<div dir="ltr"><br><div class="gmail_extra"><br><br><div class="gmail_quote">On Tue, Jun 3, 2014 at 12:31 PM, David Huberman <span dir="ltr"><<a href="mailto:David.Huberman@microsoft.com" target="_blank">David.Huberman@microsoft.com</a>></span> wrote:<br>
<blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">
<div link="#0563C1" vlink="#954F72" lang="EN-US">
<div>
<p class="MsoNormal">We had a discussion today at NANOG in the ARIN PPC about needs-basis in 8.3 transfers.
<u></u><u></u></p>
<p class="MsoNormal"><u></u> <u></u></p>
<p class="MsoNormal">I’d like to state the following, and then let’s see where the discussion takes us:<u></u><u></u></p>
<p class="MsoNormal"><u></u> <u></u></p>
<p class="MsoNormal">My team runs an AS. And yep, we’re a pretty big company. We rely on IPv4 today for most of our numbering, and will continue to do so for the next couple of years.[1] In the coming year, when we can’t get space from ARIN or other RIRs,
we have to turn to the market for our IP address needs. We may choose to buy more than a 2 year supply, because it may make business sense for us to do so. ARIN policy, however, only allows us to take the IP addresses we buy and transfer the portion which
represents a 2 year need. The rest will remain in the name of whoever sold the IP addresses to us.
<u></u><u></u></p>
<p class="MsoNormal"><u></u> <u></u></p>
<p class="MsoNormal">Why is this result good for the operator community? Wouldn’t it be better if ARIN rules allowed us to transfer into our name all the IP addresses which we now own?<u></u><u></u></p>
<p class="MsoNormal"><u></u> <u></u></p>
<p class="MsoNormal">Regards,<u></u><u></u></p>
<p class="MsoNormal">/david<u></u><u></u></p>
<p class="MsoNormal"><u></u> <u></u></p>
<p class="MsoNormal">[1] We’re working on increasing IPv6 presence in our network and our products, but large corporations move slowly ;)</p></div></div></blockquote><div><br></div><div>See, this is why I support maintaining the<br>
needs-based decisionmaking around number<br>allocations.<br><br>Because it's far too easy for a really big company<br>with a couple of billion dollars in the bank to decide<br>that IPv6 is just too hard, and it's easier to buy up<br>
</div><div>large blocks of IPv4 space, and keep their critical<br>resources on v4 addresses--which, if those resources<br></div><div>are crucial enough, could artificially drive up demand<br>for IPv4.<br><br></div><div>As a purely hypothetical exercise, let's consider<br>
the three major smartphone OS vendors; each<br>of them have over $100B in assets in the bank<br></div><div>at the moment; and each of them have a relative<br>lock on the mobile phone OS for a given set of<br></div><div>handset hardware. If they each decided that<br>
IPv6 was too hard, and they could get enough<br>IPv4 space on the unrestricted market (remember,<br>with over $100B each in assets, at $12/IP, each<br></div><div>of them could in theory be able to afford every<br>remaining IP address on the planet, with cash<br>
left over). If they kept their software update<br>servers on IPv4, they'd have a lock on over<br>a *billion* endpoints[1] that they would force a<br>requirement for IPv4 onto. That would<br></div><div>generate an ongoing demand for IPv4<br>
</div><div>resources which they would have an<br>ironfisted control over. Any mobile carrier<br>that wanted to provide service for a <br>smartphone would have to ensure the<br>phone could reach the IPv4-only software<br>
update servers.<br><br></div><div>Yes, this is a bit of hyperbole; I'm not expecting<br>the big three smartphone OS vendors to go out<br>and buy up all the IPv4 space remaining, just to<br>have absolute control over the global portable<br>
communications infrastructure. But the numbers<br></div><div>do indicate that would be a not-infeasible scenario.<br><br>Even Apple alone, with their assets in the bank,<br>could afford to buy up every remaining IPv4<br>
address at current market prices; requiring just<br></div><div>ios devices alone to speak IPv4 to get software<br>updates would force pretty much every major<br>mobile carrier in the world to have to maintain<br>IPv4 infrastructure for the foreseeable future,<br>
and could potentially put them into a situation<br>where they would have to lease IPv4 addresses<br>for their handsets from Apple, in order to be<br>able to provision their own customers.<br><br></div><div>The ability for one entity to control both ends<br>
of the connection; at the client device, and on the<br>server side, and with enough cash and assets<br>to create a monopoly on the means by which<br>those two endpoints communicate...see, that's<br>why I will continue to vote in favour of needs-based<br>
justification for resources; because the temptation<br></div><div>to have that level of absolute control over a<br>resource is too risky to leave without some level<br></div><div>of community input as well.<br></div><div>
<br>Thanks!<br><br>Matt<br></div><div>(being paranoid, yes--but also recognizing<br>that greed can drive anti-social behaviours)<br></div><div>P.S. Apologies to Apple--I used you simply as an example<br></div><div>to highlight that in responding to David, I did not mean to<br>
imply in any way that my concern was only about Microsoft;<br></div><div>any of the dominant smartphone OS vendors has the same<br>level of capability at this point in time.<br></div><div><br></div><div>[1] <a href="http://www.bloomberg.com/news/2012-10-17/smartphones-in-use-surpass-1-billion-will-double-by-2015.html">http://www.bloomberg.com/news/2012-10-17/smartphones-in-use-surpass-1-billion-will-double-by-2015.html</a><br>
<br></div></div></div></div>