5 ms·

Remember folks, the Tiobe Index is a ranking of how many web searches are being done for a language. The farther you get from that metric the more you are specu

Remember folks, the Tiobe Index is a ranking of how many web searches are being done for a language. The farther you get from that metric the more you are speculating.

truly 5y ago
I find that it is a good proxy for how much each language is being used.
- cjaybo 5y ago
  How do you find that?
  - truly 5y ago
    It seems clear that the more a language is used, the more a programmer will makes searches. Not a linear relation, but it is common sense.
    - phaedryx 5y ago
      It seems clear that the worse the documentation for a language is, the more a programmer will make searches. Not a linear relation, but it is common sense. It seems clear that the harder a language is to understand, the more a programmer will make searches. Not a linear relation, but it is common sense. It seems clear that [reason somebody would do a search] ...
- ljoshua 5y ago
  You may want to also consider other data points like the annual Stack Overflow Developer Survey [0]. Again, it's not necessarily the perfect stat reflecting actual real-world language use (since it's really more about languages used by visitors to SO), but I'd say it's a pretty decent proxy. [0]: https://insights.stackoverflow.com/survey/2021#technology https://insights.stackoverflow.com/survey/2021#technology
jstx1 5y ago
Are we going to pretend that there are really more searches for C and Visual Basic than for Javascript? I don’t know what they’re measuring exactly and how but on the surface it doesn’t pass the smell test. edit: Here's the actual method - https://www.tiobe.com/tiobe-index/programming-languages-definition/ https://www.tiobe.com/tiobe-index/programming-languages-defi.... It doesn't count searches.
- dragonwriter 5y ago
  > Are we going to pretend that there are really more searches for C and Visual Basic than for Javascript? Probably. A lot of things I would add the language name to while searching for another language I qiickly learned tagging MDN on instead gets the best JS info, and once on MDN, navigating between topics rather than doing a new search for a related topic os much more common than the haphazard set of resources that you end up with for many languages. Of course, that's a minor reason among the vast number of reasons why number of web searches explicitly targeting the language is a horrible measure of language popularity or impact.
  - jstx1 5y ago
    They don't even count searches (i.e. traffic), they count hits from searches in a number of popular engines - https://www.tiobe.com/tiobe-index/programming-languages-definition/ https://www.tiobe.com/tiobe-index/programming-languages-defi... > The ratings are calculated by counting hits of the most popular search engines. The search query that is used is +"<language> programming". The number of hits determines the ratings of a language.
- lr1970 5y ago
  Whatever they do TIOBE Index methodology is flawed. MATLAB at #13 ahead of Ruby #16 and Swift #17 makes no sense. MATLAB is a niche language popular mostly among Electrical and Mechanical Engineers while Ruby and Swift are general purpose and have much larger user and project base.
lvass 5y ago
Tiobe is great. Whenever I see someone taking those rankings at face value, I get wary of their ability or general motivation to check their sources.
ZuLuuuuuu 5y ago
Isn't it how many results there are for searching "LANGUAGE_NAME programming"? Not how many people searched for it. Or maybe I am mixing it up...
omoikane 5y ago
It actually counts number of search results. I think the intent was to use number of web pages as a proxy to measure popularity of languages, but since large chunks of the web are behind walled gardens (e.g. social networks that are invisible to bots), this metric might not be as relevant as it used to be.
CodeGlitch 5y ago
There is also: https://pypl.github.io/PYPL.html https://pypl.github.io/PYPL.html Which ranks languages based on number of tutorials searched for (via Google trends). It also has python at number 1. I lucked out and learnt python 10 years ago so feel "ahead of the game". It's a language that feels so malleable... Thre only negative is that type annotations aren't part of the core language. You have to run a separate tool (mypy) to check your types. Feels like it should be a flag on CPython or something.
GrumpyYoungMan 5y ago
I'm more inclined to believe Collabedit's stats as a measure of language use in indsutry since it's commonly used for online interviews. http://collabedit.com/programming-language-stats http://collabedit.com/programming-language-stats

Remember folks, the Tiobe Index is a ranking of how many web searches are being done for a language. The farther you get from that metric the more you are speculating.

truly 5y ago

I find that it is a good proxy for how much each language is being used.

cjaybo 5y ago
How do you find that?
- truly 5y ago
  It seems clear that the more a language is used, the more a programmer will makes searches. Not a linear relation, but it is common sense.
  - phaedryx 5y ago
    It seems clear that the worse the documentation for a language is, the more a programmer will make searches. Not a linear relation, but it is common sense. It seems clear that the harder a language is to understand, the more a programmer will make searches. Not a linear relation, but it is common sense. It seems clear that [reason somebody would do a search] ...
ljoshua 5y ago
You may want to also consider other data points like the annual Stack Overflow Developer Survey [0]. Again, it's not necessarily the perfect stat reflecting actual real-world language use (since it's really more about languages used by visitors to SO), but I'd say it's a pretty decent proxy. [0]: https://insights.stackoverflow.com/survey/2021#technology https://insights.stackoverflow.com/survey/2021#technology

jstx1 5y ago

Are we going to pretend that there are really more searches for C and Visual Basic than for Javascript? I don’t know what they’re measuring exactly and how but on the surface it doesn’t pass the smell test. edit: Here's the actual method - https://www.tiobe.com/tiobe-index/programming-languages-definition/ https://www.tiobe.com/tiobe-index/programming-languages-defi.... It doesn't count searches.

dragonwriter 5y ago
> Are we going to pretend that there are really more searches for C and Visual Basic than for Javascript? Probably. A lot of things I would add the language name to while searching for another language I qiickly learned tagging MDN on instead gets the best JS info, and once on MDN, navigating between topics rather than doing a new search for a related topic os much more common than the haphazard set of resources that you end up with for many languages. Of course, that's a minor reason among the vast number of reasons why number of web searches explicitly targeting the language is a horrible measure of language popularity or impact.
- jstx1 5y ago
  They don't even count searches (i.e. traffic), they count hits from searches in a number of popular engines - https://www.tiobe.com/tiobe-index/programming-languages-definition/ https://www.tiobe.com/tiobe-index/programming-languages-defi... > The ratings are calculated by counting hits of the most popular search engines. The search query that is used is +"<language> programming". The number of hits determines the ratings of a language.
lr1970 5y ago
Whatever they do TIOBE Index methodology is flawed. MATLAB at #13 ahead of Ruby #16 and Swift #17 makes no sense. MATLAB is a niche language popular mostly among Electrical and Mechanical Engineers while Ruby and Swift are general purpose and have much larger user and project base.

lvass 5y ago

Tiobe is great. Whenever I see someone taking those rankings at face value, I get wary of their ability or general motivation to check their sources.

ZuLuuuuuu 5y ago

Isn't it how many results there are for searching "LANGUAGE_NAME programming"? Not how many people searched for it. Or maybe I am mixing it up...

omoikane 5y ago

It actually counts number of search results. I think the intent was to use number of web pages as a proxy to measure popularity of languages, but since large chunks of the web are behind walled gardens (e.g. social networks that are invisible to bots), this metric might not be as relevant as it used to be.

CodeGlitch 5y ago

There is also: https://pypl.github.io/PYPL.html https://pypl.github.io/PYPL.html Which ranks languages based on number of tutorials searched for (via Google trends). It also has python at number 1. I lucked out and learnt python 10 years ago so feel "ahead of the game". It's a language that feels so malleable... Thre only negative is that type annotations aren't part of the core language. You have to run a separate tool (mypy) to check your types. Feels like it should be a flag on CPython or something.

GrumpyYoungMan 5y ago

I'm more inclined to believe Collabedit's stats as a measure of language use in indsutry since it's commonly used for online interviews. http://collabedit.com/programming-language-stats http://collabedit.com/programming-language-stats