When you buy a piece of chess software, do you:
(a) believe you have just paid for what has been developed
(b) funding future development
(c) a combination of (a) and (b)

If like me you believe it to be (c) then what are your thoughts around when you believe you should pay and when you should not:
(a) a user should only pay for every 'Major' release {e.g., Houdini 2, Komodo 4, etc.}
(b) a user should not pay for every 'Minor' and 'Fix' release within a 'Major' release series {e.g., Houdini 2.0c, Houdini 2.1, Komodo 4.1, etc.}
(c) a user should only not pay for every 'Fix' release within a 'Major' release series

Again, I believe it should be (c) above however I also expect the following:
(a) an engine can only go from one 'Major' release series to another 'Major' release series when some significant progress has been made adding new features/functionality
(b) an engine can only be re-released under a 'Minor' release number {e.g., Komodo 4 to say Komodo 4.1} when significant modifications to features/functions are made

A 'User' of course provides the final 'Value for Money' test and mine is certainly based on the above - I believe those that provide free chess software have made this 'Value for Money' test even more difficult for those that charge and wonder if the commercial model for chess software suppliers will need to shift where a 'User' only pays for a 'Major' release and expects to receive FOC all other releases in the series.  I also believe that those commercial suppliers who charge for their software will need to provide extraordinary things; the diminishing returns from dissatisfied 'Users' may become a problem they cannot solve without a 'radical' change in how they currently operate.

What do others think/expect from their chess software suppliers?

NB: where I mention features/functionality above then I see chess performance forming part of this and for those of you who measure performance by ELO then significant to me would be in the 100 ELO region.