I actually agree with dgs - most users are not bothered about the underlying features of the operating system. After all, many people stuck with MSDOS and crude versions of Windows for years when much more advanced options were available.
I venture to suggest that the things that matter are
1. Will it do what I want?
2. Can I afford it?
3. Will I be able to use it?
4. Can I get support when things get sticky?
5. Is it compatible with my work computer/friends etc. computers
Oh, and possibly "Is it easy to find and available to buy locally"?
"Does it look pretty?", and "How easy is software development?" are a lot further down the list for most users. I suggest that things like "what sort of multitasking does it have", and "does it have proper drag and drop" are even lower down the list.
The problems we have to address to attract more people to RISC OS are I believe primarily the five I list above. I think Windows has an advantage over RISC OS for most people in all except Item 3, and even with that one we have the problem that people will prefer something familiar, that maybe they have used at work, to something they know little about, even though it is claimed to be easier to use.
Item 1 is indirectly related to ease of software development of course, since a good development environment encourages the writing od new software for the platform. That is one that Microsoft got right!
Item 2, cost, appears to be another Windows win (whether it really is is another matter, but that's how it is perceived). And with the advantage that the size of the market gives them, Windows systems jolly well ought to be cheaper and better value for money.
And no matter how good the support is from suppliers in the RISC OS market, and in my experience it is usually very good, and from other users via the internet, if I have a problem with Windows I can ask any number of friends and colleagues, and there are any number of local computer shops, all of which may be able to help me. Most of these people will have never heard of RISC OS, so for that kind of support a RISC OS user is on his own.
I suggest that we will therefore not attract mass market users whilst this situation is as it is. RISC OS needs to attack these issues first, IMO, if we are to gain more market penetration.
Now, time for a bit of self-examination. Why do some of us prefer RISC OS despite these disadvantages? It must be that we have a different set of priorities. Anybody care to try to list them? I think for me it comes down to things like (in no particular order)
1. A love of innovation and elegant technology
2. Dislike of Microsoft/big business/monoplies
3. Grew up with Acorn, like RISC OS, (possibly tried Windows and didn't like it).
4. A desire to be different.
Not things that will concern most people who just want a computer to use I think!