I'm fascinated by JDS Uniphase (Nasdaq: JDSU), the market leader in manufacturing components for the fast-growing optical networking industry, and one of our newest holdings in the Rule Maker portfolio.

The San Jose, California company's stock has soared 308% to about $114 per share over the last year, and is up more than 13% since the Rule Maker purchased shares in February. In a year when the S&P 500 is up less than 2%, I consider a 13% gain tremendous.

It's hard to keep track of the company's acquisitions, since it has picked up at least six companies in the last year, seven if you count SDL Inc. (Nasdaq: SDLI), the rival it announced plans to purchase for $41 billion in July. The merger is expected to receive tough review from regulators, but company officials hope it will close late this year.

How does JDS Uniphase, which is helping to thread the world with glass and light, really make money? What are its most promising, profitable products? These are key questions, harder to answer than you might think because companies guard their profit centers closely. But, if you can get an idea where your favorite company's sweet spot lies, then you've got a handle on the industry's basic economics, which makes it easier to judge strategy.

I'll warn you, I don't have this figured out for JDS Uniphase, not even close. But, I can point to some areas in the product line worth paying attention to -- and, considering the depth and breadth of its offerings, this may be worth something. I'd appreciate feedback from readers more familiar with the products or cost dynamics of the optical networking industry. Over time, I'm confident we'll get a feel for the company's profit center. This is one advantage of buy-and-hold investing -- we have time to really learn the profit drivers.

Let me give you an example of why this is worth knowing. Some investors may remember that Dell Computer (Nasdaq: DELL) -- famous for applying the direct-sales model to computers -- used to sell computers in outlets such as Wal-Mart (NYSE: WMT), Best Buy (NYSE: BBY), CompUSA, and Circuit City (NYSE: CC). In 1994, CEO Michael Dell decided to exit the retail channel -- a move widely criticized at the time, since industry-wide retail sales were growing 20% annually. Many people had doubts how far the direct model would carry Dell.

Still, in 1994, Dell had been selling boxes in stores for about four years, and Michael Dell doubted the company was making any money in this venue. As he recounts in his book, Direct from Dell, he didn't think his rivals were making money either.

Here's the point: Dell ran a return-on-investment analysis on the whole company, all its divisions, and found the retail channel wasn't profitable. He didn't think it ever would be, so he pulled out. That's smart strategy, especially with critics harping that a retail strategy was essential.

It's interesting that this was the first time Dell ran an in-depth profitability analysis on all its business units. If a company as outstanding as Dell waited 10 years to measure its real returns, don't be so sure your favorite company really knows how it's making money and where it's losing out.

Anyway, every business has sweet spots. For Dell, the server market offered a lucrative payoff, and the direct model continues to thrive. Coca-Cola (NYSE: KO) makes the bulk of its profits in vending machines and soda fountains, while supermarket sales offer the lowest returns. Ford Motor Company (NYSE: F) and General Motors (NYSE: GM) are making most of their profits from trucks and sport utility vehicles.

Of course, JDS Uniphase has a broad lineup of technical products, which makes for a big job. Roughly speaking, it sells three different classifications of products: active components, passive components, and modules, which are combinations of components. Active products perform optoelectronic functions and require electrical power. Active components include source lasers, pump lasers, external modulators, wavelength stabilizing modules, and integrated laser modulator assemblies.

Passive components perform optical-only functions and don't require electrical power. Passive products include isolators, couplers, gratings, circulators, optical switches, tunable filters, amplifier modules, add-drop multiplexer modules, and switching modules.

Okay, that's complicated terrain. Where to start looking? I wasn't able to track down any margin data on the products other than very basic segment-wide information in the company's 10-K. With the industry changing as fast as it is, it may be too soon to pinpoint the company's profit center. That's one of the hazards of owning JDS -- it's hard to know.

Nevertheless, I'm intrigued by a story (free registration required) I came across on industry website Lightwave, which talked specifically about pump lasers, an active optical networking component used to help amplify light signals.

Pump lasers are a critical part of dense wavelength division multiplexing systems, which boost carrying capacity by enabling a single strand of optical fiber to carry multiple wavelengths of light. This is a big deal, considering the demand for bandwidth.

A CIBC World Markets report expects the market for 1480 nanometer pump lasers to grow to $2.7 billion in 2004, from $272 million last year -- 58% compound annual growth. It also expects the market for the 980 nm pump laser to grow to $1.7 billion in 2004, from $109 million last year -- 73% compound annual growth. These figures shouldn't be taken as gospel, to say the least, but the pump laser business is growing crisply, and JDS makes both kinds of pump lasers.

Aside from fast growth, the 980 nm pump laser business has high barriers to entry, according to the CIBC report. Why? JDS and SDL combined have a near-monopoly on the 980 nm pump laser market. Further, it would be very difficult for a new competitor to duplicate all the hours of testing the two companies have performed to create such reliable devices.

As such, the pump laser market is a good place to keep an eye on, and it will be a starting point for my research into the company's product lines. Investors should pay close attention to the JDS-SDL merger. While it's almost certain concessions will be required, a merger of the two leaders could help create shareholder value for years.

Have a great day.