Freelance developer - React Query maintainer - Occasional OSS contributor - Been putting React on servers for a decade - Father of two - Homebrewer
Stockholm
- Tweets 2,072
- Following 414
- Followers 921
- Likes 3,503
I wrote up a suggestion for supporting named/sibling <Outlet>'s in React Router v6.
We are doing this on top of v5 today to support SSR data fetching from sibling nested routes, and it unlocks some other cool things like powerful route overrides as well.
github.com/remix-run/react-r…
2
Fredrik Höglund retweeted
There was some recent concerns raised around the React `useMutableSource` hook and selector functions. The React and Redux teams just met, and it looks like `uMS` will be reworked to better support more use cases for external libs.
Notes summary here
github.com/reactwg/react-18/…
3
5
36
I'm investigating lazy loading localization without adding request waterfalls (including SSR). Also, I don't want to include translations in the build or from files, but from a backend.
How much of a world of hurt am I looking at here..? Any advice?
2
3
Fredrik Höglund retweeted
One of my recent react/ui opinions is that business logic should live inside components. State mgmt and all. I understand how our collective scars affect opinions on this, but it’s something that’s been on my mind a lot. I think I’m right, but unable to expound on it just yet.
33
9
171
Though unintuitive, placing script preloads at the bottom of body instead of in head can still have great perf advantages in some cases.
Wow - this is gold 🤯
The boot time of my #stencil sample app simulated locally on the #InternetComputer went down from 5.5s to 0.2s 🤩
Thank you @ephemjs 🙏
2
Agree, and I'm guilty as well. It seems this is often motivated with "these things are base building blocks that affect framework architecture and are hard to fix later" which is true, but so are UX and accessibility concerns.
Perf etc is also important, but balance is way off.
I'm kind of getting fed up with frameworks prioritizing performance, data-fetching and server-side rendering capabilities over basic UX necessities. It should be a high-priority that frameworks focus on making the web accessible and preventing, if not blocking, bad practices.
1
3
Nerdsniped again..
Redux+SSR users: we just had someone file a proof-of-concept PR to try to add an SSR mode to RTK Query.
None of us maintainers have SSR experience. Would appreciate any additional review / feedback on this approach!
github.com/reduxjs/redux-too…
Background:
github.com/reduxjs/redux-too…
1
Fredrik Höglund retweeted
Dear intern:
We've all screwed something up,
So welcome to the club.
"That Sinking Feeling (The #HugOps Song)"
49
863
2,604
Super excited to see Solid reach stable. 🥳 Great work everyone involved and congrats to the release!
I've said before this is a framework to keep an eye on, and right now is a good time to check it out. 😀
🚀 We are so happy to announce Solid is 1.0!! 🎉
@RyanCarniato #DEVCommunity #solidjs #javascript #webdev dev.to/ryansolid/solidjs-off…
1
7
Fredrik Höglund retweeted
I haven't reached for an IIFE (immediately-invoked function expression) since the jQuery days, but I've been using it lately for embedding a switch statement into JSX in a React component, and I quite like it:
41
58
628
Fredrik Höglund retweeted
I'm super excited to announce @GraphCDN!!! ⚡️🎉🔥
@TimSuchanek and I co-founded & built the GraphQL CDN we always wanted.
- Scale: Edge caching & smart invalidation
- Inspect: GraphQL analytics & error tracking
- Protect: Query depth limiting
graphcdn.io/blog/announcing-… 👈
74
230
1,128
It's brew day! 🍺
Sun is shining, wife took kids to visit grandparents and I'm enjoying a big pot of coffee, planning water chemistry adjustments for brewing a modern saison.
Having a hobby is good. Life is good.
1
6
Interesting tidbit, glad they removed this for now to avoid bugs and to ship 18 faster and with more confidence.
Resuming work is a nice optimization, but interrupting renders really is the superpower.
1
3
This tweet reminded me of the starvation problem, what if CPU keep being choked so it never gets around to finishing the transition? I wonder how React 18 will deal with that.
I guess not resuming old work can make this scenario more likely?
Most of the faster libraries completely choked under that and started looking liked stepped clocks. Admittedly React example at the time used `requestIdleCallback` and had the opposite effect of never updating the numbers.
2
Fredrik Höglund retweeted
If you haven’t checked what’s coming in @reactjs 18, this is your chance.
Automatic batching, transitions, and SSR/Suspense explained simply (with graphics oc, you know me!)
Watch full youtu.be/bpVRWrrfM1M?t=44
3
50
208
If you follow me on here there is a good chance you like SSR deep dives and if so, I've got two fantastic links from the new React 18 working group.
First is @dan_abramov's amazing deep dive into the new Suspense SSR architecture.
github.com/reactwg/react-18/…
1
8
Second is a kind of work in progress upgrade guide to React 18 for SSR, among other things detailing the new APIs. Don't miss checking out the example!
github.com/reactwg/react-18/…
4