Jonathan Neal
jonathanneal.bsky.social
Jonathan Neal
@jonathanneal.bsky.social
I still like you just the way you are.
Go on… 🙃
August 18, 2025 at 1:35 AM
I’ll be home tonight to package a private demo to collect feedback. Would you be interested, and would these builds cover your systems?
August 17, 2025 at 10:24 PM
I really don’t want to bundle anymore, but I want bundling performance (smaller payloads, faster starts), and I want them in my dev environment…

And I want my dev environment to closer match production.

I dunno. Maybe it will suck. Or maybe it can push the needle on local dev.
August 17, 2025 at 3:30 PM
It will use module graphs to build a Zstandard shared dictionary (CDCH), so the server can tell the client all the imports up front, using 103 Early Hints headers.

This means the browser knows what imports are coming before HTML even begins, because…
August 17, 2025 at 3:30 PM
This popped up in a search, so wanted to let you and folks know it has been a standard:
www.w3.org/TR/mediaquer...

Looks like it’s coming to Chrome (merged July 2025).
chromium-review.googlesource.com/c/chromium/s...
August 7, 2025 at 4:26 PM
The detectioj isn’t too painful, tho i wouldn’t necessarily recommend modifying the prototype, as is done here.
gist.github.com/jonathantnea...
modifierKey for KeyboardEvent — 164 bytes / 140 bytes gzipped
modifierKey for KeyboardEvent — 164 bytes / 140 bytes gzipped - modifierKey.js
gist.github.com
August 3, 2025 at 5:51 PM
I have lived this. I feel your pain. I hope the platform can addresses this eventually.

github.com/w3c/uievents...
Proposal: Way to determine if any modifier key is pressed · Issue #151 · w3c/uievents
(From whatwg/html#2856) Use Case Often times, when adding keyboard event listeners, one would like to determine if a modifier key is pressed (one of ctrlKey, metaKey, altKey, shiftKey) to create ke...
github.com
August 3, 2025 at 5:49 PM
Reposted by Jonathan Neal
Rewritten to use DSD to make the "structure" clearer.

codepen.io/paramagicdev...

:host() has some pretty annoying bugs, but this seems like the case where `:host()` doesn't get to hook into `exportparts` mappings 🤔

Also fascinating that chrome has divergent behavior from FF / Safari for me.
July 29, 2025 at 7:55 PM