JDS Uniphase
Let's Predict the Future of Broadband

Format for Printing

Format for printing

Request Reprints


By Newsman
February 15, 2002

Posts selected for this feature rarely stand alone. They are usually a part of an ongoing thread, and are out of context when presented here. The material should be read in that light. How are these posts selected? Click here to find out and nominate a post yourself!

This is long.
But at the end I actually ask for everyone else's thoughts. Plus, I think there could be no more vital topic to our JDSU investment than this. That said, here I go...

On a previous thread about Cox Communications, the gist was that the company was not really able to sign up a ton of new subscribers. "The company said sales will rise as much as 15 percent this year because of customers for newer services like digital TV."

Some may read that and think flat cable subscriber growth is hugely negative for JDSU. But the sentence I pulled out above is great news -- the only way for cable providers to grow their revenues is to offer more, premium services that customers are willing to pay more for. As long as they need more bandwidth to do that, that's a powerful incentive for them to buy the gear from our guys to enable those services.

As we try to figure out how this might shake out, let's think about how computer Internet access evolved: First a few academic institutions and the government used it. Some corporations picked it up. Then people who had come into contact with it at those places decided they would like it in their private lives, too. Those with the technical know-how went through the complicated process of buying the right gear, finding a service provider, setting modem parity bits, etc., and finally getting signed on. But the growth curve for Internet use went vertical ONLY when one company made two things happen: AOL made going online easy as popping in a free disk, and they gave people a reason to want to get online -- to chat, instant message and e-mail with family, existing friends, and new people in Toledo that golly gee they could meet online. Grandkids were getting grandma to sign up for AOL so they could all stay in touch. A girl I went to college with got her dad online so they could e-mail every day while she was out of state.

Well, here's where we are in broadband: It started off only being available at academic institutions, governmental offices and at large corporations, places that were compact enough to wire for high-speed for a large number of users to spread the cost over. Lately, many of the people who got spoiled to fast access at those places and who had the technological know-how figured out how far they were from a phone switching office, whether DSL was available, which provider made sense, got the technician out to install it, worked out all the bugs after the tech left, and finally were online. Luckier people like me called up Time Warner, asked for Road Runner cable modem service, and got a little easier time of it, and from what I understand, as long as your home is DSL capable, getting it hooked up is now much less painful than before. But getting people to pay for fast access has been difficult. Why? 1) A: You still have to be something of a techno/Internet geek to a) use a home computer and b) use it enough to be maddened by slow dial-up access 2) There are few compelling things you absolutely CAN'T do on dialup that you can on broadband.

This much of the argument is all but conventional wisdom. We're looking for a "killer app" to drive broadband growth. People have said that "audio and video" or "full-motion video" are those apps. I think that's imprecise. Some people think that means teleconferencing. There's probably some limited desire for video phones for social or telecommuting use, but it's not a must-have to your average person, especially if it's costly.

The obvious killer app to me is what I'll call universal video on demand. That means that I can sit in my living room (NOT AT MY COMPUTER DESK--we have a 19-inch monitor and a DVD drive on our comp, but my wife STILL doesn't like to sit in our sunroom and watch a movie on the computer) and be able to call up virtually any television show, movie, sporting event, State of the Union address or whatever EVER RECORDED, any time I want to. I can view that program on a high-resolution screen, run the audio through my home theater speakers and start, stop, pause and rewind the show from a remote control. A virtual Blockbuster video in my living room with a limitless inventory.

Say I'm not in the mood to watch figure skating at the Olympics? I could dial up the first episode from Survivor Africa that I missed last year. It's 3 a.m. and I can't sleep. I think I'll watch last week's Saturday Night Live. Friends are coming over. Let's all watch Office Space. I would be willing to pay a per-program charge for that ability(just like I already pay per-rental charges at the video store). Or, I could elect to virtually "buy" a movie, that is, pay a sum for the right to stream it into my home any time I want (or record it digitally for reviewing later) for a higher fee. Or if I was a total couch potato, I might be willing to pay a pretty high flat monthly fee or fees for unlimited use of all or certain categories of programming (my father in law subscribes to receive all the NFL games every Sunday on his digital satellite system, but with UVoD, he could subscribe to the ability to watch any football game ever played any time he liked).

It's not so important than I, Newsman, on the forefront of technochange, would pay for those capabilities. What is, is that I think a slob sitting in a trailer park drinking a Bud who never even found any reason to sign up for AOL would, too. He wouldn't think of it as signing up for "broadband Internet access." He'd just think of it as getting that really cool new kind of cable. Americans love nothing more than they love their televised entertainment, and if UVoD were available, people would sign up for it and people would pay. How many times have you seen a digital satellite dish on the side of a shanty? My brother's roommate had no furniture but had a DVD player.

So why the heck isn't this service available yet? Well, I don't know, and I hope you help me figure out what it will take to get there. We are seeing clear signs of an evolution in that direction: first you had basic cable, then premium channels, then pay per view, and now, my Time Warner cable offers me limited video on demand. That is, on a list limited to about 50 movies at a time, I can start, stop, pause and rewind the program whenever I want. But often I page through the list and find nothing I'm interested in.

Here are what I see as the needed elements:

1) An Internet connection to the home fast enough to carry a digitized full-screen, full-motion video feed.
2) An affordable (preferably) high-resolution, comfortably large video screen on which to watch the program.
3) An affordable set-top box that can decode the program, display a catalog of titles onscreen from which I can pick and identify me to the network so I can be billed for what I watch.
4) A remote server (or more likely, geographically diverse collection of duplicate servers) that hold all or at least the most likely sets of programs that I and my fellow customers will want to view.
5) Tons and tons of extra bandwidth. Right now, the cable company sends one river of 100-some channels past everyone's house in the entire system, and the set-top box lets you pick out the one you want. So they need enough bandwidth in their network for 100-some channels. Everyone can watch NBC at the same time and it only costs the cable company one channel of bandwidth. In UVoD, they'll still have to give everyone those 100-some fixed-time channels, but they'll also need the ability to let everyone watch a different on-demand program simultaneously. So instead of needing enough bandwidth in their network to send out just 100-some unique channels at a time, the cable company might need to be transmitting 10,000 or 100,000 different programs to different customers at any one time. The company needs enough servers to send out all those streams, and enough bandwidth between those servers and customers to get all the feeds to the right place without degrading the viewing experience.
6) A billing system and pricing structure that lets the content creation and the bandwidth-providing companies recoup their costs and make a profit but attractive enough to Joe American to make him sign up for it. We already witnessed a "universal music on demand" system in the form of Napster and its ilk. It worked stunningly, except that the only missing link was that people weren't paying for the content. If they HAD been, the seller of the content could have taken a cut for the intellectual property creation and then paid for the bandwidth the content traveled on to enable the transaction. I think THAT means that either one company has to do it all (provide the content and the bandwidth on which it is carried to the consumer, then bill that consumer --AOL Time Warner!), or there has to be some universally agreed-upon standard for making that happen (like the bank ATM system where there are standing arrangements for how much banks charge other bank's customers for using their ATM terminals). That would let me watch a Time Warner movie over my Cox cable connection, pay one company for the experience and have that company pass along part of the fee to the other party. I would in reality be paying for "broadband" every time I used it, but I would think of it as paying for the experience of watching a movie that interests me, and rather than staring at $50 a month, I'd think, "I just paid $4 to watch Die Hard." Content providers would be willing to bundle in bandwidth charges into viewing fees because it saves them from paying to duplicate media and distributing it to retail outlets that have to make their own profit.

Seems to me that 1, 2 and 3 are already here, with the possible exception that cheaper HDTV screens need to become available. But it could work on regular TV's at first. As to 3, check out, which is a Linux-based set-top box for just this very purpose. 4 is a little bit of a challenge. Time Warner Cable obviously has a server or set of them that can hold and stream 50 movies right now, but upping that to 500 choices probably gets a lot more expensive. But 5 is the big sticking point, and that's mainly due to 6, isn't it? If companies had a reasonable expectation of being able to recoup their bandwidth investment through a billing system already agreed upon, they'd build that bandwidth, wouldn't they? Or does fiber optic gear just have to get a lot cheaper to lower their risk threshold? What do y'all think will break this logjam?

Seems to me AOL Time Warner is the company most likely to implement all this first, because it can demonstrate the concept successfully using the content and cable systems it already owns (no need to make deals with third parties at first). Later, it could make deals with other companies to use their cable networks (for which it would pay a per-movie fee). It strikes me just now that JDSU and one of the network builders it supplies like Nortel or Lucent, should lend for free or really cheap all the necessary equipment for a pilot project in one metro area or one suburb to show how successful it will be, then recoup that investment in spades providing the paid systems for the rest of the metro areas later.

Of course, once this kind of system is in place, it will replace your telephone service, [and] give you a broadband jack to plug in your PC, too. Your toaster could even be on the network.

I have a lot of other thoughts, but that's enough for one post. Let's hear yours.


Become a Complete Fool
Join the best community on the web! Becoming a full member of the Fool Community is easy, takes just a minute, and is very inexpensive.