The 2020 Ember Roadmap

The aim of the Ember Roadmap course of is to rally the neighborhood round plenty of shared objectives. This publish paperwork these objectives for 2020.

Because the Ember neighborhood can not predict the longer term, we can not make certain that we’ll obtain the entire particular person objects enumerated right here. As an alternative, the aim of this doc is to offer the neighborhood a standard goal to aspire in direction of.

This yr our two headline priorities are:

• Polish the sensible and conceptual particulars of Octane (tracked properties, Glimmer elements, associated tooling, accessibility, efficiency and payload enhancements).
• Make Ember simpler to attempt to undertake, but in addition decrease boundaries for Ember builders when collaborating with the higher JavaScript mission. We’ll do that by means of enhancements and simplifications to the framework, and thru centered communication with the higher JavaScript neighborhood.

This doc was began in fall 2019. It’s a distillation of a number of sources:

1. The 2019 Neighborhood Survey.
2. Neighborhood #EmberJS2019 weblog posts authored in response to our name for posts.
3. Dialogue on https://focus on.emberjs.com, Discord, Twitter, and different public venues (see the neighborhood web page for how one can entry these communnities).
4. Deliberations among the many Ember core groups.
5. Your feedback and suggestions on this doc through the RFC course of.

The objective of the roadmap is to align the Ember neighborhood round a set of shared, achievable objectives that stability the wants of current customers with the necessity to develop and assist new customers.

Our Roadmap

Now that Ember Octane has shipped, it’s time to show our consideration to new efforts in 2020. Our objective is to construct on Octane’s launch and capitalize on that cutting-edge basis.

• Put money into Octane. Octane’s psychological mannequin and primary elements are full, however plenty of sensible and conceptual gaps stay. We’ll shut these gaps with work on tooling, by deprecating basic APIs to simplify Ember for brand new customers, and by introducing extra performance the place applicable.
• Modernize our construct system. This yr we’ll prioritize enhancements to the Ember utility construct pipeline, and to Ember itself, which is able to carry trendy optimizations like tree shaking and code splitting to each new functions and current codebases.
• Higher a11y by default. We’ll higher assist assistive applied sciences through updates to the router. Moreover we’ll present builders extra instruments for understanding and bettering the accessibility of their Ember functions. Our objective is a superb “out of the field” expertise with Ember and assistive applied sciences.
• Share Octane outdoors our neighborhood. Octane’s launch put Ember in entrance of a number of new eyes. We’ll proceed that development by means of 2020 by speaking concerning the newest version of Ember in entrance of latest audiences.

Put money into Octane

Ember Octane put the framework on a powerful footing by modernizing its most foundational APIs. Groups are already productive utilizing Octane, and from their expertise have offered a torrent of real-world suggestions. We’ll proceed bettering the developer expertise (DX) of Octane all through 2020.

Most of the tough edges in Octane aren’t on the options themselves, however within the supporting tooling. The usefulness of stack traces from Glimmer, the power to make use of TypeScript with Ember templates, how tracked properties and Glimmer elements are mirrored within the Ember Inspector, and the construct pace of our utility pipelines are all essential components of Octane’s DX. We’ll spend money on these areas of labor.

For builders who’re new to Ember, the presence of basic non-Octane APIs will be disorienting. We’ll search for inventive options that make these options trivial for current apps to proceed utilizing whereas additionally making them cheaper (in payload, efficiency, and psychological mannequin) for brand new adopters.

Lastly there are some areas of Octane options which might nonetheless profit from new characteristic work:

• The @tracked system, for instance, limits the expression of state in an utility to outlined properties on an object. Actual-world codebases typically need to preserve state as a listing or a map, and we will lengthen on the well-designed internals of Ember’s reactivity mannequin to assist these circumstances.
• Modifiers present a hook into the DOM rendering lifecycle, however Octane has no APIs for hooking into different lifecycles within the rendering and object system. We’ll create these APIs.
• We’ll proceed a push to make Ember templates higher analyzable at construct time by introducing a strict-mode template and static imports.
• We’ll make it simpler to construct ergonomic, reusable elements by delivery named blocks.

We’ll introduce new options in Ember which enhance Octane in these and different areas.

Modernize our construct system

Final yr, we began work on Embroider, an overhaul of the Ember CLI compilation pipeline. This yr, we’ll put the ending touches on Embroider and begin migrating the Ember ecosystem to this modernized construct.

Embroider integrates Ember CLI with fashionable packagers like webpack and rollup. It permits Ember apps to trivially import from any dependency revealed as customary JavaScript modules, and can unblock delivery Ember itself as npm packages within the @ember namespace.

This new method, by means of its basis on frequent packaging libraries, can even unlock new construct time optimizations. These optimizations, like tree-shaking and route-based code-splitting, will permit Embroider to supply smaller asset payloads.

Moreover, we’ll introduce a system into Ember which permits apps to drop framework code supporting deprecated options unused by an app. This can lead to smaller vendor property for functions which do not depend on deprecated options. For instance, a contemporary Octane utility might not require Ember.Element, and might profit from having the code supporting that API being dropped at construct time.

Lastly, we’ll ensure that this modernization effort supplies advantages to current functions. If a crew has been steadily upgrading their app for years now, they will not must rewrite it to get the advantages of a contemporary construct packager.

Higher a11y by default

Ember functions ought to be accessible to everybody. Sadly, even seemingly small errors could make your app troublesome or unimaginable to make use of with assistive expertise like display readers. We’ll do extra to enhance the out-of-the-box accessibility of Ember functions, and supply instruments to assist functions keep accessible as they develop.

• Repair router accessibility in order that web page navigation is accurately introduced by display readers, while not having a third-party addon.
• Incorporate accessibility checks into the built-in check helpers.
• Interact with requirements our bodies to assist fill the gaps in current internet accessibility APIs.

To contribute to this effort see RFC Concern 595 which coordinates the Ember A11y Strike Group.

Sharing Octane outdoors our neighborhood

There are extra individuals constructing internet functions than ever, and Ember should adapt to their altering wants and expectations to be able to keep related. Octane higher aligns Ember’s API with what new customers count on from a contemporary framework. We have to make the most of that change.

This yr, we’ll share with the world how Ember Octane is trendy, productive, and enjoyable. Via weblog posts, movies, social media, meetups, and conferences, we’ll share our information and experiences with the broader JavaScript neighborhood and encourage them to offer Octane a attempt.

We’ll proceed to make Octane extra enticing to new customers with a brand new documentation method, simpler web site, and with clearer communication concerning the Glimmer.js mission.

Reducing boundaries to adoption, and collaboration

Making Ember extra enticing to new customers doesn’t suggest compromising on what has made the framework so very profitable for current codebases and groups.

Probably the most primary worth of the Ember mission is that we remedy issues collectively. Whereas we intend to develop the variety of framework customers and modernize the framework in some ways, we cannot optimize for development on the expense of our current neighborhood. As an alternative, we’ll collaborate on options that include a curated adoption story.

As a result of we perceive and worth the facility of collaboration, we all know we should decrease boundaries, technical and non-technical, which not solely make new customers hesitate to undertake Ember but in addition discourage Ember builders from collaborating the higher JavaScript neighborhood.

An ideal instance of our progress in that is Octane’s embrace of native JavaScript lessons: A JavaScript developer beginning out with Ember immediately is not instantly compelled to be taught a brand new, Ember-specific class API earlier than they will get to writing code. However, a developer who begins off with Ember may also contribute to most OSS JavaScript initiatives while not having to first un-learn the Ember class system.

By sharing frequent options to frequent issues with different communities we not solely make Ember extra approachable, we additionally profit from the chance to alternate extra concepts. Everybody wins.

With our efforts to flesh out Octane, enhance the construct system and align it with the remainder of the JavaScript neighborhood, increase the baseline assist for accessible functions, and to raised talk in 2020 we’ll decrease boundaries to adoption of Ember, but in addition to our personal collaboration with the higher JavaScript mission.

Wish to become involved? Go to our neighborhood web page
and be happy to hitch us on Discord (chat) or Discourse (discussion board dialogue).