Live Blog Here at 7pm ET

by: Matt Stoller

Fri Jul 27, 2007 at 19:00


Building out broadband access to everyone in America is not a simple topic, but as Jim Baller notes, the the need for a national broadband strategy in the context of global economic trends is clear.  Just what is broadband?  Is it current DSL speeds in the US, or 100MB connections like we see in Korea?  How can we close the rural divide, and bring full broadband access to those in public housing?

There are some interesting public-private partnerships at, and we have Paul Morris of the Utah-based fiber network UTOPIA on how government policy can support such work.  And then, of course, there's the issue of money.  The Universal Service Fund, the Department of Agriculture, and HUD are all possible places where money for universal access can be found.  John Windhausen writes on full deployment and the Universal Service Fund, and Waldo McMillan at One Economy Corporation disussed his organization's 'Broadband in Public Housing' initiative through HUD. And finally, Andrew McNeill of Connect Kentucky described his  telecom-backed private-public program, and how it should be expanded to the rest of the country.

The live-blog should be interesting.  To be honest, I'm quite skeptical of McNeill's broad claims.  John Windhausen pointed out that real deployment is going to cost $2000 per household, and the idea that there's a free lunch out there, where industry will pay for full deployment, seems like a stretch.  I'd like to see McNeill and Windhausen discuss where they agree, and where they don't.

Building out infrastructure in broadband is core to America's future.  How it happens, or doesn't, is something that should concern us all.
Matt Stoller :: Live Blog Here at 7pm ET

Tags: , , , , , (All Tags)
Print Friendly View Send As Email
Live Blog Here at 7pm ET | 91 comments
Broadband discussion (0.00 / 0)
Hello.  This is John Windhausen.  I appreciate the opportunity to participate and look forward to an active discussion.
John
Welcome to night 4. (0.00 / 0)
Hello everyone. Russ Newman from Senator Durbin's staff here. We'll be joined shortly by Paul Morris of UTOPIA, John Windhausen of Telepoly.com, Jim Baller of the Baller Herbst Law Group, Waldo McMillan of One Economy, and Andrew McNeill of ConnectKentucky. Looking forward to the discussion.
Broadband discussion (0.00 / 0)
Hello. This is Paul Morris. I appreciate this opportunity and look forward to the discussion.
Broadband discussion (0.00 / 0)
Jim Baller here and looking forward to tonight's talk.
[ Parent ]
wholesale model (0.00 / 0)
Paul,
Perhaps you'll permit me to ask you a question about UTOPIA.  The question is a basic one:  can you earn a return on your investment by only providing wholesale service? John
Investment and ROI. (0.00 / 0)
John,

Excellent question!

I would like to the know what the average cost to connect a user on the Utopia project is as well as OPEX for each user on a monthly basis.

Thanks,

Ken

[ Parent ]
wholesale model (0.00 / 0)
I think that can be a real challenge for a private sector company to get an adequate ROI based on the wholesale payments. But we are a governmental entity and our ROI is economic development. We don't need an ROI. What we want is a return of our investment. In other words, we would like sufficient revenues to pay the opex and debt service but we don't need to make a "profit."
[ Parent ]
wholesale/retail (0.00 / 0)
Paul,
Some telecom providers maintain that they can only generate a return by providing retail service to consumers.  But I can think of a few examples (Level3, British Telecom) that seem to be doing quite well providing only wholesale access.  I wonder whether UTOPIA fits that description of an entity that can operate quite well under a wholesale-only model?
John
[ Parent ]
wholesale/retail (0.00 / 0)
Those type of wholesale companies are long haul/medium haul wholesale providers. We are a last mile solution. That is when it gets expensive. My sense is that it would be a challenge for those companies to run fiber passed every home and business and reap a return in the timeframe they would want on a wholesale basis only.
[ Parent ]
Hello (0.00 / 0)
Andrew McNeill here. Likewise - I appreciate the opportunity to join.
Matt's question above (0.00 / 0)
John, Matt poses an interesting question above, where he asks where you and Andrew McNeill (of Connect Kentucky) might agree or differ. Or, more likely, it's more complicated that. Curious to hear your thoughts.
reply to Matt's question about connectKentucky (0.00 / 0)
Thanks for the prod.  I am a big believer in the ConnectKentucky model as a part of the solution, but it's not the entire solution.  Connect Kentucky does a great job of working with the industry, of mapping the availability of existing broadband services, and helping the community learn about and use broadband.  However, its ability to "work with the industry" is both an advantage and a disadvantage.  To really achieve the "big" broadband that we will need in the future, we need to think big and act aggressively, and the existing industry players tend to think incrementally.  Andrew, is this a fair comment?
[ Parent ]
Reply to Matt's Question and John's Comment (0.00 / 0)
John - we appreciate the endorsement.

I tend to see the relationship more in the "advantages" column. We have mutual goals in expanded service and increased adoption.

The comment about a tendency to move incrementally is a fair observation but in my mind also constitutes prudent business practices - and not unlike what one sees across most business sectors in the country.

To think big - and John is doing a great job of prompting big thinking - is a worthy exercise. So long as the implementation of such big ideas is grounded in a sense of what is achievable over a reasonable period of time.

[ Parent ]
data (4.00 / 1)
How are you ensuring integrity of data collection on broadband deployment?  Aren't you just allowing industry self-reporting? 
[ Parent ]
Response to Matt (0.00 / 0)
We do rely upon industry self reporting as data inputs into the mapping exercise. However - we have our eCommunity teams in every county within Kentucky that works to verify where broadband is and where it isn't. Also - we run consumer surveys with statistical significance down to the county level which serves as an additional check.

Of note - in general the evidence indicates that the providers are providing very legitimate data.

[ Parent ]
good to know (0.00 / 0)
Thanks for answering.  Are these surveys publicly available?  That would be fascinating information to analyze.
[ Parent ]
ConnectKentucky Research (4.00 / 1)
Yes. Our research is in the public domain. It can be found at www.connectkentucky.org under the Research and Policy Link.

If you care to discuss more in the future - feel free to email me: amcneil@connectky.org

[ Parent ]
Broadband Discussion (4.00 / 1)
Let me echo John's endorsement and elaborate on his concerns.  ConnectKentucky appears to be doing great work in mapping broadband and aggregating demand, but I understand that CK defines "broadband" as 256K, treats provision of this level of "broadband" by a single provider to be an indicator of success, allows the incumbents to be heavily involved in its decision making process, and, most troubling to me, supports public broadband initiatives only as a last resort.  Are these statements correct?  If so, how can CK -- or its progeny around the US -- provide communities unbiased information about their options?  If a community wanted to build a network of its own that was more robust or could provide better or less expensive service than a private entity proposed to do, how would CK react? 
[ Parent ]
Using local government and coops to create competition (4.00 / 2)
treats provision of this level of "broadband" by a single provider to be an indicator of success, allows the incumbents to be heavily involved in its decision making process, and, most troubling to me, supports public broadband initiatives only as a last resort.

People assume, wrongly, that private, for profit providers are always more efficient than non-profit or government providers.  But when dealing with a natural monopoly, it's typically the other way around: introducing competition into a private market where there is a lack of competition.

We should care about there being competition, and not who supplies the service.  We should not protect corporate monopolies from competition, be it other for-profit entities, or non-profit.

[ Parent ]
Incremental profit-focused thinking (0.00 / 0)
I'd like to second John's comment about how incumbents think incrementally.  They also tend to think only about "value" they can "monetize," not broad social benefits that are much harder to fit into a price- and profit-based ROI scheme.

This raises the question of whether governments pay for the whole networks deployment, or provide subsidies and/or other incentives to motivate private sector players to do so.  The latter prospect then raises the question of whether these networks are fully open to applications and retail service providers or whether we lock ourselves back into the "network neutrality debate" situation we've got today.

[ Parent ]
Profit-driven thinking (4.00 / 2)
I would like to point out that having a critical infrastructure that is totally based on a profit has two very detrimental effects.

1.) In times of national disaster, whether we choose to look at Katrina or 9/11, when the communications infrastructure goes down, relief becomes immensely more complicated. In fact, to add insult to absurdity, we had FEMA representatives asking refugees to call in or go online to request assistance in the days immediately after Katrina hit.

From my perspective, since we, as a society, have become so dependent on the ability to communicate, we need to look at this as a service that is critical to our country's well being - not a cash cow for an entity that is more concerned with this quarter's bottom line than the lives of their customers.

2.) The second yet equally as serious an issue is that as a for profit company, it is my job to build out services that will provide me with a return. This is a mandate for me to ignore areas that cannot earn me an acceptable profit, creating a two tiered society - those that have broadband and those who do not.

At the same time I believe that based on the increased adoption rates of services like VoIP, the lower down on the economic scale an area sits, the more likely they will adopt these money saving services.

In effect, we are denying the people who need these services most the ability to take advantage of them.

[ Parent ]
The underserved are unprofitable to serve (4.00 / 1)
I think Ken has nailed it here.

We already have a market based solution for supplying high speed access.  People who are profitable to serve are being served.  The people who are not being served are much less profitable, or are not profitable to serve at all.

If we divide the country up into service areas run by a for profit entity, the incentives are still basically the same: the provider will want to prefer its high margin customers, and will neglect its unprofitable customers.  All the subsidy scheme proposed will do is reduce the providers risk, and reduce the options of non-profit entities like cooperatives or local government to compete.

I'd much rather see subsidies (especially for financing, insurance and so on) for the non-profit players, and allow private industry to enter anywhere that is profitable.  Where it is not profitable, they will not enter except as contractors of the non-profit entities.  Where it is profitable, they will supply the service.

This, I think, is the best way to prevent a Katrina-type screw up.

[ Parent ]
I like this (4.00 / 2)
Instead of giving the telcos $200 billion in a chunk to spend on a rollout their way, give it to the underserved communities and let them buy their own infrastructure with it.

That way, we don't have to worry as much about how to convince telcos to spend the money in a socially empowering manner. We give the communities the power by giving them the money.

It's simple and takes less regulation to get right.  The only catch I think is making sure local officials don't accept kickbacks from the telcos so the telcos do their own thing anyway.  Perhaps with this model we would need additional funding for anti-corruption law enforcement.  That has an upside too though: who wants TV ads run in their district saying they voted no on corruption enforcement?

[ Parent ]
Me too. (0.00 / 0)
I agree that a corrupt local government is a problem.

Perhaps the funding comes with strings attached, such as requiring an open process of allocating funds and/or some sort of referendum?

I blog on InnermostParts.org

[ Parent ]
Public/Private Partnerships And Katrina Relief (4.00 / 2)
The concept of public/private partnerships was popularized back in the early 1990 in books like Reinventing Government.  It was a good idea back then, and certainly, to the extent that local government has made use of local contracting, it's still a good one.

But it's worth mentioning that the relief effort for Hurricane Katrina was largely a story of how badly you can screw up public policy by poor use of privatization.  Large corporations can and will take over this kind of process, shutting out real competition, and taking most of the decision making power away from local government and local voters and consumers.

How do we prevent the rollout of broadband access from becoming a Halliburton-type boondoggle?

oversight and accountability (0.00 / 0)
I share this concern about the waste of government funding.  I suggested at the top the creation of a new fund for broadband - 100 Mbps that is separate from the existing universal service fund.  One of my reasons for this suggestion is that the existing USF program has very little accountability.  The money flows to companies with little reporting requirements as to how the money is used. A new funding program must have strict reporting requirements and deliverables to ensure that the targets are met and that consumers receive the service.  If not, the funding should be pulled.
[ Parent ]
I'm Here (0.00 / 0)
Hello.  This is Waldo McMillan, and I'm happy to be part of this discussion.
Broadband Discussion (0.00 / 0)
Hi, Paul.  Good to "talk" to you.  One of the folks who commented on my opening blog asked how I thought the US should define "broadband."  Care to take a crack at that?
definition of broadband (4.00 / 2)
That can be a tricky question. Some believe that it is a mistake to get locked into a discussion of speeds but rather it should be focused on applications. That is, do you have the capacity to do what you need/want to do. But i think you need at least 10 Mbps up and down to run applications that are readily available today and be happy with the experience.
[ Parent ]
adsf (0.00 / 0)
even to define functionality in terms of existing applications one can put those terms into features of the medium.  i.e. > 10Mbps down, 1Mbps upload, ave. latency < 30ms RT
(or 300ms is probably more realistic unbounded)
peak BW
latency
throughput
propagation delay maximum
RTT (round trip time)

Within Engineering and also the ITU standards are QoS metrics that are non-application specific and even not network technology or topology specific.  (CDMA, GSM, 3GPP2, Cable and so on)
These
could be used as feature definitions required for any technology...
the application itself is implied.

The ITU and some other engineering standards has  QoS quantative metrics standards defined via known perceptual requirements for communication.
These QoS metrics can be applied to any current or emerging technology.

(i.e. one cannot stream audio > 15ms latency from a video sync without it being perceived and so on).

NoSlaves.com  


The Economic Populist

[ Parent ]
Basic speeds for defining "broadband" need to grow yearly. (4.00 / 1)
Much the same way money gets devalued by inflation over time, "broadband" slips even faster.  Without regular reassessment of the concept of "what is broadband" the speeds that define (currently a ridiculously low 200kbps in a single direction) mean that what we define as broadband doesn't map onto what we generally want to do with broadband.  In the end, "broadband" as a legal definition becomes as useless for measuring the quality of life of the user experience as the poverty level is for setting how much a family needs to get by.
[ Parent ]
Broadband Discussion (0.00 / 0)
Hi, Sascha.  I know you support transparency and accountability, but how do you do that without a consensus on definitions?  Also, definitions need not be static.  They can be changed as conditions warrant.  What would you propose? 
[ Parent ]
extensibility (0.00 / 0)
In engineering stds. bodies (such as ITU-T Y.1540) one can add appendixes as more in known.  The good news is that perceptual quality for multimedia as well as many other applications is already well defined, but one can make a definition extensible.

NoSlaves.com  


The Economic Populist
[ Parent ]
Consensus vs. the Public Good. (0.00 / 0)
Hi, Jim.  Indeed, you've put your finger on an important tension running through a lot of these debates.  Unfortunately, as broadband speeds have continued to stagnate (due to a lack of consensus on how much to raise them), the public is placed in an increasingly detrimental position.  The FCC is tasked to protect the best interests in the public -- their responsibility is to guide industry towards that end.  When the FCC fails to shepherd things along, things tend to get out of whack.  And that's where we are today -- with speeds that are ridiculously slow being defined as "broadband."

Luckily, the FCC isn't tasked to build consensus (that's just a preferred tool).  When the public is being harmed, the FCC has an additional responsibility to carry a big stick and ensure that things are changed.  At this point, I think the FCC needs to take the lead, tell the stick-in-the-mud corporations that things are going to change, and dramatically increase the speed of "broadband." 

[ Parent ]
P.S. (4.00 / 1)
Upping the speed defined as "broadband" (and adding a regular speed increase schedule) would be a very good addition to any national telecommunication/broadband reform bill. 
[ Parent ]
monetary motivations (0.00 / 0)
To me the extensibility issues or the proper way to approach these sorts of definitions are well defined, coming from a body of engineering standards work....but my question is...within these sorts of standards are monetary interests, one can use technical standards for business objectives....so...
who specifically is going to be responsible to define these very metrics?

I'd say initially international engineering standards bodies, or at least the US ones, but there can be "heavy politics" for business agenda reasons in defining certain aspects in standards.  I have a hard time believing that business agendas could make it into a few basic QoS, BW, throughput, error rate sorts of metrics....but I could see it being done...

but I am bothered technically by speed being the only aspect discussed here, there are many other aspects for a high quality reliable network beyond speed capacity.

NoSlaves.com  


The Economic Populist

[ Parent ]
Public-Public Partnerships & Community (0.00 / 0)

Matt's Point (0.00 / 0)
Matt's observation is on-target. There is no free lunch. The market incentives must be structured in a way that promotes investment. Private providers are prepared to invest in rural markets if they believe that there is an opportunity to earn a reasonable return.
Been there/done that - still paying for the experience. (0.00 / 0)
Andrew,

Having lived the life of a rural broadband provider for a good number of years, I can tell to you, from personal experience, that unless you can connect to a reasonably inexpensive upstream connection no business model is ever going to make a reasonable return.

In many areas of the county T1s are the only option and especially in many rural areas the price of a T1 exceeds what a small operator can generate in revenue before the T1 becomes saturated.

Do you have any suggestions as to how these private operators can overcome this issue?

[ Parent ]
Upstream capacity (4.00 / 1)
The weakness of cable HFC and telco ADSL re: upstream capacity is notable, as Ken's comment suggests. One of the problems with this is that this provides incumbents with disincentives to offer reasonably priced tiers offering healthy upstream data rates and even to discriminate (in terms of traffic priorities) against applications that are heavy users of upstream capacity. One of these is peer-to-peer, which isn't just about pirated music and movies, but offers very efficient use of overall network capacity for a wide range of applications.

This points to the value of new networks, especially ones designed from scratch for more symmetrical traffic, which is already an area of expanding demand, with a lot more latent demand waiting in the wings.

[ Parent ]
adsf (0.00 / 0)
Skype style VoIP is based on a peer to peer network topology.

NoSlaves.com  


The Economic Populist
[ Parent ]
Private corporations won't buy you lunch (4.00 / 1)
Andrew,

Thanks for coming.

I agree that there is no such thing as a free lunch.  Someone has to pay for the (mostly) local network costs of putting in a fiber or other broadband drop.

But as mitchipd says above,


This raises the question of whether governments pay for the whole networks deployment, or provide subsidies and/or other incentives to motivate private sector players to do so.  The latter prospect then raises the question of whether these networks are fully open to applications and retail service providers or whether we lock ourselves back into the "network neutrality debate" situation we've got today.

This is what economists like to call an externality problem: the real benefits of making the investment are not readily monetized by the investor.  And I don't see a really good market solution for this that is not deeply disfunctional and against the public interest.

Most people are assuming that letting private corporations put up and run this network for profit is a good idea.  They should not be assuming this.  Governments build roads and sewers as public infrastructure, using private contractors as needed.  We don't hand these things over to corporates to supply in a "market", largely because it's difficult or impossible to get a functioning market.

There's a good chance that broadband access needs to be treated less like widgets, and more like roads. 

[ Parent ]
Public-Public partnerships & Community Broadband (0.00 / 0)
John St. Julien out of Lafayette, La here. Sorry for that slip of the finger!!

Senator Durbin in his video remarks that introduced this discussion endorses providing incentives to private providers and "looking at the roles locally provided municipal broadband can play."

Let us hope that Congress look at providing local communities with incentives as well. The easiest (by which I mean the cheapest) way to do that is to remove all disincentives and forbid the states to institute any new ones. Ideally, anything that a private entity could do with its money and and its property should be possible for its public competition. Lautenberg-McCain would be a start, but only a start, on that approach. More expensively we should be willing to learn the lessons of the last great American infrastructure success: The Eisenhower interstate system. Not widely understood outside of policy circles, that program was a public-public partnership with the states providing a 60% share and owning the roads built in their state. There is no reason outside of lacking the political will that we could not offer the same sort of deal to municipalities. I am confident that, with the right to compete on level ground with the private incumbents guaranteed, most communities would be up to the challenge and we'd see an explosion of last mile broadband to rival what we see in Northern Europe. (I've blogged on this idea if anyone would like to see an expansion of it http://lafayetteprof....)

Lafayette, alone among the models featured in Senator Durbin's introductory remarks, is a pure public play. Lafayette fully embraces the traditional progressive ideal of a public utility and its proven benefits. Those benefits are considerable: cheaper prices, guaranteed reinvestment of any profit in the local community and undiluted local control. That model ought to be available and publicly supported; it might not do for every communityâ??but in many ways it is the ideal solution to what is, arguably, a natural monopoly situation. Services offered over wireline have historically collapsed into monopolies since the days of Western Union. If that history holds true (and the recent reformation of AT&T lets us see how this process takes hold) then the best way to assure that the interests of communities and not corporations are served is to encourage communities to be owners.

I hope that we will consider providing incentives for both local communities and private corporations to participate. Community utilities have been a proven success in providing broadband in other Western economies and those relying on that segment of the economy accounts for most of the countries that have near-universal, real, broadband.

Challenges toward building such systems (0.00 / 0)
What challenges face the construction of such models, and how can they be overcome? Lafayette has faced many a legal hurdle, has it not? (I believe UTOPIA has been in the same situation?)
[ Parent ]
Re: Challenges toward building such systems (0.00 / 0)
Russ, Most of the challenges Lafayette have faced can be traced directly back to the "Local Government Fair Competition Act" that the incumbents pushed through on the heels of Lafayette announcing its intention.

Without the delays that law enabled we'd have our system now.

A stronger Lautenberg-McCain would be the tonic I'd recommend.

[ Parent ]
If you had your way, (0.00 / 0)
what would a strengthened Lautenberg-McCain look like to you?
[ Parent ]
Improving the Lautenberg bill? (0.00 / 0)
John,
Thanks for participating here.  I'm wondering, since you folks in Lafayette have faced all sorts of challenges, is there specific language you could suggest to revise the Lautenberg bill to increase the likelihood that it would really avoid incumbent-backed roadblocks?
[ Parent ]
Re: Improving the Lautenberg bill? (0.00 / 0)
mitchipd,

Honestly, if someone were to actually invest me with the power to draft a real bill I'd sit down with Jim Baller and get _him_ to draft it with the goals I've mentioned in mind.

("He who has himself for a lawyer" and all that..)

But I'd be happy to talk to anyone about it anytime. :-)

[ Parent ]
Maybe you'll get the chance... (0.00 / 0)
I'm hoping this project can evolve to the point where we'll somehow get a shot at suggesting specific language.  I'd love to see what you and Jim would come up with, and how it would differ from what's in the Lautenberg bill.
[ Parent ]
Re: If you had your way (0.00 / 0)
I'd look for a positive statement of the right of communities to build competing telecommunications systems (not simply forbidding the states to forbid).

It would also lay out a plain standard to judge whether or not a state law went to far in restricting communities through various disincentives and regulations. That standard: Anything which is legal for a private corporation is legal for the people and their communities.

(I suspect that Louisiana's Law would pass Lautenberg-McCain muster.)

[ Parent ]
Challenges (4.00 / 1)
Yes UTOPIA had its challenges too. We had to battle a lawsuit by the incumbent that tried to use the 1996 act to challenge us. We also had a delay in receiving timely funding from the USDA for our smaller member communities. But both worked out ok in the end.
[ Parent ]
Broadband Discussion (0.00 / 0)
John, welcome to the dialog. 

For readers of this thread who don't know John, let me note that he was one of the local heroes without whose intelligent blogging day in and day out the Lafayette success story might not have happened.

John, let me jump off on a point you make.  You say that Lafayette is "a pure public play," by which I think you mean that Lafayette will be owned and operated by the City.  But I would argue that this does not make it a "pure" public play.  The system was planned by the public and private sectors working together.  It was funded by private banks.  It is being designed and constructed by private firms.  The materials it uses were manufactured by private firms.  Much of the programming it airs will be produced by the private sector.  It will create thousands of new jobs that will benefit countless businesses in Lafayette and the surrounding region.  It will raise property values for everyone ... and on and on. 

In all of these respects, a project like Lafayette's is considerably more than a "pure public play" ...

So,

[ Parent ]
Re: Broadband Discussion (0.00 / 0)
Jim, Your point is well-taken. Without the private engines there'd be no building a "public" broadband network and too little to put on it.

I do think the wireline last mile a special case with a clear history of recurring monopolizaion. Having the "pure utility" model available as an alternative should be salutory in that it give a real alternative and caution the powerful private incumbents about the consequence of abuse of their position.

[ Parent ]
Broadband Discussion (0.00 / 0)
John, I've been reading your writings for years, and I'm glad to meet you at last.  Your opening blog and mine are very similar. 
Re: Broadband Discussion (0.00 / 0)
Thanks Jim, without engaging in too saccharine a set of remarks, I'd not be able to follow the scene half so well without your newsletter...and have always admired your good works. Great to see such an admirable crew here!
[ Parent ]
$2000 as an average (4.00 / 1)
I'm a former telecommunications economist that worked some years ago for Bezeq Israel Telecomm, Israel's national telecomm carrier. I've done calculations similar to the one that produced the above cited $2000 figure, and it's important to understand what the figure represents.

In all likelihood, that $2000 figure is an average hiding enormous variation in population density, how new the housing stock is, and even the hardness of the ground.  This has real implications for providing universal service.  And almost *all* of that $2000 would be spent in what telcos call the "drop": the part of the network that connects with the residence or place of business.

I don't think there's a "one size fits all" business model for broadband service.  Some urban areas will be very profitable, and competition makes great sense.

For outlying areas, I can't see a private provider behaving different from current providers: even if required to supply service, they will delay and avoid it.

This is mostly a fancy contracting problem.  For outlying areas, government is going to need to take the lead, using private contractors to do the work.  The legal and regulatory regime needs to recognize this, or expensive-to-serve rural areas will be no better served than they are now for some time to come.

build-out costs (4.00 / 1)
Verizon was quoted last year as saying its build-out of fiber to homes was coming in at around $1000 per home.  Plus, Verizon said that its fiber network had reduced operating costs.  Since the majority of the population lives in urban and suburban areas, that's what makes me think that $2000 estimate is reasonable.  Am I wrong? 
[ Parent ]
Distribution is pretty skew (4.00 / 1)
If you look at costs, it turns out they are pretty skew, with a few outliers having great influence on the average.

I suspect that Verizon's figure is accurate if they really serve everybody.  But I don't believe that for a minute: they will behave as currently do, and go for the low hanging fruit -- people who live in areas that are cheap to supply.  More to the point, if net neutrality fails and Verizon et al get what they want, they will also home in on the high revenue customers.

I do not expect that getting the big telecoms to serve the expensive, outlying areas will be any easier than it is now, and I suspect that some of the deals under discussion will lock out alternate solutions for the problem, such as cooperatives or outright municipal ownership of the plant.

[ Parent ]
Build-out costs (4.00 / 1)
Hmm, I had heard that Verizon was quoting an average of $1,800 per home connected, not $1,000.

Here is a link that confirms what I had heard.

I would like to reiterate the point that this cost is in neighborhoods that are a best case scenario. In fact, I have read the when all is said and done, fiber can cost upwards to $400K/mile in places like Manhattan - and we won't even look at the time between conception and the project finally being finished.

[ Parent ]
It's not the cost per mile that matters (4.00 / 1)
This is why taking population density into account is important.  It's not the cost-per-mile.  It's the cost per subscriber mile.

It's that expensive per mile in NYC because tearing up the streets cost an enormous amount of money.  But most of those costs are shared between very large numbers of customers.  So $400K worth of fiber gets split across a very large number of customers.

When you do the calcs, it turns out that this is less of a contribution to cost than you'd think on the service.

[ Parent ]
It's not the cost per mile that matters (4.00 / 1)
While I understand what you're saying, that was along the lines of the point I was trying to make - I probably should have worded my post a little more clearly.

However, the one fact that I was trying to convey is that it is not the cost we need to be focusing on in this one case, it is the time from the concept being proposed, through the budgetary process, then the engineering studies, to pulling the permits and all of that has to happen before any actual construction can be done.

Is there any way we can streamline this process?

On a somewhat related note, what can we do about pole attachment costs? What about make ready fees where an electric company can charge an arm and a leg to install a meter to charge for less than 1 watt of continuous power usage?

Let's face it, that can be a deal breaker in many metropolitan communities.

[ Parent ]
Verizon's FTTH costs (4.00 / 1)
Ken,
I just did an analysis of Verizon's FTTH capex based on the company's own numbers.  Different numbers sometimes show up because there are two main cost elements--"cost to pass" and "cost to connect." 

The former is largely a function of density and plant type (aerial vs. underground).  The latter is a function of what percentage of homes get connected.  The fewer homes that get connected, the lower will be the total cost per home passed but the higher will be the cost per connected home.

For example, based on VZ's numbers, their average gross cost at 10% penetration (i.e., connecting the 10% of homes that might sign up for service) would be $1,061 per home passed, but a whopping $10,608 per revenue-generating "connected" home.

This changes to $1,348 per home passed and $2,696 per home connected if 50% of homes are connected, and $1,707 per home passed and per home connected at 100% penetration.

The % of connection depends on services offered and also on whether the network operator decides to connect homes even before they sign up for service.  These decisions may very well be different between private companies and municipalities. 

For example, a community-owned network could decide to connect all homes for some "public benefit" purpose not directly tied to revenue that fully justifies the extra cost of these connections.  This could include the general goal of allowing all citizens to have access to a high-capacity communication networks, including a range of broadband-enabled services provided by the city government (e.g., interactive multimedia meetings, public hearings, etc.) healthcare providers (telemedicine, health monitoring,etc.), local utilities (e.g., automatic meter-reading, real-time-pricing).  Some of these services could be directly monetizable by end-user or other payments, but some might not.  If they're not readily monetized, then private providers would likely have no incentive to provide them.

Another factor that could add some confusion in the Verizon FTTH cost figures cited is that Verizon sometimes cites "net cost," which subtracts savings like avoiding upgrades of copper plant.  These savings wouldn't apply to a network being built from scratch by an entity other than the incumbent  telco.  For our purposes, the "gross" cost numbers would apply.

[ Parent ]
Nice analysis. (4.00 / 1)
Mitch,

I am 100% behind what you're saying and fully agree that by enabling more applications to run on a network the faster the return will be.

That said, I believe Verizon may be in for one nasty surprise. If voice service drops to a "free" service and video moves to an application like Joost, what, exactly, is going to pay for this very expensive network?

We do live in interesting times...

[ Parent ]
I agree (4.00 / 1)
You make good points Ken, which I tried to address in a report I recently wrote.  Those are real risks.

That's partly why they're fighting NN, because they want the option to discriminate in ways that insure they keep enough customers and these customers pay enough each month to pay back their investment. If too many of their broadband-connection customers use Joost etc. for video and Skype, etc. for voice, they'll be hard pressed to generate the kind of per-home revenue they need to payback network costs.

In terms of investing in the network, I think they had little choice.  If they don't, they're pretty well guaranteed to having a residential business that's in decline.  The other option would have been more like AT&T's fiber-copper hybrid network, which has its own risks, but costs a lot less per home.

[ Parent ]
How much does the backbone contribute to costs? (4.00 / 1)
Having seen a number of studies of the cost of different telephone services, I'm wondering, is the high-capacity backbone that much of a contributor to total cost?

Almost every study I've looked at show that without regard to the exact service you're looking at, the cost tend the concentrate on the borders of the network: the drop, the point of connection, and the equipment that connects the drop to whatever network you are analyzing.

We saw this with long distance: costs at the periphery fell somewhat, but at the center of the network, cost effectively fell to nothing, compared with the periphery costs.  In particular, the increased cost for capacity was surprisingly small.

Except for wireless, where capacity of the "last mile" itself is constrained, I'm unconvinced that you couldn't finance the backbone via a flat-fee "connection charge".

I haven't heard much to indicate that "growing the backbone" has been that much of an issue.  Are we sure that it actually is?

[ Parent ]
I agree on backbone costs (0.00 / 0)
mbayrob,
I agree with you on backbone costs.  As you say, the basic economics are that the farther back into the network you go, the more network costs are shared by multiple users. And, once fiber backbones are in place, the cost to dramatically boost their capacity is only in the electronics, which keeps delivering more bang for the buck.  The specifics might depend on various things (e.g., more per-user cost in a unicast service model--especially for bandwidth-hungry video--than in a broadcast or multicast model, since the unicast model would require a dedicated signal per end-user), but I think the general rule would still apply.
[ Parent ]
Your Report (0.00 / 0)
Mitch,

If you have a minute would you please send me an email and link me to the report you just referred to in your post above?

Thanks,

Ken

[ Parent ]
Can also use wireless as needed (4.00 / 1)
There's alway the option of using wireless for the final access connection if wired costs exceed a certain amount.  And if the broadcast white space spectrum was added to existing WiFi bands, that should be able to deliver fairly healthy data rates to more rural areas until demand and cost got more in balance to support even deeper fiber deployment.

And though labor costs are significant, fiber costs should probably keep coming down (as will wireless costs), so I'd think ongoing fiber extensions would become economical.  And, since the wireless supports mobility, it wouldn't become a "stranded" investment if and when fiber is extended to more rural areas.

I think this is just a matter of looking at the numbers and making some reasonable judgment calls regarding architecture.

[ Parent ]
Fiber costs to come down? (4.00 / 1)
Mitch,

I appreciate the response.

I am not sure how you would expect to see the cost of fiber drop much further, but even if we do believe that to be the case, it is the cost of labor that is the killer.

As to wireless, I am a huge fan but with respect to the lower bands (specifically the 700MHz band) I don't see there being enough aggregate bandwidth to do anything meaningful - with the possible exception of the most rural areas.

As I understand it, we are looking a ~20Mbps, which is nothing to sneeze at but at the same time in a location with any sizable population base, that amount of bandwidth will be used up pretty quickly once applications like Joost become widely adopted.

[ Parent ]
Hybrid Networks. (4.00 / 2)
I'm a huge fan of hybrid networking as a way to maximize the price vs. bandwidth sweet spot.  Certainly it makes sense in a lot of places and communities.  Two caveats are that wireless doesn't have the capacity of fiber, nor does it have the reliability.  The caveats to the caveats are that for those "last-mile" links, one often doesn't need (yet) the capacity of fiber and the reliability is often "good enough" (in Urbana, for example, our wireless network is up more often than Insight).

Wireless is still very much on the ascendancy -- so it's hard to really know what its capacity and reliability will look like in a decade (other than they'll both be a lot better).  Meanwhile, for backhaul, fiber is money quite well spent.

[ Parent ]
Costs (4.00 / 1)
I see your point, but you can always reduce cell size and reuse spectrum (as cellcos have over the years) if capacity in rural areas gets maxed out by demand.  That does cost money, but probably could be managed economically. 

One idea I've seen floated (mainly for fiber, but could apply to wireless) is for households and businesses in a given area pitching in to finance the upfront costs of network extension and/or capacity expansion, based on the value they place on it...sort of a user cooperative model.

And the wireless wouldn't need to carry the heavy capacity burden in urban and suburban areas, since these cost be cost-effectively served by fiber.

But I do agree that there are constraints and that labor is a big piece of the total costs, especially for wired networks.  But I see this as a constraint that drives architectural decisions, not a deal killer. 

If I was in a rural area, I'd much prefer 20 Mbps to 64 kbps or 1.5 Mbps.

For more on FTTH costs, see my comments above.

[ Parent ]
broadband by bond issue (4.00 / 1)
Could a city put a bond issue on the ballot to deliver broadband to, say, 95% of city residents?  Is this allowed (as it seems like direct competition with private telcos)?  If so, has this been done before?

There are no one-size-fits-all rollout plans. 
They must be different for each location, depending on geography and population density, as well as the service level desired by people.  This could look something like:
* fiber downtown and to high-desnity housing,
* wireless to the most rural of residents
* additional fiber backbones to reduce the length (and increase bandwidth) of DSL hookups.
* WiFi hubs every 2 blocks down Main Street (if there is one), or at least at every post office, city hall, and court house.

end the occupation of Iraq

[ Parent ]
Reasonably good model (4.00 / 1)
I think you suggest a reasonable model for how to finance a municipal or county service.

It's worth thinking about how to help local government pull this off.  State government can help, I think, by financing training and planning resources to help county officials better understand the issues.  There's also an opportunity here for private engineering consultants as well.

[ Parent ]
yes (0.00 / 0)
IEEE 802.16x (WiMax) standards are not only ongoing, there is now emerging standards and design involving heterogeneous network solutions (mix and match, depending on cost, availability).  WiMax esp. is considered for the last mile connectivity solution.

To Russ Newman:  I'm assuming that someone is consulting a legislative fellow from the IEEE running around the hill or alternatively, the IEEE stds. bodies chair for this sort of technical analysis on what is feasible, emerging and cost effective?

NoSlaves.com  


The Economic Populist

[ Parent ]
WiMax Capacity Woes. (4.00 / 1)
I'd be surprised if WiMax can deliver the types of speeds.  Based on the contention protocols WiMax uses, I've seen estimates that a cell's maximum throughput capacity would be 70Mbps.  Onto itself, not too shabby, but then you look at the fact that these cells are covering a 30 mile radius (2800+ square miles).  It doesn't take a genius to realize that even with minimal population density and take rates, that's not going to be a lot of bandwidth per user.
[ Parent ]
Balancing capacity and cost (4.00 / 1)
I agree that WiMAX capacity and distance claims are sometimes presented in an exaggerated way, but I still would like to think that a balance could be struck (regardless of which specific technology was used) between expanding throughput and the added cost of cell-splitting. 

And, as I said in an earlier comment, it might be feasible to employ a user co-op model in which those living in particular rural areas pitched in to help finance the capital cost of cell-splitting in their area, based on the extra value they saw from the extra capacity this would provide.  As you know, this model has been used for utility and telephone service in rural areas.

[ Parent ]
802.16m (0.00 / 0)
How up r u on this new standard?.
max 1G fixed 100MB .

There is also 802.16j which might enable better BW conservative with access points (multi user).

Do you know the cost of a data point in extreme rural areas?  esp. in comparison to the above estimates for cooper/fiber?

NoSlaves.com  


The Economic Populist

[ Parent ]
directional antennas (4.00 / 1)
This falsely assumes an omnidirectional antenna for the 70 MBps.  You can instead use directional antennas with, say, 10 degree beamspread for 36 different antennas around the tower, shroud the antennas to get just the signals you want, and then interleave channels so that even if you do have a little overlap, it's fine.

This is my understanding of how my local wireless provider uses the Motorola Canopy system running at around 6GHz, unwiredltd.com, to cover most of the East Bay (Oakland, Berkeley, etc.), as well as parts of San Francisco.  I had highly reliable service from them at 3.5 Mbps, better than the 768 kbps DSL I had before (which was the max I could get at my location in my highly urban location!).  They also offer 6 Mbps, but for either speed, you need to put an antenna on your roof with line-of-sight to the tower.

Their coverage map is here


end the occupation of Iraq

[ Parent ]
Physical Limits... (4.00 / 1)
Hi aip,

I agree that you might be able to boost WiMax capacity by some bit (the 70Mbps was the capacity that WiMax engineers were using) -- but I've never seen a 36-sector antenna array.  The physics behind building such an array would be fairly daunting (this is why, for example, you tend to see 9-sector antenna arrays for cellular phone towers).  My point is simply that given the necessary coverage area to make WiMax economical, the capacity is going to be extremely limited (particularly in metro areas or regions with even a relatively modest population density).  One of the reasons why WiFi has been successful is that it's dramatically lessened the coverage cell size -- until we can technologically fit far more data per wavelength (or utilize spread spectrum to achieve major capacity), we're limited by our technological abilities. 

WiMax doesn't appear to have the capability of delivering what we really need (though it's the perfect answer for cellular providers that want to get into data communications by retrofitting their existing infrastructure).  I wrote up a book chapter a few years back that touches on this issue in the book "The Future of Media" -- it's a bit dated, but still quite relevant to this debate.  Meanwhile, I'm looking to things like ultra-wideband, 802.11n, SDR and cognitive radios to help facilitate the necessary capacity per square mile to move things forward. 

[ Parent ]
802.11n 50 meters vs. 4 mi (0.00 / 0)
I'm not sure why you would look at 802.11x stds for the "last mile" here with a great "breakthrow" of 50 meters with 802.11n versus 4 miles.

SDR is simply almost a "wrapper" i.e. a system configurable through firmware (software defined) specification in order to flip between these various standards.  As far as I am aware it's not a standard for wireless transmission in and of itself.

Cognitive radios is somewhat similar, but a subset as I understand it,  but to efficiently sense the best network to switch to to maximum efficiency.

802.16j is a multi-hop relay technology which enhances the high bitrate throughput issue on 802.16e-2005 you are concerned about and supposedly

802.11m is not only real but a couple of years away.

All 802.11 standards are based on spread spectrum techiques, DSSS and FHSS so I'm not so sure what that statement refers to either.

So, maybe you are thinking that SDR would be the way to go generally so all systems are scalable, but that at least requires a firmware upgrade or something if a standard isn't even approved and verified yet I think.

I personally just want to see the cost estimates in a comparison contrast, a spread sheet would be really great...
  for to deploy to rural areas, the best way to go is for it to be extremely cost effective...
which from above the Verizon input above didn't look too good to me.

NoSlaves.com  


The Economic Populist

[ Parent ]
Windhausen's broadband-replacement for USF (0.00 / 0)
I'd be curious to hear folks' thoughts on John's broad description of a successor to the USF for broadband.
Would separate funding mechanism be better? (0.00 / 0)
In some respects I like that its separate from USF, which could make it a lot easier to steer clear of USF politics and inequities.  But I wonder how it would relate to the existing USF and any plans to reform it.  Having another source of funding focused on rural broadband could get messy and confusing. 

Any thoughts, John, on how the two entities and their funds could be managed in ways that didn't make the USF problems even more complicated and messy?

[ Parent ]
Universal Broadband Fund and USF (0.00 / 0)
The short answer is the same way that the USF co-exists with the Broadband Loan program administered by RUS today.  The existing USF can continue to subsidize operating costs for high-cost areas.  The Universal Broadband Fund (UBF) I am suggesting would operate under procedures similar to the RUS programs -- companies would submit applications, showing the technology they want to use and the geographic area they intend to serve.  But it would be simpler than the current RUS application process because 1) companies would not have to make a "business case" that they can pay back the money, and 2) they don't have to show that no other company is providing service in that area (since virtually nobody provides 100 Mbps service today).
[ Parent ]
Needing to jump off, but will be returning often (0.00 / 0)
Russ from Durbin's staff here: I need to sign off for now, but I plan on checking in on this thread later and throughout the weekend as it grows. We've already got a terrific conversation going, and I look forward to seeing where it goes over the next few days. Thanks to all.
back to CK (0.00 / 0)
I'd like to hear more about how the local CK teams operate and what they do.  How do they bring broadband to a community that wouldn't otherwise be served, or would be underserved?  Sen. Durbin and Sen. Inouye each have included a CK-like program in their bills.  The collection part is fine as far as it goes -- and it's only voluntary -- but the tasks the local teams perform is still somewhat of a mystery to me.
Prevent Technological Reductionism in Rolling Out Universal "Broadband" (0.00 / 0)

"Broadband" initiatives must not enable technological reductionism by network operators.

Until common carriage is established it's very important for the leading local "broadband" initiatives to make sure that they:

1) Make sure that "broadband" initiatives include a definition of the Internet based on the neutral, generic spanning layer provided by Internet Protocol, and assure that this is provided.

2) Make sure that these initiatives make a specific point of fostering the genericity and flexibility of the platform, and standards that empower end users.

3) Don't just speak of "broadband" -- this is a very undefined term that's perfectly amenable to being turned into limited service offerings by network operators.

4) Make sure that these leading initiatives include the key standards bodies, particularly IETF and the Internet Society in order to address a mission of assuring the genericity and flexibility of the platform and its conduciveness to innovation by end users.

Use a correct definition of what makes the Internet neutral, as in that of the Dynamic Platform Standards Project: http://www.dpsprojec...

My testimony to the NYC Broadband Advisory Committee for the Dynamic Platform Standards Project (http://www.dpsprojec...) is available at the following link.  My presentation is the last one, directly following Dave Burstein's:

> http://www.youtube.c...

It's important for activists to be sure to make clear that QoS and schemes like IP Multimedia Subsystems (IMS) are among the chief rationales and means being offered for *ending* net neutrality

You must state clearly that policies of treating like apps alike will not assure network neutrality, for the above reason.

These local "broadband" initiatives like to speak of providing high speed access to disadvantaged communities, and tend to pat themselves on the back with associated rhetoric.  However, these networks will not empower these communities unless they provide real Internet connectivity.  They end up with incumbent participants who don't actually want to deliver the Internet at high speed, who are only seeking a means to get local governments to accommodate their rollout of their versions of "broadband."  I've listed some key points that need to be stressed in these leading initiatives below.

See the New York City Council's Broadband Advisory Committee for one highly important local initiative:

> http://nycbroadband....
> http://wwwhatsup.com...

Here are the links to my testimony to the NYC Council's recent hearing on network neutrality that I posted before:

> http://odeo.com/show...
> http://peoplesproduc...

Some Net Neutrality Considerations:
> http://www.openleft....
> http://www.openleft....

Seth

Signing off... (0.00 / 0)
This week has been really great!  I'm heading out for a cross-country trip, but folks should feel free to drop me a line with any follow-up questions.  Thanks to tonight's discussants for a stimulating 90 minutes.

http://saschameinrat...

[ Parent ]
Broadband Discussion (4.00 / 2)
John, thanks for the compliment, but the truth is that I was heavily involved in drafting both the original Lautenberg-McCain bill and the new Lautenberg-Smith-Kerry-McCain-McCaskill-Snowe-Sevens-Inouye bill.  In developing these bills, we carefully considered and decided against including a clear and forceful federal mandate empowering municipalities to provide communications services.  The reason is that the Supreme Court (especially as currently constituted) might well view such a mandate as unconstitutional under the 10th Amendment, which prohibits the federal government from dragooning the states into embracing federal policies. 

In any event, the legal issues that held Lafayette back were not attributable to fundamental defects in the Local Government Fair Competition Act.  At the close of the negotiations on the Act in Governor Blanco's conference room, representative of BellSouth, Cox, and the municipalities, stepped through the Act provision by provision and agreed that the Act posed no barrier to what Lafayette wanted to do.  Unfortunately, BellSouth contracted a case of amnesia before the Governor's signature on the Act was dry, and it took three years of litigation and delays for the Louisiana Supreme Court to confirm that Lafayette had been right all along. 

Bottom line:  John, I share your frustration, and I strongly believe that the US must not only eliminate all barriers to public entry, but it must actively encourage public involvement by offering public entities significant incentives to step forward to do their part in keeping the US a great nation. 

Broadband Discussion (Law) (0.00 / 0)
Hi Jim,

I'm disturbed to think that my sense that the federal government is not, as a practical matter, able to coerce compliance from the states is false.--At least in regulatory matters state compliance seems a matter of course. Perhaps a back door is available there?

I'm afraid we disagree on the effect of the Local Government Fair Competition Act on Lafayette's project. Perhaps it is more a matter of perspective. I suspect that you see how it worked in the end to let the project get going (and thank you for that!) and I can't avoid seeing how it enabled delay and placed long-term disabilities on the public sector.

I think it indisputable that the legal challenges that delayed and very nearly derailed the build were all enabled by that law and cited it directly. Had no such law been in existence those particular challenges would not have been possible. There would have been no basis for challenge.

As a practical matter it may have been that a law was coming once the incumbents decided they wanted one and that this was best law possible...and it may be that the city can live with the consequences of some of the odious provisions that are still to kick in. (The worst of which use special public-sector-only regulation to directly raise customers costs.)

As you know I am a strong advocate of the repeal of this law. Perhaps any law could be similarly abused. It is harder to abuse the the presumptions of freedom. :-)

I am curious: Do you think that the new Lautenberg et al. bill would moderate Louisiana's Local Govt Fair Competition Act in any way should it become law?

Thanks for fighting the good fight.

[ Parent ]
question for Paul Morris (0.00 / 0)
Paul, that's a great project you're working on, and I hope other cities can follow your lead.

I have a few questions for you:

How much has it cost to build so far for your 20,000 connections?
How much do you expect it to cost in total for 500,000 people?
How has it been financed?  Have your finances been made public?

I couldn't find anything on the UTOPIA web site or on wikipedia, so I'm stumped. :)

end the occupation of Iraq

Framing the Issue for Mainstream Americans (0.00 / 0)
Back in January of 2002 I wrote the following commentary in an attempt to simply the significance of this issue for mainstream Americans.

It's become a national priority, and all Americans should be able to understand the long-term implications of the nation's current standing within the Global Networked Economy. However, they clearly don't.

My point: if this issue had the same PR team that masterminded the framing of invasion and occupation of Iraq, then this issue may find a much broader 'non-partisan' audience. Something to think about?

Lessons Learned: The U.S. Highway System
http://boston.intern...

Live Blog Here at 7pm ET | 91 comments
Donate to Open Left









QUICK HITS

Friends of the Earth thanks the OpenLeft community for the ideas you generate and your contributions to the progressive movement.

blog advertising is good for you
blog advertising is good for you
LIBERAL INSTITUTIONS
STATE BLOGS
SEARCH

   

Advanced Search

Home | RSS/XML Feed | Register | Contact

  © OpenLeft 2007-2009, Powered by SoapBlox