I think the julia team is still relatively small. So it might be was they have to do to make meaningful progress?

Julia's window has probably passed. Pytorch, JAX, Numba etc are eating its lunch. Which is sad because they are all pretty hacky stuff compared to what Julia could be.

Sadly i think you maybe right. With support of one the big 3 (meta,google or MSFT),it will be hard for Julia to create and sustain significant mind share. Wich is really a shame because from technical design standpoint Julia is just brilliant.

Adding to that modular mojo, and i think we might have a repeat of DART vs typescript...

As much as I love proper foundation and language design, people must accept a certain dose of realism, 80/20 approach.

JAX is the least hacky of them and my personal favorite, but ultimately I have to agree. I'm a huge Julia stan but they simply don't have the institutional support that Meta and Google can provide.

Also a lot of people left early on because their concerns and use cases were dismissed as "holding it wrong". Thus what's left of Julia audience is just the one they catered for.

The big selling point from the getgo was that Julia is general purpose. It's now over 10 years and it's still about as general purpose as MATLAB (although on a totally different level design-wise).

A big smell from the getgo was 1-based indexing. I appreciate that it's more familiar from some branches of math and physics (and MATLAB), and it's something I could live with, but more or less all general purpose languages index from zero. And there's a good reason for that. And that makes interoperability trickier.

https://www.cs.utexas.edu/users/EWD/transcriptions/EWD08xx/E...

Julia doesn't even have upper-exclusive ranges — `0:0` has length 1. (The length-0 sequence starting at 0 is, naturally, `0:-1`.) It fails on both counts!

Lol sorry but if you read it, Dijkstra's argument for 0-based indexing is literally that it looks prettier... I really wonder if anyone who has posted that link ever read the actual argument and thought about it critically.

He gives various reasons and practical examples for it.

One can give analogous reasons (literally his main reason is it looks nicer lol) and practical examples that looks nicer in 1-based indexing too. It's a terrible argument that somehow got memed over way too long a time. Probably because no one needs to read the text, just post the link and you're done.

I honestly do not understand comments like this with regard to an open source project. The language and its libraries are rather hackable as is with rather powerful packages. While I may not agree with every decision, I do not complain that someone is not implementing my favorite feature. I just do it myself.

I frankly would be concerned if the main contributors were not focused.