- Web Assembly, might just peak.
import
might just be usable in<script></script>
- Universal JavaScript solutions will continue to rise that pay homage/respect to the days of server delivered front-ends (i.e. html to the client).
- Reactive programming continues to thrive in the JavaScript scene. (see MobX and RxJS).
- React, more so the concept, will dominate. React itself will be completely re-written (see React Fiber) or evolve (see Inferno).
- Angular found SEMVER so Angular 4 (even 5) is on the roadmap for 2017.
- A return to simple websites may happen, web 1.0 retro, but with the help of 2017 tools (i.e. static site generation)
- RESTful JSON APIs will get more competition (see GraphQL)
- Could be a banner year for Vue.js.
- More devs will abandon traditional CMS solutions for static site generators & API CMS tools aka Headless CMS's.
- More people will move from Sass to PostCSS + cssnext.
- Lots more HTTP2 and HTTPS.
- Web components will continue to lurk and wait for significant traction by developers that might never come to be.
- The no framework, framework, faction will gain momentum (see Svelte).
- JavaScript will settle, and hopefully, CSS will erupt and everyone will cry fatigue until it settles.
- Hatred for apps store will grow, while the open web has no memory of wrong doing.
- Redux will continue to get stiff competition (see mobx).
- YARN will win more users.
- The idea of “front-end apps”, “Thick Client apps”, “Static apps”, “No Backend app”, “SPA's”, “Front-end driven app” might get boiled down to the term/concept called "JAM Stack".
- While bower is maintained, it's recommend to use yarn and webpack for new front-end projects.