PA-ComNet Dec. 13, 2000

Date of this meeting was changed to December 13 from the normal Dec. 6. The meeting began at 7:45 AM with Leo Creger, Telecommunications Manager of Palo Alto Utilities discussing his past experience working with Raytheon before he came to CPAU several months ago. He now is working in the fiber area. There area about 40 customers for the city fiber ring who are paying almost $1 million/year, but the business is fluid, with some customers leaving and others signing on as contracts end. He’s been meeting with customers to be sure they keep up with technology and use the fiber capacity properly. DWDM technology could drive many small users out if they aren’t careful. It’s important to have fiber companies service the smaller customers, not focus only on big accounts like H-P and Agilent. The city may consider lighting a fiber network without being a content provider. This will allow the city the ability to service smaller customers more economically and monitor the fiber network.

Ideally, the city will provide the infrastructure that will enable many uses of the fiber such as FTTH. FTTH is much the same concept, in that it is available for others to use the city fiber to deliver services. A fiber to the home network, once in place, would provide the path to the home. It is likely the city would light the fiber to deliver the content of other service providers but would not be an actual service provider beyond providing the network infrastructure support.

It is highly doubtful that Internet to the home alone will justify building a citywide FTTH network. There must be other services such as video, voice, library services, etc. to make the service viable and build an economic model supportive of a citywide FTTH network. There are very few places where a FTTH network exists that will provide content providers a platform to test their services. An objective, and hopefully an advantage of the city FTTH trial is to allow content providers to deliver and test their services to see who wants the service, how much they want it, what they will pay for it, etc.

The underlying idea is for the city to become the network, and not have to depend on the Internet to regularly transport information. A network that operates locally can provide high quality service and eliminate the need for an Internet connection to provide content within the local loop. Access to the Internet can have high interconnection costs and speed can be a bottleneck. Security is an issue, either real or perceived. We need to build users’ confidence in secure data transfer over the fiber ring. Where the city wants to provide the capacity and to operate and maintain the network infrastructure, ideally an ISP will be local, and provide locally focused support.

One of the barriers to entry for such content providers is a local presence. The idea of having services within the local network will allow users to enjoy the bandwidth capabilities available locally. To achieve this, there must be space similar to PAIX that will provide service providers a local equipment location. CPAU may have to locate a space that will provide such a facility. Once the support operations are established and functioning, they can contact a number of content and service providers and have them test their performance capabilities. There are a lot of questions to be answered. How can schools and homes be tied together effectively by fiber? What will work, and how will it be delivered? Perhaps a teacher can build a web-based exam and have students take it over time, using secure systems. Another idea is to allow students to exchange assignments and homework via the fiber network or work on collaborative projects.

There are other areas that also have networks to the home. Tacoma planned and installed a TV cable network. They built up the support staff first and ran in the red for awhile, while hoping that customers would sign up. Once proven, they moved to providing Internet access via a cable modem system. Again, they staffed up for the new venture and waited for the customers to sign up. The difference here to note however is that their network was coax-based. The Tacoma operation is in the black, so they moved to Internet over fiber, which also is successful.

Palo Alto dedicated financial resources to build the fiber network and did so with very little dedicated support staff. CPA is presently in the process of hiring the additional staff required to move forward with the city’s dark fiber program and provide resources for FTTH. Failing to adequately staff for a venture such as FTTH will likely provide marginal results at best. They also need space for equipment, and places for companies to move into so they can provide service.

There have been at least two companies that wanted to move into Palo Alto but could not locate affordable space. As a result they located in Mountain View. The loss of the PAIX II facility potentially cost the city over $1.5 million in annual revenues between electricity and fiber leases alone. Loss of that facility had a direct impact on the city’s dark fiber business. To support service providers, there has to be some type of facility like PAIX II as the PAIX downtown is predominantly full. Achieving these efforts can be time consuming and difficult.

Plans for the continued development of the FTTH trial including construction and operations will be presented to Council soon. Staff has been working with a number of equipment providers who have had great ideas over the last 2 years but has yet to see a vendor step forward to participate in the trial with CPA. Many have been contacted again now that the technologies have advanced and it is hoped that some outside support will develop.

The plan is to have this FTTH be a true trial and provide opportunities for suppliers to test their own or others equipment and determine actual and relative performance in a neutral, unbiased manner. The plan is to refund subscribers’ deposit money and make the FTTH project city-funded and a real trial of performance and satisfaction. Again we are interested in knowing what would people pay for available services, for connections, for content, for phones, etc.? We need to know if customers are interested in these kinds of bundled services? Costs for these different services are not yet established. This will be dependent on the service providers’ requirements.

Another major part of the trial is to find out the acceptance rate for FTTH, actual construction costs, operating costs, number of users willing to pay $XX for the bandwidth, etc. The trial is projected to run from May 2001 to May 2002.

Josh Wallace noted there are many people who need data service for home businesses. It may be up to 50% of residents who are interested, so there should be a big FTTH market.

In an effort to make the trial affordable to construct as a city project, the trail area was refined to provide the highest customer density in the given area. As a result, 30 of the original subscribers were no longer part of the FTTH trial area. Once the trial area is constructed and operational, it is likely that other subscribers may be added during the year of the trial. Again, determining what services are available, what ‘menu’ a subscriber would choose and what fees they would be willing to pay all factors into developing a new customer acceptance rate.

The environment for broadband service is changing rapidly. The survey conducted 2 years ago found only 5% of the respondents said they would pay for DSL (1.5Mbs service). The Utilities department plans on conducting a new survey of interest in FTTH while trying to determine what services people want. As mentioned before, it is unlikely that Internet alone will support a FTTH network. As such, the philosophy has changed from providing an ISP only, to offering anything that people want and will pay for. The majority of the content and services would be provided by others and not the City.

While someone questioned about the city being an ISP, if the City was an ISP they would have to lease the interconnection to the Internet and pass those costs on to the customers. It may be too large of a cost to justify, especially for the trial. Darwin Networks is the planned ISP at this time for the trial.

Ashland OR built an HFC system and has learned from customer responses what penetration would be likely, how high are the costs, what do people want and how much will they pay for it, etc. Other places have had to install fiber networks or FTTH themselves because no other service providers would do it for economic reasons. Palo Alto is different in the sense that many providers want to be here because of technology. The demand for high quality services is greater and there are customers now. Other cities may not have the same conditions.

Ideally all providers for the FTTH trial should be local so that we provide high speed local services at the lowest costs practical. Minimizing the dependence on the Internet may reduce costs and complexities as well. Providers of the FTTH services need to understand the systems and the customers so that they scale up as needed and provide the services that individual customers may desire.

In efforts to further expand the utilization of dark fiber in the city and ensure value for the city’s investment, CPAU may investigate providing other bandwidth related services. The original plan of multiple telecom service providers leasing large fiber counts did not happen as planned. As a result, the focus of the effort shifted to meet present needs and customer requirements. As the Staff is progressing with these plans, additional staff is being recruited to meet the demands of this venture.

Leo has spoken with other municipals involved in fiber and finds a wide variety of activity. Grant County PUD in Washington state, which has a great revenues from their electric utility. They are investing lots to set up and operate a full service fiber network partly because no one else can economically justify it. In Los Angeles, they are not planning on FTTH. They are instead looking at the idea of leasing some of their fiber as leased bandwidth. They are looking at what they can do and who they can work with. They probably will lease fiber space as Palo Alto does, and let others such as Enron run it. The idea of providing the bandwidth to a commodities market such as Enron’s effort is an exciting opportunity. This would be in addition to leasing dark fiber as Palo Alto does. Such an arrangement would allow spot buying of bandwidth as required.

Imagine if you would a business model for video on demand. A business would have equipment located in Palo Alto and once a month they would call the market for an amount of bandwidth between the movie studios in Hollywood and their Palo Alto facility. They would then download all the movies to local equipment for use. Palo Alto, Los Angeles and a long haul carrier would share in those revenues generated from the spot leasing. The service provider can now provide high-speed delivery of the latest movies to your home.

The idea of being able to spot buy bandwidth between locations should be exciting. There are numerous applications. CPAU has discussed partnering a citywide FTTH build-out with other companies like RCN or AT&T in exchange for the Palo Alto fiber and minimize network constructions. However, most business plans in place are calling for a hybrid fiber/coax solution.

Business plan development for expanded fiber delivery of services will be done in Dec. 2001, and it will be finalized for approval in Apr. 2002.

Leo has also been involved in discussions with neighboring cities and agencies to try to build interconnections between cities. There is a great interest from PA businesses in interconnecting to nearby cities. Many cities have customers and resources such as could then be interconnected. Many customers outside the city may find it easier to connect to the Palo Alto Internet Exchange than local Internet Exchanges. This in turn increases fiber leased on CPA’s dark fiber network.

The idea is for the city to become the network, and not to depend on the Internet to provide service. A one gigabit network that operates locally will provide high quality service and eliminate Internet interconnection costs for what is a local loop. That eliminates the bottleneck of the Internet, and the high interconnect costs. Security is an issue, either real or perceived. We need to build people’s confidence in secure data transfer over the fiber ring.

Palo Alto can work with SV Power to combine facilities. It’s easier for SV Power to come to PAIX than Maywest, plus we have the users and customers.

Meeting adjourned at 8:50 AM

Submitted by Bob Moss