Fred Crozatier
@fred-crozatier.dev
Experimental web frameworks, OSS etc https://github.com/fcrozatier/
Organizer - Summer of Math Exposition https://some.3b1b.co/
Founder - Sciency https://sciency.co/
Teacher - https://www.youtube.com/@captain-carre
Organizer - Summer of Math Exposition https://some.3b1b.co/
Founder - Sciency https://sciency.co/
Teacher - https://www.youtube.com/@captain-carre
I'll be at the next @svelte.dev Paris Meetup on Wednesday to talk about the new functorial reactivity primitive:
- more granular than signals (no diffing)
- more expressive (`delete` to removeAttributes...)
- more natural (you map both state and operations)
github.com/fcrozatier/f...
- more granular than signals (no diffing)
- more expressive (`delete` to removeAttributes...)
- more natural (you map both state and operations)
github.com/fcrozatier/f...
November 17, 2025 at 3:04 PM
I'll be at the next @svelte.dev Paris Meetup on Wednesday to talk about the new functorial reactivity primitive:
- more granular than signals (no diffing)
- more expressive (`delete` to removeAttributes...)
- more natural (you map both state and operations)
github.com/fcrozatier/f...
- more granular than signals (no diffing)
- more expressive (`delete` to removeAttributes...)
- more natural (you map both state and operations)
github.com/fcrozatier/f...
1. DX is comfort
2. Too much DX is magic. Hard to reason about, hard to debug.
3. Magic is the opposite of understanding
Next time you can't reason about your own code in a DX-first framework you'll know you've traded understanding for comfort.
2. Too much DX is magic. Hard to reason about, hard to debug.
3. Magic is the opposite of understanding
Next time you can't reason about your own code in a DX-first framework you'll know you've traded understanding for comfort.
November 7, 2025 at 9:55 AM
1. DX is comfort
2. Too much DX is magic. Hard to reason about, hard to debug.
3. Magic is the opposite of understanding
Next time you can't reason about your own code in a DX-first framework you'll know you've traded understanding for comfort.
2. Too much DX is magic. Hard to reason about, hard to debug.
3. Magic is the opposite of understanding
Next time you can't reason about your own code in a DX-first framework you'll know you've traded understanding for comfort.
It's wild to see AI bros freak out with AI browsers given the security risk around card credentials, while not freaking out when an AI editor controls their files and folders. 🤔
I guess that's because it's safe and will never escape the editor sandbox right?
I guess that's because it's safe and will never escape the editor sandbox right?
November 5, 2025 at 12:14 PM
It's wild to see AI bros freak out with AI browsers given the security risk around card credentials, while not freaking out when an AI editor controls their files and folders. 🤔
I guess that's because it's safe and will never escape the editor sandbox right?
I guess that's because it's safe and will never escape the editor sandbox right?
Functorial reactivity can also deal with more complex operations like item swaps inside lists in the most natural and declarative way
October 16, 2025 at 1:47 PM
Functorial reactivity can also deal with more complex operations like item swaps inside lists in the most natural and declarative way
What's the difference between signals reactivity and functorial reactivity?
TLDR: Signals reactivity is what you get when you project down functorial reactivity to a coarse topology with only one event and no associated data
TLDR: Signals reactivity is what you get when you project down functorial reactivity to a coarse topology with only one event and no associated data
October 16, 2025 at 1:16 PM
What's the difference between signals reactivity and functorial reactivity?
TLDR: Signals reactivity is what you get when you project down functorial reactivity to a coarse topology with only one event and no associated data
TLDR: Signals reactivity is what you get when you project down functorial reactivity to a coarse topology with only one event and no associated data
SoME peer review is now officially closed with 5459 votes!
The score distribution over the 464 entries looking like a skewed Gaussian shows the bar was pretty high!
Now the hard part... waiting for those winner announcements in the coming weeks.
Stay tuned!
The score distribution over the 464 entries looking like a skewed Gaussian shows the bar was pretty high!
Now the hard part... waiting for those winner announcements in the coming weeks.
Stay tuned!
September 17, 2025 at 5:32 PM
SoME peer review is now officially closed with 5459 votes!
The score distribution over the 464 entries looking like a skewed Gaussian shows the bar was pretty high!
Now the hard part... waiting for those winner announcements in the coming weeks.
Stay tuned!
The score distribution over the 464 entries looking like a skewed Gaussian shows the bar was pretty high!
Now the hard part... waiting for those winner announcements in the coming weeks.
Stay tuned!
With functorial reactivity the primitives let you build a faithful communication between your templates and the DOM:
- data and operations are transported to the DOM
- hits the highest level of granularity
- works nicely with view transitions 🤝
What do you think?
- data and operations are transported to the DOM
- hits the highest level of granularity
- works nicely with view transitions 🤝
What do you think?
September 11, 2025 at 1:26 PM
With functorial reactivity the primitives let you build a faithful communication between your templates and the DOM:
- data and operations are transported to the DOM
- hits the highest level of granularity
- works nicely with view transitions 🤝
What do you think?
- data and operations are transported to the DOM
- hits the highest level of granularity
- works nicely with view transitions 🤝
What do you think?
SoME4 peer review is open!
Skimming over the entries list, I'm really impressed by the quality and work everyone put in their entries. We've got some gems in there.
Time to review, vote and spread the word!
some.3b1b.co
Skimming over the entries list, I'm really impressed by the quality and work everyone put in their entries. We've got some gems in there.
Time to review, vote and spread the word!
some.3b1b.co
SoME — Create and discover new math content
The Summer of Math Exposition is an annual competition to foster the creation of excellent math content online.
some.3b1b.co
September 3, 2025 at 3:01 PM
SoME4 peer review is open!
Skimming over the entries list, I'm really impressed by the quality and work everyone put in their entries. We've got some gems in there.
Time to review, vote and spread the word!
some.3b1b.co
Skimming over the entries list, I'm really impressed by the quality and work everyone put in their entries. We've got some gems in there.
Time to review, vote and spread the word!
some.3b1b.co
SoME submissions are officially closed, with 470 entries this year. Good job everyone!
That spike at the end is chef's kiss, nothing says "we work best under pressure" quite like this collective data
That spike at the end is chef's kiss, nothing says "we work best under pressure" quite like this collective data
September 2, 2025 at 1:45 PM
SoME submissions are officially closed, with 470 entries this year. Good job everyone!
That spike at the end is chef's kiss, nothing says "we work best under pressure" quite like this collective data
That spike at the end is chef's kiss, nothing says "we work best under pressure" quite like this collective data
Did you know you can use the `this` keyword for slick #html one-liners?
oninput="count.textContent=this.value.length"
flems.io#0=N4IgtglgJl...
oninput="count.textContent=this.value.length"
flems.io#0=N4IgtglgJl...
August 19, 2025 at 4:31 PM
Did you know you can use the `this` keyword for slick #html one-liners?
oninput="count.textContent=this.value.length"
flems.io#0=N4IgtglgJl...
oninput="count.textContent=this.value.length"
flems.io#0=N4IgtglgJl...
Summer of Math exposition: peer review starts in 2 weeks!
August 18, 2025 at 12:29 PM
Summer of Math exposition: peer review starts in 2 weeks!
What if Signals were not the right primitive? Here are some of my observations:
1. They are reassignment-first, whereas we never reassign the DOM, which is mutation-first. So we can't transfer the semantics seamlessly (functorially)
1. They are reassignment-first, whereas we never reassign the DOM, which is mutation-first. So we can't transfer the semantics seamlessly (functorially)
August 13, 2025 at 12:36 PM
What if Signals were not the right primitive? Here are some of my observations:
1. They are reassignment-first, whereas we never reassign the DOM, which is mutation-first. So we can't transfer the semantics seamlessly (functorially)
1. They are reassignment-first, whereas we never reassign the DOM, which is mutation-first. So we can't transfer the semantics seamlessly (functorially)
TL;DR of the last 5+ years in web dev: modern #frontend development consists in building a functor from Templates to DOM that:
1. preserves the composition of structures
2. transports operations too (updates...)
Signals are just our current implementation of point 2.
That's it! Beautiful right?
1. preserves the composition of structures
2. transports operations too (updates...)
Signals are just our current implementation of point 2.
That's it! Beautiful right?
August 10, 2025 at 8:55 AM
TL;DR of the last 5+ years in web dev: modern #frontend development consists in building a functor from Templates to DOM that:
1. preserves the composition of structures
2. transports operations too (updates...)
Signals are just our current implementation of point 2.
That's it! Beautiful right?
1. preserves the composition of structures
2. transports operations too (updates...)
Signals are just our current implementation of point 2.
That's it! Beautiful right?
#CSS challenge: an expandable youtube-like description section with the following constraints:
- collapsed: clampled if needed to at most 3 lines of text
- expanded: smoothly transitions to its intrinsic size
@kevinpowell.co?
- collapsed: clampled if needed to at most 3 lines of text
- expanded: smoothly transitions to its intrinsic size
@kevinpowell.co?
August 5, 2025 at 8:49 AM
#CSS challenge: an expandable youtube-like description section with the following constraints:
- collapsed: clampled if needed to at most 3 lines of text
- expanded: smoothly transitions to its intrinsic size
@kevinpowell.co?
- collapsed: clampled if needed to at most 3 lines of text
- expanded: smoothly transitions to its intrinsic size
@kevinpowell.co?
Next time people tell you it's only about the #DX
Programming is theory building cekrem.github.io/posts/progra...
So what I want from a framework is more than convenience. I want it to:
- have strong theoretical foundations to
- help me improve as an engineer by
Programming is theory building cekrem.github.io/posts/progra...
So what I want from a framework is more than convenience. I want it to:
- have strong theoretical foundations to
- help me improve as an engineer by
Programming as Theory Building: Why Senior Developers Are More Valuable Than Ever
Peter Naur's 1985 theory of programming explains why experience matters more in the age of AI-generated code
cekrem.github.io
August 2, 2025 at 5:54 PM
Next time people tell you it's only about the #DX
Programming is theory building cekrem.github.io/posts/progra...
So what I want from a framework is more than convenience. I want it to:
- have strong theoretical foundations to
- help me improve as an engineer by
Programming is theory building cekrem.github.io/posts/progra...
So what I want from a framework is more than convenience. I want it to:
- have strong theoretical foundations to
- help me improve as an engineer by
Lost in the maze of web platform proposals?
@justinfagnani.com made a chart mapping the big picture, in particular around declarative custom elements and #webcomponents. I've turned it into an interactive roadmap for y'all. Credit to Justin
fcrozatier.github.io/wwpp/
@justinfagnani.com made a chart mapping the big picture, in particular around declarative custom elements and #webcomponents. I've turned it into an interactive roadmap for y'all. Credit to Justin
fcrozatier.github.io/wwpp/
August 1, 2025 at 2:44 PM
Lost in the maze of web platform proposals?
@justinfagnani.com made a chart mapping the big picture, in particular around declarative custom elements and #webcomponents. I've turned it into an interactive roadmap for y'all. Credit to Justin
fcrozatier.github.io/wwpp/
@justinfagnani.com made a chart mapping the big picture, in particular around declarative custom elements and #webcomponents. I've turned it into an interactive roadmap for y'all. Credit to Justin
fcrozatier.github.io/wwpp/
July 30, 2025 at 2:18 PM
My build pipeline:
1. Strip types
2. That's it
How cool is that?
Native ES modules + TypeScript = ❤️
1. Strip types
2. That's it
How cool is that?
Native ES modules + TypeScript = ❤️
July 29, 2025 at 1:08 PM
My build pipeline:
1. Strip types
2. That's it
How cool is that?
Native ES modules + TypeScript = ❤️
1. Strip types
2. That's it
How cool is that?
Native ES modules + TypeScript = ❤️
Need a simple, lightweight, spec-compliant #html parser?
I've just released a new version of HTMLCrunch
Supports:
- end tag omission
- cdata sections
- foreign elements
...and much more!
jsr.io/@fcrozatier/...
I've just released a new version of HTMLCrunch
Supports:
- end tag omission
- cdata sections
- foreign elements
...and much more!
jsr.io/@fcrozatier/...
@fcrozatier/htmlcrunch - JSR
@fcrozatier/htmlcrunch on JSR: A clean, simple, lightweight HTML parser.
jsr.io
July 4, 2025 at 1:27 PM
Need a simple, lightweight, spec-compliant #html parser?
I've just released a new version of HTMLCrunch
Supports:
- end tag omission
- cdata sections
- foreign elements
...and much more!
jsr.io/@fcrozatier/...
I've just released a new version of HTMLCrunch
Supports:
- end tag omission
- cdata sections
- foreign elements
...and much more!
jsr.io/@fcrozatier/...
Why are there so few (3-ish) rendering approaches?
Why do they all feel similar?
Rewatched @ryansolid.bsky.social's excellent video, which got me thinking about the structures at play.
Read on for a few insights, and what functoriality has to do with it 🧵
Why do they all feel similar?
Rewatched @ryansolid.bsky.social's excellent video, which got me thinking about the structures at play.
Read on for a few insights, and what functoriality has to do with it 🧵
I released another video. I think this one went a lot better than the first.
Want to know how every JavaScript framework updates the DOM? This is the video for you: youtu.be/0C-y59betmY
Want to know how every JavaScript framework updates the DOM? This is the video for you: youtu.be/0C-y59betmY
The 3 Ways JavaScript Framework Render the DOM
YouTube video by Ryan Carniato
youtu.be
July 3, 2025 at 3:15 PM
Why are there so few (3-ish) rendering approaches?
Why do they all feel similar?
Rewatched @ryansolid.bsky.social's excellent video, which got me thinking about the structures at play.
Read on for a few insights, and what functoriality has to do with it 🧵
Why do they all feel similar?
Rewatched @ryansolid.bsky.social's excellent video, which got me thinking about the structures at play.
Read on for a few insights, and what functoriality has to do with it 🧵
Beautiful video on Effect by @mattpocock.com
To be honest, I've stayed away from this disgusting syntax for years.
But now I'm feeling ready to give Effect a try. Thanks for the motivation boost!
www.youtube.com/watch?v=S2GC...
To be honest, I've stayed away from this disgusting syntax for years.
But now I'm feeling ready to give Effect a try. Thanks for the motivation boost!
www.youtube.com/watch?v=S2GC...
Effect: the unreadable library that captured my heart
YouTube video by Matt Pocock
www.youtube.com
June 30, 2025 at 11:18 AM
Beautiful video on Effect by @mattpocock.com
To be honest, I've stayed away from this disgusting syntax for years.
But now I'm feeling ready to give Effect a try. Thanks for the motivation boost!
www.youtube.com/watch?v=S2GC...
To be honest, I've stayed away from this disgusting syntax for years.
But now I'm feeling ready to give Effect a try. Thanks for the motivation boost!
www.youtube.com/watch?v=S2GC...
Damn, just realized a deep flaw in the Radish web framework. The mental model is rooted but the templating around web components is half-baked.
Working on new approach that'll grow much stronger. 💪🪴
Working on new approach that'll grow much stronger. 💪🪴
June 29, 2025 at 4:32 PM
Damn, just realized a deep flaw in the Radish web framework. The mental model is rooted but the templating around web components is half-baked.
Working on new approach that'll grow much stronger. 💪🪴
Working on new approach that'll grow much stronger. 💪🪴
The increasing AI focus in #Svelte posts has me worried. I hope the core mission of caring for the web remains the priority.
If Svelte become a vibe-coding tool for people who just want quick results, I'll know it's time to move on.
@svelte.dev
If Svelte become a vibe-coding tool for people who just want quick results, I'll know it's time to move on.
@svelte.dev
June 23, 2025 at 12:06 PM
The increasing AI focus in #Svelte posts has me worried. I hope the core mission of caring for the web remains the priority.
If Svelte become a vibe-coding tool for people who just want quick results, I'll know it's time to move on.
@svelte.dev
If Svelte become a vibe-coding tool for people who just want quick results, I'll know it's time to move on.
@svelte.dev