The number of mentions needs to be normalized by the number of job posts.<p>Example:<p>- 2021-01: posts=842, python=194, ratio = 194 / 842 = 0.23 (mentions per post)<p>- 2025-01: posts=487, python=87, ratio = 87 / 487 = 0.18<p>And then if you want to see a trend, do a moving 6 months average.<p>[2021-01] <a href="https://news.ycombinator.com/item?id=25632982">https://news.ycombinator.com/item?id=25632982</a><p>[2025-01] <a href="https://news.ycombinator.com/item?id=42575537">https://news.ycombinator.com/item?id=42575537</a>
Are posts not mentioning <i>any</i> languages? Are posts dwindling, which means Rust is actually going through the roof? Is some other language the one with rising demand?<p>Needs some kind of baseline ...
I’ve been a developer since the 90s. Mainly on the JVM, some Objective-C, and too much JavaScript garbage. I would gladly take a junior position, making junior pay if I could get an entry-level Rust gig. I know that’s not going to happen, but I sure would like it too.
I’m so confused by this graph, it seems to literally just confirm that hiring is way down across the board. The relationship between language popularity seems mostly the same (maybe less python?)
>As a proxy measure for programming language trends, let’s analyze the frequency of programming language mentions in the monthly “Ask HN: Who is hiring?” and “Ask HN: Who wants to be hired?” postings on Hacker News for a few years up until February 2025. Below are the graphs.<p>Thats terrible proxy
I write rust, c++ and python for a living. All in the same project.<p>Learning rust was painfull.<p>I tolerate c++, but I find modern c++ hard to understand. I hate gcc c++ error messages. They are the worst kind of error messages that I know.<p>Using Python for system testing is a godsend. This is where it shines .. but using poetry for package management is painful.
Those trendlines are somewhat bogus because there is a COVID induced bump [1] that is hidden here by the cutoff at 2021. Add another five years and the slopes will be less steep.<p>[1] <a href="https://hnhiring.com/trends" rel="nofollow">https://hnhiring.com/trends</a>
The chart titles say "frequency" but the y axis seems to be ambiguous about whether it's an absolute count of mentions or mentions-per-word or mentions-per-something-else.<p>The distinction is important because if it's an absolute count, I think these data would mostly just correlate with the job market overall. In that case, splitting it out by language is sort of uninteresting.
I cannot quite replicate those results myself: <a href="https://github.com/gorkaerana/hn-programming-language-trends">https://github.com/gorkaerana/hn-programming-language-trends</a>.<p>In my analysis, as a percentage of total job posts, Python jobs start to rapidly increase in 2022, even before the current AI craze started by the release of ChatGPT on November 20 2022. They peaked in 2024 and seem to have stalled since. Rust and C++ jobs (again, as a percentage of total job posts), have gone toe-to-toe since mid 2023.<p>That being said, many improvements to be done to my analysis, as it relies on simple word match.
How is the hiring experience? I'm considering Rust (or Elixir, or Go) for the next project at work. We'd obviously need to be able to hire teammates or replacements (as I'm the only current employee with tons of experience beyond C#).
If im reading this right this counts the absolute number of mentions rather than relative.<p>Seeing as the number of job postings is rapidly decreasing (and presumably "Who is hiring" is increasing), the 'rising' trends here are greatly confounded.
has anyone noticed how the hiring vs hired look like complement to each-other?<p>i.e. python is needed -> people learn it -> and then no need anymore -> lots of people with python looking for job..