Re: [DNSOP] In a vacuum, nobody can hear you scream, was On the call for adoption on Special Use Names

[hellekin]

On 10/07/2016 06:36 PM, Alain Durand wrote:
> 
> However, there is something that can be done before: provide a safe place
> in the DNS tree where people can exist without colliding with the rest of
> the tree. We can't prevent people from ignoring it and keep using whatever
> name they want, but at least we would have provided a way to play nice. In
> that spirit, efforts like .alt and friends are a step in the right direction.
> 

We have .example and example.* for documentation, yet the XMPP
documentation uses shakespeare.lit (I don't think .lit matches any SUN
or any entry in any DNS-related RFC.) FWIW, wikipedia sends .lit to some
Microsoft file extension.  One cannot say that Peter St Andre is
ignorant of IETF processes.  Use of *example* in documentation and
.invalid in free software is a good sign that developers are ready to
follow suit and respect the norms.

==
hk

___
DNSOP mailing list
DNSOP@ietf.org
https://www.ietf.org/mailman/listinfo/dnsop

Re: [DNSOP] In a vacuum, nobody can hear you scream, was On the call for adoption on Special Use Names

[Alain Durand]

> On Oct 7, 2016, at 6:51 AM, John Levine  wrote:
> 
> f someone creates popular software leaking requests for
> .PICKLE, we can grouse all we want but since we're not the Network
> Police, there's not much we can do about it.

There is not much that can be done after the fact, I agree. However, there is 
something that can be done before: provide a safe place in the DNS tree where 
people can exist without colliding with the rest of the tree. We can't prevent 
people from ignoring it and keep using whatever name they want, but at least we 
would have provided a way to play nice. In that spirit, efforts like .alt and 
friends are a step in the right direction.

Alain
___
DNSOP mailing list
DNSOP@ietf.org
https://www.ietf.org/mailman/listinfo/dnsop

Re: [DNSOP] In a vacuum, nobody can hear you scream, was On the call for adoption on Special Use Names

[John Levine]

> If someone can just start using a name and thus make it too hard
> to delegate we have a much a bigger problem.

That's been true approximately forever, viz. .onion, .belkin, .corp,
.local, .mail, and probably still .uucp that are too poisoned to allow
reliable delegation and new use.  That's why we're here.

The fundamental and I hope obvious problem, is that the IETF and ICANN
have no control over software that people write and the hardware they
embed it in.  If someone creates popular software leaking requests for
.PICKLE, we can grouse all we want but since we're not the Network
Police, there's not much we can do about it.

R's,
John

___
DNSOP mailing list
DNSOP@ietf.org
https://www.ietf.org/mailman/listinfo/dnsop

Re: [DNSOP] In a vacuum, nobody can hear you scream, was On the call for adoption on Special Use Names

[George Michaelson]

If you can come up with an efficient, "fair" and trusted process for a
unitary name space on domain principles (domains of scope. trees.)
that doesn't confront collisions over desires for labels at arbitrary
points in the tree, and of essential 'centrality' in decision making
logic over things especially the apex of the tree, computer science
would like to know.

Meantime, we have this tree, and we have a lot of documentation around
this tree, and we have a current bilateral view between two agencies
on this tree, and we're discussing this tree, in the context of one of
those agencies: we're using IETF infrastructure, IETF processes, IETF
methodologies, to discuss that tree.

I agree pejorative language doesn't help, and I share responsibility
for its over-use. I apologize for intemperate use of language.

Peer to peer, hash based, location-id separator,  all
discuss concepts which collide in this model.

It might surprise you to know, that outside of this conversation I
hold different views about social equity, and who should or should not
be vested with authority in names. I try to draw distinctions between
what I think as a consumer, and a user, and what I observe from my
training and praxis.

I hold a unitary name space as a public good in very high regard. I
think p2p models, and models of probabalistic or hash naming are
interesting, but they wind up needing to map coherently to DNS names.
What I depart from, in the conversation, is how high in the DNS tree
that coherence has to vest.

A lot of your commentary goes to procedural fairness. I won't pretend
we don't have a problem there. I think you, and others in development
of novel systems have a right to feel severely disadvantaged by
process as it stands.

-G

On Thu, Oct 6, 2016 at 7:48 PM, hellekin  wrote:
> On 10/06/2016 09:22 AM, avri doria wrote:
>>
>> As for the so-called toxic waste names (i really find that terminology
>> problematic)
>>
>
> I agree it's a problem to use that kind of vocabulary to convey a
> technical context.
>
>> the so called waste pile of usurped names
>>
>
> Therefore this is also a problem to call names-used-in-the-wild
> "usurped" or "squatted", because it says that there's a central body
> that assigns names, and it defines who can use them, with the
> exclusivity of any other approach.  I know this idea may sound funny to
> a lot of people given the missions of IANA and ICANN, and the existence
> of trademarks and so-called 'intellectual property', but to me, having
> an authority over who can use what names *in general*--as opposed to
> particular, specific cases (e.g., trademarks)--is akin to the Novlang
> Committee.
>
> Names in the DNS are sanctioned by IANA/ICANN, and those names are
> 'legitimate' in the context of Internet names.  That doesn't mean at all
> that names not sanctioned by ICANN are illegitimate, or that names
> covered by trademarks are more 'legitimate' than 'unprotected' names.
> This is all a matter of transactions and legal-firepower.  But from
> there to legitimate this transactional-belligerent perspective over any
> other (historical, cultural, incidental, ontogenetic, etc.) seems to me
> problematic and abusive.
>
> ==
> hk
>
> ___
> DNSOP mailing list
> DNSOP@ietf.org
> https://www.ietf.org/mailman/listinfo/dnsop

___
DNSOP mailing list
DNSOP@ietf.org
https://www.ietf.org/mailman/listinfo/dnsop

Re: [DNSOP] In a vacuum, nobody can hear you scream, was On the call for adoption on Special Use Names

[hellekin]

On 10/06/2016 09:22 AM, avri doria wrote:
> 
> As for the so-called toxic waste names (i really find that terminology
> problematic)
>

I agree it's a problem to use that kind of vocabulary to convey a
technical context.

> the so called waste pile of usurped names
>

Therefore this is also a problem to call names-used-in-the-wild
"usurped" or "squatted", because it says that there's a central body
that assigns names, and it defines who can use them, with the
exclusivity of any other approach.  I know this idea may sound funny to
a lot of people given the missions of IANA and ICANN, and the existence
of trademarks and so-called 'intellectual property', but to me, having
an authority over who can use what names *in general*--as opposed to
particular, specific cases (e.g., trademarks)--is akin to the Novlang
Committee.

Names in the DNS are sanctioned by IANA/ICANN, and those names are
'legitimate' in the context of Internet names.  That doesn't mean at all
that names not sanctioned by ICANN are illegitimate, or that names
covered by trademarks are more 'legitimate' than 'unprotected' names.
This is all a matter of transactions and legal-firepower.  But from
there to legitimate this transactional-belligerent perspective over any
other (historical, cultural, incidental, ontogenetic, etc.) seems to me
problematic and abusive.

==
hk

___
DNSOP mailing list
DNSOP@ietf.org
https://www.ietf.org/mailman/listinfo/dnsop

Re: [DNSOP] In a vacuum, nobody can hear you scream, was On the call for adoption on Special Use Names

[avri doria]

On 04-Oct-16 09:19, David Conrad wrote:
> As far as I know, neither ICANN (the organization) nor anyone within
> ICANN (the organization) is asking whether they should delegate such
> names. Forward motion of those names is currently "indefinitely
> deferred" pending _somebody_ (not ICANN staff) figuring out what to do
> with them. I believe the hope had been that the IETF might provide
> some technical guidance, but that didn't work. Now, some members of
> the ICANN community are asking the board that those names be delegated
> and that results in (re)opening the question of what to do with
> "indefinitely deferred" strings.

Actually I thought they were asking that work that had been promised on
further researching the problem and mitigation techniques be done as
opposed to just prohibiting things because the first thoughts turned out
to be inadequate.

As for the so-called toxic waste names (i really find that terminology
problematic) someone needs to find a solution otherwise the possibility
of adding more and more names to the so called waste pile of usurped
names over time becomes an increasing possibility.  If someone can just
start using a name and thus make it too hard to delegate we have a much
bigger problem.

avri

avri


---
This email has been checked for viruses by Avast antivirus software.
https://www.avast.com/antivirus

___
DNSOP mailing list
DNSOP@ietf.org
https://www.ietf.org/mailman/listinfo/dnsop

Re: [DNSOP] In a vacuum, nobody can hear you scream, was On the call for adoption on Special Use Names

[Jeremy Rand]

David Conrad:
> 
> The P2P crowd would like to carve out some 
> names to run their resolution scheme in parallel with the DNS, and it 
> appears they'd also like an authority they can point at. 
> Well, some do. To be honest, it feels to me that some appear to want to say 
> "we don't like ICANN" or, more generally, "Screw you, Establishment!" 

Hope I'm not stepping on any toes by joining this discussion, but I'd be
surprised if those sentiments were coming from any Namecoin people.  The
Namecoin developers are well aware that Namecoin makes different
tradeoffs from DNS, and that there are plenty of ways that Namecoin is
inferior to DNS as managed by ICANN.  We're actively attempting to
minimize those deficiencies within our design constraints, but some of
the issues are open research problems that may or may not have a
solution.  As such, I hear Namecoin people criticize Namecoin's
technology far more often than I hear them criticize ICANN.

We're trying to build something cool, not tear down ICANN.

(Maybe you were talking about some other P2P naming project, in which
case I apologize for entering this conversation.)

Cheers,
-Jeremy Rand
(Namecoin developer)



signature.asc
Description: OpenPGP digital signature
___
DNSOP mailing list
DNSOP@ietf.org
https://www.ietf.org/mailman/listinfo/dnsop

Re: [DNSOP] In a vacuum, nobody can hear you scream, was On the call for adoption on Special Use Names

[David Conrad]

Hi,

On October 3, 2016 at 5:14:24 PM, John Levine (jo...@taugh.com) wrote:
ICANN (or perhaps some people within ICANN) are 
asking whether they should delegate .corp, .home, and .mail and 
presumably other toxic waste names, and want an authority they can 
point to for the answer. 

Just a clarification:
As far as I know, neither ICANN (the organization) nor anyone within ICANN (the 
organization) is asking whether they should delegate such names. Forward motion 
of those names is currently "indefinitely deferred" pending _somebody_ (not 
ICANN staff) figuring out what to do with them. I believe the hope had been 
that the IETF might provide some technical guidance, but that didn't work. Now, 
some members of the ICANN community are asking the board that those names be 
delegated and that results in (re)opening the question of what to do with 
"indefinitely deferred" strings.

The P2P crowd would like to carve out some 
names to run their resolution scheme in parallel with the DNS, and it 
appears they'd also like an authority they can point at. 
Well, some do. To be honest, it feels to me that some appear to want to say "we 
don't like ICANN" or, more generally, "Screw you, Establishment!" 

I suppose it's flattering that everyone is looking at us, but as we are 
seeing, just because a vacuum sucks (by definition, after all) does not 
necessarily mean we are qualified to fill it. 

Not everyone. I (and I think Paul W) have been suggesting that the IETF is not 
really the best place to deal with the implications of trying to fill that 
vacuum -- too many lawyers smelling blood in the water. The new gTLD 
Applicant's Guide Book was 300+ pages for a reason and it wasn't, as some have 
(loudly) suggested, because ICANN (the organization) is evil or greedy or 
incompetent, rather the issues involved in dealing with a resource that can 
only be allocated to one entity when multiple entities might have an arguably 
valid claim to the resource, get complicated quite quickly and when money is 
involved (which names seem to attract), lawyers follow and it gets ugly fast. 

There be serious non-technical dragons here. I don't speak for ICANN, but I 
suspect ICANN (the organization) would love to have a list to point at that 
says "can't delegate these because the IETF say so -- talk to them about why" 
just as ICANN points to ISO-3166/MA when someone comes and demands their 
2-letter made up string should represent their "country." This may not be 
career enhancing, but speaking as an IETF participant (which I assume we all 
are), it isn't clear to me this would be prudent if we want the IETF (or 
rather, it's legal parent(s)) to be a viable entity in the long run. 
Particularly if the "why" turns out to be the winner of a beauty contest 
decided by the IESG as 6761 current suggests.

Regards,
-drc
(ICANN CTO, but speaking only for myself)



signature.asc
Description: Message signed with OpenPGP using AMPGpg
___
DNSOP mailing list
DNSOP@ietf.org
https://www.ietf.org/mailman/listinfo/dnsop

Re: [DNSOP] In a vacuum, nobody can hear you scream, was On the call for adoption on Special Use Names

[John Levine]

>The second problem was that CAs refused to sign certificates for .onion. Again,
>this was not an IETF problem. But somehow the tor project managed to put
>pressure on the IETF to grant them that name.
>
>I'd say this is a problem.

With the vast amount of money and effort spent on Internet Governance
you wouldn't expect to find a governance vacuum, but whaddaya know,
that's what we have.

The CAs are asking if they should sign .onion and presumably other
oddball names that come along, and want an authority they can point to
for the answer.  ICANN (or perhaps some people within ICANN) are
asking whether they should delegate .corp, .home, and .mail and
presumably other toxic waste names, and want an authority they can
point to for the answer.  The P2P crowd would like to carve out some
names to run their resolution scheme in parallel with the DNS, and it
appears they'd also like an authority they can point at.

I suppose it's flattering that everyone is looking at us, but as we are
seeing, just because a vacuum sucks (by definition, after all) does not
necessarily mean we are qualified to fill it.

R's,
John

___
DNSOP mailing list
DNSOP@ietf.org
https://www.ietf.org/mailman/listinfo/dnsop