There's a very good discussion going on about the relevance of patents on Guy Kawasaki's blog. This is a very important issue for any business that more or less relies on patenting to errect entry barriers and beat the competition.
I don't think there is a universal answer: each business is unique and I stronglya dvise my customers to have a real debate about the energy and money they want to spend in patenting something. It's quite an investment both in terms of going through national and international procedures, often involving translations in exotic languages, but also in terms of actually defending the patent once you have it. So there's a kind of balance to strike between the necessity of protecting intellectual property and the business contraints that may make protection itself irrelevant.
In Europe, for a long time and perhaps that is still the case, patenting was a sort of magic key to gaining support of governmental agencies in charge of innovation and stimulation of entrepreneurship. Universities are also extremely active in patenting their work and they are increasingly setting up organizations to commercially exploit their IP. So that creates an environment that is very much in favour of patenting and that may not be the best idea in the world.
My point is that patenting is neither something companies must do to survive, nor an unnecessary weight that can just be dropped without thought about consequences. It is neither "good" nor "bad" in itself. It all boils down to strategy and objectives: patenting is the means, not the end.