A good idea at the time

We’ve all been in a situation where someone’s seemingly plausible plan (frequently, our own) has gone awry so completely that we’re at a loss to explain what happened, or even why the plan was enacted at all. “It seemed a good idea at the time” is usually what we say.

For example, last year in this space I offered a mission statement for the optical communications industry. At the end of the editorial, I urged readers to offer comments or alternative statements that I would run in a future issue of Lightwave. It seemed a good idea at the time — but only one reader responded. To make matters worse, that reader used the column as the basis for a talk in China, where he described my effort as “too wordy and detailed; not fundamental enough.” Since I like to keep my promises, here’s what the reader came up with for his talk:

“The optical communications enterprise is in the business of harnessing lightwaves (photons) for transport and distribution of information, with low unit cost ($/bit/km) and high reliability, thus ensuring the future viability (and present survivability) of information service providers, and the advancement of modern information society.”

The reader was Tingye Le. If you compare this statement with the one I offered last August, you’ll see that Mr. Le has demonstrated why he’s a very successful entrepreneur and why I am not.

The history of optical communications records several failed ideas that seemed perfectly plausible when first espoused — and when venture capitalists and others first sank millions of dollars into them. I’m sure you can think of a technology or a company from the bubble days that attracted attention (and funding) for reasons we now can’t quite explain. If we as a community had known then what we know now, we would have saved ourselves lots of money and countless heartaches.

Yet when we look back on companies and technologies that seemed a good idea at the time, we see that in many instances the problem wasn’t with the idea, but with “the time.” In fact, there are technologies that appeared promising when first introduced, went into a period of disfavor (“Seemed a good idea at the time,” we sighed), and subsequently enjoyed a resurgence of interest. Fiber to the home is an example. Another is 40-Gbit/sec technology, which continues to seem like a good idea, even if we’re having trouble figuring out just when “the time” will finally arrive. The jury is still out on “all-optical networks,” although companies such as Infinera argue persuasively for conviction unless new photonic advances arrive like last-minute witnesses in an episode of Perry Mason. The same jury appeared to have ruled on free-space optical mesh networking technology, but companies such as ClearMesh Networks are appealing the decision.

Conversely, there are technologies whose time everyone is sure has passed, but who refuse to depart. SONET and SDH have been on death row for the better part of this century, but continue to elude the hangman. Meanwhile, technologies forecasted to replace those venerable protocols, like ATM, have found themselves adjoining cells.

The subject of this editorial came to me after a conversation with the vice president of marketing at a chip vendor who recently briefed me on his new SFP+ device. Certainly, the merits of that transceiver form factor have attracted enough attention to warrant the “good idea” label. However, I was struck by his comments about another form factor, the XFP. The company had an earlier device aimed at the XFP. The executive told me he had successfully urged his firm to spike the product because he was convinced the SFP+ would kill the market for XFP devices in the enterprise.

Now the XFP has seemed like a good idea for several years — and devices in this form factor are shipping to customers who also think it’s a good idea. Yet here is where “at the time” comes in again. Does the imminent arrival of SFP+ devices change the XFP’s status? For the enterprise market, I think so. The strength of the XFP has been smaller size and lower power than other 10-Gbit/sec form factors, with the addition of pluggability. The SFP+ takes those attributes a step further. XFP isn’t a failed idea; it’s a good idea. But for all good ideas, timing is everything.

Another good idea whose time is definitely now is nominating someone who has advanced the cause of optical access technology in North America for the FTTXcellence Award. Nominations close in the middle of this month, so go now to www.lightwaveonline.com to ensure that your candidate gets the attention he or she deserves.

In the article “Cost Drives ROADM Component Selection” that appeared on the front page of our July 2006 issue, we incorrectly identified Capella Photonics’ WSS technology as liquid crystal; it is, in fact, MEMS based. We regret any confusion we may have caused.

More in Transmission