[arin-ppml] ARIN-prop-172 Additional definition for NRPM Section 2 - Legacy Resources

Chris Engel cengel at conxeo.com
Tue Jun 12 16:39:41 EDT 2012


> [Milton L Mueller] OK, but exclusivity exists in the same sense and for the
> same reasons that exclusivity exists in the occupation of a movie theater
> chair. If I am already sitting in a chair it causes trouble for both of us for you to
> come and sit in it. (conflict) In most cases it causes more trouble than to go
> and sit in another chair. I may have no legal right to that seat (my ticket may
> not be seat-specific) but most people will avoid conflicting claims for practical
> reasons. However, when seats get really scarce (as they are, e.g., in a Hong
> Kong movie theater as opposed to a Syracuse one) ticket sales get tied to
> exclusive, defined seat assignments. Viz, my earlier point about
> "propertization." If the word 'property rights' frightens you, call it something
> else, such as "scarcity-intensified exclusivity"


Professor Mueller,

Thank your for your courteous response. I think this is where a bit of my disconnect lies with the concept being promulgated. The movie theater chair is a finite location in a finite space. This does not necessary hold true for an IP Address.  An IP address may be a finite location but it can exist in a potentially INFINITE number of spaces.  ANY number of people can potentially occupy that chair as long as they do so within their own context. Indeed, as long as the context is understood, you and I can BOTH sit in the same chair without conflict. That's why you could have a device which resides at 10.1.1.1 (an RFC 1918 address) on your network and so could I and  no conflict exists between the two as long we have some means of recognizing the context in which the address resides. In that sense, one might consider an IP address to be a movie theatre chair which exists within an infinite number of parallel universes where one can spawn ones own parallel universe at any point in time.

It's only when you ask the question, "Who's sitting in that chair?" without being able to reference the context of the question that the potential for conflict exists. The answer, of course, depends upon who you choose to use as a reference. ARIN (and the other RIR's) are ONE source for a reference and the de facto one that MOST operators tend to utilize to answer MOST queries. It is important to note that nothing actually compels any given operator to use ARIN for every or ANY given query and some occasionally don't.  ARIN only is able to act as the de facto reference for such queries because it provides a high degree of utility for most operators to generally choose to do so. If ARIN, for some reason, ceased to provide such a high degree of utility.... if for example it operated in a fashion contrary to what most operators considered useful, it would cease to have influence as reference source.

I think the concept might have a bit more traction with me if there was, in practical reality, such a thing as a homogenous "public internet". The reality seems to me a good bit more complex and the internet is a much more amorphous and heterogeneous entity then the model would suggest. It consists of a vast and ever changing array of autonomous and semi-autonomous networks run by a wide variety of public and private entities. Some of these networks are public, others private, semi-private, partially private, etc. all of which are connected to each other in a variety of complex ways and even includes networks operated by entities which don't recognize each others legal existence (i.e. China & Taiwan).

I think the crux of the matter lies in the answer to 2 questions.....

Can ARIN be compelled to update databases which it OWNS in a manner contrary to what it perceives as the interests of the constituency which it serves?

Can a vast array of private operators be compelled to utilize ARIN as an authoritative reference for queries that they find fundamentally against their interests to resolve in that manner?  Including operators in jurisdictions that may not even recognize the legal existence of the authority seeking to compel them?

Unless the answer to both those questions is "YES", I'm not sure how the property-rights model could usefully apply to IP Addresses.

Even if the answer to the above is "YES", does the internet continue to function in a reasonably useful function?

I don't often find myself agreeing with Owen on alot of issues, but in this case what he's saying makes alot of sense to me. ARIN, pending contractual agreements it's made, should have the right to update it's own listings according to it's own policies. If it starts updating it's listings in a manner contrary to the interests of the vast majority of the entities utilizing it, as a practical matter it's utility as a reference source becomes vastly reduced....hence not a good idea for it to do so...or for most entities to try to compel it to do so...and I think most entities involved actually recognize that. 

I think that also speaks to the other point which has been raised. Most entities seeking title to something aren't going to be interested in delegitimizing the authority of the Title Office that they are seeking to have recognize their claim by their actions, as it devalues the claim itself. Thus I strongly suspect entities will seek to work within ARIN's policies in conducting transfers, if at all possible as it is self-defeating for them not to do so. Conversely, I also suspect that ARIN will generally try to find a way to work with the entities seeking transfer to allow that to happen within policy if it is at all possible to do so. It's only when it's very clearly against the interests of the constituency which ARIN serves would it make sense for ARIN to do otherwise. YMMV.


Christopher Engel 


> -----Original Message-----
> From: Milton L Mueller [mailto:mueller at syr.edu]
> Sent: Tuesday, June 12, 2012 2:32 PM
> To: Chris Engel
> Cc: 'arin-ppml at arin.net'
> Subject: RE: [arin-ppml] ARIN-prop-172 Additional definition for NRPM
> Section 2 - Legacy Resources
> 
> 
> > -----Original Message-----
> > From: Chris Engel [mailto:cengel at conxeo.com]
> > spectator and someone who is also not a lawyer. I don't believe that
> > technically ARIN has or even could "hold rights over the exchange" of IP
> > addresses.  That would imply that ARIN has the authority to exclude
> > entities that it has no contractual relationship with from use of those
> > addresses and do so across multiple jurisdictions, including those which
> 
> [Milton L Mueller] Thank you. Simple, reasonable and correct point.
> 
> > It's the same reason
> > why I believe IP addresses don't really count as "property"  in the
> > traditional sense (as the holder lacks the right or means to exclude
> > others from their use).
> 
> [Milton L Mueller] OK, but exclusivity exists in the same sense and for the
> same reasons that exclusivity exists in the occupation of a movie theater
> chair. If I am already sitting in a chair it causes trouble for both of us for you to
> come and sit in it. (conflict) In most cases it causes more trouble than to go
> and sit in another chair. I may have no legal right to that seat (my ticket may
> not be seat-specific) but most people will avoid conflicting claims for practical
> reasons. However, when seats get really scarce (as they are, e.g., in a Hong
> Kong movie theater as opposed to a Syracuse one) ticket sales get tied to
> exclusive, defined seat assignments. Viz, my earlier point about
> "propertization." If the word 'property rights' frightens you, call it something
> else, such as "scarcity-intensified exclusivity"
> 
> >  From a practical sense, that would require ARIN
> > to exercise control over networks, routers, NIC's , etc. that somebody
> > else OWNS and again across multiple jurisdictions.
> 
> [Milton L Mueller] which we both seem to agree would be a Bad Thing
> 
> > At best, someone might be able to bring a case of torturous interference
> 
> [Milton L Mueller] <chuckle> tortious. This thread is approaching the
> torturous, however, so perhaps the pun was intended.
> 
> > against someone using an IP address that ARIN officially assigned to
> > someone else.  However, even in that case, I believe, there would have
> > to be some reasonable expectation that the end user had a right to
> > access those services over that particular network. That might fly for
> > most transit provider networks, but probably not for the vast majority
> > of end user networks.  For example, if as a Private Enterprise End
> > Network, I null routed Facebook's IP address(s) or even routed them to
> > an internal web-server that hosted the corporate policy against
> > accessing social media sites at work, I don't believe either ARIN or
> > Facebook would have effective legal recourse against me doing so.
> 
> [Milton L Mueller] interesting point. But it is the routing right you are
> discussing, not the exclusivity of address assignments for the public internet.
> And I believe that even Owen "Chairman Mao" DeLong believes that
> operators have private property rights over the instructions of their routers.
> 
> > What, I BELIEVE, ARIN exercises the rights over is listings within their
> > OWN databases.  Since most operators consider that the, de facto,
> > registry of record for IP addresses, that has a vast influence over any
> > IP address that you want to be widely accessible through public
> > networks. Even here, I don't believe there is anything that actually
> > compels operators to use ARIN's registry, other then common sense and a
> > desire for smooth interoperability.
> 
> [Milton L Mueller] I think you are trying to say that ARIN _owns_ the registry
> database; i.e., has exclusive authority over what goes into and out of it? Wink
> wink, nudge nudge
> 
> > None of this speaks to what ARIN'S policies SHOULD be toward transfers
> > or legacy addresses.
> 
> [Milton L Mueller] Correct! The attempt to come up with a clear definition of
> legacy has been hijacked by the foaming at the mouth caused by the
> conjunction of "IP addresses" with the word "property"
> 
> 




More information about the ARIN-PPML mailing list