Desperate? I'm not sure. The total addressable market here is every non-local computational need. That's trillions of dollars. For a market that is emerging r

Desperate? I'm not sure. The total addressable market here is every non-local computational need. That's trillions of dollars. For a market that is emerging right now, which already shows massive path dependency. So there's one time to win a slice. It can't be done later. If I was Google's position, I wouldn't be running a profit. I'd be throwing every spare copper into Cloud to grab market share. I'd be giving away credits on street corners. Driving out at least IBM and Oracle by just drowning the market with data centre capacity. Buying my way into existing relationships. Finding folks in consumer-facing divisions who've trashed the brand ("how long until this service is canceled, ha ha ha") and pulling them up sharply. Doing whatever can be done. I don't think Google really gets, at the highest levels, that this is the first business they have that can truly diversify them out of advertising. They seem to see it as a nice extra source of cash, rather than an existential imperative. But then, Google hasn't got the DNA to be afraid of existential threats. It's just not something they've dealt with since they switched on the cash pumps. Meanwhile, everyone else in this fight gets it.