This is the most sensible commentary I've seen yet. Fits the "America is pretty much screwed" articles making the rounds these days. It's actually pretty damned depressing that $20 billion dollars was spent on a legal game rather than awesome R&D.

The money's not gone, it's mostly just transferred from one party to another -- in this case, from Google to Motorola's shareholders. Of course, many man-hours are wasted in the legal wrangles, but it certainly doesn't add up to $20 billion dollars of lost productivity.

You can say that about practically any transfer payment. The cost of a transfer payment is opportunity cost. If Google invested 20 billion dollars in R&D, we'd have 20 billion dollars of R&D done in addition to someone, somewhere, having that 20 billion dollars to spend again.

Google bought a lot more than just patents.

The worst thing is that Google is now invested in the problem by $20 billion more than it was previously. The more invested U.S. companies become in patents, the more difficult it becomes to make change.

$20 billion? Did I miss a vital piece of news?

$12.5 billions for Moto, $4.5 billions for Nortel, that's actually more like 17 billions.

The other big non sequitur in this argument is that Apple and Microsoft claim that Android steals their own hard-won inventions but they're buying up other companies' IP to bludgeon Android with.

That implicitly assumes these patents are mostly forky-spoony stuff. What if they represent real innovation? In that case the system is working exactly as intended; investors are getting a return on R&D.

Whenever these ridiculously huge acquisitions take place, it's always the same. How can Google possibly think that 20 billion couldn't have created something far better than anything Motorola has to offer. They have the talent, the infrastructure, the money, the manpower.

The Skype deal felt like just the same...

I think speed is often the issue with in-house development. Sure, you could make a pretty good product by spending a few billion dollars, but you won't get it next week. An already established install count has to be worth something too.