Article Info

AT&T Project Lightspeed – Too Little Too Late

I’ve written in the past about the drawbacks of Project Lightspeed, AT&T’s ( T) program to provide video and broadband services to the home. In short, I think it’s a half measure at best and provides no competitive advantage over cable.

It looks like this opinion is starting to be echoed within the financial community. AT&T also neglected to offer any details on Lightspeed in their recent quarterly earnings call.

Investment pundits love to point out that Verizon (VZ) is being ‘punished’ for overextending themselves with FiOS capital expenditures, and praise the measured investment being made by AT&T with FTTN.

The reality of the situation is Verizon is purchasing infrastructre usable for the next 100 years with 6% long term credit. AT&T has chosen to delay an inevitable infrastructure upgrade. It’s hard to see how AT&T’s financial position will be significantly better 5 years down the road without making the upgrades now. Just like most homeowners refinanced their houses in the last few years, AT&T (to be fair, SBC) should have taken the cheap money and done the same with their residential network.

This is short-term Wall St. driven thinking at it’s finest.


Comments are disallowed for this post.

  1. Where do you get the number 100 years? It is totally unsupported, therefore the argument that there is value in it being bought with low interest rate capital has no basis in fact.

    Posted by douglas mcintyre | April 28, 2006, 9:43 AM
  2. The copper wires that used to serve my house before I had FiOS installed were 75 years old. They used to support a manually switchded 2KHz voice service. before they were cut down and replaced by fiber, they were supporting a 3Mb/s DSL connection.

    Claude Shannon would side with me on this one… the inherent bandwidth capabiltiy of the fiber AT&T has installed far exceeds the wildest imaginative uses of broadband to the home. It won’t be replaced for 100 years, if even then. The hardware at either end, yes, but the fiber, no. And the fiber install is the majority of the capex.

    Posted by Andrew Schmitt | April 28, 2006, 9:53 AM
  3. At the rate at which technology is changing an 100 year time horizon is not supportable. I have not seen a single comment from a phone company saying they plan to rely on the technology for that long a time frame.

    Posted by douglas mcintyre | April 28, 2006, 10:59 AM
  4. If Claude weren’t dead I would ask him.

    Posted by douglas mcintyre | April 28, 2006, 11:01 AM
  5. Electro-optical technology is changing very rapidly. Fiber optic technology is changing, but not as rapidly. According to a Bell Labs paper in Nature (abstract at, Lucent press release at, the theoretical limit of current-generation optical fiber is 100 Tb/s. That’s a little more than six orders of magnitude more than the best that’s currently available from FiOS, and a little more than eight orders of magnitude better than the most popular “broadband” tier sold in the US today (1.5 Mb/s DSL). It’s also at least five orders of magnitude greater than what anyone thinks will be technologically possible to pump down existing copper outside plant 1000 feet or so (if you can get 100 Mb/s on VDSL2 out 1000 feet, perhaps through some DSP voodoo you will, within the next five-ten years, be able to push 1 Gb/s).

    If you believe Nielsen’s Law (, we’ll hit 100 Tb/s in about 45 years. So, perhaps 100 years is a bit of a stretch for the viable life of the fiber that Verizon is installing today — but it’s not a stretch to say that the fiber that Verizon is putting in the ground today with capital financed at 6% has a viable life of 40+ years, and this is at the most pessimistic at least three times as long as that of the copper that AT&T is using for the last quarter mile of Lightspeed.

    Posted by DG Lewis | May 4, 2006, 10:44 AM
  6. With No HDTV and the neighborhoods that these companies are targeting their offerings are a “non starter” because the take rates will be too low one bill or not. At the end of the day, “one bill” and even a lower price cannot make up for inferior service offerings in the targeted communities.

    Posted by Light Reader | May 30, 2006, 5:01 PM