Take the 2-minute tour ×
Programmers Stack Exchange is a question and answer site for professional programmers interested in conceptual questions about software development. It's 100% free, no registration required.

I was wondering if anybody knows if there have been any surveys or studies with statistics of how much a given programming language is used in a given industry (e.g. % of code written annually per language)

Ultimately I am hoping to find/build a matrix like the one below

enter image description here

where each entry has a percentage value, and the entries in each column sum up to 100%.

The rows and columns I included above are just an example. The industry granularity can be finer or coarser than the one I used above, and the programming languages can include others too. I am aiming for a first order approximation of the full distribution of prog. languages across industries (i.e. most significant languages & industries)

I understand that the lines are very hard to draw, and that the categorization per industry is ill-defined to a large extent, but I would still like to get a sense of how predominant different languages are per industry.

More specifically:

  1. Are there any surveys or studies that tried to collect such statistics?
  2. This is more of a secondary question: Is there a way to collect this information using the tools or the API provided by StackExchange? Has anybody done it?

Why I am interested in this:

On one hand, I am curious to know how much we know about the distribution of programming languages across industries. The distribution exists whether or not we have ever measured it.

I also believe we can learn much about the way we use programming languages by looking at which programming languages are popular across different industries and problem types.

share|improve this question

closed as not constructive by Jarrod Roberson, Jim G., maple_shaft Jul 14 '12 at 20:14

As it currently stands, this question is not a good fit for our Q&A format. We expect answers to be supported by facts, references, or expertise, but this question will likely solicit debate, arguments, polling, or extended discussion. If you feel that this question can be improved and possibly reopened, visit the help center for guidance.If this question can be reworded to fit the rules in the help center, please edit the question.

    
You can find this kind of informations on careers stats websites (if a company in finance is looking for a C++ developer, they are likely to use C++, and so on): it will be a bit biased, but unless you do a proper statistical survey I can't see any way you could get unbiased data to work on. –  Uboonto Feb 13 '12 at 15:12
4  
How is this obvious? How do you know that web dev is "likely split between Java and C#"? I would like to see some data behind the claim. For example, I would say that a large share of the code written for web development is also written in RoR or PHP. Another example, I believe most desktop/notebook OS's are written in C, but I don't have any data to back this up. –  user815423426 Feb 13 '12 at 16:28
1  
@Kevin - There is C++ used in mobile phones - Nokia used to use Qt, Samsung's Bada platform uses C++ as well. Plus, many mobile platforms allow for C/C++ use for computation intensive applications (such as Android and WebOS) –  Jetti Feb 13 '12 at 18:41
1  
@roseck - Quick question: Assume you got all this information from an oracle, then what? What do you plan to do with this information? Why is it useful? This should tell you whether it is indeed worth it and probably why it hasn't been done (Cost of information gathering >> value of information) (probably) –  PhD Feb 13 '12 at 19:10
1  
@comingstorm Your assertion is only true if the self-selected population is actually different from "all programmers". It would be more accurate to say that he wouldn't know if it was accurate. –  Fomite Feb 13 '12 at 19:30
show 9 more comments

3 Answers

You hit the nail in the head by listing the difficulties. See more detail below, but, more generally, the question is ill-formed as it stands, meaning that any answers which are not based on the immediately obvious, will either not contain much meaningful information, or be easily manipulated to show any point is valid, or a combination of both.

Specifically:

  • Programming language use - how do you measure use?

How many programmers are hired with this skill, in the given field? How many lines of code are written? How complex are the lines of code? Which goals is the language used towards?

  • Industry - where do you draw the boundaries?

Do you count consulting work done in the industry? Do you count development efforts put into supporting tools (ie, do you count development work on Excel as applying towards all industries which rely heavily on Excel to do their heavy lifting?) Does research done at the MaRS centre count as industry work for the sponsors, or as academia work?

  • Relevance

What are you trying to measure - a snapshot in time (how much work is currently being done in a given industry) or a historical measurements (how much work has been done so far in a given industry)? Has all the work in multiple languages been relevant/useful to the field (ie, maybe loads of work was done in COBOL, but no one uses the results anymore for anything)?

share|improve this answer
add comment

The closest "study" I know that relate to your question is the Tiobe Index. I'm really concerned by the accuracy of this ranking as the methodology is based on search engine results.

As an alternative to get a better estimate of what is on demand in a given market, I scan job offers.

share|improve this answer
1  
Thanks @Pierre - this study is close in spirit to what I was looking for! –  user815423426 Feb 13 '12 at 20:13
add comment

Not what you are asking, but here is a link with a chart comparing language popularity on GitHub vs. StackOverflow.

share|improve this answer
add comment

Not the answer you're looking for? Browse other questions tagged or ask your own question.