I choose boring technology because it less likely to waste my time. If I choose exciting technology, I will inevitably run into a cryptic error, even if I follow "Getting Started" perfectly, and it will be either difficult or impractical to resolve properly. That happens to me all the time, and I imagine it's due to insufficient testing and developers relying too heavily on their local environment.

Better to work on something boring, but was carefully constructed over time and doesn't require 1/10 the dependencies of today's newfangled thing.