Very cool. We had a similar experience when implementing the Simplex method for linear optimization, except we evaluated PyPy versus Julia instead of Cython. Its similar in that it benchmarks "just" the language, not BLAS, which is common pitfall I see when people compare the two. I guess at this point, a year or so later, it isn't surprising to me anymore that Julia is as fast as it is for these sorts of computational tasks - its been demonstrated pretty comprehensively at this point.
That's certainly not true. Take the matter at hand; if I have performance requirements that can be met using Julia, but not by using Python, I'll go with Julia over C because ease of use also matters, though it's not always the driving requirement.
Actually, performance evaluation is one of my first considerations for picking a language if I already know what kind of operating environment/performance I need to achieve.
I would argue that my personal choice is 1) familiarity with the language (ie, pick the most comfortable language for the task at hand) 2) performance required
Slides: http://www.mit.edu/~mlubin/informs2013.pdf
Paper: http://arxiv.org/abs/1312.1431