I'm Rewriting IBM SPSS Statistics in 100% Safe Rust, Alone

Does safety matter for this kind of software? It’s mostly internal tooling used by researchers and statisticians. I get it rust is cool but safety isn’t the prime concern here I think.

Sorry but the problem is not primarily software, it’s community, teaching and standardization.

We have R/tidyverse which does 90% of everything and has an amazing community and procedures eg. around documentation. For everything else you have excel or python. So I don’t see the use case.

But why?

[flagged]

I'm (well, my LLM is) Rewriting IBM SPSS Statistics in 100% Safe Rust, Alone (for a given value of alone).

You even let your LLM write the README.

Look, we nearly all use LLMs these days, but the phrasing of "I am... ...alone" is particularly weird when it's obviously not just you, it's you with LLMs.

Maybe title it as "A rewrite of IBM SPSS Statistics in Rust with no unsafe" - the 100% safe claim is something an LLM makes, it's not something a reasonable human would say, as it implies the Rust is totally safe, whereas you're trying to communicate that your Rust code uses no unsafe blocks, and these are two different things.