First it was "the desktop." Now it's "the netbook" -- as in, what's the big proving ground for Linux vs. Windows going to be? And the latest hotly-debated bit of conventional wisdom is whether the Linux-based netbooks just don't cut it compared to their Windows cousins. The real problem seems to be who's willing to do more to bring regular users in.

Keir Thomas at ComputerWorld has his own take on the whole issue, in which he asserts that ordinary users run into too many little barriers with Linux-based machines to really stick with them.

What happens is that the software problems presented by Linux, combined with the hardware problems presented by smaller computers, push users over the edge. Pretty soon they've just had enough. They return their netbooks, and write off the concept as a bad idea.

If Windows is installed on their netbook, the user has a far easier ride. This isn't down to superior software. Far from it. This is quite simply because the software side of things is more familiar. They (or somebody they know) will be able to fix up the software side of things in a jiffy with just a few downloads. They just haven't got to worry about that side of things.

The reason they don't worry is because for a long time, Windows was the PC. If your PC had a problem, odds were it was a Windows problem, and so a whole culture of troubleshooting sprung up over years and then decades. Even most non-technical users now still can chant the "Did you defrag/reboot/clean the Registry?" mantra, because they've spent so much time hearing it from the very experts who came to their rescue.

Linux doesn't have this advantage. You aren't steeped in the Linux culture of how to fix things unless you've been running it yourself -- and on top of that, there's the fact that each Linux distribution often presents its own specific problems. Linux may represent choice -- and choices are great! -- but a choice that is in effect no valid choice at all (for all of the above-mentioned reasons and more) isn't one worth making, or even offering.

There is a solution to all of this, and it involves a degree of humility. The people making Linux distributions -- and open source in general, really -- have to be willing to go to the very non-technical community that they claim to want to win over. You have to go to them, not wait for them to come to you. Don't just put the software somewhere and expect people to come to them, but grab people from the street and perform usage testing and run focus groups and do all of the things that, yes, Microsoft spent tons of money and time on.

What we need is an open source usability foundation -- an organization that can be hired or enlisted by an outfit that produces an open source product, and have that product rigorously tested against the very people they want to have using it. This would not be free -- this kind of effort costs money, there's no way to get around that -- but it would mean that many less experiences where designers second-guess, imitate (badly) or otherwise trip over their own feet when creating something that's meant to be used easily.

Who's up for that job?

Learn more about all the latest products and technologies at TechWeb's Interop Las Vegas, May 17-21. Join us (registration required).

Follow me and the rest of InformationWeek on Twitter.