The people who buy 10.6 (assuming they don't give it away). Same as the people who pay for the intel side of development.
In terms of cost, the question is whether the amount of cash raised by selling copies to PPC owners is more than it costs to update the PPC version.
Point 1: Apple announced the Intel transition at WWDC 2005. Intel Macs started shipping in January 2006.
Point 2: Leopard support will be dropped at the earliest with the introduction of 10.7. If 10.6 is released in a year (lets say WWDC 2009-ish). that means 10.7 might be introduced sometime during 2010 or 2011 and 10.5 support dropped. That's if Apple doesn't decide to provide long-term support because of this exceptional circumstance.
Given those two points, the number of PPC users is going to decrease dramatically because some are going to be replacing their aging machines ANYWAY with new ones in what..the 5-6 years since the intel transition for the poor idiots who decided to buy a mac right then. Even more years for the people who bought older machines. That's a
MINIMUM of 5-6 years if you bought your mac at the worst time ever, even longer if you bought it before. (And if you bought it AFTER the intel transition was announced well then. I have no words for that. Caveat emptor. Really.)
So with a smaller and smaller user base, should Apple just decide to throw more money at work and people to keep supporting PPC? ffs, they won't even backport a lot of new things in Leopard to Tiger. Just supporting Tiger and Leopard for longer makes a lot more sense. I'd appreciate it if those engineers and that money went to R&D to improve the stuff I saw at WWDC this year, not to provide support for an architecture they switched away from back in 2006! I don't give a *&#$ about smaller installs and no old PPC stuff in Snow Leopard, really. I just don't want the time and effort spent on making all these things work on PPC when they could be spent working on, say, (insert every single thing discussed about Snow Leopard at WWDC) or (anythign about the new iPhone OS).
Not only that, but the other reason to switch to a new OS are your apps. Leopard and Snow Leopard in particular have a LOT of developer candy, even more than end-user candy (Time Machine? Spaces? psh, please, Core Animation alone is jawdroppingly awesome). I still can't get over some of the amazing stuff I saw at some of the WWDC sessions earlier today. And I'd love to use them, and it's easy for me because I don't have that many apps and legacy code to deal with. But for those people who have customers still on the likes of Panther?!?! I can't imagine what they would do if they could just do whatever they want. Prolly snow leopard only apps..even leopard itself..lots and lots and lots of improvements for developers.
All this complaining because either some people bought their PPC Mac at the wrong time or don't want to purchase a new Mac in 2010 or 2011? OR, they just don't want to keep using their old OSs even beyond then? I still can't believe the number of people I know who still use Jaguar or Panther, really. It's so silly I can't believe it. Half the arguments are specious and the other half are just bitter.
Let me just say now that I have nothing against PPC as an architecture. I still have PPC macs, but the wear and tear and the **** performance (compared to intel for most tasks) and the high cost of maintenance (new batteries, more memory, fixing issues..) is making me really reconsider keeping them and just replacing them with Intel Macs. Not only can I run alternative operating systems (a must-have for me), but also it's just a lot cheaper than keeping the old ones.
I guess my main point is that Apple can't support these forever. There's always going to be people bitching about it even if it's 10 years from now (why the HELL you'd want to be using a PPC mac 10 years from now is beyond me...) or if it was right now.
