Go Survey Shows Show Continuing Preference For Go |
Written by Janet Swift |
Monday, 01 April 2019 |
The result of the 2018 Go User Survey show that half of respondents now use Go as part of their daily routine, with web development remaining the most common domain. Go users are increasing in expertise with the language and express strong preference for it. The third Go User Survey was conducted during November 2018 and attracted responses from 5,883 developers in 103 countries. In order to identify trends the Go log report by Todd Kulesza and Steve Francia includes data from the previous surveys in 2016 and 2017. There has been a steady increase in the proportion of "Gophers" who are paid to write Go at work with 72% claiming to do so in 2018. The proportion who use it outside work has also increased in 2018. On the other hand the proportion programming in another language at work has increased - and 69% of respondents claimed expertise in 5 different languages, which as the report points out means that their attitudes towards Go are influenced by familiarity with with other programming languages. The survey findings with respect to users preferences among languages and their relative expertise in Go, and also Rust and Kotlin, compared to other languages are worth quoting:
With regard to what developers use Go for, the survey reveals a high level of consistency, with Web development dominating followed by Dev Ops and Systems programming.
Developers using Go tend to be happy with, as the 89% agreement with this sentiment included for the first time in 2018, and the fact that 90% would recommend the language. However, while 85% would prefer to use Go for new projects and two-thirds of respondents this the language works well for their teams on 44% see Go as critical to their companies' success. The 2018 survey included additional question probing into developer satisfaction which are summarized in this chart: Interpreting this chart the report states: Survey respondents were very satisfied with Go applications' CPU performance (46:1, meaning 46 respondents said they were satisfied for every 1 respondent who said they were not satisfied), build speed (37:1), and application memory utilization (32:1). Responses for application debuggability (3.2:1) and binary size (6.4:1), however, suggest room for improvement. The dissatisfaction with binary size largely comes from developers building CLIs, only 30% of whom are satisfied with the size of Go's generated binaries. For all other types of applications, however, developer satisfaction was > 50%, and binary size was consistently ranked at the bottom of the list of important factors. Debuggability, conversely, stands out when we look at how respondents ranked the importance of each aspect; 44% of respondents ranked debuggability as their most or second-most important aspect, but only 36% were satisfied with the current state of Go debugging. Debuggability was consistently rated about as important as memory usage and build speed but with significantly lower satisfaction levels, and this pattern held true regardless of the type of software respondents were building. The two most recent Go releases, Go 1.11 and 1.12, both contained significant improvements to debuggability. We plan to investigate how developers debug Go applications in more depth this year, with a goal of improving the overall debugging experience for Go developers. So the survey should contribute to improvements in Go and as expertise in the language increases, so should future satisfaction.
More InformationRelated ArticlesGo 1.10 Adds Automatic Caching To be informed about new articles on I Programmer, sign up for our weekly newsletter, subscribe to the RSS feed and follow us on Twitter, Facebook or Linkedin.
Comments
or email your comment to: comments@i-programmer.info |
Last Updated ( Monday, 01 April 2019 ) |