I am making the Python Hype Survey Results Publically available. Also I have emailed the two survey winners and am waiting for them to respond to see if they want to be publically announced. The prize is a dinner where they may invite a guest. I will join them and pick up the bill.

So are you are excited to hear the results?

Overview

The design of the survey is to help predict Python Programing languages trajectory within years to come. The survey takes three major paths based on experience. That functionality built into Survey Monkey was the missing element required to ask the right people the right questions. The group are:

New to Python: This group we care about the Learned and Migrated aspects. Most interesting here are first impressions; however, it's assumed they will not know enough of the history of Python to project its projection. Python Programmers: This is the critical mass and we care about what problems were Addressed and what was building a Platform in Python. It assumes these users build something with Python day to day. Python Experts: These are those who are most intimate with the language. The focus is what was Retained (maintained) and Promoted. As experts, their opinions on the direction of the language are most valuable. We even break it down to individual package analysis.

The distribution between the groups was near normal:

The distribution of the groups are about the same (group 1 shown), the largest areas are Web, Science, and Software Development:

3 groups compared, showing a slight increase in education and science for more experienced users, but overall about the same:

Results

How was Python Learned and Migrated from (group 1)?

2008-2009 The highest used languages (other than Python) were JavaScript and Java although all other languages followed. This tells us Python programmers came from all over

2010-2013 We saw a 40% across the board but with the distributions of languages about the same.

after 2013, we saw a 35% increase with the distribution about the same. Languages like Go and Swift were added but did not play a strong role.

44% heard of Python from Word of Mouth!

What was is the general trend of reactions of those who are Python Users (group 2)?

For this metrics we look at the critical mass and ask they to evaluate their initial, secondary, and current reaction to Python.

First Reactions (76% very positive, 19% positive)

Second Reactions (76% very positive, 19% positive)

Current Reactions (70% very positive, 26% positive)

How much to experts promote Python (group 3)?

77% of the experts are considered heavy promoters using the Net Promoter Score VS 21% are Passive, and 3% are Detractors. This is within the group of experts. The Group 2 also scored high with 66NPS ( 73%, 20%, 7%, respectively). Group 1 had a 54NPS. As expected, those who use the language longer generally promote it more.

Among experts surveyed, Python 3 is nearly as used as any other version of Python, including Python 2.7.

Jython, and IronPython barely used, but PyPy seeing some usage. Likewise a large amount of experts are using some (50%) Python 3

Python Trends

The remaining questions were asked to all three groups. The objective is to determine if the known or published trends are supported by those who took the survey.

Gartner Hype Curve

The users surveyed 1/3 to 1/2 support that the Python has indeed followed the Hype Curve and is now entering the "Plateau of Productivity." According to Gartner, that is defined by reaching High Growth adoption where 20%-30% of the potential audience has adopted Python. I wrote a blog about Python Hype Curve and this is the data to help validate. Disclaimer: I am not affiliated with Gartner and this is not a Gartner provided study, just using the concept to help survey takers understand this sort of trend.

This would mean that the downward trend (4, 5, 6) has already happened. The results are not strong however. In other words, as seen in 2011-2014, the agreement rate did see an increase of only 3% that the curve reached a peak. The total observation during that same time was only about 10% of those surveyed. In other words, only 1 in 10 think that the language peaked.

TIOBE

The TIOBE Index uses top search engines according to Alexa (Google 7%, YouTube 7%, Baidu...) and combines the results.

When did Python Peak:

2007: 1%

2010: 28%

Never: 46%

Other: 23%

When compared to the Hype curve rating, we did see we did see about a 19% - 15% agreement in the peak in 2009 (see column 2 in Hype Curve). Around 10% difference in opinion.

Meanwhile, still a 40% and 46% opinion that no peak has been seen in either model.

Github Trends

This graph is simply number of repositories on Github of a certain language.



The shifting of github pages:

Python score in 2008-09: 7.8

Python score in 2010-13: 8.6

Python score after 2013: 10.11

The downward displacement in 2011 was unsupported by the survey. Github blog stated that raise of Java projects were related to the increase of development on Android.

PYPL

This PYPL index shows how often languages tutorials were searched for on Google .

When compared to JAVA and PHP:

When compared to many other languages:

The majority 46% see that the steady incline is not following any Hype curve. 18% felt that there was no disruptive technology built with Python (like Java and Android) to cause a drastic shift.

The same goes with the spike due to the introduction of new languages. Swift and Go showed a common spike as what would be expected to large initial hype of a new language. Python has been around a long enough that that peak was very initial or never seen due to the computing climate when Python was introduced.

Conclusion

Python will not experience any drastic decline in popularity any time soon.

The majority of data either points to:

Python have already peaked and now are steady

Python is on a slow and steady upward trend

Certainly the popularity of Python has been affected by trends like increase in use of Data Science (KDNuggets reporting that 49% of analytics and data mining developers have used R, and 35% have used Python) or when Web Development flourished (Django Web Framework is the most Forked on Github):

There also is a steady demand for a Python developer. For instance, in Chicago, it was presented at Chicago Python User Group that there is a high demand for Python jobs:

Beyond all the statistics on the hype, one thing that remains true for users new or old:

Python programmers really love Python

Of those surveyed 53% think Python Very high quality, 39% High, less then 9% Natural or below.

THANK YOU

A couple of thanks to those who helped make this happen: