>Facebook may very well end up in the majority's good graces with this move.

Really? They are doing the minimum possible to prevent an exodus from their stack after a public outcry.

There was no sign of an exodus. That’s ludicrously hyperbolic.

Not at all. There was an exodus brewing. For starters:

https://ma.tt/2017/09/on-react-and-wordpress/

https://hacks.hyperspacer.com/app/items/15256532

An HN thread with a dozen responses? Hardly an exodus.

A quarter of all websites? An exodus.

Those websites weren't leaving, they never arrived..

They were close to arriving but they backtracked, all 25% of them. I'd call that an exodus before you even arrive.

If that's the definition then let me tell you about my exoduses from almost every country on the planet...

...of sites that posted on HN.

Projects that I work(ed) on that use React have no plans to switch away and aren't represented in that 25%.

That's not what I meant.

25% of sites on the Internet run WordPress. WordPress was rewriting its editor in React, but decided to switch away due to BSD+Patents.

Right so even then, its a bit weird to say that 25% of sites decided to not go with React, but actually 25% of sites (unknowingly) use software that decided not to go with React.

You're splitting hairs to reduce the significance of WP's stance.

No, he's not. He's pointing out that calling it an "exodus" implies that many individual projects have decided to abandon React. In this context WordPress is just one project, not "25% of all websites".

WordPress runs 25% of all websites. That is huge. You're trying to trivialize it as being "just one project" when it's really one of the most widely used and significant projects out there. One decision that WordPress makes is worth tens of thousands of decisions from much smaller projects that aren't used very widely. WordPress is a many billion dollar a year industry in itself, and whether it uses React or not will directly contribute to many thousands of web developers out there learning it or not.

You are trivializing the difference between "using" and "choosing". One is transitive, the other is not.

I'm saying that it doesn't much matter. WordPress is the largest player in the web platform ecosystem. The choices they make have huge consequences. Them choosing to use one library over another directly contributes to thousands of more developers using that library. The choice doesn't need to be transitive because developers using the platform don't have a choice of library -- they go along with whatever the platform people are using.

Nobody is disagreeing with you, you're just talking about something different than everyone else in this thread.

All of that is true but that's not what we're talking about here. We're talking about individual projects who decided to abandon React -- not the influence of said projects.

You can't die if you were never alive to begin with

And yet a project can die before it's born

Does the decision to use React or not affect end users? Each WordPress site is not making an independent determination.

CNCF had begun the process of moving two internal projects (that will be open sourced when they hit beta) from React to Vue.js. We'll now re-evaluate.

FWIW, why we prefer Apache 2.0: https://www.cncf.io/blog/2017/02/01/cncf-recommends-aslv2/

You must have missed a lot of news. A personal example: in my company we were comparing React vs Vue internally and the hint of licensing issues committed us to Vue.

This Facebook post gives us hope for GraphQL moving away from BSD+, as it is currently in our stack, but without doubt we are considering a move away due to uncertainty.

The same thing happened at my company (Fortune 150 company). We settled on Vue too, and it has been amazing. But if this had happened back when we were considering things we probably would have gone with React.

Same here, this conversation is happening right now since we've built a pretty large application using React. I would much prefer to use React so this couldn't be better timing.

>In the wake of uncertainty about our license, we know that many teams went through the process of selecting an alternative library to React.

Evidently FB saw enough people leaving to prompt this change. No doubt they were also afraid of an exodus of their internal developers on these projects. High-profile developers will move on to companies where they know they'll have greater impact if that's seen to be at risk.

That's a strange claim given how many major projects have appeared on the front page of HN because of it, everything from the Apache foundation to Wordpress.

> everything from the Apache foundation to Wordpress.

Those were the only two.

WordPress accounts for something like 19% of all websites, more so than any other single platform. Do not underestimate the importance of their decision.

A bit higher, according to those who monitor these sorts of things.

Some 28.7% of sites use Wordpress.* Curiously, I've seen the 19% number being cited as 19% of all new websites use Wordpress. I'm not sure how 19% becomes 28.7%, but it may just be that sites using WP stick around longer.

* https://w3techs.com/technologies/details/cm-wordpress/all/al...

Only 2 that you know of. I work for a Fortune 150 company and we settled on Vue almost solely due to the licensing issue.

I too work for a Fortune 150 company and we've gone all in on React.

I'm told that one reason companies might have chosen to avoid React was that they could have imagined themselves be in competition with Facebook at some point, and thereby in the danger zone with their old BSD licence.

They explicitly said in their FAQ that you could use React even in direct competition. None of this matters now, of course, because it's been relicensed. The only impact the patents license ever had on anyone was iff they initiated a patent lawsuit against Facebook.

The point is that if you're in direct competition, the chances of Facebook violating one of your patents is much higher, and then your cost of suing them is significantly increased (since you also have to rewrite your app).

If you're ok with trusting a direct competitor not to misbehave, then sure, use React under those terms while competing with FB. But large companies consider their patent portfolio to be a significant strategic asset, so it shouldn't be surprising that their legal counsel advises against accepting such a clause.

Would the previous license have become relevant if Facebook initiated a patent lawsuit against you?

Even being on the back-foot when having to defend your patents, by refactoring React out of your application, was apparently damning enough for people to migrate.

Yeah, obviously not all lawyers see eye to eye.

Drupal was also close to picking a new library to replace or supplement Backbone, and React was just dropped out of the running (with a strong favor for Vue) a week or two ago.

We get it, React would have been fine with out them.

But it did matter.

It set a strong precedent.

Huh? Where have you been for the last month or two?

React is pretty popular already. FB only really needs to change the license to keep the growth.

Exactly.

React would have been fine, but to say the big players jumping ship didn't matter is just wrong.

This. FB is intentionally leaving a ton of their ecosystem with the PATENTS virus.

Theorizing, but I suspect FB will begin targeting and picking off IP of apps built entirely on React stacks in a few years.

Please don't go back to them. Just contribute to Vue's ecosystem like crazy. Worst case scenario it'll keep FB in check.

When has this ever happened?

there was no exodus, React is the number one javascript framework in the job market in Boston and probably other tech hubs too

Yeah, we've already settled on Vue (because our lawyers forbade us from using React), but now I'm sure they would approve React. So... do we abandon our work on Vue (which so far is amazing) to switch to React since the React ecosystem is so large, or do we stay the course? Sigh. I wish Facebook had made this change a year ago.

I would stay with Vue. While the react world is bugger it's mostly made up of competing solutions instead of distinct tasks.

That said, this healthy Vue/Angular/React competition has been great for us devs.

"great" is relative to how much free time you have evaluating all the various alternatives and finding the bugs/missing use cases nobody cares to tell you about up front..

> the bugs/missing use cases nobody cares to tell you about

In the case of fewer frameworks, you would still have these. Arguably more due to lack of competition.

> evaluating all the various alternatives

Pick one of the "Big 3" above. You'll be fine. Largely due to the previous point.

I agree that competition is generally good, but I hate choices so much. I don't even care if Vue is 10 times better than React, I've just spent the last year diving into React and React Native. I've finished 3 big projects and now I'm feeling more confident. And now I get the feeling that I have to throw out a whole bunch of experience and switch to the next big thing.

I think I'm going to dig in my heels this time. React is fine. It's more than fine, I absolutely love working with React, Redux, redux-saga, redux-observable, and a whole bunch of other libraries and patterns that I've finally begun to master after over a year of headaches and frustration. I'm not going to switch to Vue, not even for my next side project.

I didn't personally care about the BSD+Patents license, but you definitely have to keep your eyes and ears open with backlash like this. Even if I think it's a total issue, it might affect my ability to hire employees or even sell my company. So I'm really glad that React is now released under the MIT license.

I don't think anyone's really claiming that Vue is a 'next big thing' to switch to from React. Vue/React/Angular are all competitors of the same generation.

> over a year of headaches and frustration

Seems like an argument against the react ecosystem

A thousand times this. We'll know the JavaScript UI toolkit has started to mature when we can pick up a new tool and not have to abandon all of the other tools we already learned and spend a year getting on our feet in a new "ecosystem".

Don't get me wrong, it was the good kind of struggle where you grow and push yourself. Like going to the gym.

It was the same as any other programming language, framework, or ecosystem. Exactly the same as my experience with Ruby on Rails when I first started doing web development.

If you don't mind sharing, what problem did you have before that was solved by adopting React? (I'm not referring to React Native. It's obvious to me what problem that solves.)

The main thing is having the state flow top-down through your app, so you never have to worry about manually keeping things in sync in the UI. Redux and redux saga have really taught me a lot about programming, and then RxJS and redux-observable completely blew my mind.

I've never found manually keeping things in sync with the UI to be overly burdensome. I guess I've just been lucky.

Many things going forward will happen in reacts petri dish. Lots of ideas seeking to get consensus like the evolution of styles, animations, native performance, async scheduler, easy-to-create native renderers. Vue is nice and all, but it's rather complacent, catering to old Angular-like roots and web artefacts. If you believed the hype then Vue was supposed to be the be-all-end-all of frameworks, the reality of course is quite sober: http://www.npmtrends.com/angular-vs-react-vs-vue-vs-@angular...

The point has been made before that npm downloads are a poor measure of framework popularity because the architecture of the framework and the typical distribution channels can skew the number of npm pulls in a way unrelated to the number of projects utilizing it.

These numbers track actual, daily usage in production environments. You can't explain these numbers away, not if they're both competing in the same field. Both are distributed in the same way (npm, unpkg, various cdns), both websites show you how, both do not need build tools to function, both are better off with build tools. It would be pretty disingenuous to claim suddenly the only actual numbers we have are skewed.

I don't understand the link. We don't get Vue from NPM. I assume others don't either. That implies to me that the numbers implied in that chart are inaccurate at best.

Why do you say the statistics are inaccurate. They show a clear picture, actually the clearest we have, because if it weren't for actual usage stats we would have to rely on github stars, google trends and hear-say. Both react and vue release through the same channels, npm and cdns. Of course npm is the more important channel for modern environments and businesses. The point is that they both are subject to the same rules. The portion of users that like script tags use script tags with react as well. Those would fall out, but that happens on both fronts obviously.

The comment i made above, that most things going forward will happen in the react eco system, is supported by these statistics. If you have lots and lots of vue users that are still using script tags, and let us - without any data to confirm it - assume that there are less react users doing this, then that would speak volumes about the userbase at large, not the most forward leaning to say the least.

where do you get it?

No way! Vue is way better than react. Ok, it depends on personal opinion, but I think it’s s much nicer framework to work with.

Out of curiosity, what stage is your company at that your lawyer has an influence on your tech stack? Was this part of due diligence for fundraising?

Um, I work for a fortune 150 company. I'm not sure what stage that makes us, but we definitely don't do fundraising ;-)

I... well, I kinda expected it. Facebook wanted their products to be used, but it was obviously their legal department that made them put the patent clause in. The more people who develop using their libraries, the more contributions they get back, and the more benefit they get from it.

I almost guarantee you none of the devs wanted that weird-ass patent clause in there, but I'm guessing that none of them thought that it was a big deal in the long run, just CYA shit you'd expect from a legal department.

They were proven wrong, and it impacted their market share, they raised a stink about it and they got it changed.

>I wonder how some of the bigger players will react (no pun intended) after their abandonment of React in lieu of Vue or some other framework

Competing frameworks will now need a new "Reason #1 to use us instead" for their pitch-decks.

As I see it this shows that you can't trust Facebook. And that Mark still is the same dude that once said "They trust me. Dumb fucks".

perhaps making a big deal out of it on mainstream media made it less unexpected? Odds are, it was a data-driven (shift in hype/marketshare) decision.

> I wonder how some of the bigger players will react (no pun intended) after their abandonment of React in lieu of Vue or some other framework

What do you expect them to do, shrug and come back? I doubt anyone who left based on concerns with their stack being owned by Facebook is likely to regret weaning themselves off of it.

The react ecosystem is massive, and includes impressive projects like the various backends/renderers: react-native, react-sketch.app, react-vr, react-pdf, ink, that have no equivalent in other frameworks (or at least to nowhere near the same level of maturity), so yeah people ditching it may well regret it.