Microsoft vies for budget laptop market with XP price cuts (ars technica)

So?  The cost of providing the implied support that is provided with it is non-zero.
 
Nor is (r&d
cost)*(1/probability-of-success)*(1/number-of-units)*(inflation-factor)*(reward-for-risk-factor).
 
ie the required capital return for an investor to bankroll the development.
 
Its very difficult to make a business case for an investment in free software development that
isn't essentially monetising something already done by wrapping it up in fancy pants.

(If you know how to, please do let on!)
 
Try it, and bear in mind that providing services has a high cost of sales, a high service
provider cost, a delay in revenue (so there is a time decay in the effective value), and a
problem with competition from low-cost outfits who didn't foot any r&d bill.
 
Lets NOT sneer at pay-for-ware, until we actually have a coherent answer that doesn't see
software developed as a hobby, as a side effect of hardware sales, or as a P&R exercise by
insurance salesmen (sorry, 'enterprise support providers').