The 2023 Julia User & Developer Survey was presented Wednesday, July 26 during JuliaCon. The presentation is available here.
This was the fifth annual iteration of the survey, which is conducted by JuliaHub on behalf of the Julia community.
We are grateful to the 1,329 Julia users and developers who participated, community members who provided comments on the draft survey, shared and promoted the survey, and the translators who translated the survey into Spanish, Chinese (Simplified and Traditional), Japanese, and, for the first time this year, Korean. We are also grateful to Roger Luo who moved the survey to GitHub this year.
The survey was promoted online using Slack, Discourse, Forem, Twitter, LinkedIn, email, JuliaLang.org and JuliaHub.com. Survey responses were collected May 27-July 4, 2023.
Summary
Julia and the Julia community continue to grow, expand and improve. As they do, Julia users and programmers find they can do more and more in Julia, and need other languages less and less each year.
Findings
- Julia use has increased among Julia users since last year: The use of Python and most other languages is down. Julia users are using Julia more and other languages less, and finding Julia more useful for more things as the language, package environment, tools and capabilities grow and improve.
- Julia users are planning to use more Julia next year: Julia users plan to use less Python, less MATLAB and less of most other languages next year. There is an uptick for Rust - potentially as a companion language to Julia.
- 80% say Julia is one of their favorite languages: This is the highest percentage for this question in the 5 year history of this survey.
- Most popular technical features of Julia: Speed, performance; ease of use; open source; multiple dispatch; solves the two language problem; composable; ease of installation; type system.
- Most popular non-technical features of Julia: Free; Julia community of developers is talented and active; Julia community of developers is warm and welcoming; easy to create packages.
- Biggest technical problems with Julia: Cannot generate self-contained binaries or libraries such as .exe, .dll, .so, etc.; slow compile times - too long to generate first plot; confusing or overly verbose error messages; require packages or tools that are only available or are more complete in another language; debugger is too slow and/or not fully featured.
- Biggest non-technical problems with Julia: Not enough Julia users in my field or industry; online tutorials and documentation are outdated or insufficient.
- Reasons for choosing Julia: Seems like language of the future; elegance; faster for the work I am doing; solves the two language problem; like learning new languages; preferable syntax to other languages.
- Public Julia packages: 39% have created or contributed to registered open source Julia packages.
- Julia use is growing: Most do at least 40% of their programming work in Julia, including 30% who do at least 80% of their programming work in Julia.
- Julia community is helpful and collaborative: 62% say the Julia community is helpful and collaborative - highest ever.
- Nearly half of survey respondents (47%) are professionals: 58% are academic users. Note that respondents can be both professionals and academics.
- Most Julia users and developers are age 23-40: But 36% are age 41+, and that share is growing.
- Julia community is global: Respondents come from over 82 countries or regions, live or work in over 75 countries and regions and speak 60 languages fluently.
- 21% identify as underrepresented in science or computing: Reasons include age, race, ethnicity, national origin, education level, parents’ education level, income, socioeconomic status, language, gender, politics, sexual orientation, religion, disability or gender identity.
Comments
If you have any comments on this year’s survey or would like to provide input for next year’s survey, please contact andrew.claster@juliahub.com.
