Detailed introduction to JavaScript developer survey report (picture)

Detailed introduction to JavaScript developer survey report (picture)

Our survey on JavaScript developers has also ended. The survey received more responses than expected. I'm so excited to share the results with you all!

With over 5,000 responses, 5,350 to be exact, I can’t wait to share the details of the survey. But before that, I want to thank everyone for participating. This is a great moment for the JavaScript community, let's look forward to it!

I didn't expect such a fierce response, I will make some improvements next time. That said, I would first put the survey on GitHub so that the community could have a few weeks to assemble relevant questions and options before launching the survey. This will produce more accurate results and avoid "I'm surprised you didn't include Emacs!" complaints.

Now, it’s the turn of the investigation results. I will remain impartial when announcing the results so that everyone can draw their own unbiased conclusions.

1.What type of JavaScript are you writing?

97.4% of those surveyed write JavaScript for web browsers, and 37% of these people also write mobile web applications. More than 3,000 people - about 56.6% - write server-side JavaScript. Of those, 5.5% use JavaScript in some kind of embedded environment, such as Tessel or Raspberry Pi.

Some participants said they use JavaScript in other places, most notably when developing CLI and desktop applications. Several also mentioned Pebble and Apple TV. These are classified in the Other category, accounting for 2.2% of the votes.

#2. Where do you use JavaScript?

As expected, 94.9% of voters use JavaScript at work. However, a large portion of the statistics - 82.3% - also use JavaScript in side projects. Other responses included teaching and learning, as a hobby, or for a non-profit organization.

#3. How long have you been writing JavaScript?

Over 33% of those surveyed have been writing JavaScript code for more than 6 years. Additionally, 5.2% started writing JavaScript a year ago, 12.4% two years ago, and 15.1% three years ago. In other words, 32.7% of people started writing JavaScript in recent years.

#4. Which language do you use that compiles to JavaScript?

85% of people responded that they compiled ES6 to ES5. Meanwhile 15% still use CoffeeScript, 15.2% use TypeScript, and a pitiful 1.1% use Dart.

This is one of the questions I hope more people can participate in, because 13.8% chose the answer "Other". Of that 13.8%, the vast majority of answers were ClojureScript, elm, Flow, and JSX.

#5. What JavaScript format option do you prefer?

The vast majority of JavaScript developers who voted seemed to prefer semicolons, at 79.9%. In contrast, 11% said they dislike using semicolons. When it comes to commas, 44.9% favor using them after an expression, while 4.9% want comma-first syntax. There's also space indentation, with 65.5% preferring the space bar, while 29.1% prefer using tabs.

#6. Which ES5 features do you use?

While 79.2% of the respondents agreed with the Array method of the function, 76.3% said they were using strict mode, 30% used Object.create, and only 28% used getters and setters.

7. Which ES6 features do you use?

It’s worth noting that arrow functions were the most commonly used ES6 feature in this survey: 79.6%. The use of Let and const keywords is 77.8%, and promises are also strong at 74.4%. Unsurprisingly, only 4% of respondents use proxies. Only 13.1% of users said they use symbols, and more than 30% said they use iterators.

8. Do you write tests?

Although 21.7% said they never write any tests, most people have written some tests, and 34.8% often write tests.

9. Do you run continuous integration tests?

The same is true for CI, although more people don't use CI servers - more than 40%. Nearly 60% of respondents use CI at least occasionally, with 32% frequently running tests on CI servers.

10. How do you run the test?

59% prefer to run automated browser tests with PhantomJS or similar, and 51.3% prefer to perform manual tests on the web browser. Server-side automated testing received 53.5% of the votes.

11.What unit testing library do you use?

It seems that most respondents preferred either Mocha or Jasmine to run JavaScript tests, with Tape getting 9.8% of the vote.

#12. What code quality tools do you use?

It seems that those surveyed either choose ESLint or are in the JSHint camp, but JSLint is surprisingly strong after so many years, with nearly 30%.

13. How do you deal with client-side dependencies?

npm became the first choice for client dependency management system with 60% of the votes. Bower still has 20% share, while the old