[151477] in North American Network Operators' Group
Re: last mile, regulatory incentives, etc (was: att fiber, et al)
daemon@ATHENA.MIT.EDU (Owen DeLong)
Thu Mar 22 16:42:42 2012
From: Owen DeLong <owen@delong.com>
In-Reply-To: <CAKnNFz-hY0yCjxVMRi03X1_pmGqpH9h+YWu-GEx01ZbmjkaJjA@mail.gmail.com>
Date: Thu, 22 Mar 2012 13:40:27 -0700
To: chris <tknchris@gmail.com>
Cc: nanog@nanog.org
Errors-To: nanog-bounces+nanog.discuss=bloom-picayune.mit.edu@nanog.org
On Mar 22, 2012, at 10:12 AM, chris wrote:
> On Thu, Mar 22, 2012 at 12:26 PM, Jared Mauch <jared@puck.nether.net> =
wrote:
>=20
>>=20
>> On Mar 22, 2012, at 11:05 AM, chris wrote:
>>=20
>>> I'm all for VZ being able to reclaim it as long as they open their =
fiber
>>> which I don't see happening unless its by force via government. At =
the
>> end
>>> of the day there needs to be the ability to allow competitors in so =
of
>>> course they shouldnt be allowed to rip out the regulated part and =
replace
>>> it with a unregulated one.
>>=20
>> I think this partly captures the incentive case here, but there is =
also a
>> larger one at play. Over the years the copper infrastructure was =
installed
>> and extended through various incentive programs. You can see the
>> modern-day reflection of that in the RUS (used to manage rural
>> electrification act, part of USDA) and NTIA (Department of Commerce).
>>=20
Yes, I find it quite "amusing" that I am paying additional fees on all =
of my telecommunications services to subsidize high speed PON networks =
in rural bumf*ck while I can't get anything like it in San Jose, =
California.
>> The barriers to entry are significant for a new player in the =
marketplace.
>> The cost is putting the cabling in the ground vs the cost of the =
cable
>> itself. One can easily pick up hardware for $250 to light a single =
strand
>> of 9/125 SM fiber @ 10km for a 1Gb/s ethernet link. That's low =
enough you
>> could likely get a consumer to buy the hardware. The real cost is =
the
>> installation per strand foot/mile.
>>=20
Yes, at some point, we need to recognize that LMI (Last Mile =
Infrastructure) is and likely always will be a natural monopoly in all =
but the most densely populated areas (and actually even in many of =
those). THe market simply won't support the costs of deploying duplicate =
infrastructure installed by multiple providers. Given this fact, the =
only way to ensure competition in the services arena is to divorce the =
infrastructure from the services and require an independent operator of =
the infrastructure to make it available on an equal basis to all service =
providers.
>> In the past this has been subsidized for copper plant. There is no =
reason
>> in my mind that the fiber plant should be treated differently from =
this
>> standpoint. I can find fiber optic cabling for $0.25/ft. The =
problem here
>> is a multi-dimensional one that I've seen play out in a few markets:
>>=20
One reason the fiber plant should be treated differently is that we =
should learn from the mistakes we made with copper and we shouldn't =
continue to subsidize corporations to build out infrastructure that =
extends their ability to block competitors and should, instead insist =
that subsidized infrastructure is deployed in such a manner as to =
benefit all and support healthy competition for the services market.
>> It is my firm belief that without a regulatory regime it will not be
>> feasible to connect many communities robustly to modern =
communications
>> infrastructure. This could clearly change if the carriers involved =
see fit
>> to replace this infrastructure, but with their current debt loads, I =
think
>> it will be challenging to say the least.
>>=20
WHile I agree with you, the situation is already somewhat inverted in =
the US in that the existing USF subsidies have now made it more cost =
effective to build advanced networks into rural low-density subscriber =
bases than into moderately populated areas.
>> I do think we need a new last-mile regime in many areas, be it more =
"fair"
>> access similar to pole attach fees or the removal of local barriers =
to
>> build this infrastructure.
>>=20
The mechanism I have described above has been deployed in Sweden for =
some time now and is working out quite well from what I hear. It's also =
being tried in Australia now, much to the consternation of Telstra, but, =
it seems to be going well for the residents and businesses.
>> Some school and other governments here in Michigan would love to
>> sell/lease their excess fiber capacity to the private sector, but are
>> worried about turning a profit when it was built with taxpayer funds =
and
>> problems associated with that. I'd like to see these barriers =
removed. If
>> it's there, lets make it of value. If the school system turns a =
profit on
>> their enterprise, that's fine, it can lower the tax burden elsewhere.
+1
I do not understand this aversion to government having other sources of =
revenue besides direct taxation. If government can earn money from =
infrastructure it built with taxpayer money by leasing it to =
corporations or others, so long as it doesn't interfere with the =
original purpose for which the taxpayers funded its construction, I =
think this should absolutely be allowed and even encouraged.
>> Me? I'd be willing to pay $2500 to have Fiber built to my home. I =
might
>> even pay more. At this point, my research continues on building the =
fiber
>> and arranging my own easements for where to place it. I suspect you =
just
>> need a few geeks that are willing to part with some extra $ for fiber
>> bragging rights and one can build it.
There was a project in New Zealand that started out not too far off from =
what you are describing above and resulted in a fiber run that now =
stretches from one end to the other of one of their islands IIRC. It was =
presented at PacNOG in American Samoa a couple of years back.
Owen
