I have to think that HP could just throw $100m at the problem and build an embedded OS that's far better than Palm, without having to take on the assets of a long-since failed corporation.
By buying Palm HP knows that it's getting a modern mobile OS that's already in use on shipping products. For a company like HP a known quantity is better than spending money on R&D it may not be able to pull off.
Tell them what you need. Not interms of specifications, you as a big company are not good about it probably, everything will be too much formal and the resulting code will suck. Just tell your start up the final result, the user experience.
I think this stuff is doable, just can't understand why companies don't do it. I mean if you are going to acquire a company and this is a successful company in its market this makes sense. But with Palm what they are getting is more or less only the technology and the people.
Even with what could be considered infinite money, it takes time to build a Palm (and then there's the risk).
Think of it as $100m for a great mobile OS and $1.1billion for a time machine.
But you're right, HP is incapable of building anything that is competitive in the marketplace.
Maybe they're incapable of building a cellphone that doesn't make phone calls, or of putting out silly little web products, but HP is massive, and has built some incredible network and server hardware.