The problem I have with any of these types of rankings is always the basis, which is so nebulous that it never holds up under scrutiny. In this case, popularit

The problem I have with any of these types of rankings is always the basis, which is so nebulous that it never holds up under scrutiny. In this case, popularity is measured in terms of the number of programmers using a language, the number of training courses, and the number of third-party vendors. I use about 5 different languages on this list, and another 3-4 that aren't. Which group am I counted in? Training courses = popularity? Hmm, maybe. That also represents a sales angle, so it's skewed. As pointed out elsewhere, Javascript is incredibly popular but I don't find a lot of training courses available for it. It must not be popular! Third-party vendors? Guess it depends on what you're selling (again, sales skews the line.) Some languages don't formulate well to products being sold (think dynamic languages that can't be compiled/obscured). And lastly, what purpose does a popularity list not driven by job descriptions serve?