The time is right for a DOM templating API

Summary

TL;DR: I want to propose adding a declarative templating API to the web platform. Here's why...The web platform is the most successful application runtime of all time. While the largest reason for this is the web's reach, it wouldn't be possible without the DOM API, which turns a mostly static document viewer into a highly dynamic and expressive runtime.For as much hate as the DOM sometimes receives (some of that deserved, but some really not!) the DOM is undeniably a very powerful API. This is easily shown by all the amazing and highly dynamic web applications out there - even an app as complex as Photoshop is available as a web app, and its entire UI is built with web components!Yet the DOM is missing one of the most important and popular features in modern web development: templating. Currently there is no ergonomic way in the standard DOM APIs to create a chunk of DOM nodes, from data, event listeners added, properties set on elements, safe against XSS attacks; and then to efficiently update that chunk with new data.This kind of templating is the cornerstone of all modern web frameworks and rendering libraries these days, all of which let you declaratively combine markup with data. This includes React, Vue, Angular, Preact, Lit, Svelte, SolidJS, Stencil, Quik, Ember, FAST, Polymer, Marko, and just about every other framework out there.The reasons that declarative templating are so popular and foundational may be obvious, but let's recount them anyway:The ergonomics are far superior to imperative DOM APIs. Declarative interpolation - plus features that go beyond plain HTML, like event listeners and property setting - help preserve the locality of behavior, making templates much easier to write and read than imperative code.It's easier to be secure against XSS attacks, since templates can automatically escape interpolations.Performance can be faster than all but the best hand-written code. Good template systems take great care to update only the DOM that needs to ...