Mr. Wirbel has written an interesting article regarding the dark fiber payback.
However, the concepts he espouses - Ethernet in the First Mile, Metropolitan Ethernet Forum and Resilient Packet Ring Alliance - all relate to architectures and technologies in the metro network, not the last mile access networks that we are concerned with in our discussions. His assertion is that if you solve the last mile bottleneck, possibly by use of Ethernet over Passive Optical Networks, then you could start lighting up all that dark fiber in the ground profitably.
The most telling line is this one:"Just as the metro fiber becomes ubiquitous, broadband last-mile access stops dead in its tracks."
As MikeM would say, with all the embedded twisted pair copper and coax cable already in the ground and FTTC/HFC buildouts already done or planned, Wirbel's view of Ethernet over PON to solve the last mile bottleneck is utopian at this point at best. Video over IP is just now being studied, we can't do it yet. There is not enough excess fiber in the ground to do a total PON as he suggests without placing fiber much deeper into neighborhoods. There just isn't the fiber in the last mile to do it. In the metro networks, yes. In the access networks, I don't think so.
As far as the reliability of current HFC based telephony goes, numerous real world studies have concluded that AT&T is meeting or exceeding the telco industry benchmark of 99.99% availability, with less than 53 minutes/line/year downtime average in the networks. The ILECs typically struggle to achieve this benchmark. Some AT&T market network are approaching 99.999% availability, or less than 5.3 minutes downtime/line/year. The powering solution that has pretty much been standardized on is the battery backup at the node which provides a minimum of 8 hours backup time. After that, portable generators are deployed to keep any node areas still without power up and running. There are a few older systems that still rely on local AC powered backup (local supply plugged into the customer's outlet) but these are small areas now that are being converted to "network" power (power down the coax center conductor or off composite twisted pair wire).
Penetration rates in established markets are in excess of 20% nationwide, with some markets in the 35% range. But it takes time, Ray, to roll out the equipment and services. It takes lots of time to order, build, ship, install, provision, test and put in service all the equipment required in the headend, at the homes, at the switches, along with setting up interconnection agreements, to make Armstrong's vision and strategy happen. When you add all the things that need to be put in place to do digital video and high speed cable modems, it takes lots of time to roll it out.
Time that WS was not willing to give. Times may be a'changing, but there are some telco truths that remain self-evident. One of those, I believe, is that ubiquitous last mile access to homes is a precious thing that should not be squandered.
Finally, to answer your question about whether AT&T is nimble enough as a giant bureaucratic org to compete with the ILECs and CLECs out there. My gut answer is that they are just as nimble as the ILECs (no great compliment there, though) and not as nimble as the CLECS. But they know what they are doing to provide the services. Part of the strategic plan by T is to stem the erosion of the LD business through bundling with the local service. And it is working (but too slow, I'm afraid). Of those customers who take AT&T Digital Phone (cable phone over HFC), over 85% take a nicely priced package that includes LD service. Less than 15% stick with their other LD supplier like Worldcom or Sprint. The price is right, they get 1 bill, and customer service for both local and LD at one place. Many of these same people take cable modem service for the same reasons and also have their cable service provided by AT&T. Ray, I truly believe the strategy works. Unfortunately, the numbers are still too small to make any appreciable dent in the overall scheme of things like the eroding LD margins and market share. They have really just started. Although it's been 2 years or so since it was announced, because of the planning, lead time on orders, personnel training required, call centers, operations centers, etc. etc. that it takes to do this right, it has really just begun. And they want to dismantle it just as it is beginning to take off.
All the best,
DT |
