Hacker News Reader

Comment by hising | Hacker News Reader

Maybe I read this with the wrong glasses, but I find some of the answers a bit high-horse? "How many components do you have" etc.

View on HN

I'm on the React team and momentarily read this the same way you did. It probably could have been clearer, but I think I don't think number of components here is meant to be an accomplishment. Rather, it just means that we've done the work of testing compatibility already on a gigantic codebase so there's a good chance that most other apps will work out of the box.

How does testing work within FB, and how much of the frontend is tested? I don't think I have seen much on how FB does testing out there, and I have a good friend who complained about lack of tests at FB (he got called about his code change breaking the spam filter at midnight, and was pissed because there were no tests so he wasn't able to guarantee something like that wouldn't happen).

I only ask out of curiosity, it'd be good to know how FB approached testing as a data point on a major company's approach (vs. say Google)

I can't say much about Google, but at FB we have a combination of unit tests and integration tests. We don't strive to have 100% coverage, but the amount of tests is always growing, and we are investing effort into efficient tooling for running thousands of front-end tests: http://facebook.github.io/jest/.

My point being, you are doing a great job, if a person is frustrated over some tech and writes a blog post or whatever about it, maybe you should just leave it with that? Coming out defending design decisions etc just leaves a foul taste. React is awesome, but maybe it is not for everyone?

It's definitely not for everyone, but it would be nice for people to make that decision based on accurate information.

There is unfortunately a lot of FUD that has been spread around React in general, the React ecosystem, and the upcoming React Fiber rewrite. It's a combination of longstanding complaints like "HTML in my JS? EWW!", concerns about build tools like Babel and Webpack, badly written articles and headlines like the recent TechCrunch post that claimed "React Fiber is a complete change that Facebook has never talked about", and of course lots of arguments in comment sections.

If someone has taken time to evaluate React and determined that it's not for them, that's totally fine. But, when poor articles get upvoted and spread widely, it doesn't help anyone.

Thanks for feedback! I edited the article to remove those sentences, as indeed I can see how it can be seen as aggressive (that wasn't my intention, and Ben's interpretation is correct).

Regarding the motivation, it wasn't to defend any decisions. I'm sorry I didn't make this clear enough. The motivation for replying was to highlight some factual inaccuracies that have started to gain traction. I see posts like this every other week, and I've noticed a few incorrect points shared between them, so I felt this is a good time to jump in and help clarify them. I could've done it in a better way though.

I added these two paragraphs:

>Your post includes some misconceptions commonly held in the React community, so I wanted to take a moment to clarify them for everyone else who has the same concerns.

>That is not to say that React works well for everyone, or that the issues raised are not valid. But there are a few facts that I think are important to get right before considering those problems.

I agree. In general, I aim to never use emojis within a post like this because that leaves too much up in the air for interpretation. I did find it a bit high horse, and subtle jabs, albeit still jabs at the original article's author.

Left a bad taste in my mouth when I finished reading it.

I can understand the author's frustration when the user's criticism is unfounded in many areas but his points could've been made without the tongue in cheek comments.

It's a double standard for sure, but it's the same concept as when you're responding to a potential troll on social media. You don't want to add fuel to the fire.

Thanks for sharing! I edited the post in response to the negative feedback here, so I'd appreciate if you could give it another look and let me know if there's something else jumping out.

It looks quite good!

In all fairness, I wanted to make sure your points were heard clearly and I felt the tongue in cheek comments distracted from the clear explanation of how the author was offbase and incorrect.

Looking forward to the future releases of React!

Sounds like the original post he was replying to was a hatchet job to begin with. The author did an admirable job of not getting down in the mud, I'm sure it was difficult not to take the original post personally and respond in kind.

Why respond at all if so?

Because, as I'm reading the response, the original post had a number of inaccuracies and FUD that it's not a good idea to leave out there unanswered since people will then assume they are true.

Yep, that was the main motivation. I should've made it clearer (I hope some edits to the post helped). It's far from the first post like this but I noticed some of the factual misconceptions seem to be picking up, so I wanted to set them straight on the record. But I didn't consider the tone enough, and I regret this. I hope it's a little better now. :-)

The responses felt very defensive.

Is that surprising? I have nothing to do with react and even I felt like ordering a shield after reading that original post. It wasn't well thought-out criticism, it was full of attacks and poorly researched complaints.

I'm sure Dan would have done nothing with it, had the HN community not upvoted the original post through the roof.

Admittedly, Dan does make it a point to search out confusion and concerns about React and try to offer help and address problems. But yes, that kind of high-profile post was definitely worth a response (particularly given the odd arguments and confusion in the post itself, much less the comment threads).

Thanks for feedback! I certainly made a mess there. :-)

I edited the post, is it better now?

Reply on news.ycombinator.com