ZDNet has an article titled Linux Ready for the Desktop.
While I do enjoy Ubuntu quite a bit (at least on one desktop) and I would never doubt the usefulness of Linux on a wide variety of machines, my concerns over open source systems have become more widespread with issues about forks.
Linux is certainly a different scenario than MySQL but consider this - mySQL has been considered a viable alternative for commercial RDBMS' for how long? It's been around since 1994 but has only really become a force to reckon with in the past 10 years.
Certain incarnations of Linux are certainly ready for the desktop (OpenOffice is arguably a good alternative to MS Office - bloat and all) - but then so are an entirely cloud or web-based OSes.
My #1 concern with any OS that I'm going to place on my desktop is "does it make me more productive?" (an offshoot of that would be "does it make any of my clients more productive?")
I'm a firm believer that people will pay for things that make them more productive. I believe Vista has shown that people are NOT willing to pay for upgrades that don't. The security argument doesn't completely wash with me because too many people gun for Microsoft since it is the leader - no OS is impervious to attack (except the one that isn't connected to anything).
Now I've been playing with the Windows 7 beta and so far, it isn't bad - but it still annoys me a bit more than Windows XP but far less than Vista. I've been living in a Visual Studio 2008 project with third party controls and while some of the results are nice, I still rely on Visual FoxPro to do a great deal of the background work.
On the same project, I work with SQL 2005 but I also have SQL 2008 on my office machines - I can safely say SQL 2008 is far better than SQL 2005 in making me more productive. Yet I still use VFP to write automated SQL scripts - it's simply better at text handling than any other tool.
Some people love the Windows Explorer - others find alternatives. The only real way any new operating system is going to make it is by making people more productive, not by a little but by a lot.
The iPhone isn't a great phone - but it does open up the entire world of smart phones to consumers in a way that Windows Mobile simply hasn't. Both phones have features that the others don't - but cell phones are a consumer product - they rely not just on impression and productivity but also "coolness".
Desktop operating systems do not rely on the coolness factor - they are there to get a job done. I don't want an operating system that simply replaces another - I want one that makes me do more with less.