ARIN proposal

[Tim Russell]

Steve Putz (putz at parc.xerox.com) writes:

> I have read the information on http://www.arin.net/, and would like to submit
> my request that the proposal *not* be implemented.
>
> In particular, the proposal does not explain why the proposed high fees are
> needed or how the money will be spent.  Also the proposed non-profit
> organization does not have a published charter, goals or mission.

    Now this message gets my vote for the best summary of my (and, I suspect,
many other peoples') feelings on this matter.

    Allow me to sum up the position I'm in, to (hopefully) put a little
perspective on this matter.

    I run Probe Technology, an ISP which, though small for many other areas,
is definitely one of the three largest in the Omaha, Nebraska, US area.
We started up almost exactly two years ago, and in that time have expanded
from a 256k connection and 16 dialup lines to a full T1 that's somewhat
rapidly nearing capacity and 120 (soon to be 144) dialup lines, 10 ISDN
dialups, and a plethora of dedicated connections via frame relay and ISDN.
We're expanding even more quickly as of late.

    Since usage on our MCI T1 has been climbing, we want to stay ahead of
the game and get the ball rolling on a second T1 before the need becomes
pressing, rather than after.  Also, rather than add another T1 to MCI (who,
I might add, has been an outstanding carrier for us), we'd like to take
the plunge and get a second provider and multihome.

    This may seem like a big move for a small ISP; in our opinion, it isn't.
While MCI has been great, there have been a handful of small periods where
their network has, understandably, "melted down" and we've been off the net.
This is unacceptable to us, if there's a way for us to avoid it.  MCI has
already told me that the cost of adding a second T1 to them will be double
the cost of our current connection, i.e. no price breaks.  Given that, it
seems patently silly for me to add a second connection to them, adding to
our dependence on their network and IP addresses.

    So, we'll be getting a bigger router and going to BGP.  This will cost
money, but it's money we're ready to spend to increase our reliability and
bandwidth.  We'll also want to get a block of provider-independent addresses
to avoid future renumbering for us and our customers.  We've gone through
it once, early on - it wasn't fun, and I don't even want to think about
it now.

    Yes, we'll be adding a CIDR block and the corresponding route in
routers all over the net.  I think this is a small price to pay to increase
net availability to a large portion of Omaha. I'm certainly not willing
to forgoe said reliability in the interest of saving a few bytes in
a router, which will be taken up in any case by the next guy along who's
more than willing to compete with us on the basis of our not being multi-
homed.  I'm willing to take someone else's entry and pay the expenses
of memory and CPU cycles, I would expect the same from them.

    Even though $2500 a year may not seem like a large sum of money to
some people on this list, it's a good chunk of money to us.  That's not
to say that we can't afford it; we certainly can.  Nevertheless, it's
not an amount that I'm willing to just blindly hand over to an organization
that looks to me like it's wide open for Internic control and not especially
accountable to its members.  In fact, it looks like CIX all over again to me.

    Forgive me, but I simply don't see what it is in all this that makes
$2500 an acceptable figure to charge for a block of addresses.  IN-ADDR
service, unless I'm mistaken, doesn't take any more resources to provide
than a domain name, and neither does database maintenance on contacts
for address blocks.  Routes themselves will be maintained by the seperate
backbone providers, not ARIN, so that's not a factor, again, unless I'm
mistaken.

    I have no delusions that I'll get much of a say in the day-to-day
operations of ARIN with a "small" provider membership, and as for the rest,
it doesn't seem to me that the services rendered are more than that for a
domain name, and are in fact less since updates will be less frequent. I'm
/not/ willing to pay the overinflated salary of a CEO for an organization
that has no competition and no need to compete.  In fact, this seems to me
to be one of those situations where anyone who wants the job will most
likely be the worst candidate.

    So, given that, if someone can explain to me exactly what it is that
requires such a large sum of money, I (and, I suspect, many others) will
be much more willing to get behind this proposal.

    My main point in all this is that there are many areas of the country
that rely on "small" ISPs, and simply brushing the issue off by saying
that we'll only be paying an extra $10 to our current provider doesn't
cut it.  It offends me, in fact, because it isn't the issue.  The issue
is the future of the net as a whole, and whether it will be enhanced
or not by accepting the current proposal.

    I hope my comments have added to the discussion.  Please feel free
to correct me on any points; I'm always open to (constructive) criticism,
and although I've used the net daily since 1988, I know I'm definitely
not the most knowledgable person out there.

--
Tim Russell      System Admin, Probe Technology      email: russell at probe.net
   "It has become appallingly clear that mankind's technology has finally
    surpassed his humanity."                            - Albert Einstein