The "modern" way seems to require tools that are jack-of-all-trades rather than be the best they can be at one thing.
But this isn't the philosophy behind a video capable DSLR. The philosophy is "Since we have liveview for our DSLR still shot cameras, we can add value to our products by simply capturing that liveview output and allow users to video thru their current investment in glass." The whole thing is a byproduct of liveview for stills, which was designed in an effort to improve stills. It just so happens that the byproduct turned out to be very good.
As for me (enthusiast, not pro), I'll be buying both videocams and still cams for the forseeable future because they are still way too different. I'm not giving up the anti-shake of the CX550 nor the stills IQ of a 50d and will only want to improve on both. But I'm very pleased with the idea of being able to do both functions with either device (well at least when I get a 5d3
) so I can choose which function is "primary" for the day without giving up the other entirely or resorting to toting both devices. Bottom line is I'm all for improving the stills on a videocam and the video on a still cam and this doesn't mean I want a "jack of all trades". It only means I find value in the secondary function when it is good quality.