[mrweasel]

There's certainly value in frameworks, but like with so many things we build today, most of them are build for a scale that most people don't need.

I'm sure React is great, Angular as well, and VueJS, but they have grown to be all encompassing, so a developer is never stuck and need to rewrite everything. Many frameworks starts up being small, so people can understand them and easily learn. Then as time goes by the frameworks grown until they become to big and someone once again feel the need for a lighter framework and the cycle repeats.

Currently I'm trying to build a VERY small web app. It does need a tiny bit of JavaScript to pull in new data every minute or so. It's much much easier to just forget about using a framework. It would take more time learning how you start a React or Angular project, than just learning the bits of modern JavaScript I need. Then I also don't need to bother with npm, webpack or any other packaging and build non-sense a framework would try to impose on me. That save even more time.

The article is correct that something like React will save developer time, if the project is large enough. If not, simply figurering out which framework to pick and learn how to set it up is going to constitute the majority of the time spend on the project.

[incrudible]

> It would take more time learning how you start a React or Angular project…

That time has has already been spent, because you hired React developers. They will be fastest by pulling in React. Really, the argument could end here. Yes, there is overhead in using it, things can get out of hand, but they do not have to. If they do get out of hand, would vanilla web technology have prevented that?

[ragnese]

> That time has has already been spent, because you hired React developers. They will be fastest by pulling in React.

I'm not entirely convinced that this is true, either. At least not in some general, a priori, way.

Doing stuff "the right way" with React or any other framework probably still involves a fair amount of ceremony, boilerplate, and testing. Would your hypothetical React dev be faster with React than some other framework? Probably--if that other framework is similar in scope and complexity. But, otherwise, I just wouldn't assume that an expert in FrameworkX will always do better using FrameworkX without regard to the specifics of the project.

[incrudible]

It is true in context. They will be faster than if they did not use a framework for anything that has even a tiny amount of repetition or interaction. That is why we have frameworks in the first place. There is no right way. React itself is not much of a framework anyway. It is a fairly small library for reactive UI. You do not need to be an expert to be productive in it.