I’ve heard many people complain that Apple is charging too much (or that they shouldn’t be charging at all) for the “point releases” (10.2, 10.3, etc.). I completely sympathize with these complaints. The last two point-releases, 10.2 and 10.3, each costs $129 US ($179 Canadian).
However, I have to stick to the position that I’ve long since taken on software upgrades and pricing. I would love it Adobe would stop cramming crap new features into Photoshop and Illustrator. Rather, I would like them to spend a few months without adding any new features and just make everything better. Take the startup time down a few seconds, subtly refine the UI, fix bugs. I would pay.
This is what Apple has done with OS X 10.3. There are some new features, but the important improvements are subtle and all over the place. The end result is that the system just feels better. That’s worth the money for me. I wish more developers would focus on solidifying and simplifying rather adding more and more features.
My favourite part of the design process is the latter stages, even post-launch/release, when a slew of tiny improvements that seem independently insignificant add up for make the end result seem more mature and mysteriously better. For example, I would love to see the subtle changes that Matthew Haughey has suggested for the new A List Apart site implemented.
In the early stages of development of Mozilla Firebird there was a rule that each release (and there were frequent releases) had to be smaller (file size of the download) than the previous release. This forced the developers to keep simplicty and efficiency in mind and encouraged the optimization of existing features as much as adding new features.
I think where this gets lost is when the marketing department (curse them!) starts to get control over the feature list. I’m convinced that Microsoft Office changes its “skin” with each version just to look like something new and worth paying for.