Chaz, ever seen "Crimson Tide"? This Cisco vs. Nortel thing reminds me of a conversation Gene Hackman and Denzel Washington had near the beginning of the movie. They're having dinner in the officers' quarters, when Hackman questions Washington regarding whether he would've chosen to drop the atomic bomb on Hiroshima were it his decision. Washington gives a long, intricate answer regarding why he would've chosen not to. In turn, Hackman replies, "I would've said, 'Drop the %$&*ers, twice.' That's the difference between you and me. The Navy wants men like you to be complex, while they want men like me to be simple." You see, Hackman was an admiral, while Washington was a lieutenant, and thus naturally, the former felt that he had to be decisive and straightforward in his actions, and thus "simple."
Of course, it was that "simple" mindset of Hackman's that nearly started World War III later on in the movie, and likewise, it's Cisco simple mindset, the product of all those years of success in the data networking field, that's going to cost them dearly when dealing with the "complex" Nortels, Sycamores, and Cienas that inhabit the optical networking market.
To see how this is true, let's first look at Cisco's decision to outsource the production of all optical components. In the switch and router market, while everyone makes their own network processor and switch fabric chips, most other components get outsourced to the likes of Vitesse, Applied Micro, PMC-Sierra, etc, as these companies have always made sure that the performance of these chips never become a bottleneck in regards to the performance of a high-end switch or router. So now Cisco, with its simple mindset, has it ingrained within them "outsourcing is good, in-house production is bad." But optical networking's a whole different game. Much of Nortel's success can be attributed to its possession of bleeding-edge components that it obtained via in-house R&D, and has chosen not to sell to any competitors. Cisco, by depending entirely on JDS Uniphase and its ilk, won't have access to any optical components that its competitors won't also have access to.
Then there's Cisco second "simple" error: the assumption that they can get by with a somewhat-inferior product by means of its sales and marketing clout. This has been a standard practice of theirs in the enterprise networking market, and in the carrier/ISP router market, but they're forgetting two things. First, the enterprise networking market's far more political and bureaucratic. The value of customer relationships, brand name, and sales expertise runs a lot deeper there. As for the high-end router market, Cisco's always had IOS to depend on to limit market share losses (although these losses are beginning to accelerate a bit nonetheless). On the other hand, not only does Cisco have no proprietary lock on its customers in the DWDM transmission and optical switching markets, its customers, depending on their networks for their revenues, tend to go for "best of breed" solutions, even if this means turning down Cisco for some Mountain View garage startup. The astonishing lack of success they've had with their DWDM systems unit, which they acquired from Pirelli last year, is reflective of this fact. It would've been smarter for Cisco to simply acquire Ciena, who has a far more competitive product (albeit still not as good as Nortel's), even if it cost them a bit more.
Finally, there's the issue of how the data networking industry's evolved. Which is just that: it's simply evolved over time, nothing more. In terms of its function, the products offered in Cisco's high-end GSR aren't much different from the first routers the company offered 15 years ago. Sure, they're infinitely more powerful, and the software they run is much better, and the industry's moved from IP to IPv2 to IPv4, and is now set to go to IPv6, but at heart, these devices act in the same manner that they always have. The same with Layer 2 switches. While the protocols have changed and evolved (Frame Relay, ATM, Ethernet), the switches still generally act in the same manner. As a result, Cisco has found that it's best not to go for any "rash innovations," and simply listen closely to what its largest customers want.
Optical networking, like the semiconductor market 25-30 years ago (yes, I know it's an overdone analogy, but it's really on the mark), is still rapidly changing. The industry got turned upside-down a few years ago by DWDM, and another such industry shift is set to come via all-optical switching, which should cut network equipment costs tremendously. Cisco, with its "let's listen to what out customers want right now" approach, is missing the boat here completely. All-optical switching is still a couple of years away, and the company's switch/router customers are still primarily using SONET add/drops and cross-connects, with a few of them buying "hybrid" switches, such as those that Sycamore, Corvis, and Ciena are offering.
Cisco, with its "simple" mentality, figures that this is the way to go, and shells out $10 billion for Cerent, with its leading-edge SONET cross-connect, and another $700 million for Monterrey and its not-so-leading-edge hybrid switch. Although the response for Monterey's products has been lukewarm at best, Cerent seems to have one hell of a cross-connect, by many accounts far superior to the ones Lucent, Tellabs, and Nortel have (although a startup known as Cyras supposedly has an even better cross-connect), and Cisco's set to generate over $1 billion in revenues from this product this year. Which is great, except for the fact that the product will be completely obsolete within three years thanks to all-optical switching, a market for which Cisco appears to have made no attempts so far to develop a product for. Lucent, meanwhile, has poured hundreds of millions into creating it s WaveStar LambadaRouter, while Nortel recently bought out Xros, which has developed the world's most scalable all-optical switch.
It's very important to remember that when we buy shares of a company, we're not just investing in the company, but the people that run it as well. And that's what they are, people, just like you and me. And like all people, they carry a certain mindset in regards to the manner that they approach a given situation. And as the case of Cisco shows, it's also extremely important to make sure that you're able to "tune in" to this mindset, so that you can get a better understanding of the future success that their company will have within specific markets.
Eric |
