2018 Node.js User Survey Report

This is the third annual Node.js User Survey conducted by the Node.js Foundation. The survey was fielded in English and Chinese from October 2017 through January 2018, yielding 1,626 respondents. The primary objective of the research is to profile Node.js users, understand how they use Node and gauge satisfaction with and anticipated future use of Node and other related technologies. To ensure data integrity and unbiased interpretation, data analysis and reporting was conducted by Research Collaborative, an independent market research firm.

Node.js is continuing to have a positive impact on users particularly around developer productivity and satisfaction; when asked to describe Node.js, respondents use mostly positive terms like – “fast”, “easy”, “awesome”, “powerful”, “flexible” and even “fun”.

• The largest percentage of users deploy on Amazon Web Services followed by on-premise infrastructure. Though a significant number of users deploy to Heroku, DigitalOcean, Google Cloud Platform and Microsoft Azure.
• Web Apps remain the most popular use case for Node, with 85% of respondents indicating they use Node for this work. 43% of respondents indicate they use Node for enterprise applications.
• Integration with databases, front-end frameworks and load balancing are at the forefront of the range of technologies integrated with Node.js.
• Most users expect to increase their use of Node.js, particularly in Latin America and EMEA.
• Babel is the leading transpiler, but back-end, full stack and ‘other’ developers are increasing their use of Typescript. Among Module Bundlers, Webpack seems to be consolidating its lead across most regions and development areas.
• is becoming increasingly important to users to manage different packages for multiple environments. npm is by far the most widely used package manager, but Yarn is gaining in popularity in many segments.

Node.js continues to see it’s popularity grow on every continent and in a very broad set of use cases due to its flexibility and utility for a wide variety of use cases.

• The typical Node.js user is male, age 31, college educated and white.
• Respondents are slightly older in this wave, but gender and education are unchanged.

• As in the prior study, respondents come from across the globe, but the mix of countries has changed slightly with fewer respondents from the US and China, and more from India and Canada.
• Fewer than half consider English their primary language.

• The typical respondent is a developer in a small (<100 employees) firm, with 5 years professional experience,
• Although many have 10+ years total development experience, respondents are somewhat less experienced overall in this wave.

• There are notable differences in business and personal profile by region.
• Those in the US/CA are older, more experienced and work in larger companies than their peers around the globe.
• EMEA respondents are particularly highly educated.
• The profile of APAC respondents has changed in many respects vs. last wave – perhaps not surprising given the drop in China as a percentage of APAC respondents this year. APAC respondents are relatively new to development.

• There has been no change in the Node.js usage profile since last wave.
• The typical respondent has been using Node for just over 2 years, and spends more than half of his development time with Node.

• Three in four Node.js users are focused primarily on back-end or full stack development.
• There has been a slight drop in this wave in those who have any focus on back-end development.
• US/CA respondents are more focused on full stack than back-end.
• Ops/DevOps is really only a focus in US/CA – although even here, it is not widely focused on.

• There are some differences in users’ business and personal characteristics based on development focus.
• Full stack developers have been using Node the longest, and along with back-end developers, spend the most time with it.
• Those outside of the three traditional development areas tend to be older and more highly educated than others.

• AWS is the primary place where respondents deploy code for production, and it seems to be growing for use in development.
• On-Premise infrastructure is also widely used, but has dropped for use in production since last year.

• EMEA respondents are less likely than others to use AWS, preferring on-premise infrastructure. US/CA respondents are also likely to be deploying via on-premise infrastructure.
• Heroku is growing in both APAC and Latin America, and is one of the top choices for deployment for development in Latin America.

• Deployment also varies somewhat based on development focus.
• AWS is widely used by back-end, full stack and front-end developers, but less so for others.
• Heroku is relatively popular among full-stack developers.

• As before, the vast majority of respondents are spending time developing web apps, particularly those in full stack or front end positions.
• A notable minority also engage in enterprise and/or hobbyist work.
• Those outside the traditional development areas are more likely to be working on embedded systems.

• Respondents are using a range of tools with Node.js including, primarily: databases, libraries and Node.js frameworks.
• There is considerable variation based on primary development focus – with back-end and full stack developers most likely to use a range of tools.
• Messaging systems and CI are less commonly used than other tools – and usage has dropped since last year.

• Respondents are using a range of tools within each category, although in most categories, one or two tools dominate.
• There has been a drop in usage for many specific tools – but some have increased since last year, including Angular 2, Vue, GraphQL, Docker and Kubernetes.

• Most respondents use a transpiler and module bundler – particularly front-end and full stack developers.
• Use of transpilers has risen since last year; Babel is most common, but use of Typescript is on the rise.
• Webpack seems to be solidifying its notable lead among module bundlers.

• Ubuntu is the primary OS/Distro used in production, while MAC OS is primary in development
• Use of Windows – in both production and development – has increased since last year, particularly in US/CA, EMEA and in smaller companies (among other segments).

• Distro use varies somewhat by region, with Ubuntu more popular in APAC and Latin America, and MAC OS more popular in US/CA.
• Debian-based Linux, while not widely used anywhere for development, is somewhat popular in EMEA and Latin America in production

• Primary distro varies somewhat by development focus.
• Ubuntu is most popular among back-end and full stack developers, while Windows is more popular among front-end and “other” developers (where it is the #1 choice for both production and development).

• Respondents are using at least 3 other languages on average besides Node – typically JavaScript and then Python, Java or PHP.
• More than a third of respondents are using ES2017 or above.

• Language usage varies somewhat by region and development focus.
• Among other differences, PHP is less popular in US/CA and among those outside of traditional programming areas.
• Despite being less likely to use JavaScript and PHP, those “other” developers use more languages on average (closer to 4), including C++ and C.

• Three quarters of respondents say they plan to increase their use of Node over the next 12 months – up from last year.
• The rise in increased usage is attributable to back-end developers and those in EMEA, although Latin American respondents are most likely to say they will increase their usage.

• Use of other languages will also increase among current users – including Rust, Go, and JavaScript.
• Usage of PHP and Ruby appears to be on the decline – although many of those in Asia/Pacific plan to increase their usage of PHP.

• Planned usage does not vary considerably across the three main developer segments – except that back-end developers are particularly likely to decrease their use of Java and front-end developers are less likely to decrease their use of PHP.

• Those who plan to increase their use of Node.js will be increasing their use of a number of other languages as well.
• Many of those who will just maintain/decrease with Node will be increasing their focus on Go or Swift.

• NPM is, by far, the most widely used package manager – but Yarn is gaining in popularity in many subgroups.
• Respondents search for packages almost entirely on npmjs.org or through Google/search engine, which are gaining popularity in APAC and among other developers.

• is becoming increasingly important to be able to manage different packages for multi environments – the rise most evident among full stack developers and those in US/CA.
• These rises notwithstanding, managing different packages is particularly important to those in APAC and Latin America.

• The availability of multiple registries is not widely seen as important – at least not outside Latin America.
• EMEA respondents, and those in companies with fewer than 100 employees are least likely to value access to multiple registries.

• The primary way respondents are gaining informal coding education is via online courses without an instructor – particularly outside US/CA.
• While not widely used, tutoring is more popular in APAC than in other regions.

• Nearly all respondents learned Node in English – which is non-native for more than half of respondents (up since last year).
• certain regions – including EMEA and (especially) Latin America – a vast majority of respondents learned Node in a non-native Language.

• About half of respondents say it was generally easy to learn Node.js – up slightly from last year. Very few complain that it is difficult.
• Surprisingly, those in Latin America – who are most likely to have learned in a non-native language – are particularly upbeat about ease of learning.
• Newer Node users are less enthusiastic than those using it 2+ years – suggesting that more could be done to improve the learning experience.

• There have been notable improvements in access to and/or quality of learning resources – particularly for General Node.js programming and Asynchronous programming.
• Still, more needs to be done to improve ratings for resources around managing node in production and Node.js and security – for both areas, high negative scores are a red flag.

• Perceptions of availability of resources vary considerably by region – with those in Latin America generally more upbeat, but those in APAC and EMEA having concerns in some specific topic areas.

• A similar pattern exists with regard to quality: Latin America respondents are more pleased with quality while APAC and (less so) EMEA have concerns.

• Respondents in mid-size companies perceive greater access to some learning resources than those in larger and smaller firms, but perceived quality is only marginally better.
• To the extent there has been an improvement in perceptions of availability of resources, it’s coming from midsize and smaller companies.

• Newer Node users give lower scores than others on availability in several topic areas.
• While still low, their higher negative ratings are worthy of note.

• Respondents use many resources when learning a new language – documentation and Stack Overflow chief among them.
• Tutorial videos are also widely used – more than in last year.
• While documentation and Stack Overflow are still top, newer Node users are more likely than others to also use free & paid online courses and tutorial videos.

• To the extent users want more learning resources, documentation, free online courses and tutorial videos top the list.
• Consistent with their current usage, new Node users are particularly likely to want free online courses and tutorial videos.

• Latin American users are particularly open to new learning resources – particularly around free online courses, tutorial videos, conferences and (increasingly) conference talk videos.
• Front end developers are less likely than others (and than last wave) to want more meet-up events.

• Three in four Node users say they use a Node.js version manager – typically NVM.

• More than half of users rely on LTS release line – but that number is slipping.
• Current is particularly popular among small companies, and newer Node users.

• is important to most users to have LTS for Node.js.
• Users in EMEA and smaller companies are less likely to see it as a priority, but even here, more than half say it is important.

• While many say the LTS Schedule/Support timeframe is clear, that number has dropped significantly since last year – overall and across multiple segments.
• APAC users, front-end developers and those in small companies are least likely to see LTS schedule/support timeframe as clear.

• Newer Node users are less likely than their more seasoned peers in their use of version manager, and are more likely to use Apt-Get.
• While it is not as important to them to have LTS support, most novice users still do want it, and their understanding of the schedule/support timeframe is weak.

• their own words, respondents used mostly positive adjectives to describe Node.js
• They particularly like that it is fast, simple, easy, yet powerful and flexible.

• As with last year, many users say that Node.js has had a positive impact on their business – chiefly through increased developer productivity and satisfaction.
• Reduced development costs and increased application performance are also important outcomes tied to Node.js.
• Perhaps not surprisingly, longer-tenured users are far more likely than novices to report an impact from Node.js.

• Users in Latin America are particularly likely to note positive impacts – particularly around productivity, application performance and uptime.
• Node has helped with recruiting in the US/CA more than other regions.

• This report presents selected findings from the 2017 Node.js User Survey.
• The primary objective of the research was to profile Node.js users and identify potential areas of improvement. The findings will be used for program development, marketing and PR/external communications.
• The study was conducted online from Oct 5, 2017 to January 7, 2018 via a self-administered survey.
• The survey was fielded worldwide in English and Chinese to encourage maximum response.
• The survey link was distributed by the Node Foundation through a number of channels including email, Twitter, conferences, blogs and word of mouth (meet-ups).
• A total of 1,626 individuals responded to at least some questions in the survey.
• To ensure data integrity and unbiased interpretation, data analysis and reporting was conducted by Research Collaborative, an independent market research firm.
• Where appropriate, statistically significant differences are noted as follows:
  indicates a score is higher/lower than one or more other subgroup score(s) at 95% confidence level
  indicates a score is higher/lower than prior wave (last year) score at 95% confidence level (prior wave data not always shown)
• Additionally, significant differences may be noted in callout boxes:
• Numbers may not total to 100% due to rounding