9 ms·
While not 100% reliable, more questions in stackoverflow does indicate higher probability of the language being more popular, if you think in Bayesian network t
by raz32dust 8y ago
While not 100% reliable, more questions in stackoverflow does indicate higher probability of the language being more popular, if you think in Bayesian network terms. And they call out all such caveats very clearly. Any "ranking" of programming languages is going to be flawed in some way or the other. Using only the github data might skew it towards more recreational languages and hide the enterprise languages, for example.
- Insanity 8y agoNot to forget that many students will use stackoverflow to ask questions. Which means that languages being taught at uni have a higher ranking (Java, Python and JS come to mind). both GH and SO are probably not representative of the industry.
- ksec 8y agoBut for things like Rails, most of the questioned were answered already and likely wont grow in any place. I wonder if PageView for questions ( How many people were checking in for each PL ) and a mix of number of questions would be a better indicator for popularity of languages.