Hacker News Reader

Comment by cranium | Hacker News Reader

I loved Elm as soon as I had a running app that I could refactor and be (reasonably) sure it would work once the compiler was satisfied.

However, I don't like the way the project is run. It's one Benevolent Dictator that has a particular vision for Elm and he's not really sharing with the community. For context, the latest Elm version (0.19.1) was released in October 2019. It kind of feels like a nice tree house left to the elements because the person grew out of interest, but kept a lock on the door.

View on HN

Have there been any efforts to fork it?

0.19 also prohibited interaction with non-Elm code which I recall a controversy about. But can't seem to find out if anyone forked 0.18 because of that.

Not quite a fork, but Gren is a descendent: https://gren-lang.org/news/220530_first_release/

Gren is a fork. Their first commit is the head of Elm 0.19.1 right?

The faq in the book says this about it:

> Gren started as a fork of Elm. This is mostly considered to be an implementation detail, a way to speed up initial development.

> It's not a goal of Gren to replace, or stay compatible in any way with, Elm.

A little vague but I read it as they're planning on diverging strongly and immediately with a clean break. Which doesn't make it not a fork I guess, but isn't what I normally think of when someone says a project is a fork of another.

Yeah I guess "stay compatible with" and "cover the same use cases" are the connotations for a fork and it really doesn't hit those, fair enough.

Fork means "to the originating project's source code and develop it further in the context and governance of a new project" in my understanding.

So Gren is a fork.

I don't think it's a fork, but prominent Elm developer/author Richard Feldman has been working with and hyping up Roc: https://www.roc-lang.org

He in fact works on Roc full time now! As does one other Roc contributor whose name escapes me. Exciting times.

It is my understanding that roc aims to "bring Elm to the backend", so it does not at all look like a substitute for Elm.

Ah my mistake. I wonder if Roc plans to have first-class support/integrations for Elm frontends. Or maybe they’ll stay completely agnostic and just mention Elm in docs and examples?

From what I know from skimming the docs and listening to Richard's podcast, their focus currently is on achieving great compiled performance. Plus, from what I can tell, Richard is still a fan of Elm, so I'd guess he thinks there's space for Roc to contribute to the frontend (that last part is pure speculation though).

Meant to say "... there isn't space ..."

Interesting! From a surface language level it feels a lot like F#, but with purity and a managed effects system--which is something that I've kinda wished for/had to build patterns for in F# codebases... I'll have to keep an eye on it, thanks! (reading through the FAQ, there are some interesting design decisions! I appreciate the pragmatic focus. Didn't realize that it was non-curried lang until the FAQ)

It's such an easy restriction to work around that it just doesn't come up a ton on code that's always been written for 0.19. I imagine it was a monstrous hassle if you depended on it with a big 0.18 codebase, but the new JS interop (ports but more importantly custom elements) are very nice to work with tbh.

There's a third inspired-by-elm language Derw (after Gren and Roc that are mentioned elsefork) but I don't know much about it.

Indeed the releases are less frequent and Evan works in batches to avoid breaking changes every year. But he didn't left, he still works on private branches and on security patches when needed. Yesterday he gave this presentation:

https://gotoaarhus.com/2023/sessions/2529/elm-on-the-backend

Why has his work not turned into a public bug fix update in 2.5 years?

This feels a lot like Patrick Rothfuss on his Kingkiller Chronicles. The creator has a fan base he doesn't want to disappoint and who will defend his honor as long as he can give them any evidence he's actually making progress, but all the evidence suggests he's actually lost interest or doesn't know where to go from here.

Not that there's anything wrong with losing interest in a hobby project. The problem with Elm is that it was pitched as more than a hobby project but then held so close to Evan's chest that no one else has been able to take over now that he's not maintaining it.

Too many people are concerned with the language's creator, how much he is working on, how many people love or hate him, what his motives are, what he wrote in a comment four years ago, and what he had for breakfast this morning.

These comments are always the same: "Elm is great, but..." Then they dig up some drama about its creator with a years-old post. If the language is great, then maybe they should use it and stop fretting about the person who made it.

I'm not talking about any specific drama, and I didn't go digging anything up. I'm just talking about what's actually happened to the language: it has been 2.5 years since a single bug was fixed. That suggests that the creator is (contrary to the person I replied to) not working on it any more.

I couldn't care less about the creator or his drama, but I am a bit sad that a language with so much potential has been dropped instead of being handed over to the community.

I am talking about this:

>The creator has a fan base he doesn't want to disappoint and who will defend his honor as long as he can give them any evidence he's actually making progress

So yes, it's fair to say that you do care about the creator and his drama.

You're reading way more emotion into that statement than I meant to write. I meant it strictly literally and apathetically: both creators have fans who will jump in to reassure onlookers that things are actually moving forward.

Knowing who the BDFL is and how a project is run are very important considerations for long term projects. It's not something you can just discard as trivial. Doing an AoC in a language you know nothing about is totally fine, but not in any serious environment.

Seriously...the person who controls everything about the project, irrelevant and immaterial! Sure, like I'll buy that.

The only way to watch this is on a mobile app??

Yeah, partially because of how the project is run I chose to go with elmish for my own project.

Reply on news.ycombinator.com