Bob Rice has had many careers. He was an attorney with the U.S. Department of Justice, a partner at law firm Milbank Tweed Hadley & McCloy, C.E.O. of a tech startup, and now runs merchant bank Tangent Capital, which he founded in 2005.

Today, he's squeezing in some blogging. One day. One place: Portfolio.com.

Google is our version of the original Sears catalog. These two revolutions in advertising and retailing were driven by exactly the same precept: a breakthrough idea that gave consumers easy access to goods, price transparency, and lower costs.

Back in the day, Mr. Sears, who worked as a train-station agent out west, observed that farmers (who then made up a huge chunk of the retail market) were forced to purchase all their supplies at the nearest general store. This not only limited the menu; it put the price at the whim (and negotiating power) of the local merchant...who, in fact, usually had a very high cost of goods because he had obtained them through a long chain of resellers.

The Sears mail-order catalog provided an incredibly convenient way to find what you wanted and get it at a price you'd know in advance. Of course, people first had to learn to trust the company, since they couldn't physically handle the goods before buying, a nice echo of the early hurdles in e-commerce. Learn to trust they did, paving the way to Sears' position as the world's dominant retailer.

Of course, all good things eventually come to an end, so there's no need to cry for Sears. But in our "red-shifted" business environment, the end of your heyday comes a lot faster than it used to. And, as I see it, the summer solstice has passed not just for us, but for Google too. Here are three reasons why:

1. Google will certainly remain the dominant general search engine for a good while, maybe many years. The recent, rather pathetic, unveiling of Cuil just underscores the point: "Google Killers" are very unlikely to succeed.

So, what's the problem? Specialization. Über search engines, like the general Sears catalog, are inefficient if you know what you want. They bring up (and will always bring up) a huge amount of noise in response to a request. The future of search will be specific engines that cover particular categories, so that the nuances of the request can be better understood, more relevant search criteria can be requested of the user, and the response database of sites can be tuned to the information sought. Niche engines will spring up to serve the most valuable parts of the search world (for example, Kayak for airplane flights), limiting Google's growth in search to the noncommercial, nonpaying segments of the market. Remember, general catalogs have lost out to specialty catalogs.

2. "The Dream" will have limited success. Folks tend to get a little too excited about Google's technology expertise, and seem to think it'll automatically translate into the world of cell-phone operating systems. Not so fast! There's an enormous difference between cataloging the internet and not dropping 3G connections between cell towers. Moreover, just making an OS doesn't do much by itself (there are scores of them out there). By contrast, the iPhone combines a great OS, beautiful design, the content the user wants (music, podcasts, etc.), and, most importantly, the ability to iterate all these elements very quickly and seamlessly. Every "standard by committee" I can remember has either failed completely, or operated at a crawl to develop new features and address problems; and that's what Google will essentially have when its putting out the OS and several other companies are making the hardware and/or providing the content.

So even though Google has the right idea in developing an open OS and inviting developers in (as Apple already has done), a lack of expertise, product integration capabilities, and decisionmaking control will be extremely difficult to overcome. Don't bet on this Dream.

3. Management. Hey, M.I.T. frat parties are a blast; they just tend to be hard to control. And that's sorta what we have here: a vast pool of talent having a lot of fun doing just about whatever they want. I've got nothing against "Do No Harm," but I suspect it's not quite enough of a coherent idea to keep people on the same page for long. And eventually, that will be an enormous problem.

The company has grown so quickly that it has had no time to develop intelligent internal management structures. Stories of new hires being given a cube without anything in particular to do for weeks are common. The culture is exciting and attractive for talent, but there is very little sense of developing priorities around potential profitability, or even of a need to execute on ideas with which one disagrees.

Call me old-fashioned, but it's hard to believe that this sort of culture will be able to buckle down when the inevitable problems arise (such as the ones mentioned above). I love randomness and unpredictability--it's fun. But it's hard to invest in.