I wouldn't take that as accurate by any means, anyone with Windows 8 installed is an enthusiast, so it is a bit like asking how many atheists there are at Sunday School.
This is an Mojave type experiment, where they are Preaching to the Choir.
Wow, did you miss the point!
Please note that I specifically pointed out that the straw poll was non-scientific and that you could argue about the audience skewing the results. That WAS the point. Rather than arguing for or against the point, you chose to argue about the audience skewing the results. Brilliant!
Let me try and make the point (more of an observation) a little more clear. C9 is full of people who have installed Windows 8 and are obviously enthusiasts. Yet, if you ask how many of the C9 followers are "athiests", err, Windows 8 lovers, I suspect the results would be exactly opposite of what the WDS straw poll showed. That's why the results shocked me.
As @cbae pointed out, yes, the audience and the low numbers of the poll certainly means the results are going to be skewed and not represantative of the population at large, but the direction in which those results would be/should be skewed is very debatable. Based on my experiences here and elsewhere, I expected the skew to be in the other direction, as especially developers tend to be the most vocal and resistant to change.