Is picking a a "boring" for the sake of "boring" stack really all that sensible? How about we choose things because of tangible reasons, like test coverage, inf

Is picking a a "boring" for the sake of "boring" stack really all that sensible? How about we choose things because of tangible reasons, like test coverage, infrastructural/operational simplicity, availability of development resources, industry/enterprise adoption. IMHO these are the things that make sense when choosing a framework/language. Choosing something just because it's old and boring is illogical.