Optical networking, in a broad sense, has precipitously fallen from its position as the hottest show in town. Of course, nobody is suggesting that optical networking isn’t the future of communication technology. Of the recent period of exuberance, Deshpande says, “It’s like having a party, and this is the hangover time. But the long-term stuff looks pretty much intact.”

But what really happens next? Sycamore is trading at around $5, down from $125. Corvis is at $2, down from $90. Investors might as well have invested in WebVan. Or is it a little premature to talk about a meltdown?

The New Reality

No networking company can say with a straight face that it has gone through the correction unscathed (although that’s what many companies claim). Carriers haven’t stopped spending, so there is still a huge revenue opportunity, but even in the exciting metro and edge markets, the vast array of vendors fighting for the same pie cannot be supported. That much is obvious. The question is, which companies will be around when spending returns?

Juniper CTO Pradeep Sindhu helped build his company from upstart to major player in less than five years. But from his new vantage point, he doesn’t see that happening much anymore. “Customers are going to be more conservative. Companies that come into the market will have a more difficult time,” he explains.

Of course an army of well-funded startups protests that point. Sprint deployed the first CIENA systems in 1996. CIENA was an upstart, but it had a compelling enough value proposition that implementing its products was a no-brainer. Companies like Metro-Optix, Luminous Networks, and others feel their metro area products provide a significant enough value proposition to generate sales even in a down market, since metro area networks still need to be built out. The new marketing pitch is not that a product is better than what the competition offers, but that carriers can’t afford not to deploy it. Kris Shankar of Metro-Optix explains, “You have to enable more customers per square inch.”

According to a number of observers, any startup company that builds products that are essentially similar to competitive offerings from market leaders will die on the vine.

ONI founder and CTO Rohit Sharma admits, “It’s not a fair process. Companies that have good products and technologies will get suffocated in the next 18 months.”

Jagdeep Singh, founder of blockbuster CIENA acquisition Lightera, now building a startup called Zepton, echoes, “If they can buy the same thing from Cisco, Nortel or Lucent than they can buy from the startup, then that’s a tough sale for the startup.”

Fighting for the Market


Ashok Jain’s track record speaks for itself. He was a co-founder of NexGen Microsystems, which AMD bought in 1995, and then Internet Devices, which Alcatel bought. His latest venture, TeraBurst Networks, is an optical switch maker that’s still fairly stealthy.

What is clear is that TeraBurst will have significant direct competition from public players Sycamore, CIENA, Tellium and others. And the obvious question is how in its wildest dreams, if Sycamore can’t get customer traction and Tellium has the same problem, TeraBurst can possibly hope to find success? Why not just return the $30 million-plus the company has in the bank to investors.

To give TeraBurst credit, optical switching is still a young market. And optical switches will remain an important network element. Surely disruptive solutions could find a real market opportunity in the space over the coming years. But a product coming out today has highly questionable timing.

Nobody doubts that TeraBurst’s products are advanced. The question is if there is any real differentiation between TeraBurst and the competition, and if it’s possible to ramp up sales as a startup in a timeframe that’s feasible for a venture-funded company in a closed IPO market.

Jain sees his differentiation in features like “carrier hotels,” which enable different carriers to share bandwidth. TeraBurst, with products currently in beta, will experience sales cycles with tier one customers that will probably take a year. For now, according to Jain, the company hopes to generate revenue earlier than that by approaching tier two carriers with shorter sales cycles.

The company has more than 18 months of cash in the bank, but will sales have picked up by then? And will big carriers really be looking for a new optical switch solution by then? Jain expects things to pick up in six months, but the company is wisely conserving cash. However you spin it, it’s not an easy time to be an optical startup.

If the market recovers within the year, companies like TeraBurst could be the success stories of 2003. “There’s a lot of demand for just optical switching, and there’s no clear leader,” Jain points out. He admits that there will be no real traction for the company in the near future.

Riders on the Storm

The famous Doors tune fits the market situation pretty well right now. Deshpande at Sycamore may be a big fish compared to competitors like TeraBurst, but he faces impatient investors waiting for a positive quarter, which Sycamore has been unable to deliver recently.

If there’s one good thing about the market bubble, it’s that companies were able to raise huge IPO cash. Sycamore still has over $800 million in its coffers. Deshpande explains that the company will invest in R&D as it continues sales efforts and waits for the market to get back on its feet. Juniper, CIENA, Corvis, others like them and of course mighty Cisco have the balance sheets to persevere, assuming they have what’s needed when the market picks up.

Even as the analyst downgrades continue, the public players are working to strengthen their technological and financial position and make it through on top. Cisco has reorganized itself (see sidebar).

Some ambitious startups are still betting on real growth in select segments. Vivek Raghavan, Siara founder, and former Redback CEO, is building Atrica, a startup focused on developing carrier class optical Ethernet in the metro network. This is clearly a segment where the network is just being built out, but it’s a risky play. Raghavan admits that it is hugely more difficult to build a company now, compared to his experience with Siara, which was bought for more than $4 billion in 1999.

Former Qwest-exec-turned-VC Vab Goel, Partner at Norwest Venture Partners, insists that we “are in year five or six of a 20- to 25-year optics build-out.” On paper that’s true. Those private companies trying to make it right now will suffer through a shakeout that will leave only the very best on their feet. Public players will feel the pressure from Wall Street. Entrepreneurs starting out right now, and willing to spend the next 18 months doing some serious engineering, may emerge to a kinder, and potentially very lucrative, landscape. si