OxCaml - a set of extensions to the OCaml programming language.

https://oxcaml.org/

284•lairv•17h ago

Comments

legobmw99•17h ago

The first feature that originated in this fork to be upstreamed is labeled tuples, which will be in OCaml 5.4:

https://github.com/ocaml/ocaml/pull/13498

https://discuss.ocaml.org/t/first-alpha-release-of-ocaml-5-4...

debugnik•16h ago

Immutable arrays were ported from this fork as well, and merged for 5.4; although with different syntax I think.

andrepd•16h ago

Anonymous labeled structs and enums are some of my top wished-for features in programming languages! For instance, in Rust you can define labelled and unlabelled (i.e. tuple) structs

    struct Foo(i32, i32);
    struct Bar{sum: i32, product: i32}

But you can only e.g. return from functions an anonymous tuple, not an anonymous labelled struct

    fn can() -> (i32, i32)
    fn cant() -> {sum: i32, product: i32}

tialaramex•16h ago

Hmm. Let me first check that I've understood what you care about

    struct First(this: i8, that: i64)
    struct Second(this: i8, that: i8)
    struct Third(that: i64, this: i8)
    struct Fourth(this: i8, that: i64)
    struct Fifth(some: i8, other: i64)

You want First and Fourth as the same type, but Second and Third are different - how about Fifth?

I see that this is different from Rust's existing product types, in which First and Fourth are always different types.

Second though, can you give me some examples where I'd want this? I can't say I have ever wished I had this, but that might be a different experience.

cAtte_•15h ago

they're not asking for a structural typing overhaul, just a way to make ad-hoc anonymous types with named fields and pass them around. a lot of times with tuple return types you're left wondering what that random `usize` is supposed to represent, so having names for it would be very convenient. i don't see why, under the hood, it couldn't just be implemented the exact same way as current tuple return types

zozbot234•15h ago

> they're not asking for a structural typing overhaul, just a way to make ad-hoc anonymous types with named fields and pass them around.

And their point is that the two boil down to the same thing, especially in a non-trivial program. If switching field positions around changes their semantics, tuples may well the most sensible choice. As for "what that random usize is supposed to represent" that's something that can be addressed with in-code documentation, which Rust has great support for.

tialaramex•14h ago

Also, if it's not just a "random usize" then you should use the new type paradigm. In a language like Rust that's not quite as smooth as it could possibly be, but it's transparent to the machine code. Rust's Option<OwnedFd> is the same machine code as C's int file descriptor, but the same ergonomics as a fancy Haskell type. We can't accidentally confuse "None, there isn't a file descriptor" for an actual file descriptor as we so easily could in C, nor can we mistakenly do arithmetic with file descriptors - which is nonsense but would work (hilarity ensues) in C.

If these aren't "random" usizes but FileSizes or ColumnNumbers or SocketTimeouts then say so and the confusion is eliminated.

munificent•14h ago

In Dart, we merged tuples and records into a single construct. A record can have positional fields, named fields, or both. A record type can appear anywhere a type annotation is allowed. So in Dart these are both fine:

    (int, int) can() => (1, 2);
    ({int sum, int product}) alsoCan() => (sum: 1, product: 2);
    (int, {int remainder}) evenThis() => (1, remainder: 2);

The curly braces in the record type annotation distinguish the named fields from the positional ones. I don't love the syntax, but it's consistent with function parameter lists where the curly braces delimit the named parameters.

https://dart.dev/language/records

afiori•12h ago

How do you distinguish a tuple with both positional and named fields from a tuple that has a record as a field

Like how do you write the type of (1, {sum:2}) ? Is it different from (1 , sum :2)?

munificent•10h ago

The syntax is a little funny for mostly historical reasons. The curly braces are only part of the record type syntax. There's no ambiguity there because curly braces aren't used for anything else in type annotations (well, except for named parameters inside a function type's parameter list, but that's a different part of the grammar).

int_19h•11h ago

It's interesting that languages which start with purely nominal structs tend to acquire some form of structurally typed records in the long run. E.g. C# has always had (nominally typed) structs, then .NET added (structurally typed) tuples, and then eventually the language added (still structurally typed) tuples with named items on top of that.

munk-a•11h ago

PHP has it all!

I think the main dividing line here is whether you want to lean into strict typing or whether you prefer a more loose typing structure. The extremes of both (where, for instance, the length of an array is part of its type definition or there are not contractual guarantees about data) are both awful. I think the level of type strictness you desire as a product is probably best dictated by team and project size (which you'll note changes over the lifetime of the product) with a lack of typing making it much easier to prototype early code while extremely strict typing can serve as a strong code contract in a large codebase where no one person can still comprehend the entirety of it.

It's a constant push and pull of conflicting motivations.

aseipp•15h ago

Yeah, pretty excited about this one even though it seems minor. A paper and talk by the author of this particular feature from ML2024, too:

- https://www.youtube.com/watch?v=WM7ZVne8eQE

- https://tyconmismatch.com/papers/ml2024_labeled_tuples.pdf

munchler•15h ago

> Because sum:int * product:int is a different type from product:int * sum:int, the use of a labeled tuple in this example prevents us from accidentally returning the pair in the wrong order, or mixing up the order of the initial values.

Hmm, I think I like F#'s anonymous records better than this. For example, {| product = 6; sum = 5 |}. The order of the fields doesn't matter, since the value is not a tuple.

rwmj•14h ago

Isn't that just the same as the ordinary OCaml { product = 6; sum = 5 } (with a very slightly different syntax)?

munchler•14h ago

The difference between { … } and {| … |} is that the latter’s type is anonymous, so it doesn’t have to be declared ahead of time.

rwmj•13h ago

Oh I see, good point. I'm wondering how this is represented internally. Fields alphabetically? I've also desired extensible anonymous structs (with defaults) from time to time, but implementing that would involve some kind of global analysis I suppose.

reycharles•3h ago

One reason why they're not the same is because the memory representation is different (sort of). This will break FFIs if you allow reordering the tuple arbitrarily.

IshKebab•17h ago

So this is "oxidized" because it tries to achieve the same features as Rust (e.g. "fearless concurrency" is mentioned, and avoiding GC)... Not because it actually uses Rust in any way right? Slightly confusing.

debugnik•17h ago

Correct, Jane Street has been publishing a series of blog posts titled "Oxidizing OCaml" for a while.

zozbot234•16h ago

Rust will probably become usable with custom tracing GCs (which is helpful if you're dealing with general graph-like data but still want the highest performance as far as practical) way before this effort reaches genuine feature parity with Rust. Not seeing much of a point in this, unless they perhaps intend to focus on the lowest-hanging fruit and have big O(x)Caml codebases that they care about.

greener_grass•15h ago

OxCaml takes a different approach to encoding locality to Rust. Rust (arguably) overburdens the type-system with this information whilst in OxCaml this is orthogonal to the return types. In that sense it's a bit like algebraic effects. Personally I'm quite bullish on OCaml these days.

aseipp•15h ago

> Not seeing much of a point in this

OCaml is a good language and these extensions are very welcome to existing OCaml programmers and programs, as well as many of the other extensions Jane Street has added. I don't understand what you mean here.

zozbot234•15h ago

> > Not seeing much of a point in this

> ...I don't understand what you mean here.

Yeah it seems there was a mistake there, it looks like my comment got cut off while you were reading it. Try reloading the page maybe?

aseipp•15h ago

Your comment remains completely stupid regardless of how much of it is in the quote. OxCaml's goal is to be upstreamed. "I don't see the point in this unless people have programs using OCaml" and it's like, yeah, that's the point of adding features to a computer program. Any computer program. Because people already use the computer program and it will be useful to them.

aseipp•15h ago

Yes, they've used this terminology for a while, even the recent technical paper on this effort was titled "Oxidizing OCaml with Modal Memory Management", though the word "oxidize" itself is never actually referenced or defined in the paper. A bit strange, I agree, though it's kind of catchy I admit.

john-h-k•14h ago

It's ironic of course because Rust (the language) is named after the fungus called Rust, rather than iron-oxide

avsm•17h ago

If anyone's trying out the new opam switch, I found it helpful to use:

env OCAMLPARAM="alert=-unsafe_multidomain,_," opam install cohttp-lwt-unix

Because alerts are promoted to errors, they break existing package installs unnecessarily. The OCAMLPARAM environment variable just forces that alert to be disabled and allows the package installation to continue.

Lyngbakr•17h ago

The Janet Street folks, who created this, also did an interesting episode[0] of their podcast where they discuss performance considerations when working with OCaml. What I was curious about was applying a GC language to a use case that must have extremely low latency. It seems like an important consideration, as a GC pause in the middle of high-frequency trading could be problematic.

[0] https://signalsandthreads.com/performance-engineering-on-har...

enricozb•15h ago

Haven't looked at the link, but I think for a scenario like trading where there are market open and close times, you can just disable the GC, and restart the program after market close.

spooneybarger•13h ago

It is a common strategy.

rauljara•15h ago

GC compactions were indeed a problem for a number of systems. The trading systems in general had a policy of not allocating after startup. JS has a library, called "Zero" that provides a host of non-allocating ways of doing things.

jitl•14h ago

Couldn’t find this after 6 seconds of googling, link?

jallmann•14h ago

The linked podcast episode mentions it.

notnullorvoid•13h ago

There's no mention of a library called zero, or even JavaScript.

garbthetill•12h ago

Im assuming the JS refers to Janes street

notnullorvoid•10h ago

That makes sense, I guess I've got web tunnel vision.

sheepscreek•6h ago

I was bit by the same spider that gave you web tunnel vision. In any case, I find OCaml too esoteric for my taste. F# is softer and feels more..modern perhaps? But I don’t think GC can be avoided in dotnet.

debugnik•50m ago

You can avoid GC in hot loops in F# with value-types, explicit inlining, and mutability.

Mutability may not result in very idiomatic code however, although it can often be wrapped with a functional API (e.g. parser combinators).

jallmann•11h ago

> This is what I like to call a dialect of OCaml. We speak in sometimes and sometimes we gently say it’s zero alloc OCaml. And the most notable thing about it, it tries to avoid touching the garbage collector ...

mardifoufs•15h ago

You just let the garbage accumulate and collect it whenever markets are closed. In most cases, whenever you need ultra low latency in trading, you usually have very well defined time constraints (market open/close).

Maybe it's different for markets that are always open (crypto?) but most HFT happens during regular market hours.

dmkolobov•15h ago

Is that really a viable solution for a timeframe of 6+ hours?

logicchains•14h ago

You can just add more RAM until it is viable.

jitl•14h ago

Sure, if you know how much you allocate per minute (and don’t exceed your budget) you just buy enough RAM and it’s fine.

mayoff•14h ago

(this comment was off topic, sorry)

robertlagrant•14h ago

Is this relevant to OCaml?

mayoff•13h ago

ha ha oops I got confused

ackfoobar•13h ago

This will decrease performance because of reduced locality. Maybe increased jitter because of TLB misses.

jitl•6h ago

Compared to what, running a garbage collector?

dmkolobov•1h ago

Probably? Locality becomes fairly important at scale. That’s why there’s a strong preference for array-based data structures in high-performance code.

If I was them I’d be using OCaml to build up functional “kernels” which could be run in a way that requires zero allocation. Then you dispatch requests to these kernels and let the fast modern generational GC clean up the minor cost of dispatching: most of the work happens in the zero-allocation kernels.

spooneybarger•13h ago

Yes. It is a very common design pattern within banks for systems that only need to run during market hours.

iainctduncan•12h ago

I talk about doing this in an audio context and get met with raised eyebrows, I'd love some references on others doing it, if anyone can share!

mardifoufs•7h ago

I think it is, but to be clear I think (from my very limited experience, just a couple of years before leaving finance, and the people with more experience that I've talked with) that c++ is still a lot more common than any GC language (typically java, since OCaml is even rarer). So it is possible, and some firms seem to take that approach, but I'm not sure exactly how besides turning off GC or very specific GC tuning.

Here is a JVM project I saw a few years back, I'm not sure how successful the creators are but they seem to use it in actual production. It's super rare to get even a glimpse at HFT infra from the outside so it's still useful.

https://github.com/OpenHFT

amw-zero•5h ago

Are you aware of how many allocations the average program executes in the span of a couple of minutes? Where do you propose all of that memory lives in a way that doesn’t prevent the application from running?

great_wubwub•11h ago

*Jane Street

esafak•7h ago

It's a great name for a competitor :)

AdieuToLogic•4h ago

> What I was curious about was applying a GC language to a use case that must have extremely low latency. It seems like an important consideration, as a GC pause in the middle of high-frequency trading could be problematic.

Regarding a run-time environment using garbage collection in general, not OCaml specifically, GC pauses can be minimized with parallel collection algorithms such as found in the JVM[0]. They do not provide hard guarantees however, so over-provisioning system RAM may also be needed in order to achieve required system performance.

Another more complex approach is to over-provision the servers such that each can drop out of the available pool for a short time, thus allowing "offline GC." This involves collaboration between request routers and other servers, so may not be worth the effort if a deployment can financially support over-provisioning servers such that there is always an idle CPU available for parallel GC on each.

0 - https://docs.oracle.com/en/java/javase/17/gctuning/parallel-...

pjmlp•1h ago

Java is like C and C++, there isn't the one implementation.

So if you want hard guarantees, you reach out to real time JVM implementations like the commercial ones from PTC and Aicas.

pjmlp•1h ago

The problem is not a GC language per se, people keep putting all GC languages in the same basket.

The real issue is being a GC language, without support for explicit manipulation of stack and value types.

Want a GC language, with productivity of GC languages, with the knobs to do low level systems coding?

Cedar, Oberon language family, Modula-3, D, Nim, Eiffel, C#, F#, Swift, Go.

jaennaet•23m ago

Does Go have much in the way of GC knobs? It didn't some years ago, but I haven't kept up on latest developments

debugnik•16h ago

I wasn't aware that this fork supported SIMD! Between this, unboxed types and the local mode with explicit stack allocation, OxCaml almost entirely replaces my past interest in F#; this could actually become usable for gamedev and similar consumer scenarios if it also supported Windows.

TheNumbat•16h ago

Yeah, this would be great! Currently only 128-bit SSE/NEON is working but AVX is coming very soon. There's also nothing blocking Windows, but it will require some work. (I added the SIMD support in OxCaml)

aseipp•15h ago

FWIW, the "Get OxCaml" page actually says that SIMD on ARM isn't supported yet. If it actually works it would be worth removing that from the known issues list https://oxcaml.org/get-oxcaml/

TheNumbat•15h ago

Indeed, it says that because we don't have a library of NEON intrinsics (like ocaml_simd_sse) yet, but the extension itself works.

aseipp•15h ago

Ah, that clears it up a bit. Thanks! Looking forward to all of this. (For the interested readers, here's the SSE library: https://github.com/janestreet/ocaml_simd/tree/with-extension...)

debugnik•15h ago

Cool to hear there aren't any technical blockers to add Windows support! You just convinced me into giving OxCaml a try for a hobby project. 128-bit SSE is likely to be enough for my use case and target specs.

avsm•14h ago

David Allsopp had an oxcaml branch compiling on Windows a few months ago, so it’s in the queue…

croes•16h ago

So OxCaml is the extension of the extension of a dialect of ML.

Can’t wait for the next level

munchler•14h ago

Can I interest you in F#?

jasperry•9h ago

I had a similar thought to this, but then I thought, who is worse: programmers who keep bloating up existing languages with new features, or programmers who create yet another new language to add to the already crowded field? (I'm in that latter category.)

I guess programmers are just genetically incapable of leaving their tools the way they are.

debugnik•2h ago

This did actually halt my hobby programming language project, whose features overlap a lot and is advancing slowly anyway.

At the very least I'll give OxCaml a try to compare. Best case I drop mine and use this, worst case I learn what works and what doesn't.

anthk•14h ago

OcaML's micro cousin:

http://t3x.org/mlite/index.html

iLoveOncall•13h ago

The sunk cost fallacy at Jane Street is strong.

ackfoobar•13h ago

Maybe it's sunk cost fallacy for them. But without them there wouldn't be a language that gets me 80/20 benefit/effort of Rust.

fuzztester•10h ago

what is the sunk cost fallacy in the Jane Street case?

ackfoobar•10h ago

I said "maybe" as granting the root comment's premise - I don't believe that.

abathologist•12h ago

What are you talking about?

wk_end•9h ago

The implication here, I think, is that "Jane Street has foolishly invested heavily into Ocaml and, rather than sensibly change course and migrate away, they continue to invest heavily into it in the hopes that one day the investment will pay off."

Which is a ridiculous take: Jane Street has done extraordinarily well for themselves, possibly at least in part because of their investment in Ocaml, and any sort of migration would be a huge undertaking, and an absurd and unnecessary risk.

abathologist•8h ago

Ah yeah, that is absurd, given how successful they have been, and how much they attribute that success to their technical decisions.

People love to sulk in their parochial pits, I guess?

Thanks for clarifying :)

ng12•8h ago

Maybe Jane Street succeeds because of the people who are good at finance in spite of the people who like OCaml.

debugnik•1h ago

I do think Jane Street fell into a sunk cost fallacy when they first stuck to OCaml. But after having invested further into an OCaml/OxCaml codebase, teams and tooling, switching tech stacks has lost most of its value.

Also, their efforts benefit me greatly as an OCaml (now maybe OxCaml) user, so I'm glad they took that path.

umanwizard•12h ago

Wait til you hear about PHP at Meta...

agumonkey•11h ago

did they talked about it publicly ? i stopped following around the hiphop vm era

toolslive•1h ago

You can also argue it's "Stockholm syndrome". Anyway,it's funny how opinions differ: I still consider Rust to be a poor man's OCaml.

phplovesong•3m ago

Why? They have one of the largest Ocaml codebases in the world. It has clearly work out VERY well for them, in terms of productivity and pure ROI in features and revenue.

Ocaml has so much good going on, even being a less known language.

I assume you are a rust fanboy, and i want to see just how long your smirk holds when you compile a 20M LOC project that takes 45 minutes.

dingdingdang•10h ago

Probably spoilt here but being used to the excellent vscode plugin (well vscodium in my case) for Golang but... any plans to integrate with vscode ecosystem? Makes setup so straightforward!

jasperry•9h ago

The OCaml vscode plugin seems to have already integrated a lot of new syntaxes (dune, menhir, reason), so if OxCaml gains traction it should only be a matter of time.

(can't really speak for that myself, though, I use emacs)

clancon•5h ago

If you follow the installation instructions on oxcaml.org, you’ll get a patched Merlin with LSP support etc. It’s not perfect, but does mostly work out of the box with VSCode and the OCaml Platform extension.

debugnik•2h ago

Error squiggles and ocamlformat stopped working for me after the first time I restarted VS Code, compared to regular OCaml. They do work fine through the Dune CLI though.

bhamta•4h ago

What are the chances that they are releasing this so that LLMs can index this information for free and they can use public models in their codebase rather than finetuning public models?

debugnik•47m ago

Given how poor LLMs are at regular OCaml, which has so much more training data that OxCaml, probably none. An MCP for docs would have been more productive for that purpose.

reverseblade2•4h ago

Just use F# for sake

debugnik•2h ago

I actually see this as an escape opportunity for F#-ers. I love the language, but the runtime and C# ecosystem are not only holding it back from having cleaner semantics (until the C# team sees the value in them), they've been actively kicking it down the ladder with breaking changes and new incompatible APIs.

reverseblade2•1h ago

F# is mostly liberated from .net thanks to Fable. Although the compiler strictly is .net

debugnik•31m ago

Fable is the complete opposite direction to .NET and OxCaml, though. Nice for people doing web I guess (I'd rather do TypeScript than deal with interop), but the JS runtime and ecosystem are pretty much useless for my use cases.

Maybe a pure WASM target could actually liberate F#.

Push is Faster [using std:cpp 2025] [video]

Solving Super Agentic Planning

Saab achieves AI milestone with Gripen E

Free Digital Revolutionary zine: 24 hours only

Debian Package to Comply with the Spanish National Security Framework (ENS)

How multiplication is defined in Peano arithmetic

AI Slop YouTube Channels Are Telling Google About Motorcycles That Don't Exist

Pizza delivery monitor alerts to secret Israel attack

KDE Plasma finally gets Wayland PiP

Link Aggregators

Nvidia and Deutsche Telekom Partner to Advance Germany's Sovereign AI

Strace Tips for Better Debugging

Leading EU Telcos Build AI Infrastructure with Nvidia for Regional Enterprises

Orange Business and Toshiba to launch commercial quantum-safe service in France

Everything old is new agAIn

Roof Painters Reflect on Their Role Fighting Climate Change

Hypershell: A Type-Level DSL for Shell-Scripting in Rust Powered by Context-Gen

Show HN: Plugin-jj (fish abbreviations plugin for jj)

Show HN: I built a library of free Google Sheets and Excel templates

NoVNC, the open-source VNC browser client

Just shipped LandscapioAI – AI-generated landscape plans with ±15% cost estimate

What's New in Chromium 137.0.7151.104

Meta-analysis of three different notions of software complexity

Dance Captcha

Night and Fog (2003)

Facebook's Relay for React v20 released

RAG Anti-Patterns

M3-Fleu Instrumental

Local AI for Private, Uncensored Chat on iPhone, iPad, and Mac

Bamboo-Copter

Push is Faster [using std:cpp 2025] [video]

Solving Super Agentic Planning

Saab achieves AI milestone with Gripen E

Free Digital Revolutionary zine: 24 hours only

Debian Package to Comply with the Spanish National Security Framework (ENS)

How multiplication is defined in Peano arithmetic

AI Slop YouTube Channels Are Telling Google About Motorcycles That Don't Exist

Pizza delivery monitor alerts to secret Israel attack

KDE Plasma finally gets Wayland PiP

Link Aggregators

Nvidia and Deutsche Telekom Partner to Advance Germany's Sovereign AI

Strace Tips for Better Debugging

Leading EU Telcos Build AI Infrastructure with Nvidia for Regional Enterprises

Orange Business and Toshiba to launch commercial quantum-safe service in France

Everything old is new agAIn

Roof Painters Reflect on Their Role Fighting Climate Change

Hypershell: A Type-Level DSL for Shell-Scripting in Rust Powered by Context-Gen

Show HN: Plugin-jj (fish abbreviations plugin for jj)

Show HN: I built a library of free Google Sheets and Excel templates

NoVNC, the open-source VNC browser client

Just shipped LandscapioAI – AI-generated landscape plans with ±15% cost estimate

What's New in Chromium 137.0.7151.104

Meta-analysis of three different notions of software complexity

Dance Captcha

Night and Fog (2003)

Facebook's Relay for React v20 released

RAG Anti-Patterns

M3-Fleu Instrumental

Local AI for Private, Uncensored Chat on iPhone, iPad, and Mac

Bamboo-Copter

OxCaml - a set of extensions to the OCaml programming language.

Comments