Ranking the popularity of programming languages

Posted: December 9th, 2010 | Author: | Filed under: outliers | Tags: , , | 75 Comments »

How would you rank the popularity of a programming language? People often discuss which languages are the best, or which are on the rise, but how do we actually measure that?

One way to do so is to count the number of projects using each language, and rank those with the most projects as being the most popular. Another might be to measure the size of a language’s “community,” and use that as a proxy for its popularity. Each has their advantages and disadvantages. Counting the number of projects is perhaps the “purest” measure of a language’s popularity, but it may overweight languages based on their legacy or use in production systems. Likewise, measuring community size can provide insight into the breadth of applications for a language, but it can be difficult to distinguish among language with a vocal minority versus those that are actually have large communities.

Solution: measure both, and compare.

This week John Myles White and I set out to gather data that measured both the number of projects using various languages, as well as their community sizes. While neither metric has a straightforward means of collection, we decided to exploit data on Github and StackOverflow to measure each respectively. Github provides a popularity ranking for each language based on the number of projects, and using the below R function we were able to collect the number of questions tagged for each language on StackOverflow.

The above chart shows the results of this data collection, where high rank values indicate greater popularity, i.e., the most popular languages on each dimension are in the upper-right of the chart. Even with this simple comparison there are several items of note:

  • Metrics are highly correlated: perhaps unsurprisingly, we find that these ranks have a correlation of almost 0.8. Much less clear, however, is whether extensive use leads to large communities, or vice-a-versa?
  • Popularity is tiered: for those languages that conform to the linear fit, there appears to be clear separation among tiers of popularity. From “super-popular” cluster in the upper-right, to the more specialized languages in the second tier, and then those niche and deprecated languages in the lower-left.
  • What’s up with VimL and Delphi?: The presence of severe outliers may be an indication of weakness in these measures, but they are interesting to consider nonetheless. How is it popular that VimL could be the 10th most popular language on Github, but have almost no questions on StackOverflow? Is the StackOverflow measure actually picking up the opaqueness of languages rather than the size of their community? That might explain the position of R.

We Dataists have a much more shallow language toolkit than is represented in this graph. Having worked with my co-authors a few times, I know we primarily stick to the shell, Python, R stack; and to a lesser extent C, Perl and Ruby, so it is difficult to provide insight as to the position of many of these languages. If you see your favorite language and have a comment, please let us know.

Raw ranking data available here.


  • http://twitter.com/jonskeet Jon Skeet

    Why does the graph show C# has having fewer Stack Overflow questions than Java, C++, Objective-C, Python etc?

    http://stackoverflow.com/tags suggests that C# has more questions than any other tag. Given the location of C and C# (same horizontal line as far as I can see) I wonder whether this isn’t a URI-encoding fail…

  • http://topsy.com/www.dataists.com/2010/12/ranking-the-popularity-of-programming-langauges/?utm_source=pingback&utm_campaign=L2 Tweets that mention dataists » Blog Archive » Ranking the popularity of programming langauges — Topsy.com

    [...] This post was mentioned on Twitter by Sara Chipps, Drew Conway, Brian Lesser, Michael Dewar, dataists and others. dataists said: Ranking the popularity of programming languages by @drewconway and @johnmyleswhite http://bit.ly/i9W6K5 [...]

  • Kent Johnson

    VimL is the name GitHub gives to the language used to program Vim. If you search for Vim on StackOverflow you get 2651 questions which puts it up with R and Scala. This probably over-counts because it includes questions about the editor itself but it may be better than your use of 0 for VimL.

  • Anonymous

    Yeah I agree with @JonSkeet – this looks wrong, C# is #1 on Stack Overflow and not looking so hot here.

  • Anonymous

    Excellent point!

    Changed the code, fixed the chart, and updated the data set. Thank you.

  • http://khurtwilliams.com/ Khürt Williams

    So … perl isn’t dead after all. I wonder how many of those perl projects are using Perl 6?

  • http://twitter.com/mik3cap Mike Caprio

    How about measuring popularity by number of job reqs looking for particular language skills? Does anyone publish good data on that?

  • http://twitter.com/skilldrick Nick Morgan

    Now that C# is showing in its correct position, it’s interesting to see how much of an outlier it is. That doesn’t surprise me though, as

    (a) Stack Overflow started with a big C# bias, due to its initial audience, and
    (b) github’s always had an open source bias, also due to its initial audience.

  • Leonardo M. Ramé

    > despite seeing very little Delphi activity elsewhere.

    This could be true if you search for “Delphi”. Remember that Delphi is *just* an IDE/Compiler for the Object Pascal language, you’ll be surprised with what you get if you search for “Pascal”, “Object Pascal” , “FreePascal” and/or “Lazarus”.

    BTW, drewconway, shouldn’t you sum all Object Pascal related TAGS (Delphi, ObjectPascal, FreePascal, Lazarus) and replace Delphi?.

  • http://twitter.com/mathaix mathaix

    It would be awesome if this could be plotted Hans Rosling style (bubble motion) over time. but I guess stackoverflow and github have not been around that long.

  • http://twitter.com/davidnwelton David N. Welton

    My own site, http://langpop.com uses job requests, books at Powell’s, delicio.us and a few other things as well, and also gives you a chart where you can change the relative weights of the various data sources.

  • jjnguy

    If you are ever doing more data analysis on Stackoverflow, it might be easier for you to use their API. http://stackapps.com/questions/1/api-documentation-and-help

    In particular, this method would have probably been super helpful in this post: http://api.stackoverflow.com/1.0/help/method?method=tags

  • http://godheadatomic.myopenid.com/ king

    I can’t speak for other shops, but perl ehn’t dead here. We use it primarily for automation and large set data analysis/transformations. We’re currently still using 5, but I’m looking forward to getting started with updating to 6.

  • http://flossmole.org Megan Squire

    My project FLOSSmole collects this sort of data (including programming languages) since 2004 for ALL the projects on 10 different source code repositories. We currently have data for Sourceforge, Freshmeat, Rubyforge, Objectweb, Github, Google Code, Free Software Foundation, Savannah, Tigris, Launchpad, etc etc The raw files are available here: http://code.google.com/p/flossmole/downloads/list and our web site is flossmole.org if anyone wants to save themselves the step of collecting the data. We also have a mailing list for discussion purposes.

  • Anonymous

    Yes. Chalk this up to measurement error on the part of the measurer, not the device, since—as I said—my programming language toolkit is rather shallow and thus I did not know about the distinctions.

  • http://twitter.com/IORayne Anthony Simpson

    I saw that Io is pretty high in your graph. I’m assuming you looked at the io tag? Most of those questions aren’t about the Io *language*, but about io in general. Either iolanguage or io-language is the tag you’re looking for there.

  • http://www.facebook.com/thiago1982 Thiago Alves Silva

    I think that the chart pretty much reflects the reality. Indeed, the more common languages are in the upper right corner

  • http://stevelosh.com/ Steve Losh

    Searching SO for “vimscript” results in 76 matches, which is probably closer to the real number.

    The line is kind of blurry though, because “how do I do X in Vim?” implicitly means “how do I do X in VimL?”, because of the existence of “exec” and “normal”. It’s hard to separate questions about the scripting language from questions about using the editor.

  • Nick

    Lots of people use github to version their vim dotfiles. That’s probably why it shows up so highly there.

  • http://www.michaelbarton.me.uk/ Michael Barton

    Think there might be a typo? “langauges”

  • Greg Lindahl

    Over at blekko, we implemented our search engine and NoSQL database in perl. We aren’t having trouble finding good engineers, either.

  • Asdf

    Do you have the ability to correlate this relationship over a span of time? For instance, per month or quarter, allowing us to visualize growth of popularity.

  • http://www.worldspinner.us/stack-overflow-at-line-0 World Spinner

    dataists » Blog Archive » Ranking the popularity of programming ……

    Here at World Spinner we are debating the same thing……

  • http://twitter.com/jessykate Jessy Cowan-Sharp

    It’s funny when I googles viml a stack overflow link was the first hit :)

    It’s possible viml as a community has a different place they tend to ask questions… I wonder about comparing other social metrics like twitter mentions or google trends or even number of results on a google search for the language name. Also page hits for different questions might hint at different distributions– eg. there may be more questions that are less commonly shared by others in certain languages (used for more diverse purposes? a more complex language?), and other languages might have a smaller set of questions that are asked more frequently– those would be qualitative factors, as opposed to a popularity measure, though might easily be misconstrued as the latter.

  • http://tech.puredanger.com Alex Miller

    Both of these sites accumulate data over time with timestamps. Therefore you should be able to do this kind of analysis with a time dimension and generate an animation over time. WANT.

  • http://twitter.com/jergason Jamison Dance

    The reason Javacript is so popular on Github is that it is included in many projects that are not really “Javascript” projects. Javascript lacks a proper distribution system for libraries, so if you write a small Rails app and include jQuery, your project will show up as mostly Javascript and very little Ruby.

    Another problem is that GitHub was started as a Ruby on Rails application. I don’t know what their architecture is now, but for Ruby developers, GitHub is the “cool” place to host your project. I believe this results in Ruby showing up as more popular on GitHub than it is in real life.

  • http://morgansutherland.myopenid.com/ Morgan

    I’m surprised to see supercollider ranking!

    Please note that “supercollider” and “sclang” are separated in your graph when they represent the same thing. SuperCollider is an environment for audio synthesis that consists of two processes: scsynth and sclang. They communicate with each other using OSC messages over UDP sockets. Extensions to scsynth are written in C++ and extensions to sclang are written in sclang.

    I’m also surprised to see pure data in the ranking. Pure data is visual data-flow programming environment for audio, video, control, etc. Extensions for pure data are written in C.

  • Scott

    Github isn’t “cool” because of the Rails pedigree. It’s cool because it emphasis community productivity. Pull Requests are really invigorating for projects. Forking is no longer a sign of poor project leadership, but rather the health of a project.

  • Anonymous

    The reason for the VimL outlier is that, as of semi-recently, GitHub has become (somewhat by convention) THE place to host development versions of vim plugins. In fact, nearly if not all vimscript plugins available on vim’s official script repo are at least mirrored on GitHub even if not by their primary maintainers. There are several reasons for this:

    * Vim’s site is poorly organized, difficult to search, and only allows downloading incremental snapshots of scripts. There’s no quick way to determine at a glance which scripts satisfy which roles and which are actively maintained. Everybody using github improves this. It’s easy to track development and see which scripts are forks of others (vimscript forks and forks of forks are super commonplace).

    * Unless your distro maintains packages of, and has someone manually paying attention to these, you have to do that yourself for every plugin.

    * Especially via vim pathogen, maintaining your collection of plugins tracking development is simple. Basically github acts as a package repo for vim plugins. (It still sucks to use the odd plugin whose developers aren’t using GitHub though like I say, some people are mirroring those anyway.)

  • http://www.neurosoftware.ro/programming-blog/facebook-web-design/web-resources/ranking-programming-languages-by-size-of-community-and-number-of-projects/ Ranking Programming Languages by Size of Community and Number of Projects | Programming Blog

    [...] using a new system: the size of the community and the number of projects. In Conway’s blog post about the results, he admits that there’s no perfect way to find data about either measure. [...]

  • http://thefinalcastle.com/1447-ranking-programming-languages-by-size-of-community-and-number-of-projects-2 Ranking Programming Languages by Size of Community and Number of Projects | thefinalcastle.com

    [...] using a new system: the size of the community and the number of projects. In Conway’s blog post about the results, he admits that there’s no perfect way to find data about either measure. [...]

  • http://j.mp/jetxee Сергей

    An observation: languages below the diagonal are more popular in the Open Source community, producing more more free code with the community of the same size (the same level of activity). See Lua, D and the whole Lisp family. And above the diagonal are the languages of choice for proprietary, custom and enterprisey software (with active community and little code on github). See C#, Java, C++, Visual Basic… The diagonal itself shows the balance between complexity (the number of questions asked) and fun (the quantity of open source code).

  • http://www.facebook.com/geoff.langdale Geoff Langdale

    1. The graph would be greatly improved with data points rather than just names.

    2. What correlation function did you use? I think Spearman’s rank correlation coefficient or Kendall’s Tau would be appropriate but my memory of this stuff is pretty vague…

  • Craig

    What, no Cobol?

  • Roymorien

    It would be interesting to find out Why a particular programming language is popular ot not. Why do developers choose Labguage A over Language B? What are the selection criteria that are applied? Also, what about contemporary use as distinct from legacy use? As one question was asked, What about COBOL? Apparently it is still the most popular language if exisiting lines of code are measured (or similar measurement), but that may be because of a lengthy period of legacy use, with little new development.

  • Rherzog

    Where is Matlab in the Table?

  • Christian Sciberras

    Did PHP just surpass C++? C, old as it is, still remains in the lead.

  • http://twitter.com/SterlingCamden Sterling Camden

    However, using GitHub alone as a measure of open source joy seems a bit arbitrary. Not all of us open source developers use GitHub. What about BitBucket, for instance?

  • http://twitter.com/SterlingCamden Sterling Camden

    Uh, and that’s different from BitBucket how?

  • Santa

    So if a language so simple to use you don’t need to ask questions about it and it is not used in any open source projects it can’t possibly be popular?

  • Rick

    The results are provocative and VERY misleading. The methodology of using # of projects and community size are both fundamentally flawed. The sources, by their very nature, attract interest in just a subset of the languages. For example, GitHub started as a Ruby on Rails app and appeals to Open Source programmers. By using the number of questions tagged for each languages the community measures ignores all those programmers that do not ask questions. Instead it, it focusses on the more vocal users who may be contending with a more difficult kanguage. Like it or not, the communittee measure doesn’t even indicate langauge popularity within StackOverflow, let alone generalizing beyond that source.

  • http://j.mp/jetxee Сергей

    Yes, on one hand, it would be intersting to see stats for other DVCS and hosting services, but as far as I know Bitbucket doesn’t provide language usage statistics. On the other hand GitHub is big and popular enough to be representative of the community as a whole.

  • Pawel Glowacki

    What about Delphi?

  • izzy

    Correct… And if every java developer spontaneously combusted this afternoon, its demise would not be accounted for in this chart. The search for hypothetical extremes is useful when choosing your data set, but is not necessary when graphic that data set.

  • http://twitter.com/jergason Jamison Dance

    I didn’t mean to knock GitHub in my comment. I agree that GitHub is a fantastic product that encourages community development. I was just pointing out that lots of Ruby projects get hosted on GitHub because everyone else’s Ruby projects are hosted on GitHub. Ruby is nowhere near the most popular programming language [1][2], yet is still the second-most popular language on GitHub.

    [1] http://langpop.com/
    [2] http://www.tiobe.com/index.php/content/paperinfo/tpci/index.html

  • Dan

    Delphi is there, as an outlier. I’m guessing, however, that the popularity of comments in StackOverflow (at least a few of them are mine) and lack of projects in GitHub may be due to a number of factors, such as the relative uncertainty of the language (Borland…Embarcadero) and the current lack of a free “try-it-at-home” version. Embarcadero eliminated the “Turbo” versions of Delphi (unlike M$ Express versions of their popular languages), leaving hobbyists high and dry for small open-source projects, and the versions are very specific, so seeking community support requires that one not only know Delphi, but have the correct version installed, and have the correct (and often not free) libraries installed.

    I love Delphi, and think it’s a great language, at least as of v2005. However, outside of programming for legacy applications, I personally see very little support for it, moving forward.

  • http://about.me/khurt Khürt Williams

    Good to know. I’ve been coding in Perl for 13 years. I still do most of my scripting in Perl but PHP has become my default web application language.

  • http://about.me/khurt Khürt Williams

    I use it primarily to manage/analyze large GB log files. I haven’t used 6 — not sure if the language changes can be described as “Perl” — but look forward to it.

  • http://www.facebook.com/people/David-Intersimone/537732571 David Intersimone

    There are many additional sources of programming language related “popularity” data. You might consider the TIOBE index, University/School programming courses use of languages, newsgroup activitries, SourceForge/CodePlex/etc projects and the languages they use, Industry standards approved, Amazon books, Dice/Monster/etc job boards, and other “global” based (not just US) sources of data pertaining to programming languages.

  • RichWa

    Trying to decide what language is best is like trying to decide which tool is best; it all depends on what you’re trying to do. Solve the problem properly and the solution should tell you which language/tool set is best for this particular problem.

  • Leandro Nunes

    Excelente iniciativa! nós estudantes e desenvolvedores precisamos de pesquisas como esta para decidirmos qual rumo tomar.

  • Sammy

    There are probably some 100 folks in our building doing some type of software develpment. The company designs computer chassis systems.
    Most of the developers are writing embedded code using C++.
    10 folks use VHDL.
    5 folks use Labview.
    5 folks use TK/TCL
    3 folks are using Matlab.
    2 use PHP.
    1 uses Javascript/jQuery.

    No one seems to be using Ruby or Python.

    I think that I am the only one using Perl on a large scale.
    At least 6 other folks purchased Perl books because of the work I have been doing. They think that Perl does Magic on its own. Their books are collecting dust.

    Perl is great for writing simple throw away scripts, automation tools, and heavy-duty applications.

    One of my applications is an ETL database collector. It consists of some 25 modules of code spanning some 5000 NCL lines. Another application in Perl uses threads to run diagnostics in parallel and other tests in a 10 hour test.

    I have also successfully use Perl within Labview shaving development time and having Perl calls do all the parsing.

    We are a hardware company and so the breakdown of programming languages does make sense.
    Every language has its role. I do not think that for what I am doing, that there is a better language.

    One needs to find the right tool for the right job!

  • uchan

    maybe, its time to move to java or php ^_^

  • http://www.facebook.com/kentlbeck Kent Beck

    Would it be possible to use the same methodology to rank testing frameworks? (He says, having written JUnit).

  • http://www.facebook.com/organizedlight Andres Castañeda

    This is for all you Developers and Hackers! 

  • http://www.bigcholebags.com/ Chole Bags

    It’s funny how we adopt words and adapt our lexicon to the times. This is a very useful slant on things.

  • Isaac Gouy

    afaict the “popularity” of most of those languages is being grossly distorted when you convert the “# of Tags” and “# of Projects” data to rankings.The range in rank value for the stackoverflow tags is from 1 to 56, but the range in “# of Tags” that rank is based upon is from 0 to 82,923 and the data is so skewed that only 11 of 56 languages have above average “# of Tags”.

    Haskell is well below average for “# of Tags” and Java is well above average for “# of Tags” –

    #56 Java = 82,923
    >>> mean = 18,770 <<<
    #40 Haskell = 1,896
    # 1 F# = 0

    (The story seems to be the same for the github "# of Projects" rank numbers — see http://www.r-chart.com/2010/08/github-stats-on-programming-languages.html)

  • http://www.extremeprogrammer.de/programmiersprachen-trends-201/ Programmiersprachen Trends 2012

    [...] tecosystems nachgegangen haben ein klasse Diagramm dazu erstellt. Die Idee stammt ursprünglich von Drew Conway und stellt ein Ranking der verschiedenen Programmiersprachen dar. Uploaded with [...]

  • http://code.girishrao.me/?p=26 Coding Ansible
  • http://techcrunch.com/2012/09/12/javascript-tops-latest-programming-language-popularity-ranking-from-redmonk/ JavaScript Tops Latest Programming Language Popularity Ranking From RedMonk | TechCrunch

    [...] The rankings are based data collected from the open source project host GitHub and the programming questions and answer site StackOverflow, a measurement invented by Drew Conway and John Myles White in 2010. [...]

  • http://woblia.com/2012/09/13/javascript-tops-latest-programming-language-popularity-ranking-from-redmonk/ JavaScript Tops Latest Programming Language Popularity Ranking From RedMonk | Woblia

    [...] The rankings are based data collected from the open source project host GitHub and the programming questions and answer site StackOverflow, a measurement invented by Drew Conway and John Myles White in 2010. [...]

  • http://atif-az.org/javascript-tops-latest-programming-language-popularity-ranking-from-redmonk/ JavaScript Tops Latest Programming Language Popularity Ranking From RedMonk | Arizona Technology Investor Forum

    [...] The rankings are based on data collected from the open source project host GitHub and the programming questions and answer site StackOverflow, a measurement invented by Drew Conway and John Myles White in 2010. [...]

  • http://www.shorewoodlocal.com/javascript-tops-latest-programming-language-popularity-ranking-from-redmonk/ JavaScript Tops Latest Programming Language Popularity Ranking From RedMonk | Shorewood News

    [...] The rankings are based data collected from the open source project host GitHub and the programming questions and answer site StackOverflow, a measurement invented by Drew Conway and John Myles White in 2010. [...]

  • http://www.vestfolio.com/2012/09/13/javascript-tops-latest-programming-language-popularity-ranking-from-redmonk/ Latest Stock Market and Wall Street stories » JavaScript Tops Latest Programming Language Popularity Ranking From RedMonk

    [...] The rankings are based on data collected from the open source project host GitHub and the programming questions and answer site StackOverflow, a measurement invented by Drew Conway and John Myles White in 2010. [...]

  • http://jp.techcrunch.com/archives/20120912javascript-tops-latest-programming-language-popularity-ranking-from-redmonk/ GitHub/StackOverflowの人気上位言語はJavaScript, Java, PHP, Python; Scalaが着実に成長

    [...] このランキングの基(もと)となったデータは、オープンソースのプロジェクトホスティングサイトGitHubの利用実態と、プログラミングのQ&AサイトStackOverflowにおける会話の動向だ。ランクの算出方法は、2010年にDrew ConwayとJohn Myles Whiteが作ったものを使用している。 [...]

  • http://kupku.com/?p=106 JavaScript Tops Latest Programming Language Popularity Ranking From RedMonk | kupku

    [...] The rankings are based on data collected from the open source project host GitHub and the programming questions and answer site StackOverflow, a measurement invented by Drew Conway and John Myles White in 2010. [...]

  • http://tedibunigevo.wordpress.com/2012/09/13/javascript-tops-latest-programming-language-popularity-ranking-from-redmonk/ JavaScript Tops Latest Programming Language Popularity Ranking From RedMonk « tedibunigevo

    [...] Industry analyst firm RedMonk published today its latest quarterly programming language popularity ranking. JavaScript came out on top, followed closely by Java, PHP, and Python. The rankings are based on data collected from the open source project host GitHub and the programming questions and answer site StackOverflow, a measurement invented by Drew Conway and John Myles White in 2010. [...]

  • http://adaptivepatterns.net/blog/?p=605 Time to Get Hip to the JavaScript. « ClintonNash

    [...] The rankings are based on data collected from the open source project host GitHub and the programming questions and answer siteStackOverflow, a measurement invented by Drew Conway and John Myles White in 2010. [...]

  • http://tm.durusau.net/?p=31398 Revisiting “Ranking the popularity of programming languages”: creating tiers « Another Word For It

    [...] the post: In a post on dataists almost two years ago, John Myles White and I posed the question: “How would you rank the popularity of a [...]

  • Daniel Haskin

    For my part, When I have seen vim scripting questions on stack overflow, they are tagged with ‘vim’, not (usually ever) ‘viml’, even though they have to do with the scripting language. This confusion of SO tags might be what we see with viml. Even I have my .vimrc file on gitHub, but i don’t think of it as a ‘viml’ repo, but a ‘vim’ repo.

  • http://www.hilarymason.com/blog/getting-started-with-data-science/ » Getting Started with Data Science hilarymason.com

    [...] Third, put your projects out in public. Share them on Github, your blog, and Twitter. Explain why you thought the question was interesting, where you got the data (and good data is everywhere), and how you came to a conclusion. It doesn’t have to be perfect. A couple examples of data projects motivated by nothing more than the author’s curiosity are  Yvo’s TechCrunch analysis and Drew and John’s Ranking the Popularity of Programming Languages. [...]

  • http://www.digitaldiva20.com/blog/web-news/getting-started-with-data-science/ Getting Started with Data Science ‹ DigitalDiva 2.0

    [...] Third, put your projects out in public. Share them on Github, your blog, and Twitter. Explain why you thought the question was interesting, where you got the data (and good data is everywhere), and how you came to a conclusion. It doesn’t have to be perfect. A couple examples of data projects motivated by nothing more than the author’s curiosity are  Yvo’s TechCrunch analysis and Drew and John’s Ranking the Popularity of Programming Languages. [...]

  • https://www.kickstarter.com/projects/1747096622/crow-hill-the-feature-film?ref=live rubidium
  • alaanile

    There are many additional sources of programming language related “popularity” data. You might consider the TIOBE index, University/School programming courses use of languages, newsgroup activitries, SourceForge/CodePlex/etc projects and the languages they use, Industry standards approved, Amazon books, Dice/Monster/etc job boards, and other “global” based (not just US) sources of data pertaining to programming languages.

    شركة عزل اسطح بالرياض

    شركة عزل خزانات بالرياض

    تنظيف فلل بجدة

    شركة رش مبيدات بالمدينة
    المنورة

    رش مبيدات مكة

    نقل عفش مكة

    شركة كشف تسربات المياه بالدمام

    شركة تنظيف موكيت بالدمام

    here

    here

    here

    here

    here

    شركات تنظيف خزانات بالدمام

    شركة كشف تسرب المياه بمكة

    شركة عزل اسطح بجدة

    اسعار ايجار السيارات فى مصر

    شركة تسليك مجارى بمكة

    here

    here

  • naught101

    The reason R is so far above the line is probably for two reasons: 1) there are tonnes of scientists using R who have no formal coding training, and so need more help, and 2) there are tonnes of scientists who write tonnes of exploratory analysis in R, and that code is often signle-use, and not really worth sharing.