Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Github and Stackoverflow programming language popularity

redmonk.com

Post reply on HN
Print view View original

31–40 of 48 posts

Re: Github and Stackoverflow programming language popularity

#31

Post by sogrady »

mertd wrote: post #21
somid3 wrote: post #19

is it possible to get numbers and units on the axis? what does the 80 mean, I think this is incredible

I'm having trouble parsing the SO axis as well. Python (145K tags) and PHP (300K tags) are right next to each other around 90. R (19K tags) and Delphi (17K tags) are just slightly below around 80.

the numbers for both axes are their respective rankings, not actual tag counts. their positioning, therefore, isn't directly proportional to the actual tag volume, but how they rank relative to one another. if we had actual numbers on the github side rather than just the rankings, we'd account for this by introducing a logarithmic scale, but we're constrained by what the data we have access to.

Re: Github and Stackoverflow programming language popularity

#32

Post by GlennS »

anonymfus wrote: post #10
Toshio wrote: post #3

My interpretation of this is that Common Lisp is a computer scientist's language (lots of code written in it, very few questions, presumably these are very advanced level questions). On the other hand, C# and Visual Basic are weekend coders' languages (reasonable amounts of code, massive amounts of very basic level questions).

>On the other hand, C# and Visual Basic are weekend coders' languages Or they simply have anti-git bias in developer's culture. Like F#, which is not popular among beginners but placed far away over median too.

I'd suggest it's just as likely that Stack Overflow has a slight bias in favour of .NET. It's written in .NET, its creators blogs talk about .NET, and a lot of the higher profile members of the .NET community regularly answer questions on there.

Re: Github and Stackoverflow programming language popularity

#33

Post by JesseAldridge »

smoyer wrote: post #13

"a few surprises on this list – the continued traction of Java" I'm not surprised by this at all ... while looking for work recently, I was a bit shocked at how much of the doom and gloom being spoken about Java was wrong. It's not cool to say you're a Java developer, but most of the corporate world runs on it, even if no one admits to be the person developing those applications. There were even a few projects where…

> I will admit that I love Python and Coffeescript too, but I don't generally write large projects in either one.

Why not Python? Just wondering.

Re: Github and Stackoverflow programming language popularity

#34

Post by halvsjur »

teacurran wrote: post #20
smoyer wrote: post #13

"a few surprises on this list – the continued traction of Java" I'm not surprised by this at all ... while looking for work recently, I was a bit shocked at how much of the doom and gloom being spoken about Java was wrong. It's not cool to say you're a Java developer, but most of the corporate world runs on it, even if no one admits to be the person developing those applications. There were even a few projects where…

Java doesn't get a lot of press or cool points, but the enterprise world is massive and anyone in it knows there's still a lot of spending going on for java projects. It's also worth noting that several other languages on the list only run on the Java Virtual Machine. (Scala, Groovy, Clojure, Gosu, Ceylon). Jobs for these languages are commonly filled by people who identify as Java developers.

There is a Scala.NET (http://lampwww.epfl.ch/~magarcia/ScalaNET/) project by the way.

Re: Github and Stackoverflow programming language popularity

#35

Post by jff »

This is a good measurement of what languages people on Github use, and which languages people on StackOverflow are having trouble with.

They mention Go, for instance. I write a lot of Go. I put it all on Bitbucket. I've also found that questions about Go are best answered on golang-nuts or in the IRC channels, or in the documentation, rather than StackOverflow.

It's a pretty cool plot, as long as you take it for what it is: a representation of how many repositories exist on Github, and how many people are asking questions on Stackoverflow. In terms of measuring language popularity, it's about as useful as TIOBE.

Re: Github and Stackoverflow programming language popularity

#36

Post by igouy »

mertd wrote: post #21
somid3 wrote: post #19

is it possible to get numbers and units on the axis? what does the 80 mean, I think this is incredible

I'm having trouble parsing the SO axis as well. Python (145K tags) and PHP (300K tags) are right next to each other around 90. R (19K tags) and Delphi (17K tags) are just slightly below around 80.

1) The "analysis" is just as broken as it was in September (the last time it was posted here).

The "popularity" of most of those languages is being grossly distorted when sogrady converts the "# of Tags" and "# of Projects" data to rankings.

The range in rank value for the stackoverflow tags was from 1 to 56, but the range in "# of Tags" that rank is based upon was from 0 to 82,923 and the data was so skewed that only 11 of 56 languages had above average "# of Tags".

Haskell was well below average for "# of Tags" and Java was well above average for "# of Tags" --

  #56 Java = 82,923
  >>> mean = 18,770
(The story was the same for the github "# of Projects" rank numbers.)

2) Which gives rise to this kind of bad-math "analysis" --

"Go jumping from #32 in 2010 to #30 today, a number that sounds modest but means that in that time it has improved more in popularity than Scala or Haskell and as much as Java, at least from a rankings standpoint (obviously growth becomes more difficult the more popular the language becomes)."

3) 90% of Tags were for just 10 languages.

50% of Tags were for just 3 languages.

The cumulative bottom 1% of Tags were for 31 different languages (including Haskell and Go).

Re: Github and Stackoverflow programming language popularity

#39

Post by KMag »

city41 wrote: post #24

These are curious statements: * CoffeeScript is a simplied version of JavaScript that infuriates technologists with its technical compromises What compromises and how is it simplified? If anything it's more complex than JavaScript. * while Assembly is as close to the bare metal as most developers today are likely to get. The only thing closer would be machine code. Not sure I want to develop in that :) (btw, hijackin…

Arguably VHDL and Verilog are closer to the bare metal than assembly. Of course, everyone I personally know who are coding VHDL or Verilog professionally are either working on latency-critical financial systems or defense systems.

Edit: ... and my guess is that VHDL and Verilog are likely to be under-represented in public data.

Re: Github and Stackoverflow programming language popularity

#40

Post by smoyer »

JesseAldridge wrote: post #33
smoyer wrote: post #13

"a few surprises on this list – the continued traction of Java" I'm not surprised by this at all ... while looking for work recently, I was a bit shocked at how much of the doom and gloom being spoken about Java was wrong. It's not cool to say you're a Java developer, but most of the corporate world runs on it, even if no one admits to be the person developing those applications. There were even a few projects where…

> I will admit that I love Python and Coffeescript too, but I don't generally write large projects in either one. Why not Python? Just wondering.

The last "big" Python project I wrote was in 2007 or 2008 and involved the import of a lot of data, processing that data and then a relatively simple web application for managing jobs and displaying graphs of the resulting data.

I used TurboGears since it seemed to be Python's equivalent to RoR and while the web application development was great, the ability make use of all the processors I provided was effectively blocked by the GIL (and yes ... I tried several different strategies to get around it). I think the class of problems I was trying at the time were prone to hitting locks, but after three or four projects like that, I just decided it wasn't worth it.

Post reply on HN