This guy understands the pain of working with backend developers who think it's much better to just slap jquery and some basic css on frontend and leave it at that.
My colleague who is a node backend developer was always praised by the boss, manger, client, etc.. for completing a feature in just 2 days. Me, a front end developer, I was blamed of taking a week to implement the same feature in front end. But one day, it happened to me to take a long leave for about a month, because my hand was fractured. During that time, that backend developer was asked to do the front end development also and he took the whole month but cannot finish. When I came back the next day, I finished it before the end of that day. I earned my front end developer respect that day. Now everyone in my office know what a frontend developer is.
@@codemaster9513 And then everyone clapped, am I right? ... I'll tell you my story working with frontend developers. We complete our backend work in a few hours. By the end of the week the frontend devs are still fiddling with their UI and adding more npm bloat to their SPA. One day we got so frustrated we rebuilt the whole thing with just Alpine js. It was done in 1/10th of the time and the customers were just happy they didn't have to load a 50mb React app just to read an article. Maybe if YOU knew what backend development entailed (data structures, security, performance etc) you'd know that frontend complexities are just artificial hurdles created by all those silly, pointless, bloated frameworks.
Thanks for your talk and for your excellent library "XState". With your JavaScript library and with the help of react XState library, I am working on a project to visually design the react application logic. This is extremely inspired by Unity3D's visual scripting asset library "Playmaker".
I'm struggling a bit with how it ties in with the slide at 7:19. I see the Ryan Florence and The State Machine pattern using setState in components but is this at the top level or, I mean is your concept of 'finite state' a application wide thing or do components each have their own FSM? or both? so how does it help with the problem shown in the slide if you need to somehow get state from one sibling to another without going through the parent?
You don't go with the bottom up approach in your child component: your state is still passed by props from the parent. But the intersection of different possibilities is reduced and pre-defined.
It's kind of hilarious that if you go to a Vue conference and ask if developing user interfaces is easy, you'll see almost every hand go up, but you go to a React conference and ask that literally not a single hand goes up.....
Talk not clear: How is this better than Redux or Elm architecture? Redux and Elm are complete solutions while this (JSON object with static transitions) can only handle fixed/static transitions: what do you do if you have actual data (user-input strings, etc.)?
Redux doesn't solve the *state management* problem. Redux solves the props-drilling problem - but after releasing the React Context API, we don't need Redux any more.
Elm is not a solution for React. This is a React talk. I agree u can totally do the same pattern with Redux, but u use Redux for global state. Redux encourages you to put as much stuff as possible into a local state. This is where xstate comes in as it can help with component local state.
Thank you for bringing this topic in React!
when you see a 4 years old comment by one of your favourite UA-camrs 👌🏻
when you see a 4 years old comment by one of your favourite UA-camrs 👌🏻
@@hugodsa89 haha nice 🙌🏼
Excellent talk!!
This deserves way more views.
Very inspiring talk
This guy understands the pain of working with backend developers who think it's much better to just slap jquery and some basic css on frontend and leave it at that.
I used to be one of those guys until I start writing frontend code
My colleague who is a node backend developer was always praised by the boss, manger, client, etc.. for completing a feature in just 2 days. Me, a front end developer, I was blamed of taking a week to implement the same feature in front end. But one day, it happened to me to take a long leave for about a month, because my hand was fractured. During that time, that backend developer was asked to do the front end development also and he took the whole month but cannot finish. When I came back the next day, I finished it before the end of that day. I earned my front end developer respect that day. Now everyone in my office know what a frontend developer is.
@@codemaster9513 And then everyone clapped, am I right? ... I'll tell you my story working with frontend developers. We complete our backend work in a few hours. By the end of the week the frontend devs are still fiddling with their UI and adding more npm bloat to their SPA. One day we got so frustrated we rebuilt the whole thing with just Alpine js. It was done in 1/10th of the time and the customers were just happy they didn't have to load a 50mb React app just to read an article. Maybe if YOU knew what backend development entailed (data structures, security, performance etc) you'd know that frontend complexities are just artificial hurdles created by all those silly, pointless, bloated frameworks.
@@codemaster9513 sorry it had to take you fracturing your hand lol
Thanks for your talk and for your excellent library "XState". With your JavaScript library and with the help of react XState library, I am working on a project to visually design the react application logic. This is extremely inspired by Unity3D's visual scripting asset library "Playmaker".
'if it's a week old it's a legacy app'
LOL! Configuring Webpack xD, fortunately I use FuseBox :)
I'm struggling a bit with how it ties in with the slide at 7:19. I see the Ryan Florence and The State Machine pattern using setState in components but is this at the top level or, I mean is your concept of 'finite state' a application wide thing or do components each have their own FSM? or both? so how does it help with the problem shown in the slide if you need to somehow get state from one sibling to another without going through the parent?
You don't go with the bottom up approach in your child component: your state is still passed by props from the parent. But the intersection of different possibilities is reduced and pre-defined.
I wonder who is talking at 2:11 (?)
It's kind of hilarious that if you go to a Vue conference and ask if developing user interfaces is easy, you'll see almost every hand go up, but you go to a React conference and ask that literally not a single hand goes up.....
Not using FSM in development? what?
Talk not clear:
How is this better than Redux or Elm architecture?
Redux and Elm are complete solutions while this (JSON object with static transitions) can only handle fixed/static transitions:
what do you do if you have actual data (user-input strings, etc.)?
Redux doesn't solve the *state management* problem. Redux solves the props-drilling problem - but after releasing the React Context API, we don't need Redux any more.
Elm is not a solution for React. This is a React talk. I agree u can totally do the same pattern with Redux, but u use Redux for global state. Redux encourages you to put as much stuff as possible into a local state. This is where xstate comes in as it can help with component local state.
It would be even nicer if you used typescript union and sum types to represent your state transitions, no more magic strings :).
or just a constant object with the transition names haha
Talk starts around @8:30.
No! It started at 0:05 and every bit of it was great