Re: [sig-policy] Prop-050-v003 and reversability/modification?
I can certainly understand the urge to see action -- Geoff's
impressive collection of train wreck photos is always hovering in the
back of my own mind -- but I think the "humbled by the complexities"
sentiment is even more appropriate at this particular moment. Although
full privatization may not be the intent of this policy, it's
certainly one possible outcome (and one that has been contemplated for
a long time, c.f., http://tools.ietf.org/html/rfc1744). If things did
turn out that way, I'd speculate that the results would be at least as
profound as other recent major privatization initiatives -- e.g., all
of the assets in the most of the post-Soviet world (esp. Russia,
Poland, etc.). Those episodes teach that things rarely turn out as
Basically, my concern is not that we cannot capture the idea of limits
and/or reversability in policy-speak and incorporate that into a
transfer proposal; it's about whether such language would or could
have any binding effect after the fact. It seems to me that once a
transfer transaction has occurred, the result will necessarily
intersect will all kinds of property-related rights, obligations,
enforcement mechanisms, etc. In that context, conflicting RIR rules
might be interpreted as a kind of "prior restraint", albeit from an
alien (or perhaps inferior/subordinate) legal jurisdiction. In such
circumstances a transfer recipient might seek and have a chance of
receiving local injunctive relief from "foreign" RIR requirements.
IANAL, but that's what I'd do if for whatever reason I wanted to have
free and clear title to address space.
Anyway, I've copied an excerpt from a forthcoming paper on this topic
Maybe we could get some corporate attorneys from different national
jurisdictions around the region to assess this risk...?
" How then would IPv4 come to achieve the status of private property?
This would not need to happen all at once or as a result of a specific
action to “privatize IPv4 addresses.” Rather, it could happen
incrementally, beginning with something as simple as the material fact
of a private, bilateral exchange of IPv4 addresses for money. If IPv4
continues to be an essential, non-substitutable input for Internet
service delivery, then sooner or later transactions like this that are
not clearly mediated by a non-interested third party (i.e., something
like an RIR) are likely to occur with some regularity. Whereas
transactions in the context of such mediation might retain enough of
the RIR’s former special status to resist attributions of de facto
privatization -- whether made by buyers, sellers, or third parties –
unmediated transactions may retain few features meriting continued
special status or immunity from such claims. For example, will
accountants and tax officials regard such transactions as capital
expenditures or operating expenses? Will IPv4 purchased in one
jurisdiction and assigned in another be subject to tariffs and/or
transfer pricing restrictions? Will they be subject to sector-specific
direct investment requirements and/or restrictions? Will privacy
restrictions preclude the publication of WHOIS data, or mandate the
publication of data deemed to be privacy infringing in other
jurisdictions -- or alternately just immunize resource recipients from
registration requirements? Resolving uncertainties like this will
almost certainly result in many legal battles, any of which might
expose additional "latent ambiguities" that previously had been
obscured by the novelty of Internet technologies and the non-property
status that once kept address policy-related matters invisible, or at
least orthogonal, to most national legal systems. Eventually, at least
some rulings are likely to favor private property status, and in turn
those decisions could become precedents for future IPv4 private
property claims. Finally, given the freedom to act unilaterally in
ways that previously have been demonstrated to tip IPv4 into the
private property category, which status are commercial network
services providers more likely to prefer?"
On Aug 28, 2008, at 4:35 AM, Erik Kline wrote:
If there is a credible expectation that meaningful fine-tuning
possible after the fact, then that might make a radical undertaking
this somewhat easier for current dissenters to embrace...
This is essentially what I was trying to ask. If it's possible to run
it as an "experiment" where everyone gets to "taste test" and yet
retain a safe means to call a halt to it if it's all becoming too
bitter somehow (or too sweet?) then it's possible to measure things
that can only be conceived of in theory at this point.
I'm *naively* imagining the following:
1. The market would be open for a period of, say, 15 months (to
pick a random amount of time).
2. At the end of this time all transfers would stop and
registrations would remain fixed in the state on the date of the
closing of the market.
3. All those participating agree to be bound by these terms, i.e.
knowing full well that the market will close and may never reopen.
Certainly there's an argument that the promise of a market that will
eventually be "closed for renovation ... indefinitely" is so small
that it will not be attractive for players to participate at all.
However that too would be a useful data point to have collected: even
the opportunity of trading netblocks for X months wasn't interesting.
That might mean folks aren't (yet) desperate enough or ...? And if
they did start swapping, then that would invalidate the argument that
the value of trading netblocks is solely in the lack of oversight by
Being mostly or entirely clueless I don't know that I have much of a
stake or a vested interest in the outcome. I'm just a bit of an
anarchist and wanted to ask about taking an experimental, and
therefore data-driven, approach. (Not to imply that previous
approaches have not involved data or anything derogatory at all; I'm
not privy to anything that went before this.)
Ignorant beyond measure of the vast complexities,