Hacker News Reader

Comment by thumbuddy | Hacker News Reader

I think Julia is one of those researcher languages like Haskell so it's probably just someone having fun, writing a paper, or whatever.

View on HN

Julia is still niche, but it's definitely used in production?

The NY Fed have their main model of the economy in Julia (ported from Matlab)

this https://juliahub.com/case-studies/ny-fed/ ?

This:

https://frbny-dsge.github.io/DSGE.jl/latest/

Sounds like research to me? I dunno...

I don't agree with this. Julia does not have new ideas the same way Haskell does, and Julia has done a lot of work to be practical, like having a good way of calling Python and R code from early on in its life. If Julia has done anything new, it's probably related to being fast, but that's not exactly a novel idea.

Julia is unique in using multiple dispatch pervasively. It did not invent multiple dispatch, but it certainly popularized it.

Sure it does.

Have you seen the type system used to generically dispatch matrices to GPUs or cpus.

Or the dispatching to give you auto-differentiation?

Maybe it's not your definition of "new ideas", but they are really useful and original.

How come you never see anything practical made in Julia in production somewhere that isn't a pile of research mathematics? I personally can't think of a single tool or company. It's not like it's a young language either? I have seen dozens of niche papers written using it though.

Being fair, I have seen dozens of not hundreds of products and mass deployed projects written in similar languages. Even Haskell made its way to a lot of desktops in the Linux world, and Haskell is super niche research territory imo.

The clinical analysis for the COVID-19 vaccines was pretty practical and that was Pumas used by Moderna. There's a JuliaCon talk about that from the Director Head of Pharmacometrics at Moderna. There's Formula 1 usage, there's a some satellites running Julia onboard, etc. Some are documented here https://juliahub.com/case-studies/ though some people post what they are doing on the Discourse and Slack and so those end up being the most comprehensive source. Of course you can always argue whether 30 public examples is enough, or 50, or etc. but it's far from 0. And of course, most examples are never for public release.

Fifteen years ish and there's maybe ten-thirty production use cases that kinda can't be verified because they all sound like research projects doesn't sound super great to me. No offense, but I do wonder if all of those cases could have been accomplished more easily/sustainably with another tool and if they exist simply because someone was selling the language to a company as a contractor as a form of vendor lock in or something.

I guess I'll be curious once I install some software or an OS and see it brings in Julia in as a dependency or something. Otherwise I worry it's Matlab 2.0 with less of a mindshare...

Reply on news.ycombinator.com