Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Netflix: Removing client-side React.js improved performance by 50%

twitter.com

Post reply on HN
Print view View original

31–40 of 172 posts

Re: Netflix: Removing client-side React.js improved performance by 50%

#31

Post by christophilus »

I was at the All Things Open conference this week, and Yehuda Katz gave a talk on Glimmerjs[0].

It was enlightening. The size of your front-end application is generally dominated by view code. So, they precompile views into a super simple set of binary VM instructions (making your views very compact). These views don't go through the JS compile / parse phase on the client (saving hundreds of ms, up to seconds on slow devices). They are instantly executable by a small VM written as a JS library, and can be streamed in and rendered as they arrive (the way plain ol' HTML pages do). Turns out to be hella fast, but still gives the benefit of rich client-side rendering capabilities.

[0] https://glimmerjs.com/

Re: Netflix: Removing client-side React.js improved performance by 50%

#32

Post by carry_bit »

falcolas wrote: post #27
brokentone wrote: post #20

Earlier quoted context omitted.

Because it provides a robust, consistent, isomorphic framework which many people have decided that certain benefits (for us common tooling, maintenance, consistent implementation across the org, write once run twice isomorphism) outweigh the costs (interactable time, download size, etc). Everyone has to make their own choices, but this news was not surprising and did not change our opinion.

Playing devil's advocate for a moment: So, the developer experience trumps the user experience, in an industry where fractions of a second of load time can cost a company customers and conversions?

Bad UX and buggy implementation can also cost a company customers and conversion. Sometimes the trade off is worthwhile.

Re: Netflix: Removing client-side React.js improved performance by 50%

#33

Post by bobsman123 »

_reaction_ wrote: post #15

My experience with React has not been good. A gullible manager bought into "react everywhere" hype, flew in consultants for training and then proceeded to have a dozen pages on the website written in React. It's important to note that all but one was either completely static or a simple form with less than 5 fields. This project did not go well. It took months to develop, has 0 test coverage, numerous bugs and doesn'…

It sounds like the development team was incompetent and deserved to be fired. Outsourcing by any chance?

6 months to develop a single page is something I’ve only seen out of India.

Re: Netflix: Removing client-side React.js improved performance by 50%

#34

Post by shepardrtc »

jinfiesto wrote: post #24
_reaction_ wrote: post #15

My experience with React has not been good. A gullible manager bought into "react everywhere" hype, flew in consultants for training and then proceeded to have a dozen pages on the website written in React. It's important to note that all but one was either completely static or a simple form with less than 5 fields. This project did not go well. It took months to develop, has 0 test coverage, numerous bugs and doesn'…

Don't mean to be rude, and it certainly seems like React was overkill for your project, but 6 months to generate a react site like you're describing seems like the problem might not have been react. Maybe it was engineering, or maybe it was management, but cranking out a react app that's "completely static or a simple form with less than 5 fields" is something one developer should be able to do in a day or two after…

> cranking out a react app that's "completely static or a simple form with less than 5 fields" is something one developer should be able to do in a day or two after reading a basic react tutorial

I've only done very basic web development during my career, and absolutely no javascript. But, I've been learning React for a week or so and that's something I could do in a few minutes. 6 months is insane, hyperbolic, and probably untrue.

Re: Netflix: Removing client-side React.js improved performance by 50%

#35

Post by Gaelan »

falcolas wrote: post #27
brokentone wrote: post #20

Earlier quoted context omitted.

Because it provides a robust, consistent, isomorphic framework which many people have decided that certain benefits (for us common tooling, maintenance, consistent implementation across the org, write once run twice isomorphism) outweigh the costs (interactable time, download size, etc). Everyone has to make their own choices, but this news was not surprising and did not change our opinion.

Playing devil's advocate for a moment: So, the developer experience trumps the user experience, in an industry where fractions of a second of load time can cost a company customers and conversions?

Better DX = more time to work on UX

Re: Netflix: Removing client-side React.js improved performance by 50%

#36

Post by littlestymaar »

falcolas wrote: post #27
brokentone wrote: post #20

Earlier quoted context omitted.

Because it provides a robust, consistent, isomorphic framework which many people have decided that certain benefits (for us common tooling, maintenance, consistent implementation across the org, write once run twice isomorphism) outweigh the costs (interactable time, download size, etc). Everyone has to make their own choices, but this news was not surprising and did not change our opinion.

Playing devil's advocate for a moment: So, the developer experience trumps the user experience, in an industry where fractions of a second of load time can cost a company customers and conversions?

Life is full of trade-offs, gaining 1% more customers for a 10% increase in your engineering costs is not always a good deal. Being customer-centric is good, but not at all costs.

Re: Netflix: Removing client-side React.js improved performance by 50%

#38

Post by shepardrtc »

_reaction_ wrote: post #15

My experience with React has not been good. A gullible manager bought into "react everywhere" hype, flew in consultants for training and then proceeded to have a dozen pages on the website written in React. It's important to note that all but one was either completely static or a simple form with less than 5 fields. This project did not go well. It took months to develop, has 0 test coverage, numerous bugs and doesn'…

> Everyone on the team was either fired or quit less than 6 months after it launched.

If that team took 6 months to develop such a simple site with a framework that basically does everything for you, I'd fire them all, too. I guarantee the problem was not with the framework.

Re: Netflix: Removing client-side React.js improved performance by 50%

#39

Post by dustingetz »

If you get your SSR right, you can just not serve javascript for landing pages and readonly dashboard type things. This is what Hyperfiddle does - all readonly things work 100% with javascript disabled (try disabling it in the dev tools)

http://hyperfiddle.net

Re: Netflix: Removing client-side React.js improved performance by 50%

#40

Post by chasd00 »

_reaction_ wrote: post #15

My experience with React has not been good. A gullible manager bought into "react everywhere" hype, flew in consultants for training and then proceeded to have a dozen pages on the website written in React. It's important to note that all but one was either completely static or a simple form with less than 5 fields. This project did not go well. It took months to develop, has 0 test coverage, numerous bugs and doesn'…

I've see this go down before. React was the wrong choice but I'd put the blame right at the feet of the consulting firm. Let me guess, the budget was under $5M but the firm was one of the large ones probably being used elsewhere in the company right? I work for a very very large consulting firm, anything under $30M is not worth getting out of bed for them. Projects like this get staffed by nobodies and run by nobodies, unless you have a PM and one dev who actually care they will always go down in flames.
Post reply on HN