3 ms·

According to TIOBE, C has lost 60% of its popularity between 2015 and 2017, and then doubled next year. What's more likely: that a 40-year-old language C had su

by pornel 4y ago

According to TIOBE, C has lost 60% of its popularity between 2015 and 2017, and then doubled next year. What's more likely: that a 40-year-old language C had such a massive sudden swing, or that TIOBE data is garbage and measures search engine's algorithms, not language relevance? https://blog.nindalf.com/posts/stop-citing-tiobe/ https://blog.nindalf.com/posts/stop-citing-tiobe/ BTW: According to the TIOBE horoscope Rust is way ahead of TypeScript and Bash.

taylodl 4y ago
I don't see the 60% drop in popularity between 2015 and 2017 as you claim (see https://www.tiobe.com/tiobe-index https://www.tiobe.com/tiobe-index). C has never not been in the top 5 languages used since the index was created 35 years ago. There have been numerous cycles where it's been #1. There's been very little volatility in the top 10 for the past 10 years, as opposed to the 25-50 ranked languages where Rust resides which has considerably more volatility in their rankings. That's why a lot of people consider anything beyond the top 25 to be just noise.
- pornel 4y ago
  The Ratings % line for C: Nov 7 2015 = 17.15%, Aug 2 2017 = 6.48% C# has jumped from 0.77% in Mar 2020 to 4.75% in Apr 2022. One month, 6x increase. Even if you just take relative rankings, they're also total nonsense: TypeScript is 40th, behind Prolog, Visual Fox Pro, Dart, and Rust.
  - taylodl 4y ago
    Yeah, that's why many people treat everything beyond the top 25 as noise. There's a lot of churn there.
    - pornel 4y ago
      TypeScript is the 8th most popular language according to Redmonk. TIOBE has issues even in the top 10, and +/- 60% errors bars on their top 2 language! It's not a problem of noisy data for the long tail languages. It's just wrong data. It's a poor unstable proxy (subject to proprietary query interpretation and anti-spam algorithms that vary over time, query terms don't reflect colloquial names, and name mentions aren't usage), and it's a misrepresentation of the data (searches include historical documents, which is not a reflection of current usage). It's all noise, no signal.

by pornel 4y ago

taylodl 4y ago

I don't see the 60% drop in popularity between 2015 and 2017 as you claim (see https://www.tiobe.com/tiobe-index https://www.tiobe.com/tiobe-index). C has never not been in the top 5 languages used since the index was created 35 years ago. There have been numerous cycles where it's been #1. There's been very little volatility in the top 10 for the past 10 years, as opposed to the 25-50 ranked languages where Rust resides which has considerably more volatility in their rankings. That's why a lot of people consider anything beyond the top 25 to be just noise.

pornel 4y ago
The Ratings % line for C: Nov 7 2015 = 17.15%, Aug 2 2017 = 6.48% C# has jumped from 0.77% in Mar 2020 to 4.75% in Apr 2022. One month, 6x increase. Even if you just take relative rankings, they're also total nonsense: TypeScript is 40th, behind Prolog, Visual Fox Pro, Dart, and Rust.
- taylodl 4y ago
  Yeah, that's why many people treat everything beyond the top 25 as noise. There's a lot of churn there.
  - pornel 4y ago
    TypeScript is the 8th most popular language according to Redmonk. TIOBE has issues even in the top 10, and +/- 60% errors bars on their top 2 language! It's not a problem of noisy data for the long tail languages. It's just wrong data. It's a poor unstable proxy (subject to proprietary query interpretation and anti-spam algorithms that vary over time, query terms don't reflect colloquial names, and name mentions aren't usage), and it's a misrepresentation of the data (searches include historical documents, which is not a reflection of current usage). It's all noise, no signal.