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
Huge thanks to @tannerlinsley, Niek (github.com/boschni), @pseudo_su and @kamranayub for reviewing and valuable input. In fact, this PR became the most commented in React Query history! Not sure if that’s good or bad.. ^^
github.com/tannerlinsley/rea…
2
🚀 Release alert 🚀 - React Query now has improved support for serverside rendering! 🥳
Being able to de/rehydrate the cache between server/client is probably my largest OSS contribution so far. This also supports persisting to/from localstorage.
react-query.tanstack.com/doc…
2
14
133
I enjoy working from home, but today I really miss being able to celebrate in person with all the colleagues who did such a great job on pushing this through.. 🥰
1
1
Today I'm celebrating a major win at work. An incremental refactor around data fetching that I designed is finished and the old abstraction is gone.
It's a lot of work setting up backwards compatible solutions, and it's easy to loose momentum, but it feels so good when it works!
1
8
Back at work and hence the computer after a long summer. With two kids and a new house, this has been the time in my life with the least computer-time ever which has been a nice change, but also glad to be back!
1
4
Fredrik Höglund retweeted
Coming soon to Next.js: you no longer have to provide `href` and `as` for dynamic routes, only `href` will be enough 🚀
In Next.js 9.5 we made some big changes to the Next.js runtime that made this possible.
This is available on next@canary if you want to try it out.
62
172
1,365
Fredrik Höglund retweeted
@ephemjs is killing it with SSR react across a bunch of different projects, I keep stumbling across his name in issues/PRs (react-query, react-lightyear, react-helmet-async 🤯).
I'd love the discussion about SSR react to be less about frameworks and more about this. @dan_abramov
1
1
3
Fredrik Höglund retweeted
I know some awesome doc writers looking for work. My team - the entire MDN writers team and remains of dev rel - got laid off.
26
56
106
Fredrik Höglund retweeted
I just discovered @code Search Editors, and they're completely off the hook awesome. Find across workspace, edit results inline with multi cursor, cross-file undo stack, you name it 🤯
5
70
373
Fredrik Höglund retweeted
Introducing Rome: A linter for JavaScript and TypeScript. The first in a suite of tools for frontend languages that includes bundling, compiling, testing, and more. romefrontend.dev/blog/2020/0…
36
526
1,826
Fredrik Höglund retweeted
I am happy to announce that my translated book React Deep Dive is now available for sale on Amazon (almost) world-wide!
amazon.com/React-Deep-Dive-M…
amazon.de/React-Deep-Dive-Ma…
amazon.ca/React-Deep-Dive-Ma…
amazon.co.uk/React-Deep-Dive…
Grab your copy while it's hot and fresh from the press!
2
10
15
Fredrik Höglund retweeted
sweden eschewed lockdown
some have claimed it is a disaster and the horror of scandanavia
others that this was wise and that it's mostly the more aggressive swedish counting for COV that has driven result divergence
all cause deaths can give us insight into which is correct.
54
486
1,216
I know it's probably bad, but on a scale 1-10, exactly HOW bad is reading the content of a DOM-node inside render if you memoize it..? 😇
What kind of trouble am I asking for?
1
Fredrik Höglund retweeted
SCOOP with @MikeIsaac, Facebook employees are staging a walkout today for the first time in the company's history. They are protesting the decision by Zuckerberg to allow Trump's post to stand: nytimes.com/2020/06/01/techn…
160
2,607
7,066
I had a hunch that this was too good to be true, and it turns out that was correct, we had different gzip compression levels in different environments.. Real numbers are around 7% uncompressed and about half that after gzip.
By targeting Safari 12 capabilities and up we seem to gain an extra 10-15% decrease in JS-payload compared to targeting es modules-capable browsers, without moving very many users to the "legacy"-condition.
We can do this because we use UA instead of the module/nomodule pattern.
2
1
"Some of you have gotten really good at solving this, and have become expert selector, caching, and memoization developers"
I feel seen.
"..these are all solutions that are all but necessary due to problems that are caused solely by using a global, atomic store."
Oh. Right. 😓
Both these posts are very, very good and on point. 👏
I once used to be a global state proponent and wish I would have read these a few years ago so I could have skipped that phase in my life.. 😊
✍️ I wrote another half of a post on @ThePracticalDev, discussing global vs. local state and comparing Redux to yet another pattern, the Actor model:
dev.to/davidkpiano/redux-is-…
1
3