Language popularity absurdity

Looking at April’s TIOBE index for language popularity, it got even more absurd. As they write: “Other interesting moves in the TIOBE index this month can be found outside the top 20. This is due to the fact that the index uses 7 different search engines as of this month.”

As a result, many languages have dropped out the 20 to 50 range, Groovy and Scala are two examples. And some surprising items entered. While I have better things to do than to analyse their search procedure, two simple indicators show how ridiculous this index has become:

  • Progress 4GL : TIOBE position 45 – Google for “Progress 4GL programming language”: 10,400 results (“4GL programming language” 42,000 results) – not listed on Github
  • TOM : TIOBE position 49 – Google for “TOM programming language” : 21,500 results – not listed on Github
  • SIGNAL : TIOBE position 50 – Google for “SIGNAL programming language” : 194,000 results – not listed on Github

In comparison:

  • Scala : TIOBE position > 50 – Google for “Scala programming language” : 739,000 results – Github position #17
  • Groovy : TIOBE position > 50 – Google for “Groovy programming language” : 39,100 results – Github position #22

This is an index about popularity, right? Well maybe the results are distorted by searching over all time. Let’s search in google with a restriction to posts within the last year:

  • “4GL programming language” – 124 results
  • “TOM programming language” – 59 results
  • “SIGNAL programming language” – 870 results
  • “Scala programming language” – 45,000 results
  • “Groovy programming language” – 644 results

Ooops. Conclusion: The TIOBE index is a complete joke, and it’s sad that they seem to be the main index to ask. At least the folks at the Transparent Language Popularity Index seem to be still halfway sane.