So is optics really just hype after all?

By Thomas Nolle
Network World, 11/20/00

The recent meltdown of fiber-optics companies on Wall Street raises troubling questions about one of the most fundamental assumptions of our industry: Bandwidth needs are exploding, and networks will have to change to respond to this trend. While many may scoff at giving Wall Street bean counters a deciding vote in the future of networking, we have to remember that these guys are going to have to finance any major changes in infrastructure and any new service providers. Are they right that optics is overblown?

One indication that Wall Street is right is that optical components and fiber are being promoted as higher-growth opportunities than the equipment that would incorporate these components and the service networks that would use that equipment. Service opportunity equals demand. Demand equals new devices. New devices equals new components. Well, service provider stocks have generally been tanking, and the major equipment players such as Nortel Networks, Lucent and Cisco have also fallen. Does the optical market fail this very first test of financial sanity?

Not really. The problem is that we're looking in the wrong place. There is a service provider segment that's doing quite well - the regional Bell operating companies (RBOC). RBOC data services are growing at more than 30% per year at a time when the most data-centric of the interexchange carriers, WorldCom, has reduced its data earnings estimates. What's more, this service provider segment is buying fiber equipment. This year alone, thousands of new fiber remotes have been deployed. In all, the fiber remote market will generate more than $30 billion in sales. Who got the deal? Primarily Alcatel - which has its own fiber-optics components unit.

Alcatel and Marconi, two offshore equipment giants, have quietly eaten the lunch of our domestic equipment players in the fiber remote market. Alone among the big name fiber players, these two firms have focused on fiber in the outside plant or access network, rather than the fiber core. As DSL deployment by the RBOCs expands, remote fiber concentrators for DSL lines explode. The international players have been ready for this, as domestic players missed the signals of their own buyers.

If our equipment vendors have been dumb, we can't claim as a market to have been much smarter. We neglected another truth: You can't get onto a high-bandwidth network through a tin can and string. Access bandwidth reform is a precursor to any major expansion in core bandwidth requirements because without a fast delivery infrastructure, you can't sell high-bandwidth services even if the core can create them.

What we're seeing now isn't the death of optics, it's the birth of optics. First, improvements in the access network will increase the deliverable quantity of bandwidth. We can expect to see access bandwidth explode by a factor of 100 in the next four years. During that time, most of the optical investment will go to fiber remotes.

Access modernization will enable service modernization, which will enable core network bandwidth expansion and technology change. Forget the Internet as a driver; even for WorldCom it was only about 10% of revenue. We don't really have a handle on what the service drivers will turn out to be; VPNs, perhaps content - more likely, both. Eventually they'll increase our core bandwidth by that same factor of 100. As this happens, core optical capacity growth will drive considerable deployment of new optical gear.

So what's the problem? Our demand for instant and continuous gratification. We want the technology equivalent of the storm of the century every year. RBOC fiber isn't exciting to us because the RBOCs are the same old telephone companies we've always known. Venture capitalists don't want RBOC fiber either, because they don't fund RBOCs. Core fiber explosions in four years aren't exciting either and don't promote a fast profit.

Here's the choice. We can entertain ourselves with fancy stories of new-generation networks built on a financial house of cards, with no credible revenue behind them, and watch as Wall Street pulls its money out of technology, or we can face reality and recognize that we're in the process of modernizing an industry whose scale dwarfs the gross domestic product of some countries. This is going to be done smart, right and slow.

Taking the long view used to be respectable in financial and technology matters. All we have to do to recognize the value of optics is to make long views respectable again.