Submitted by acohill on Mon, 10/31/2005 - 08:11
The Kansai Electric company in Japan has deployed new equipment that enables them to transmit 1 terabit of data per second over their company fiber lines.
That is 100 times faster than most of the fiber transmission electronics currently in use, and shows why fiber, while somewhat more expensive on the front end, is such a safe bet. Research in laboratories has actually achieved even higher speeds, but the Kansai equipment has actually been deployed in the field. Fiber can be upgraded without replacing the fiber and duct lines in the ground or on poles, which is a big advantage, and it makes fiber future proof.
Submitted by acohill on Fri, 10/14/2005 - 10:27
The Rural Telecommunications Congress 9th Annual Meeting is over, but I'm still catching up on presentations. Matt Wenger of PacketFront, a company that specializes in the network hardware and software needed to manage communitywide networks, presented an interesting model for promoting innovation and paying for the network.
Wenger did not say "fiber should be free" in precisely those words, but that is what I took away from his remarks. Wenger argued that it is services that people are interested in--VoIP, video on demand, security, network backups, etc.--and that a connection-based business model (what everyone uses now) actually penalizes both users and service providers.
In a connection-based model, where you pay for a connection of a certain amount of bandwidth (e.g. a T1), if you use services that require a lot of bandwidth, you have to pay more, both on the service provider side and the customer side. So success in marketing your services, or using lots of broadband services, is discouraged.
Wenger pointed out that it is particularly bad for service providers; if they successfully sell lots of services, their costs go up, unlike practically any other business on the planet, where costs typically decrease as business volume increases.
Wenger insists that the way a communitywide broadband network should work is to charge a small fee (e.g. 5%) on the revenue of service providers. In other words, customers and service providers can connect to the network for free, but fees paid to the community network manager increase based on demand, rather than on bandwidth.
Sound crazy? It might be, except PacketFront already has it working in a community in Sweden, where more than 60 service providers are selling services over a community broadband network, and doing so successfully.
Submitted by acohill on Mon, 10/10/2005 - 13:10
Pete Johnson, the Federal co-chair of the Delta Regional Authority (the Mississipi delta of several states and 10 million people) spoke at lunch abou the importance of infrastructure to the health and vitality of communities. He made several points in the early part of his talk:
Johnson gave a brief history of the Delta region starting with the War Between the States, and up to the present day. He has been heavily involved in dealing with the aftermath of hurricanes Katrina and Rita, which caused heavy damage to infrastructure in the Delta region as far as 200 miles inland.
Johnson enumerated four areas of technology and telecommunications infrastructure investment that he sees as critical for rural communities.
Submitted by acohill on Thu, 10/06/2005 - 08:28
I was told recently that the MSAP (Multimedia Services Access Point) was outdated and no longer needed. The MSAP is a public peering point that we pioneered in Blacksburg in 1999. It is still in operation today, and vastly improves network performance within the community.
The MSAP is just a network data exchange point, which is old as the Internet. But what was new and different about what we did in Blacksburg was the concept that communities and regions needed to provide public peering points--for a whole variety of reasons, most related to lower costs for bandwidth and greatly improved quality of service for things like voice and video services.
But another reason is to make sure private interests don't have monopoly control over the network, as illustrated perfectly in this item.
Briefly put, two major backbone Internet providers (Cogent and Level 3) are squabbling and Level 3 has stopped allowing Cogent's traffic to cross its network. This means, in some cases, that you cannot send IP traffic from here to there. Sites lose traffic, businesses are affected, and performance is degraded. And businesses in affected communities can't do a thing about it.
Public peering points like MSAPs and RNAPs (Regional Network Access Points) won't stop the squabbles, but they can help mitigate the effects, and give communities and regions some control over their destiny.
Put another way, imagine if all roads out of your community were private toll roads that could be shut down at any time by a private company? Would that be good for business? Would it be good for the community? Would it help attract business and industry?
The answer, of course, is an emphatic no. But that is exactly the situation we have today with the Internet. As the Internet becomes more and more important to commerce, governance, and daily life, communities and regions cannot keep ignoring these issues.
Submitted by acohill on Thu, 09/01/2005 - 08:20
Buried in several different news stories are brief mentions that the only communications working in the storm-ravaged areas of Louisiana and Mississippi are satellite phones. In New Orleans, apparently the only working telecom facility is the phone company central office (colocation facility), which was designed specifically to withstand storms and flooding. But it does not help much since all landlines to and from the facility are out.
It is a sober reminder of the power of nature and the need to have disaster recovery plans in place. FEMA and other agencies have been designing "instant communications" trucks for these kinds of disasters, but there probably not nearly enough. Picture one of the mobile TV station trucks with one of those extendable booms that rise up out of the truck, but instead of TV broadcasting equipment, the truck can provide cellular phone service, can connect to a working landline to act as a local phone switch, and can provide an instant WiFi hotspot so that data can be exchanged between laptops, as well as use other wireless to try to connect to the Internet via other trucks or working wireless access points.
Submitted by acohill on Wed, 07/13/2005 - 11:44
The city of Wellington, New Zealand has created an MSAP service they call CityLink. It is exactly the MSAP concept, and like Blacksburg, which began offering MSAP service in 1999, ISPs have flocked to it because it lowers costs and enables them to provide better services.
Submitted by acohill on Wed, 07/13/2005 - 11:29
BC.NET, a project of the British Columbia provincial government, is deploying what they call Transit Exchange Hubs in communities throughout the province.
The Transit Exchange Hub is similar in concept to the MSAP (Multimedia Services Access Point), which I designed and implemented in 1999 as part of the Blacksburg Electronic Village. Both the Hubs and the MSAP build on a standard piece of Internet systems architecture called a NAP, or Network Access Point. What NAPs do is allow different networks (remember that the Internet is a network of networks, not a single contiguous thing) to exchange data.
What is different about the MSAP/Transit Exchange approach is that it pushes the NAP down into communities and regions. The benefits can be dramatic--network performance can increase substantially and the cost of intracommunity network traffic can go down significantly. So MSAPs reduce costs and improve network performance. For local applications that are sensitive to time delays like videoconferencing, voice telephone calls, or gaming, the MSAP can be critical to performance and usability by keeping local data local and not requiring Internet Service Providers to haul large amounts of data across their very expensive Internet backbone connections.
Submitted by acohill on Thu, 07/07/2005 - 09:34
One of the common arguments against running fiber to every home and business goes like this: "Once we all have a cellphone with data service, we won't even need a landline."
From a certain squinty distance, it sounds very reasonable, and some invalid data to support it usually goes like this: "And I know several people that don't even have a landline phone anymore."
Here are a couple of contrarian data points. While at the beach last week, in a flat area with few tall trees and in visual site of two cell towers, neither my phone nor my wife's phone worked reliably. And by that I mean you could not place a call consistently, if you did place a call it usually dropped in less than a minute, and you could not reliably retrieve voicemail. The landline phone became indispensable. And we had two different phones from two different manufacturers, so it was not just a device problem.
A more sobering example is the tragedy in London this morning. After the bomb blast, cellphone circuits became so jammed that the system essentially stopped functioning entirely.
All wireless communications suffer from the same unavoidable problem that gets down to basic physics, rather than system design--any wireless system uses a certain amount of bandwidth, and you can only spread that bandwidth among so many people. After that you run out. That limitation does not exist with fiber networks, as you can add more fiber as needed to provide nearly infinite (perceived) bandwidth.
Wireless vendors are always finding ways to squeeze a little more traffic onto a wireless network, and there are new systems like Ultra Wide Band (UWB) and frequency hopping that promise to improve bandwidth availability. But we are all going to want reliable communications, especially in emergencies, and that means we need both very capable wireless networks AND solid, high reliability fiber connections.
Submitted by acohill on Thu, 05/19/2005 - 10:13
If your community is looking at Broadband Over Powerlines (BPL) as a cheap way to get broadband out to neighborhoods or rural areas, you should read this article over at NewsForge, which says BPL still has some issues that have to be worked out.
Among the problems this article raises are relatively high costs, the need to deploy a fiber backbone to support neighborhood level BPL, and radio interference in frequencies used by public safety (fire, police, rescue).
In short, BPL is no shortcut, and may not even be a bargain, compared to other entry level broadband systems like wireless.
Submitted by acohill on Tue, 05/17/2005 - 12:21
Loma Linda, California, a community of 20,000 people, may be the first town in the country to require broadband infrastructure in new housing. This article from the May, 2005 issue of Broadband Properties (scroll down to get the PDF file) details the ordinance that requires builders to install structured Ethernet (broadband) cabling in every living space in new homes, as well as run fiber to the homes in the development, and to provide neighborhood colocation space for network equipment (what I call an NSAP, or Neighborhood Service Access Point).
This kind of approach future-proofs the community and reduces the cost of broadband access. Builders install the neighborhood infrastructure and turn it over to the town when the development is complete, just as they turn over other infrastructure like streets, sidewalks, water, and sewer. The article cites a study that shows homes with fiber to the home (FTTH) sell for $4,000 to $14,000 more than the same home without broadband access. So the builders easily recoup the additional cost, and the increased value of the home provides tax benefits to the town (which helps pay for maintenance).
Submitted by acohill on Thu, 04/28/2005 - 12:22
I'm a big fan of microduct and blown fiber, and Emtelle is one of the world leaders in the technology. I think it is an ideal solution for community and neighborhood fiber projects, as it works with both passive and active optical network equipment, it's easy to install, and easy to repair--essential qualities for community-managed systems. But it's always been hard to explain without actually seeing it. This movie on the Emtelle site is short and illustrates how it works end to end (you need a Flash player plug-in for your browser).
Note that I have no financial connection to Emtelle; I just think they make terrific products. Emtelle studies show that microduct systems are as much as 44% less expensive than traditional fiber cable. Microduct systems need fewer or no pedestals, no handholes or pullboxes, the duct requires no special handling (cheaper to install), and you typically spend less on fiber.
Submitted by acohill on Tue, 04/19/2005 - 10:46
The Monday afternoon keynote was by Keith Wilson, the CEO of Dynamic City, which has the contract to design, build, and operate the Utah UTOPIA project (an 18 community fiber project serving 300,000 homes).
The U.S. has the most expensive broadband in the world; the per megabit cost of broadband in Japan is ninety cents. In Korea, it's $2.50. In the U.S., it averages $25-$30 per megabit, or thirty times higher than the lowest. Clearly, the current reliance on incumbents to provide broadband is not working.
Wilson identified four characteristics of a viable communitywide network:
A wholesale business model that allows for many service providers (as opposed to just one voice provider, one video on demand provider, etc.) reduces the risk for the network owner--if a service vendor fails or pulls out, the financial health of the network is less at risk.
Networks are like airports--a shared facility built by the community and used by multiple service providers (airlines) to offer a variety of services. Airports are good for communities because no airline would come to a community and build their own airport.
Communities need a "communications utility," and no less than the future of the community is at stake. A successful network must have widespread availability, must be affordable, and must offer customers choice. A closed network cannot offer all three, because the incumbent providers don't want competition. Private buildouts (the current situation with incumbents) capture the future of a community because no other provider will come, so the community becomes hostage to a single company.
If regulated monopolies have not worked in the past in terms of affordability and choice, why do we think unregulated monopolies (what we have now, in effect) will work better? What is best for a single company is not necessarily best for the businesses and residents of a community.
Submitted by acohill on Thu, 03/03/2005 - 08:51
James Carlini has a must-read article that has some solid data on the value of municipal investments in broadband, as well as some fascinating historical data that shows community investments in "new" infrastructure pay off.
Carlini has new data on the Waterloo, Iowa and Cedar Falls, Iowa comparison (here is a one page summary--look for the handout titled Case Studies). Waterloo, Iowa decided to let the cable and telelphone company decide what kind of broadband the community had. Twenty-five miles away, Cedar Falls, with less developable land and some other economic development disadvantages, invested in community fiber. Five years later, Cedar Falls lowered taxes slightly, and Waterloo had to raise taxes. Why? Because business investment in Cedar Falls boomed because of the community fiber, and economic development in Waterloo stalled.
Carlini also provides some historical data on St. Louis and Chicago. At the end of the Civil War, St. Louis was the major gateway to the West because it dominated water-based trade. Chicago was much less prosperous, and decided to really push railroads. St. Louis decided to pass laws that discouraged railroad development, and tried to protect water routes by discouraged railroad bridges across the Mississippi. The result--Chicago's population boomed from the economic development the railroads brought, while St. Louis barely grew at all.
We're at the same place today. The telephone and cable companies are trying to get our legislators to hold back the railroads to protect canal barge traffic. Does your community want to be St. Louis or Chicago? Waterloo or Cedar Falls?
Submitted by acohill on Tue, 02/01/2005 - 09:56
Back on October 31, 2003, I wrote about supercomputers as the economic development infrastructure. I suggested that regions that wanted to have a real marketing edge invest in a modest supercomputer cluster and rent it out to businesses that wanted occasional access to such equipment but could not justify the cost of owning it.
Today, Sun Computers had a full page ad in the Wall Street Journal advertising their supercomputer cluster for hire, at a rate of $1/hour/CPU. That's a good bit higher than what some universities like Virginia Tech are charging for business access to their supercomputer facilities, but it shows that there is a market out there.
How about your region? Are you still building steel-sided shell buildings that are sitting empty, or are you ready to enter the Knowledge Economy with some investments that businesses really want?
Submitted by acohill on Thu, 01/27/2005 - 07:56
Take rate is an industry term for the number of customers that agree to buy a service. Take rates are notoriously hard to predict, and historically, take rates for services like telephone and cable service have been very low (e.g. 10%, 15%), meaning it takes years to get most households connected to a new service.
The town of Nuenen, Holland recently installed a blown fiber to the home, open access network, and had a remarkable 96% take rate. This means that essentially, every household that is likely to be a customer became one as soon as the service became available.
This is the global competition.
While U.S. incumbents are gingerly sticking their toes in the waters of *real* high performance broadband by grandly promoting one or two trial projects, overseas, communities are just going ahead and doing what needs to be done. Nuenen's open access network means customers have a choice of providers for their services. Nuenen is proof that not only can it be done, but that there will be customers waiting when the duct goes by the house.
Emtelle, which provided the microduct for the project, has a short video and a four page description of the project. There is some sales stuff in both, but I believe microduct is an excellent approach to implementing community broadband networks.
Submitted by acohill on Tue, 01/04/2005 - 10:53
Yet another former third world country has broadband projects underway that leave U.S. efforts in the dust. Andra Pradesh, a state of India, has embarked on an ambitious but entirely doable project to build a statewide network consisting of a 10 gigabit per second backbone, 1 gigabit Ethernet trunks to a thousand locations, and 100 megabit fiber connections to every town in the state. More than 40,000 government offices will get fiber connections, and will be able to deliver government services via town kiosks and other public Internet locations.
Even more interesting, the official tourism site offers 24 hour chat service to online visitors and potential tourists. What about your community? Unfortunately, in the United States, we have the telcos busily trying to usurp the right of communities to develop community infrastructure, with the legislation in Pennsylvania as a perfect example--PA towns now have to ask Verizon's permission to chart their own destiny.
Submitted by acohill on Mon, 10/11/2004 - 19:23
The second speaker is James Baker, from central Pennsylvania, with the Council of Governments--an 11 county area with 300 local government entities of one kind or another. Most communities are under 2500 population, and many are under 1000. Generally a very low density area--20-40 households per square mile as an average.
Broadband services are expanding in the area. However, 98% of Pennsylvania urban areas have some form of broadband, but only 25% of rural areas have some kind of service. Providers view rural areas as not good markets.
The state of Pennsylvania has funded a GIS system that provides service maps for various kinds of services available (i.e. DSL, cable modem, etc). Good tool, but data quality varies, some limitations in granularity of data.
Wireless services were considered for expansion in one county by swapping tower space on an EMS tower with space on a commercially-owned tower in another part of the county. EMS would get better radio coverage, and residents and businesses would get more access and choice in broadband.
Murphy's Law kicked in...the six inch square antenna which was to be put on the county tower would require a $5000 engineering study to make sure it would not add significant wind loading to the 200' tower. No one would pay for the study, so the project got slowed down while a variety of funding sources were pursued. The ARC came to the rescue, but the $5000 grant application required almost the same amount of paperwork as a $150,000 grant.
After the engineering studies were done, it was discovered that the county did have legal control of the tower, and that has required additional effort. Testing by the service provider has shown that nearly the entire anticipated service area will be covered.
In the meantime, the government fiber project is using wireless to expand coverage beyond the ends of the fiber. Some nonprofits are getting service.
Issues include legal problems--one person, the county lawyer, has the power to stop these projects dead in their tracks. If the cable company expands service, the wireless provider may feel it is not worth it to continue expansion--it becomes very important for government to be able to move quickly to help private businesses.
Submitted by acohill on Mon, 10/11/2004 - 18:59
Dave Nelson is the first speaker, from Chelan County, WA. The PUDs in the state can provide wholesale broadband (not retail). Chelan decided to do a PON (Passive Optical Network) pilot project.
The project turned out well--high take rate, costs were close to estimates, and technology worked well. This led to a broader build out.
Take rates for the optical service is between 25% and 50%. By 2003 fiber had been placed by more than 3000 homes, and dial tone services were added to the network. In 2004, an additional 5000 drops are being added. Cable TV services are being studied--primarily a policy and administrative issue, not a technology issue.
By 2008, goal is to have 75% of county with fiber, or about 30,000 homes and businesses. This is an open access network with 12 ISPs offering access on the network. One provider offers fully E911 compliant telephone service. Television should be available in 2004; content and franchise issues have made this more difficult. Alcatel's B-PON system is being used.
The cable companies have been offered access to deliver programming, but so far, they have not been interested. Service providers pay the PUD for each port (Ethernet, phone line, etc) for which they deliver a service. PUD operates on a nonprofit basis. DSL and cable modem service is available in most areas of the county.
Take rates are ahead of projections. The PUD maintains the right of way and fiber distribution. The biggest problem is not being able to build out fast enough--"a million phone calls a day" about when fiber is getting past someone's house. The PUD had to design a real time Web site that shows construction progress on a daily basis to help ease the phone calls.
Wireless has been provided in some areas where it was going to be some time before fiber arrived. Customers have actually cried (true) when the fiber arrived, they were so happy. The PUD has a good relationship with the service providers (12!).
BPL pilot is also underway; there are some distance/repeater issues.
Submitted by acohill on Tue, 09/07/2004 - 08:52
PCWorld has a nice summary of fiber to the home projects, mostly from the telephone and cable company perspective.
Some of the key ideas in the article:
Submitted by acohill on Thu, 06/24/2004 - 09:16
In a widely carried AP report, AT&T has announced it is getting out of local dial tone and long distance in several states, and may abandon most other states shortly. There are two things going on here, and only one of them was discussed in the article.
The article correctly notes that the proximate cause for the AT&T pullback is the FCC ruling that allows the local phone companies to charge higher wholesale rates for their antique copper telephone lines. AT&T has been leasing these in bulk to provide local dialtone. The higher rates make it unprofitable for AT&T to do so.
On the face of it, this looks bad for local communities, as there seems to be less competition, and puts the local phone companies back near their previous monopoly status for dialtone.
What was not covered well in the AP article is the fact that AT&T is making a major push for Voice over IP local and long distance services. The company has wisely decided to abandon the antique phone service market and concentrate on selling what is going to count in the future. It's a smart move.
Some of the phone companies are not standing still, however. SBC has announced it will spend billions on fiber to the neighborhood and fiber to the premises, although the latter will be done only in new neighborhoods for now. The new system will have the capacity for a single channel of HD TV--much higher capacity than existing DSL lines, but still not what will be needed in the future. But the fiber has the carrying capacity--SBC is reluctant to put in the electronics, probably because of cost and because they are trying to control access.
Communities getting these new systems may breathe a sigh of relief that they don't need to do that telecom planning after all, but their headaches are simply being deferred to the future. A monopoly is a monopoly, and it does not matter much if it is a legal monopoly (the old, pre-1996 approach) or a de facto marketplace monopoly.
Design Nine provides visionary broadband architecture and engineering services to our clients. We have over seventy years of staff experience with telecom and community broadband-more than any other company in the United States.
We have a full range of broadband and telecom planning, design, and project management services.
Free Fiber to the Home
Save NC Broadband
Blandin on Broadband
Intelligent Community Forum
FCC Broadband Blog
KGP Broadband Stimulus
Ars Technica Tech Policy
Bill St. Arnaud
Stop the Cap
Broadband Policy Watch
Lafayette Pro Fiber