last mile, regulatory incentives, etc (was: att fiber, et al)

Mailman List Mirror (Read Only)
21

Hmm even most urban environments aren't worth deploying in or are probably marginal profit. So I would expect 30-45% of population of the US to not be worth or marginally worth deploying. I am assuming most urban less than 250k and probably spread out. Not to mention to provide transit without services to residential is a margins game to begin with and without at least a 20-30% take rate it probably isn't worth the cost of l3 infrastructure. On the other hand for actual dense urban environments it makes perfect sense as long as the are willing to maintain it.

I see the possibilities, but have a gut feeling it would become a political mess and unreliable, not to mention cost us more than we pay now.

22

wiring center you enable all technologies. GPON today, direct GigE
or 10GE where necessary, and all future technologies.

yep, agreed - much more sensible, much more resilient to failure and only
marginally more expensive.

It'll never be done though. Too much to lose by creating a topology which
allows you to unbundle the tail.

Nick

23

In a message written on Sun, Mar 25, 2012 at 07:15:47PM +0100, Nick Hilliard wrote:

It'll never be done though. Too much to lose by creating a topology which
allows you to unbundle the tail.

Only if it is your capital building the tail.

Today's Internet companies are still trying to achive penetration
to the 30-40% of households that are cheap to reach, and profitable
as customers in commercial timescales (3-5 years). As we reach
saturation in that market (less than 10 years from now, I think)
they will have to look to the "unprofitable" customers as the only
real source of new business.

The economics then change. While it's better to have it be your
asset and create a customer lock-in, when the risk is high enough
it will be seen as better to have a municipality or other take on
the risk even if it means unbundled competition.

Phone provides the history here; telephone companies started out
with only the most profitable companies. To reach the commercially
unprofitable ones they turned to government, in the form of things
like the Rural Electrification Act (governemnt backed loans to rural
providers) and the Universal Service Fund. These government subsidies
were also a _major_ driver in the argument that copper local loops
should be unbundled since in a lot of cases government had paid for
them, not private companies.

Politically the makings of a similar situation already exist.
Goverment has swung the USF funds to fuel rual broadband, strongly
favoring FTTx where it makes sense. While companies like Verizon
enjoy not having to share their fiber lines now, these same forces
will conspire to drive unbundling in fiber, just as it did in copper.
What they are getting now is simply a first mover advantage.

Government at the end of the day will fund the 20-40% of America
which is profitable in the long run, but not in commercial time
scales. They will also fund the 10% of America which will never
be profitable, no mater what. It happened with Electricity and
Telephone, and I suspect the societal drivers to do the same with
the Internet will be even stronger. Companies will have to accept an
unbundled tail to get access to this 30-50% of the market; and while
they aren't interested now, they will be very soon.

24

In a message written on Sun, Mar 25, 2012 at 11:47:58AM -0400, Jay Ashworth wrote:

Well, for my part, /most of the poiny/ of muni is The Public Good; if /actual/ bond financed muni fiber is skipping the Hard Parts, it deserves to lose.

It doesn't matter if it's a bond-financed project or a privately funded (privately owned) project - they are using a public resource (the street/poles) to lay their lines, and usually also using the power of the municipality's right to eminent domain to put in or use poles (or underground conduits) to run lines across private properties. As part of the Public Good contract to use these public resources, they should be required to service both the the easy parts and the hard parts, no matter the source of the financing or the ownership of the lines.

If a commercial company goes in to serve folks with fiber they
expect a relatively short ROI, 3-5 years typically. This is why
rural customers aren't "profitable"; they can't get money from a
bank or wall-street for a longer time so they are trying to spread
out the build costs over too short of a recoupment period.

Fiber has a 20-50 year life.

The biggest problem is determining how certain that lifespan is. Remember how Netflix looked like an awesome business to deliver DVDs by mail in 2002, and had one of the most successful IPOs of the era? Less than 10 years later we have widespread broadband and companies can deliver that same content by copper/fiber/802.11. Now Netflix is in the position of being in direct business conflict with the companies they rely on to carry their product to their customers (e.g. Comcast) and their future is very uncertain. Can you promise that fiber has a *feasible* lifetime of 20-50 years? Maybe in 5-10 years all consumer data will be transferred via wireless, and investment in municipal wired data systems (fiber and copper) becomes worthless.

This is why most modern build-outs have to show a ROI of under 5 years. We just don't know what new technology breakthroughs might happen, which could make a project that requires a 10-30 year payback schedule go bankrupt when a new technology makes the prior one obsolete.

jc

25

Nick Hilliard wrote:

wiring center you enable all technologies. GPON today, direct GigE
or 10GE where necessary, and all future technologies.

yep, agreed - much more sensible, much more resilient to failure and only
marginally more expensive.

You should suspect cost figures provided by those who want to
keep their monopoly.

At least, if population density is below some threshold, SS is
less expensive than PON, because the expected number of
subscribers to share a fiber with reasonably short drop cables
is small.

It'll never be done though. Too much to lose by creating a topology which
allows you to unbundle the tail.

It is still possible to unbundle PON if regulators want to do so.

See our paper:

  Competition Promoting Unbundling of PON
  http://dl.acm.org/citation.cfm?id=1914349

            Masataka Ohta

26

And that would be using what spectrum and what technology? Consider what the
release of one Apple product did to the associated carrier's wireless net.
Then consider the current tendency for "unlimited wireless data" to mean 2-3G
per month.

Where's the economic incentive for all these carriers to build out enough
capacity to move "all consumer data" (or a large fraction anyhow), and lower
their prices to match?

Sure, it may happen *eventually*, but for it to happen in 5-10 years, it would
have to be in motion *now*. So who's already in motion?

27

Nick Hilliard wrote:

most of the expense of laying fibre is associated with ducting + wayleave.

Another important expense of FTTH is at the last yards of
dropping cables fro the laed fiber, where SS needs simple
closures and shorter dropping cables than PON.

            Masataka Ohta

28

In a message written on Sun, Mar 25, 2012 at 12:37:24PM -0700, JC Dill wrote:

their future is very uncertain. Can you promise that fiber has a
*feasible* lifetime of 20-50 years? Maybe in 5-10 years all consumer
data will be transferred via wireless, and investment in municipal wired
data systems (fiber and copper) becomes worthless.

You have offered a two part problem.

The initial question is, will fiber put in the ground today still
be able to do something useful in 20-50 years. I believe the answer
to that is yes. There is fiber that was installed in the early
1980's that is still in use today. It's predecessor technology,
copper wires to the home, has been in use far longer and with today's
DSL technolgy has done far more than ever intended. High quality
transmission media in the ground has long life, and new, well
designed fiber would be no exception.

The second part of your question is really "might fiber be replaced
with some disruptive technology?" That is always a risk, but I
actually think the avenues for advancement are few. Wireless of
some type is probably the only viable competitor, and it's anything
but cheap at scale.

The real way to address the second part is to look at the outgoing
technology, copper/dsl. Even though phone lines were designed to
just carry 8khz voice, we've found it far cheaper and easier to
design DSL technology around those properties rather than replace
it with fiber or wireless. The reason? Build cost mostly. Diging
to bury new fiber is expensive, and even with wireless permitting
new transmitter locations and spectrum are very expensive.

Can I _guarantee_ no better technology will come along? No. However
I would posit even if it does come along the life span of fiber is
still 20 years just due to the build cost and timeframe of the new
tech. It's if it doesn't come along the timeline grows to more
like 50 years.

There's risk in any technology investment, however I think having
a high bandwidth, high reliability, cheap to operate pipe into the
home will always have enormous value, and right now fiber is the
best tech to that and thus the best place to invest.

29

Most rural GPON deployments I see today are homerun back to the CO or a hut
-- there's few that have passive splitters in a cabinet. They also want
their GPON to be future-proofed.

Frank

30

These enclosures (including all electronics but SFP) are around $350.
SFP is another $110 for 10km bidi optics w/ DOM.

The cable cost to reach the home with connectors can run around $1-2/meter, excluding pole attach or burying costs.

The cable cost quickly comes to rival the cost of the enclosure.

- Jared

31

But they also deserve to have or enjoy the benefits that comes with living in the big cities

32

Jared Mauch wrote:

Another important expense of FTTH is at the last yards of
dropping cables fro the laed fiber, where SS needs simple
closures and shorter dropping cables than PON.

These enclosures (including all electronics but SFP) are around $350.

What?

What do you mean "including all electronics" inside closures
of PON or SS?

SFP is another $110 for 10km bidi optics w/ DOM.

The cable cost to reach the home with connectors can run around $1-2/meter, excluding pole attach or burying costs.

What costs is not material but installation.

              Masataka Ohta

33

Active Ethernet solution outdoor enclosure sfp+2xGE+2xPOTS is about 350 without optics

Inside device is closer to 150-160.

... Certainly agree on install costs.

Jared

34

Well, "deserve" is a strong word... but the underlying thought is my
primary reason for believing that municipal fiber is a good solution, and
I'll expand that thought one more layer:

The Public Good is not often all that cost effective; sometimes, it's a
money loser. That's why corporations can almost always be depended on *not*
to be working in its interest, absent regulations to force them to do so,
such as the Universal Service Obligation, imposed on AT&T in one form or
another all the way back to the Communications Act, and expanded in TCA96.

This is one of many things that seems to militate in favor of municipally
owned and operated layer 1 fiber builds -- is *is* the obligation *of a
municipality* to operate in favor of the Public Good: it *is the Public*,
in a very real sense.

And the members of that body politic, properly informed, can make sure that
such a build will be, by direction, equally accessible to all in their area:
it will be a bond issue, and such items are generally ballot questions.

Or at least, they can try; you can't make people vote.

Cheers,
-- jra

35

I grew up in a rural area served by dialup for the first 15 years of my life, so please don't misunderstand what I'm about to say. No, they don't.

Living in a rural area is a different set of value propositions than living in the Big City, and we shouldn't pretend otherwise. Do people living in the big cities reap the benefits of living in the country? No ambient noise, no air pollution, low crime rates, neighbors you know and can trust your children with? No, they don't.

That isn't to say that broadband technology won't (or shouldn't) find ways of serving people in rural areas with increasingly usable levels of throughput while decreasing jitter and loss; it already is (and should), and the situation is constantly improving. But I think it's a mistake to say that people who have made the decision to live in the Big City should expect to enjoy the same benefits as people who have made the decision to live in rural towns, and vice versa. They'll never be the same, and unless I'm very much mistaken, that's actually OK.

Nathan Eisenberg

36

Nathan Eisenberg wrote:

From: joshua.klubi@gmail.com [mailto:joshua.klubi@gmail.com]

But they also deserve to have or enjoy the benefits that comes with
living in the big cities

I grew up in a rural area served by dialup for the first 15 years of my life, so please don't misunderstand what I'm about to say. No, they don't.

Living in a rural area is a different set of value propositions than living in the Big City, and we shouldn't pretend otherwise. Do people living in the big cities reap the benefits of living in the country? No ambient noise, no air pollution, low crime rates, neighbors you know and can trust your children with? No, they don't.

That isn't to say that broadband technology won't (or shouldn't) find ways of serving people in rural areas with increasingly usable levels of throughput while decreasing jitter and loss; it already is (and should), and the situation is constantly improving. But I think it's a mistake to say that people who have made the decision to live in the Big City should expect to enjoy the same benefits as people who have made the decision to live in rural towns, and vice versa. They'll never be the same, and unless I'm very much mistaken, that's actually OK.

There's truth to what you say, but there is another side that often gets missed. The rational for universal telephone service isn't just that rural residents need access, but that folks in denser areas need to be able to reach them - the value of a network connection lies not only in who can reach you, but who you can reach.

A similar argument applies to broadband. In today's economy, supply chains are spread all across the map - extending networks into rural areas, is not just for the benefit of those who live in those rural areas.

Miles Fidelman

37

A municipality hasn't much to lose; they can declare a monopoly.

Which was rather precisely the point.

Cheers,
-- jra

38

From: "JC Dill" <jcdill.lists@gmail.com>

> In a message written on Sun, Mar 25, 2012 at 11:47:58AM -0400, Jay
> Ashworth wrote:
>> Well, for my part, /most of the poiny/ of muni is The Public Good;
>> if /actual/ bond financed muni fiber is skipping the Hard Parts, it
>> deserves to lose.

It doesn't matter if it's a bond-financed project or a privately
funded (privately owned) project - they are using a public resource (the
street/poles) to lay their lines, and usually also using the power of
the municipality's right to eminent domain to put in or use poles (or
underground conduits) to run lines across private properties. As part
of the Public Good contract to use these public resources, they should
be required to service both the the easy parts and the hard parts, no
matter the source of the financing or the ownership of the lines.

Yup; that's what I said. But it cannot be privately financed; *it must
be the property of the municipality*, legally. I don't care if they
sub out the actual trench and splice, or even the operation of layer 1...

but they have to own it; that's the whole point.

> Fiber has a 20-50 year life.

The biggest problem is determining how certain that lifespan is.
Remember how Netflix looked like an awesome business to deliver DVDs by
mail in 2002, and had one of the most successful IPOs of the era? Less
than 10 years later we have widespread broadband and companies can
deliver that same content by copper/fiber/802.11. Now Netflix is in the
position of being in direct business conflict with the companies they
rely on to carry their product to their customers (e.g. Comcast) and
their future is very uncertain. Can you promise that fiber has a
*feasible* lifetime of 20-50 years? Maybe in 5-10 years all consumer
data will be transferred via wireless, and investment in municipal
wired data systems (fiber and copper) becomes worthless.

His assertion wasn't economic life, it was *functional* life; I think we're
pretty close to 50 years from the first deployment of optical fiber, and I
think it's still serviceable.

The question here is: did you design layer 1 properly, so as to make it
cost-competitive for a long time (see the other thread on this).

Cheers,
-- jra

39

True, but it's the one monopoly where you get a vote.
I'm not sure it's fair to call a municipality a monopoly ... but that's
just me.

40

I wasn't clear (again; I have to work harder on that -- it made sense to
*me* :-)...

A municipality can declare a monopoly on the installation of fiber within
its jurisdictional bounds, and *require* anyone who wants to connect its
residents to use its fiber; it *owns* (or has easements on) all the spaces
necessary to do subterranean fiber (and I believe it leases such easements
to power utilities to erect their poles, and may therefore have control
over that as well, though I'd have to research that point.

Clearly, I think that's a feature, not a but (if you've been following the
thread)...

Cheers,
-- jra