// the take
A research project from Domenic Denicola (then on the Blink team) that tried to re-implement native HTML elements like <a>, <video>, and <hr> using the then-new Custom Elements spec, specifically to find gaps between what native elements can do and what the platform exposes to JS. It's for people interested in web platform history or how Custom Elements v1 and Shadow DOM ended up shaped the way they are, not for anyone wanting a usable component library.
The gap-analysis methodology is genuinely clever: rather than just writing custom elements, it builds them against native element behavior as a spec and documents exactly where the platform falls short (see docs/accessibility.md). It's backed by people who had real influence on the spec, so this wasn't idle tinkering — it fed into actual Custom Elements v1 and Shadow DOM discussions. The IDL files for each 'Custom' element make the delta explicit and readable.
Dead since 2015 and built against the pre-v1 Custom Elements API, which changed enough afterward that none of this code runs correctly against the modern spec. It's explicitly exploratory — only a handful of elements (anchor, media, div, hr, span, blockquote) are implemented, nowhere near full HTML coverage, so you can't lift anything here for actual use. The encapsulation problem it raises (membranes around custom element internals) was never solved within this project and still isn't fully solved by Shadow DOM today, so the 'this will become real browser engine code' framing in the README never materialized.
We read the README, the directory tree and a key source file in October 2026. Nobody pays to be listed. How we judge →