JetBrains makes very interesting surveys, and "Python Developers Survey 2017" (https://www.jetbrains.com/research/python-developers-survey-2017) keeps with that tradition. It provides a very informative overview on the Python Landscape, namely: who is this community; what development they do with Python and a view on the technologies they use.

I found the following facts to be the most interesting:

  • from the surveyed people (9500 developers from 150 different countries), 79% of the developers use Python as their main language. This is quite impressive number. One may argue that mostly this survey was responded by Python developers... but given the increased adoption observed over the last few years, I would not be surprised on seeing large numbers of developers using Python as their primary language. This result may also be a reflection of the profile of the developers that responded: about 50% are Web developers (where python has reasonable popularity) and 50% are Data Scientists (where Python is by no shadow of doubt the most popular programming language).

  • another interesting fact about the respondents is that most of them have a combination of roles, namely: data analysis + machine learning (24%); Web development + data analysis (24%), among others. As pointed out on the survey: it is rather new to see the mix of Web development and data analysis on the same person. To me this just shows that the mix of Data Science and Web development is reaching "production Web applications" at large scale, which is a natural step given the increasing importance of bringing data science to a wider range of applications.

  • on the technology part, the most interesting fact for me was to see Django (a Python Web development framework) with a percentage of usage similar to Numpy / pandas / Matplotlib / scipy / etc. - which are popular data science libraries/frameworks. Again this shows the alike popularity of Python among Web developers and Data scientists.

  • I was also positively surprised on seeing IDEs (such as PyCharm) being on the top of most used editors for Python. Python has traditionally stayed on the realm of "editors" (not IDEs). However, the adoption of Python to build increasingly more complex applications seems to justify the wider adoption of IDEs for Python development. Nevertheless, I (and probably a wide range of people) also like to quickly jump into Vim/Atom/VSCode to do quick tests... so if the survey would allow multiple choice on the IDEs/Editors used, I would say the result would probably different combinations of IDE+editors.

Thank you for the survey @JetBrains. Looking forward for version 2018.