Every framework is a bet about what stays the same
A short piece on why frontend churn is not fashion.
The standard complaint about frontend is that it churns for the sake of it. I have never found that convincing, having watched several of the transitions from close enough to see what drove them.
A framework is a bet that some part of the platform will hold still long enough to build on. jQuery bet that the DOM APIs would stay incompatible. Backbone bet that the server would keep rendering. React bet that the expensive part was keeping a tree in sync with state. Every one of those was true when it was made.
The churn is what it looks like when a bet stops paying. The DOM APIs converged, so jQuery’s reason to exist evaporated — not because it was bad, because it won, and the platform absorbed the thing it was compensating for. That is the best possible outcome for a library and it still reads as obsolescence.
What this means practically is that the useful question about any framework is not whether it is good. It is which platform gap it is standing in, and whether that gap is closing. Nobody asks this, because it requires knowing the history, and the history is not what gets taught.