Jason Miller

Would love to get performance numbers for this change. The size decrease is awesome, but we'll want to be confident about the perf of this (and of setTimeout fallback in...

This is a tricky one: while Brotli is certainly the more logical choice from a purely technical standpoint, changing the compression algorithm we use to track size invalidates all prior...

@jridgewell This seems reasonable and I understand the logic for splitting things up. However, I'm wondering if it would be simpler to do the reverse, and create a wrapper module...

Another option: `preact/compat/dom` (for the package.json resolution).

@jridgewell re `preact/compat/dom` - yes, that would require another package.json at `compat/dom/package.json` and additional exports in the root package.json. From a consumption standpoint, it would be as if there was...

Probably considered a duplicate of #1899. We should be patching this in compat, in core the recommendation is to use an event that either produces a render or to use...

The trick here is that we currently use a "proxy" handle in order to avoid constantly removing and re-adding handlers when developers use inline functions in render methods. If we...

Definitely! I can throw a PR together that should at least show the rough outline of what we are proposing here, then we can go from there?

Just an update since I stumbled on this in my neverending list of tasks: I did get a chance to prototype full support for object event listener values in Preact,...

That is a bug in Babel and doesn't affect createElement(), it only affects jsxs()