Everything else? Not so much. jQuery has become the kitchen sink. I use an awesome library called “when.js” for promises and lots of other specific-purpose tools even though jQuery has a promises implementation. I’ve written my own stuff for a lot of UI things like tooltips, underscore-type utility functions, and so on. I’ve basically got my own framework, pieced together from my own work and the work of others. It’s not that well organized, but I like my own choices or implementations of each feature better than the ones included in jQuery. I did some work to make a jquery plugin I wrote zepto compatible a while ago, and found it took only about one screenful of polyfills to cover all the non-core jQuery methods I was using. I just don’t use most of jQuery.

I think the important part of the evolution is adoption of standards for how our code interoperates like CommonJS. The rise of Node has also done a lot to pave a path for this, and if people start adopting design standards that allow for trouble-free plugging of modules, things will be awesome. But I don’t think that a big, universal framework is in the future (what jQuery has kind of become) because one size just doesn’t fit all when you have a hundred different possible platforms your code could be running on.

Though I do believe that frameworks will always serve as an extension to native behavior and as javascript gets more powerful, so too will the frameworks, I agree with you that the role of frameworks is shifting. People need to learn what’s natively supported.

That said, awesome read. Thanks.

(I’m going out on a limb with this counterargument, because HTML5 contains loads of things that most people who don’t understand semantic or multiplatform development won’t use, and will continue structuring their pages like it’s 2002, because right now that still works. So, one might say that the W3C made another misstep in building in a bunch of speculative features that will go the way of the marquee tag. I don’t agree with that idea. So! Counterargument!) HTML5 is *also* the product of most of the web dev community balking at the state of HTML4. We demanded a toolset that was better suited for what we hoped to accomplish. We got it. jQuery’s existence is another success story, as I understand it: a small set of developers responded to demand for change, and change happened.

We’re going to see this real soon now with media queries, I think: their format is unsustainable. It is hacky and awful. Writing them makes me cringe. Something else will be the media targeting hotness in a year or less. Dependence on whatever the toolchain is will not be the problem, because the tools are iterable — we can keep demanding (and contributing to the development of) better ones. Having to go back to our code that we wrote a year ago and decide whether to gut it or rewrite it will be the problem. I would argue very strenuously against duplicating and freezing library code out of a belief that it will make your future code perform as you like. Remaining vocal and influential on the tool-development community seems like a way more responsible practice.