Most of us know about the "Windows Tax" -- the extra cash you shell out to pay for the cost of a Windows license when you buy a new PC. But what about a (so-called) "Linux Tax," the cost incurred by an ordinary user switching to Linux from Windows?
The concept: An average user's switch from, say, Windows to Linux will involve some kind of cost, whether that cost is in the form of time, effort, money (i.e., paying an expert, buying a commercial-grade Linux distro), or what have you. This might also include the effort involved to find a reputable manufacturer who'll sell you a machine with Linux preloaded, although that's become far easier as of late. Dell and IBM come to mind, of course, although there has been a slew of smaller manufacturers who have done this for a long time -- just not as visibly.
The word "tax" -- with all of its negative connotations -- comes into play because it's assumed that the switch will in fact come at a cost. This is generally true if you're an entrenched Windows user, but far less so for people who are not specifically Windows- or even PC-centric. Every year thousands of people sit down at a computer for the first time ever, and do so with no preconceptions about how a computer is supposed to behave. For them, Windows, Linux, and Macintosh are all on theoretically equal footing; the only tax for them is the time they're willing to invest in learning how to use a computer, period.
This brings us to the other big assertion: that Linux is markedly more complex than the other operating systems, and therefore requires that much more effort to use. That's become far more debatable than ever, and not just because Linux GUIs are that much more polished than they used to be -- although that's probably one of the biggest reasons. The whole concept of a flat "Linux tax" only works if Linux presents everyone with that much more of a challenge.
So is there a Linux tax after all? For people switching away from proprietary software, yes -- since they have to perform discovery, so to speak, to find out what they can and can't do. Sometimes the losses are nothing they needed anyway, like the lack of protected HD content support on Linux. But the process of finding out, the investment of time and effort -- that's effort in their eyes that might be better spent actually getting work done. If they see a switch to Linux or a switch to open source as an extended process of re-educating themselves, they'll be that much less likely to ever do it. It's perceptual, but perceptions mean a lot. The lower the bar for everyone, expert and amateur, business user and code wizard alike, the better.