[arin-discuss] [arin-ppml] ARIN as a public interest business
Luke S. Crawford
lsc at prgmr.com
Thu Feb 23 16:45:03 EST 2012
On Thu, Feb 23, 2012 at 09:16:32AM -0500, Martin Hannigan wrote:
> Larger providers want fees to stay the same or higher? You're very
> wrong about that. Very wrong. Noone wants to pay higher fees,
> especially when ARIN has $30 million in cash sitting in the bank not
> working for the members in a way that we want it to work for us.
Hm. I am both small and low-margin. I recently obtained my first /20.
I do a lot of consulting for slightly larger (but still quite small
in the scheme of things) companies.
When I got my own /20 after five years of working to get enough users
to justify it? my per-ip costs immediately began falling as I
returned space to my upstreams, and I'm in a much stronger position to
negotiate new bandwidth contracts. From what I see from others
towards my end of the market? they'd be happy to pay quite a lot
more if it meant they would get their own allocation sooner (rather
than getting small blocks piecemeal from your upstreams, then
getting a direct block, then renumbering out of your upstream IPs.)
I mean, I'm sure other companies have different cost structures; some
of them may even have less revenue per IP than I do. But the thing I worry
about is "can I renumber out of all my PA space before runout?" relying
on PA space is an extremely frightening thing, especially as providers
even now are using runout as an excuse to raise prices.
I'm just saying, for me? I'd be quite happy paying ARIN 2x or 3x as
much if it meant, say, that some space would be reserved for when
I could justify it. Using IPs you don't have direct from ARIN
is a frighteningly expensive proposition. The cheapest PA /24s
I have cost me a grand a year. The most expensive PA /24s cost
me $384. And this is the line item on the invoice; I believe I'm
paying more than I need to for the rest of the services I get from
those providers because they know it's a huge pain for me to lose
those IPs before I finish the painful process of getting everyone to
move. Nearly all of those blocks were free with the bandwidth
when the contract started.
Certainly, not everyone feels this way, (and certainly, it's more
difficult for me to renumber than for most people, and my current
difficulties are largely unrelated to anything but some poorly-considered
promises I have made to my own customers.) but I can't tell you the
number of consulting clients (that were not large enough to justify
a direct allocation) that just wanted to write a large check to get
a large block from ARIN.
If anything, with v4 runout approaching, I'm glad they have something of a
war chest to help smooth runout. I mean, I don't claim to know what is
going to happen, but I'm pretty sure that if ARIN no longer has address
space, it's going to be a /whole lot/ more difficult for those of us
who came of age after CIDR and therefore don't have huge class B blocks
to compete in spaces that require lots of low-cost IPv4 addresses, like
the virtual private server market.
So yeah, if anything? I'd vote to charge me more if it means ARIN is
more prepared for runout. (I don't know if they can use money to help
solve that problem, but they are in a position to do something more than
anyone else is.)
More information about the ARIN-discuss
mailing list