It sometimes takes me a while to wake up to a trend -- it's probably a "forest for the trees" thing -- and it really only occurred to me this last week when I was scanning a few hundred press releases looking for a particular one that I wanted. Maybe six companies, all relatively new, that I had started to track had suddenly disappeared from my radar. Not all at once, but over a period of a couple of months.

The strange thing (or it was strange until I woke up) was that all of these companies had been really heavy on their public relations work, which for a working editor means being plagued by repeated coverage requests from PR agencies. Most of those requests are very polite. But, uniquely, I had not run any stories about their products, although a couple of my colleagues had. No stories because the products did not met my criteria: "real" products that will make the company money. "Real" to me is that the product is at least sampling, it is priced and there is a data sheet for it.

What has happened to those companies? They have been bought out. And it got me thinking about the logic of starting an analog or mixed-signal semiconductor business in the year 2000. It is generally accepted -- and I believe it -- that beginning a new manufacturing company for analog is not affordable and probably not financially backable, so all the new companies follow the fabless model: a model that is fatally flawed, of course, as independent fab capability dries up.

So why would you even start a fabless company in the year 2000? Certainly investors seem not to have yet twigged on to the dicey nature of the model, and money is available to marketers and designers with a track record . . . but what game do you envisage when you start such a company? Maybe, as has happened, you look to one of the established companies for seed money and some fab capability -- or at least the available pressure that can be brought to bear on an outside fab. In that form you are probably expecting to make a product that will be marvelously successful and your seed investor will absorb you as a hero. Maybe, as has also happened, you so utterly believe in yourselves that you intend to go to IPO.

Or maybe you don't have any intention of staying in business.

When a company sets itself up to design the next "killer" product and then samples it with no real sales and marketing team, what does that suggest to you? To me it means they are selling themselves. Deliberately creating a very attractive sticky flypaper for a corporate insect to come and adhere to.

So, if you are a really good designer and feel you're not getting to be a millionaire fast enough, this is what you do. Think of a really creative product, with really creative architecture that has lots and lots of hooks built-in (forget the market for now, that's almost irrelevant), find a good shill (one comfortable in a suit), and set up shop. Design and make sure that those press releases start hitting the snail- and e-mail-boxes and fax machines of all the editors your agency can find, as often as you can. Wait -- not too long -- and they will come to you.

The going rate to get you off the street seems to be averaging about $150 million. They'll make you sign a contract so you'll still have to work on this product for your new bosses, but you can get it trimmed back to something realistic, work the hours you want, drive to work in whatever you like, and you really will laugh all the way to the bank.

Did I say I liked this trend? I don't blame the individuals for a quick trip to Nirvana in exchange for their skills; I don't like time-wasting PR work for such a cause; but I do think it is incredibly funny to see established manufacturers so thrown into a shopping frenzy by the possibility of missing something.

Analog Main | Product of the Week | Columns | Editorial | Tech Notes

		Click here to get your listing up.