Tuesday, January 20, 2009

Roll your own

Personally, I have never rolled my own cigarettes. Being on a jet plane for half my life I am privy to the wonders of duty-free stores, where I spend my hard-earned pay on Dunhill International cigarettes. These fine English tobacco products are perfectly rolled, inexpensive, and meet the requirements of my not-so-sociable habit.


But there are others who do not have the luxury of purchasing these cigarettes at duty-free prices. In England, where cigarettes cost upwards of $100 per carton, the population is forced to "roll their own." Although this habit died out somewhat in the United States in the late 1970s, there are those in England who still perform the miraculous act of purchasing a packet of Woodbine tobacco and their own roll-up papers, and manufacturing their own unfiltered cigarettes.


A friend of mine in England swears that this process results in a much more satisfying smoke and one that costs far less than paying $10 for a pack of 20 Dunhills. Although the intentions of such addicts may be dubious, the potential cost savings are enormous.

After smuggling five cartons of Dunhill cigarettes through customs on my recent visit to VISION 2008 in Stuttgart for my wayward twin brother, I began to ponder the situation further. However, it was not until I visited the largest machine-vision and image-processing tradeshow in the world that I realized the potential of the "roll your own" phenomenon.

Rather than purchase fine ready-made camera products from a host of vendors, certain companies at Stuttgart seemed to thinkthat offering developers the potential to develop their own camera products -- albeit with a little help -- was a much better idea. Indeed, this year's Vision Award prize winner, Supercomputing Systems (Zurich, Switzerland; http://www.scsvision.ch/), proposed a model for its leanXcam intelligent color camera based on the concept of open source computing for intelligent cameras that would offer developers a means to develop their own camera systems for very little cost.

Supercomputing Systems, however, was not alone in the idea of "roll your own" cameras. At the Kamiera booth (Hod Hasaron, Israel; http://www.kamiera.com/), Yuval Nahum, vice president of sales and marketing, was also making a pitch for the concept. As a spin-off of GigaLinx, the company plans to offer what it calls Open-Cam, a business approach that allows customers to cut costs and increase flexibility by manufacturing cameras on their own rather than purchasing them off-the-shelf.

All this talk of lowering costs and offering manufacturers a way to develop their own specialized camera products certainlycaused a stir, especially among established camera vendors offering high-performance Camera Link cameras. A number of thesecommented on the validity of this business model and wondered how such companies could profit from these offerings.

Others, however -- most notably Kerry Van Iseghem of Imaging Solutions Group (Rochester, NY, USA; http://www.isgchips.com/) -- were less skeptical. Although the company doesn't offer business models such as those from Supercomputing Systems and Kamiera, the company does tailor its cameras for the needs of specialized applications.

According to Van Iseghem, the level of interest in tailoring cameras for specific industrial, medical, and military applications at VISION was very high, possibly endorsing the "roll your own" model. What may prove to be an initially successful business model, however, may or may not play out in the long term. Although the concept of open systems is good, one must question how companies that offer this type of product can compete in the long term, not just on price alone.

By endorsing such approaches, camera customers must then be forced to consider the availability of OEM components and product life cycles that may need to be supported for a number of years.

Although off-the-shelf cameras may seem more expensive and not completely tailored to a specific purpose, the "roll your own" approach may, although initially less expensive, prove a little riskier.

No comments: