We take OpenOffice, we hire top committers to it, and we sell support packs to businesses at 10% of what they're paying for the other guys. We donate the templates and other code we create to the community. We make a ton of money.
Does this strike you as a tad naive? My theory about open source and the mass market is that there needs to be a compelling reason to switch to open source software *and* it has to be cheaper. It's simply not good enough to be either/or. Especially when we're talking about desktop software used by businesses - they'll pay top dollar for more features if they're over and above everything else that's available. What might actually work is if Dana employed his plan with a couple of changes:
1. include every feature that customers/users expect from the other stuff, *and*
2. charge less than they do.
Human beings can get very comfortable with an existing environment, and getting them to switch can be a pain, especially when you force them to make tradeoffs that don't necessarily work in their favor.
The primary problem with mass market open source software is that the perceived benefits of participating in a software development community doesn't really apply to the vast majority of people that use a computer. I think it's fun to participate in these communities, mind you, but I doubt a majority of people whose day job entails pecking around on a PC feel the same way. In fact, I would say this is the realm of a very small majority. If they get no joy from being a community participant, and they don't feel that all of the features are there, then they're not going to feel compelled to switch just because it's cheap. Look at OpenOffice - it's dog slow, it doesn't include all of the features of Microsoft Office, and oh sure, it's much cheaper, but in the grand scheme of things, paying around $200 less per person in an organization is basically peanuts. The OpenOffice experience is fundamentally broken and simply does not offer organizations a compelling reason to switch. Now if someone came along and made it at least faster and more responsive than Office... and fixed all of the buggy behavior... and kept it free. Well then, that might be a compelling reason for some people to switch, but there's still the matter of fewer features (as a sometime book editor, I can testify to the severe limitations of OpenOffice for this task).
Now let's contrast this with something like Mozilla - better user experience, no lack of features, and even then, it struggles to overcome the 10% user base barrier. This means that 10% of people have found a compelling reason to switch, and the rest have not.
Basically, what it boils down to is that in order to get mass market acceptance, you have to offer *more* than what already exists. At this point, "more" just doesn't apply to open source desktop software. Data center and enterprise software vendors can overcome this limitation because their customers enjoy the participatory process. Their users are geeks, and they thrive on the back-and-forth with ISV's that value community. Not to mention, the enterprise software market isn't necessarily about features - there's plenty of room for vendors that provide a solid platform on which to integrate other stuff in the data center. Not to mention, many IT managers and sysadmins follow a "less is more" principle, further diminishing any feature gap that may exist.
...and many people wonder why I detest Red Hat. Red Hat makes its living from the soft belly of old Unix users. Red Hat seems to think it doesn't have to worry about converting old Windows users and giving them a compelling reason to switch. Red Hat also seems to think that they can simply make do from the enterprise data center sale... but I would like to point out that this strategy has definite limits and does no good in terms of priming the pump of converted Windows users. In fact, Red Hat is not giving young Windows desktop users a compelling reason to switch. One could argue that Red Hat's potential market has hard limits and that, as long as they see no reason to push desktop usability, their potential market will not grow. Oh yeah, and guess what - OpenSolars ain't going away, and Microsoft's not stupid, either. It's going to be very interesting to watch Red Hat get eaten from both ends.