Are the Intel chips better? Probably not. Is there going to be some radical improvement? I doubt that too. What there is, however, is a low-power, low-heat mobile chip that has a more viable roadmap than the G4, and is readily available. I can't speak to pricing on the chips, but I'd be hard pressed to believe that IBM isn't charging a premium on the G5s - they can't make enough to keep up with demand currently (ever wonder why there are regular delays on the G5s?). And evidently there's little to suggest that IBM had much incentive to agressively pursue R+D on a mobile variation, given that Apple is such a small piece of their chip volume...and about to get smaller now that the game consoles are all moving to PPC. And Apple needs to see some movement on their portable line. (Note that this is also at least one reason for no AMD - they don't have a viable portable chip that compares to the Pentium-M...and overall power consumption and heat on the AMDs is higher, IIRC).
It's nice to see confirmation from Mr. McNally that the transition to fat binaries looks to be relatively painless (thank god for abstraction layers, eh?), though I expect that the transition will not be without its hiccups. However, unlike the 68k to PPC transition, where the OS continued to run legacy code in emulation, I suspect that the OS will be fully optimized to run on both processors.
As many have pointed out, a G5 (or a G4 laptop) will not suddenly stop working when the Intel-based Macs start rolling out. Nor will the OS (or applications, for that matter) stop working. Heck, I'd be surprised if any G4-based machine, and probably most of the modern G3 machines, weren't supported to some degree well into the future. (I say this as someone who runs 10.3 on a 5-year-old iMac DV SE at home...).
And I'd be surprised if the Intel line that Apple's planning on using isn't 64-bit, so I suspect that the optimizations to take advantage of the 64-bit processor will still occur, since as I understand it, it'd have to happen either way.
I've been planning to do a refresh of my creative department's machines. Did this announcement cause me to rethink my plans? Yes. Am I still going ahead with the refresh? Most likely, yes. Is there a risk involved? Perhaps, though I think it's negligible; the machines would be replaced in the product line in a year no matter what the hardware, so that's not a big concern. And I need the machines now. The cost to me to replce these machines with Windows boxes (in terms of increased maintenance costs, user re-training, software investment) is much higher than the cost of replacing the machines in 3 years once the Apple line is fully migrated to Intel.
What I am constantly amazed by is the "wait and see" attitude that so many in the Mac world take when it comes to hardware ("I need a new machine, but I'm going to wait and see what gets announced at Macworld/WWDC." "Oh, there's new product in the pipeline? I'm going to wait and see what shakes out..."). If you need a new computer, buy one. If you don't, then don't. You can be assured, that Apple will NOT be abandoning support for the current products anytime soon.
Remember folks - This is just a change in processor. That's it. This is closer in nature to the shift from G3-G4, or G4-G5, than it is from 68K to PPC.
(Omitted here is discussion of the fact that the real loss in market share for Apple during the 68K to PPC transition was with the release of the 2nd generation PPC hardware, a bloated and convoluted product line, shortages and high costs on the products that consumers wanted, hundreds of thousands of underpowered overpriced "consumer" Macs sitting in warehouses - something I saw first hand working for an Apple contractor at the time - and a general lack of focus, direction, and innovation. Oh, and the licensed clones, which cannibalized existing sales and did nothing to generate new interest in the platform. There was a lot more going on at Apple, and in Apple's product line, than just a processor transition.)
If Apple's hardware line dies because of this transition (and I don't think it will) - it'll be because we Mac users decided to kill it by saying "screw you Apple, I'm not going to buy your hardware any more."
Apple Computer: Software Only in Five Years