FTTH: Back to the Future

HALF MOON BAY -- Drew Lanza yesterday unveiled an interesting business plan for a fiber-to-the-home (FTTH) startup. The company has a commitment from an East-coast RBOC and a projection showing revenues growing to $5 billion in about ten years.

Lanza, a general partner at Morgenthaler, shared the documents on Monday during a morning panel session at Light Reading's Links 2004 Executive Summit, decribing a bullish scenario in which the startup would ride the wave of explosive FTTP growth.

The only problem? In Lanza's own words: "It's a fraud, and I have misled you." Just like Dan Rather, Lanza had cooked the books.

The punchline? The startup was Raynet, which Lanza had helped found -- in the 1980s. "Every slide I've shown you is more than 15 years old."

Lanza's point underscores a theme noted by his fellow panelists: FTTH is hung up on the hurdle of demand. The eventual takeoff of FTTH buildouts "has nothing to do with technology," he says. "The technology has been around for 15 years."

Lanza articulates the issues in a recent column on Light Reading (see Fiber's Sticky Wicket). His conclusion is that if the applications in use don't require much bandwidth, there's no real reason for FTTH to storm the nation.

The outlook from the lone carrier representative on the panel -- Mark Kaish, vice president of next-generation solutions for BellSouth Corp. (NYSE: BLS) -- wasn't much more promising. "There's not a tremendous amount of demand growth," he says.

In a sense, the move to the converged IP/MPLS network is in the same situation. What's lacking, Kaish says, is the "compelling event" that drives carriers to change to a new network. Frame relay, for example, benefitted from the shift to LAN/WAN architectures from mainframes. Nothing like that is pushing customers to IP and fiber. "There's no driver that's forcing people to change their networks," Kaish says. "It's a customer-by-customer or almost industry-by-industry move towards this type of network."

Another barrier is the politics within carriers, says Bert Whyte, CEO of Network Equipment Technologies Inc. (net.com) (NYSE: NWK). Whyte characterizes the telecom world as being run by old guys reluctant to radically change the network they grew up with. "They want that telephone network model moved to the broadband generation," Whyte says.

Whyte notes that it's up to equipment vendors to do the hand-holding to help carriers get over their fears. Moreover, as IP hasn't been a moneymaker for carriers to date, he declares it imperative for the equipment community to help carriers work out the equation. "We've got to work hard trying to develop services and applications so they can make money on the network," he says. "We haven't done that. We want to sell them equipment and tell them about the 'converged network.' "

Carriers also have to find ways to retain customers as more broadband options become available, including wireless. Bundling is a key differentiator for service providers, but it's an intermediate step, says Kevin Walsh, vice president of marketing for Calix Networks Inc.

The endgame is to have services integrated, to the point where a single-end device handles voice, data, video, and whatever else is coming down the pipe. Some of this is beginning already, in the form of cell phones surfing the Web or TV sets showing caller ID information. "Obviously the service provider that's able to do that will reduce churn," Walsh says.

BellSouth's Kaish agrees on that point. "Ultimately, seamless integration is where the excitement is," he says.

— Craig Matsumoto, Senior Editor, Light Reading

For more on this topic, check out:

For further education, visit the archives of related Light Reading Webinars:

Page 1 / 30   >   >>
dwdm2 12/5/2012 | 1:15:50 AM
re: FTTH: Back to the Future Lanza's own words: "It's a fraud, and I have misled you."

"The technology has been around for 15 years."

Indeed it is misleading... because:

"SBC's Kaish agrees on that point. "Ultimately, seamless integration is where the excitement is," he says."

The technology does not exist.

It is a problem with many converted VC wannabe's. Ironic thing is that many people actually 'reads' the LR. Manipulating numbers is not a business plan. Read Gates's "The Road Ahead" or many other famous business plans available before you confuse/mislead more people.

Drew Lanza 12/5/2012 | 1:15:49 AM
re: FTTH: Back to the Future Well, thanks for calling me a "converted VC wannabe". That makes the 121'st insult I've received on Light Reading. Is anyone else keeping score?

I was trying to be funny when I said to the audience at the Light Reading Summit that I'd 'pulled a Dan Rather' on them. It got a big laugh. Afterwards, dozens of people came up and thanked me personally for saying that the emperor wasn't wearing much.

You can throw all the sticks and stones that you want at me. It won't change the fact that I've been in the fiber-to-the-x business for almost 20 years now. That entitles me to engage in the debate. Period.

When I said that the technology has been around for 15 years, I was correct. Raynet deployed the world's first PON system about 15 years ago. It used integrated optoelectronics from Lucent. It cost about the same as today's PON systems do. And you can't argue that point. The product worked and it was deployed around the world. End of story.

There have been many improvements made to the PON architecture (with many more to come). And there are many new and exciting optical components to add to PON systems. Heck, the world wide web didn't even exist when we started to ship our PON systems.

So what? I have been consistent in my posted views on Light Reading for three or four years:

The basic science is pretty much done. The effort now falls to the engineers. I'm not denigrating anyone. I'm an engineer. I love to do engineering. But it's different from science.

Science was inventing optical fiber. Science was designing the DFB laser or the first OEIC, the AT&T ODL-50 (which Raynet used). Science was Mick Reeve coming up with the PON architecture.

Innovation was Raynet pulling all of these pieces together into the first commercial PON product 15 years ago. And innovation was taking a handful of existing protocols and technologies and creating the world wide web.

I do not know over what media my home will be connected 10 years from now. It may be copper (coax or twisted pair), or it may be fiber, or maybe even wireless.

But the architecture and components of that network are already here. They have been put together by large and small companies and they have been trialed at carriers around the world.

I am not a pessimist. I am an optimist. We have ended one phase and are set to begin the next. As I have said many times before, it is now time for us to wire the planet. Not with wishful thinking, but with real, cost effective, well engineered products.

Sorry if you don't like my message. Rather than shooting the messenger, why don't you engage in the debate?

rbkoontz 12/5/2012 | 1:15:48 AM
re: FTTH: Back to the Future Drew,
Congrats on your successes to date. But let's be clear - the Raynet product didn't work. I rememeber - I was there. We took tremendous pressure from Nynex over the dismal reliability and performance of LOC-2. If I recall, you bailed out of Raynet before the product ever was mass deployed in BQB or Germany. The business case failed to develop over time because the product did not perform as promised and had no capability to support data or video over copper. It merely replaced legacy voice services with a lower reliability, higher maintenance headache. Raychem was lucky to sell a burning bag of dung, called Raynet, to Ericsson in 1995.

After Raynet, it was on to EO Networks for you wasn't it? Fiber to the Farm? Another failed product that merely replaced legacy services and had no migration capability to support IP data or IP video bandwidths.

Yes FTTx optics technologies have been around for 15 years, but never before has anyone surround the optics with ultra-broadband, high performance systems for near price parity with legacy. In spite of your cynicism, we are about to embark on a tremendous capex spending cycle from the incumbent carriers which will enable them to offer truely integrated services. Will you be watching from the sidelines?
rjs 12/5/2012 | 1:15:48 AM
re: FTTH: Back to the Future Drew, let me pose you this question:
Was the Model T science or technology/business?
When Henry Ford decided to make a cheap car for the
masses, the railways and waterways were the only means of transport. Roads as we know them today were the result of mass acceptance of Model Ts because the average middle class family could afford one in their lifetime (2 years pay to be exact)
and soon the local roads were jammed with these vehicles.

Automobiles were already a known science when Henry Ford realized that what was needed to replace the horse buggies was a low-cost vehicle.
He made a business case for it in the Ford Model T.

I see a clear analogy between the Model T and the horse buggy and todays Fiber and Copper.

I don't see a model T in the fiber world though.
I am not going pay 1000 dollars for a stupid 10G optical transceivers. We need a Henry Ford and a Model T to make a business case for it. And that is what has been lacking. Make OE ICs as cheap as
silicon ICs and I bet you will have some sort of FTTH or FTTx get a foothold.

And Oh, making a 50 thousand dollar Cadillac for
2 thousand dollars takes a lot of science!

If we were to go based on what you are telling us right now, Sony which commercialized the first low-cost red lasers was barking up the wrong tree in the mid-eighties. They are currently one of the largest vendors of these commodity lasers that they use in-house for their own consumer electronics. And they showed that there was a market for it.

You want another case, look into the history of smelting Aluminum and the Hall process. There were
many like you who asked the need for large quantities of Aluminum. History has proven otherwise. Aluminum was available and there was nothing new in the science of isolating Aluminum.
However, the science of making Aluminum cheap was still to be invented by Charles Hall.

My two cents ...


niblick 12/5/2012 | 1:15:48 AM
re: FTTH: Back to the Future Mark Kaish said in the article that converged IP/MPLS networks similar to FTTx will see very slow growth.

I thought most Carriers are moving towards IP/MPLS based core networks not only to save on Capex and Opex but also to enable new services such as VoIP, Video conferencing, gaming, etc. According to a recent Yankee report, 85% of Carriers have already made a decision to deploy IP/MPLS based networks. Mr. Kaish's remarks seem contrary to this view.

Drew Lanza 12/5/2012 | 1:15:47 AM
re: FTTH: Back to the Future Most Wanted:

Linda Seale told me at the Light Reading conference that the Raynet systems installed in BQB are still there and operating just fine. For that matter, products that I helped to design at E/O Networks are still being manufactured and sold today, ten years after we first introduced them. Neither company became an AFC, but I think it's a little harsh to describe them as 'failed products'.

The Raynet system did offer video and data services, too. The data services were frame-based, because the world wide web had not yet been introduced, so we did not base them on IP. The initial video system was based on FM transmission, because we could not find optical sources with enough power and linearity to transmit the signal in an AM format.

We did have a lot of troubles with the early installation of LOC-2. I've never known a company who built products for the outside plant that didn't.

The modern systems (like those from Wave7, where we are investors) are vastly superior to the products we built at Raynet and E/O Networks. I acknowledged that in my talk yesterday. But they are not substantially different. The modern automobile is not substantially different from the Model T.

Since you were there at Raynet, then you know that I left in the early 90's after 7 years with the company because I disagreed with the approaches we were taking to product cost and product reliability. We could have done better and I regret that we didn't. I was in my twenties when I helped to start Raynet. I am almost 50 now. I have learned a lot in twenty years.

But even if we had built the greatest system in the world and given it away for free, we still would have gone under.

It is not architecture nor technology that has held us back for the past decade. It is politics. Sometimes those are the politics inside of an RBOC. And sometimes those are external politics coming from the FCC or the state PUC's.

Those same politics have not held back countries like South Korea. They adapted the technologies that Raynet and other startups developed in the early 90's to build the world's greatest network.

As Ivan Seidenberg, the head of Verizon, said in late 2003, "We're no South Korea." He was right.

photonsu 12/5/2012 | 1:15:47 AM
re: FTTH: Back to the Future Who is Mich Reeve?
deauxfaux 12/5/2012 | 1:15:46 AM
re: FTTH: Back to the Future Most wanted

Great post. Thanks for stripping away the revisionist history.

Drew Lanza 12/5/2012 | 1:15:46 AM
re: FTTH: Back to the Future Mick Reeve is one of the CTO's of British Telecom.

You can see a picture of him and the band at:


Back in the mid-1980's he and some of his mates invented the PON system. If you search the British Patent Office you can find the original PON patent there.

At the same time, Mick and team also invented blown fiber. These guys were on a roll.

As I recall, they had someone (Fujitsu?) build them a trial PON system that Telecom Australia installed outside of Sydney. I flew down a couple of times to inspect it. Then they went to work convincing us.

Mick ultimately convinced Raynet to shift to a PON architecture. Jim Triplett, who was the head of technology at the time at Raynet (and I just had lunch with a few weeks ago along with Marcus Nebeling), introduced Mick to the rest of the company. We all got along great and worked with Mick to build the world's first commercial PON system that we deployed about 15 years ago.

I ran into Mick five years ago at a conference in Florida. He said he was doing great. Still playing in the band and still enjoying working for British Telecom.

If anyone out there reads this and bumps into Mick, please tell him I said 'hi'.

I said in a previous post that I had earned the right to engage in the debate. Mick started the debate. Enough said.

rjs 12/5/2012 | 1:15:44 AM
re: FTTH: Back to the Future PON makes sense when the cost of transceivers and
other optical components/switches is large.
If the price of transceivers and end optics comes down to commodity levels and still leave enough margin for the surviving vendors to make a decent
business case, the case for PON will vanish.

Consider the good old ethernet. You don't see any more CSMA/CD based coax-cables. Ethernet switches and transceivers are cheap enough to build ethernet based networks point-to-point, user-to- switch. PON is somewhat similar to the coax based ethernet.

With low cost opto-electronics, point-to-point or
user-to-switch (central office) architectures make
more sense.

Our good old POTS is a point-to-point architecture
on copper. The end-user is connected to the central office(the switch) on a dedicated copper line. One can replace the copper with fiber as and when permitted, provided the cost of the optics is small enough to be insignificant.

All this boils down to needing a model T for optics. BTW, the model T was immensely profitable for the FORD motor company. Just because it was low-cost did not mean it did not have a good profit margin.

The problem with optical components industry right
now is that the margins are very low or even negative. They are squeezing blood out of rock!
I repeat, we need a model T.

Page 1 / 30   >   >>
Sign In