Ask HN: Why don't people see Rust replacing Ruby/Python etc.? | Hacker News Reader

Ask HN: Why don't people see Rust replacing Ruby/Python etc.?

3 pointsasattarmd4 comments

Rust is quite expressive. Isn't that where Ruby/Python shine?

I'm assuming the context of the question is "People say they see it replacing C++ all the time".

People actively want a C++ replacement, and a lot of the features of Rust specifically target the pain points of C++. The pain points are painful enough, and the differences in Rust are soothing enough, that people are willing to invest in Rust as a future replacement.

People don't actively want a Ruby or Python replacement; they're good enough for now. There isn't much in Rust that a Ruby or Python programmer will look at and say "that problem has caused me issues for years". It doesn't offer enough over what they already have to invest in such a young language.

If Rust succeeds, I wouldn't be surprised to see it at least sitting alongside Ruby and Python, but by that point the ecosystem will have built up enough around it that you're not losing so much by switching.

In what time frame? It takes a decade or more for one language to replace another. Look how long the Python 2 to 3 transition has taken, and those two languages are close enough that it's possible to build complex code using just their intersection.

One big part is inertia. There's a lot of existing code that works in Ruby or Python. There's also a lot more people who know how to use them, so it's easier to ask for or find help about a specific problem.

A related part is the number of existing libraries. If I want to do cheminformatics (my field) in Rust, I would basically have to write the entire underlying packages by hand. While for Python there's 6 or so packages available to me, each taking at least 10 years of developer time to build.

The main point for any programming language out there isn't the language itself, no matter how beautiful or awkward it is.

What counts is the the ecosystem around the language, more precisely: the availability of modules/packages/frameworks/libraries.

Ruby has RubyGems with an impressive set of stuff for nearly everything thinkable, and a bunch of frameworks for different stuff, like Rails or Sinatra or Shoes or ... .

Javascript/Node has NPM, with a staggering amount of modules for anything (Atwoods Law...), plus all the browser goodness (bower anyone?), full stack stuff like meteor and unique projects like node-webkit and phone gap to build anything "appy" in no time.

Of course there is also Perl's CPAN, the all-mighty JVM, ...

Rust has... an alpha release of a package management system? (saw it here on HN just now).

I would add a point to the ecosystem: the availability of developers in your area. If you want to develop a new product based on X, but you can't even assemble a team, you are going to have a hard time. This is a big contributor to the inertia of any technology. (That could be used as a competitive advantage, but I am talking about the broad adoption of a new technology)

Reply on news.ycombinator.com