,
and
what will that achieve?
Stick to the single root, stand firm on this policy and new.net wont
get
far!
Steve
I'm tempted to simply blackhole new.net -- just to keep from having to
fix the problems that are sure to arise from customers using the blasted
plug-in...
Mark Radabaugh
Amplex
(419) 833-3635
It is stupid and irresponsible to setup a new DNS root. End of story, read
2826.
Is this some scripture to quote? 
And.. if you let new.net do it then every other capitalist in the world
will start doing it and then the Internet will become disfunctional, and
what will that achieve?
I think that you would find that capitalists tend towards effective market
solutions(that is in fact why most of them are capitalists.)
Stick to the single root, stand firm on this policy and new.net wont get
far!
Suddenly, I feel like I should be grabbing your hand and engaging in a
rousing rendition of "Kumbayah my lord."
Rather you or myself like it or not, given the existence of a player
that has the capital to make this idea go, the market now has an
opportunity to decide for itself.
Perhaps next you might wish to stamp your feet and threaten to hold your
breath until they go away?
On or around Tue, Mar 13, 2001 at 11:37:26AM -0800, Patrick Greenwell may have written:
Suddenly, I feel like I should be grabbing your hand and engaging in a
rousing rendition of "Kumbayah my lord."Rather you or myself like it or not, given the existence of a player
that has the capital to make this idea go, the market now has an
opportunity to decide for itself.
Unfortunately, "the market" tends to consist in large majority of 1) users, and 2) management. And we all know how bright those two particular segments of the population tend to be.
Perhaps next you might wish to stamp your feet and threaten to hold your
breath until they go away?
let's not forget what mailing list this is - the operators in this forum can
have a very real and significant impact on the direction "the market" takes.
If, as a group, the NANOG readership decides to take a single position on
anything (ha!), then we could very likely effectively determine in which
direction "the market" will go. After all, if _nobody's_ customers can access
new.net's non-sanctioned gTLDs, they can't very well go to another provider
for such access, and new.net will die the quick death that it deserves.
(yes, I'm obviously idealistic and naive to think that even a significant
majority of NANOG readers could even agree on which way is up, but I think
enough people agree on this issue that we don't necessarily have to sit back
and let "the market" make decisions that will have real operational impact
for the foreseeable future. We can make those decisions ourselves.)
It is stupid and irresponsible to setup a new DNS root. End of story, read
2826.
Such reasoned and well thought out discussion points thrown out on NANOG?
I'm in shock.
And.. if you let new.net do it then every other capitalist in the world
will start doing it and then the Internet will become disfunctional, and
what will that achieve?
Uhhh, so what if they do? Nobody is forcing you to use new.net, or ORSC
or TINC or whoever, or even setup your own roots with the "pick of the
bunch" from ORSC or TINC or whoever.
If you don't use the alternative roots, you're not affected by their
existance... so you can just pretend they don't exist and live life in
your 2826 compliant happy world.
(yes, I'm obviously idealistic and naive to think that even a significant
majority of NANOG readers could even agree on which way is up, but I think
enough people agree on this issue that we don't necessarily have
to sit back
and let "the market" make decisions that will have real operational impact
for the foreseeable future. We can make those decisions ourselves.)
Yes we can make those decisions. So set up your own root server (maybe two
or three) on your piece of the net, pick a source for an inclusive root
zone, and start slaving it (be sure to edit the glue to point to your own
roots). That takes care of any doubts you might have about the resiliancy
of the inclusive roots, since you are running the roots your caches use.
Better yet, start out with the root.zone from ftp.internic.net, add
delegations for the expanded set of TLDs, install in private root server.
This can all be automated to happen twice a day. Point your caches at it,
tell your users about it, and be happy.
What is the big deal? It's a 60K text file, it's not like the old
hosts.txt, it doesn't need to list every host on the Internet, just the TLD
glue, and there's only 253 of those in the ICANN root zone. Is that so hard
to deal with? We're all flummoxed by a 60K text file? The internet comes to
a halt because of a 60K text file?
Right now, you depend on a single party to provide 13 servers to start
recursing from. Is that a risk-free arrangement for the long run? Can you
imagine something more resilient? Do you have an SLA with Verisign for root
service? Think you're going to get one? You have one with your upstream,
perhaps, and with other vendors that are essential to your operations.
So, what if an inclusive root server operator offered a SLA for root zone
service? Is that at all interesting? There are any number of things people
can come up with, if they use their imagination, and shed for a moment the
idea that there must be one root, and only one. If the root zone is so
important, why do people not demand the same level of service that they
demand elsewhere?
If I may borrow a line from The Matrix: Free Your Mind.
(that was kinda corny, I apologize
It is stupid and irresponsible to setup a new DNS root. End of story, read
2826.
End of story for you. Others here would appear to prefer to discuss the
issue.
And.. if you let new.net do it then every other capitalist in the world
will start doing it and then the Internet will become disfunctional, and
what will that achieve?
If you let BigCompany operate a top-level Usenet hierarchy, then every other
capitalist in the world will start doing it and then Usenet will become
disfunctional, and what will that achieve?
Oh, that's right, Usenet is still usable, with fairly comprehensive
guidelines on how hierarchy and group maintainers can keep things running
smoothly. Even in the face of different news servers having different
namespaces. Funny, that.
(Usenet has it's problems, but distributed management of a hierarchy isn't
one of them, and hasn't been for a very long time.)
Stick to the single root, stand firm on this policy and new.net wont get
far!
What exactly scares you so much about this? The fact that individual
nameserver operators are realizing that they are the ultimate authorities
over their DNS infrastructure, rather than a political commitee? (Why am
I suddenly reminded of the "Grassroots DNS" effort that someone put forth a
few years back?)
For the record, I don't think this is being handled well, by any of the
parties involved. But I'm not concerned at all about new roots; in fact, I'm
very interested in seeing how this plays out. I see three possibilities:
a) The new alt. root on the block loses its funding, and we keep going under
the same strained system. Until next year's entrant to the alt. root camp
(the pattern is becoming fairly obvious at this point).
b) The alternative roots start gaining ground, and we begin seeing
individual ISPs operating "." zones for customers, tracking top-level
delegations themselves ala Usenet to ensure maximum availability of
names for their customers.
c) DNS shows its age and melts down, making way for more effective resource
location schemes. <URL:http://advogato.org/article/109.html>
Unfortunately, "the market" tends to consist in large majority of 1)
users, and 2) management. And we all know how bright those two
particular segments of the population tend to be.
Bright or not, these not-so-bright people have a direct impact on your
bottom line. Up to you if you want to ignore that, though.
If, as a group, the NANOG readership decides to take a single position on
anything (ha!), then we could very likely effectively determine in which
direction "the market" will go. After all, if _nobody's_ customers can access
new.net's non-sanctioned gTLDs, they can't very well go to another provider
for such access, and new.net will die the quick death that it deserves.
I read this and immediately remembered Michael Dillon; is he still around?
There were always interesting discussions about collusion and price-fixing
back in the good old days.
(I'm referring to the idea that the core Internet operators would collude
to drive a relatively new entrant to the field (new.net) out of business.
Yes, it's an anal way to look at it. No, IANAL. No, I don't have a point,
just reminiscing. 
John,
folk, life can be pretty simple. i recommend all the folk who want to play
multi-root games please do so. being unable to communicate with you will be
no problem to me. darwinian effects will do far better than all the blather
on mailing lists.
randy
It is stupid and irresponsible to setup a new DNS root. End of story, read
2826.End of story for you. Others here would appear to prefer to discuss the
issue.
talk is cheap. please go do it. and use it for the in-addr.arpa of your
mail smtp sender, please. or better yet, do your own special zone for
inverse. at least i won't have to read the blather any more.
randy
Oh, that's right, Usenet is still usable, with fairly comprehensive
guidelines on how hierarchy and group maintainers can keep things
running smoothly. Even in the face of different news servers having
different namespaces. Funny, that.
Usenet is awash in hierarchical disorder. Different servers have different
groups, at the whim of the server operator. Please let's not make DNS work
like Usenet. Thanks.
> And.. if you let new.net do it then every other capitalist in the world
> will start doing it and then the Internet will become disfunctional, and
> what will that achieve?Uhhh, so what if they do? Nobody is forcing you to use new.net, or ORSC
or TINC or whoever, or even setup your own roots with the "pick of the
bunch" from ORSC or TINC or whoever.If you don't use the alternative roots, you're not affected by their
existance... so you can just pretend they don't exist and live life in
your 2826 compliant happy world.
Oh yes I am (or may (will ?!)) be, because it does not depend on which root
I am using, but which root anyone wanting to send me email is using, as with multiple
roots there is potential that email sent by them (or though them) is either not going
to make it (becausethey or an intermediate site cannot resolve my email address)
or to someone else (because there is a duplication of domains).
Same goes for web etc.
The larger damage is going to be the legal fallout when multiple 'owners' of the
same domain (owned by different roots) start fighting for the right to 'their' name.
It will take many courts many years to decide that, with decisions even worse and diverse
than are right now coming out of the UDRP. Moreover, it will take just to decide on
which courts have jurisdiction (or if a court declares itself competent to make owners, registries
etc in other jurisdictions comply).
This is a can of worms we should *not* open. The only ones profiting by it would be lawyers,
and I think even they would not really welcome this.
Mathias
Oh, that's right, Usenet is still usable, with fairly comprehensive
guidelines on how hierarchy and group maintainers can keep things running
smoothly. Even in the face of different news servers having different
namespaces. Funny, that.
DO you think so? It *is* acceptable (to some, not everyone) that an article posted
in a group may not make it all the way around the world because my server's
copy of a NG is different from one with the same name over there.
But I don't think it would be at all acceptable if my email address had a different
owner over there. There *is* a difference in a basic naming scheme that is the basis
for a large multitude of communication systems and one that is specific to one
non-mission critical service.
What exactly scares you so much about this? The fact that individual
nameserver operators are realizing that they are the ultimate authorities
over their DNS infrastructure, rather than a political commitee? (Why am
I suddenly reminded of the "Grassroots DNS" effort that someone
put forth a
few years back?)
I don't know what scares him. What scares me is that a decision made by my
brother's ISP back in Germany might impact whether email he sends to me
could go to a different person altogether. And he would not have an easy choice
which ISP to chose, because each one might have a different view of the DNS and
namespace, thus making it impossible to decide which ISP to use at all.
The fact that services like email perform DNS lookups at every hop along the way
(from the initial MX lookups via the reverse lookps, sender-domain checks to
fight SPAM etc) would mean that no-one could say whether the emailevenhad a chance
reaching me.
And this is only email. Other services will introduce other chances for amiguities etc altogether.
The upshot of all this would be that no-one can rely on email and the Internet as a medium
for communication (and e-commerce etc) would be finished.
And don't think once this mistake is realized it would be easy (or fast) to turn bak the time
to a sane state, as AFAIK there is no complete record of the current state and disputes over
what is and is not to be part of the restored Internet would be legion.
Mathias
On or around Tue, Mar 13, 2001 at 11:37:26AM -0800, Patrick Greenwell may have written:
>
> Suddenly, I feel like I should be grabbing your hand and engaging in a
> rousing rendition of "Kumbayah my lord."
>
> Rather you or myself like it or not, given the existence of a player
> that has the capital to make this idea go, the market now has an
> opportunity to decide for itself.Unfortunately, "the market" tends to consist in large majority of 1) users,
and 2) management. And we all know how bright those two particular
segments of the population tend to be.
Well, those are the people defining your paycheck, sure you want to write
them off so quickly?
If, as a group, the NANOG readership decides to take a single position on
anything (ha!), then we could very likely effectively determine in which
direction "the market" will go. After all, if _nobody's_ customers can access
new.net's non-sanctioned gTLDs, they can't very well go to another provider
for such access, and new.net will die the quick death that it deserves.
You might want to take a long, careful, hard look at who has been doing
the sanctioning and how they've been making those decisions before jumping
on the bandwagon. Just a friendly suggestion.
I'm afraid you are fairly wrong there. 90% if the differences are in the
alt.* hierarchy which is explicity and area set aside for groups to be
created at random by anyone. Even in that area sites will have most of the
popular groups.
Most other hierarchies are either local or fairly well controlled (rec,
nz, uk) and groups are added and removed by pgp signed crontrol messages.
But this isn't really the place to discuss Usenet hierarchy policy.
On Tue, Mar 13, 2001 at 08:47:05PM -0800, Patrick Greenwell had this to say:
>
> Unfortunately, "the market" tends to consist in large majority of 1) users,
> and 2) management. And we all know how bright those two particular
> segments of the population tend to be.Well, those are the people defining your paycheck, sure you want to write
them off so quickly?
the very reason they pay my (all our) paycheck is for technical expertise -
if Joe Q. User had technical expertise sufficient to make informed decisions
on this type of matter, why would he need to hire a network operator?
I'm not saying that users, clients and management don't have their place -
but I _AM_ saying that place is _not_ in making critical _technical_ decisions
that will have a significant, possibly severely detrimental, effect on the
future of the networks they have hired _us_ to operate for them.
You might want to take a long, careful, hard look at who has been doing
the sanctioning and how they've been making those decisions before jumping
on the bandwagon. Just a friendly suggestion.
This whole matter boils down to one question - that being, what way is the
Right Way to operate DNS or its equivalent? It seems to me (and a few others)
that, logically, any hierarchical system _must_ have an ultimate authority -
not 2 or 3 or 27, which is essentially what new.net is trying to do: create
an alternate ultimate authority. How exactly will a user know which site
foo.com takes them to, if new.net's response and the rest of the Internet's
response a la *.root-servers.net don't jibe? The concept of unique and separate
domains breaks down when you have conflicting responses to the question, "Where
does this domain actually point?"
What some of us are saying is the new.net concept in its current forms is
_guaranteed_ to create exactly that kind of confusion, all arguments about
politics or alternate addressing possibilities aside.