Never bought into rust (have studied, have a (mostly AI-generated app in rust).
Wrote some Zig but Odin is even less overhead for me. I first loved Zigs built-in build system but having tried to wrap/use C libraries from both, I must say I prefer Odin.
Wrapping some sqlite 3 API’s for my first little Odin program - just because I need so little of the API that it seems easier this way - and speaking to C from Odin is a pleasure.
That is, imho, where Rust fails the most - the second part is the C++’ish approach to memory management (RAII) - that’s not how systems programming or games (I’m told) tend to work.
To each their own. I had some fun with Rust too, but for me, Odin seems the most appealing :)
That is, imho, where Rust fails the most - the second part is the C++’ish approach to memory management (RAII) - that’s not how systems programming or games (I’m told) tend to work.
Casey Muratori and Jon Blow have pushed this concept frequently.
They largely don't deeply elaborate, which is sad because I am a professional game developer who is interested in precisely presented knowledge so I can apply it to my work.
My interpretation is that games want to allocate large pools of resources, like GPU buffers, cpu memory, etc, and reuse that memory over and over. Ie: Reinitialize it.
In my experience, RAII is my preferred pattern for certain things (std::lock_guard), and you can almost certainly express the majority of these "uber game dev patterns" using RAII/smart pointers/etc, but the c++ implementation of these "uber game dev patterns" tends to be more complicated (and imo esthetically ugly) compared to really well written C.
These new languages (zig, odin, jai) appear to be attempts to improve C to allow an alternative to C++ that doesn't have the ugly baggage that C++ has.
From what I understood, their critique of RAII is twofold: coupling of allocation and initialisation, and enforcement of deallocation. The ease of use of smart pointers makes it tempting to allocate/free of temporary structures even within one single function. Given enough number of such occurrences, it kills performance by a thousand cuts. Also I remember they mentioned it’s not necessary to free memory if you’re about to close your program, because the OS will take the memory back. Obviously you need to gracefully deinitialise some things, like audio or other devices, but that’s beyond the discussion.
As on some references, Ryan Fleury did an episode on Wookash podcast on RAD debugger showing ECS like approach.
IMO, their RAII critique is a but nuanced, but because of their personality the discourse often gets polarising.
Edit: the sibling comment just proved my last point.
Casey Muratori and Jon Blow are both hyper-dogmatic "my way or the highway" types, and, crucially, neither of them has built any of the super high fidelity types of game that would require that level of optimisation. They're basically influencer types.
I think those guys both hate C++ so much that they want to dismiss everything about it instead of using all the features that work for them like most people.
My interpretation is that games want to allocate large pools of resources, like GPU buffers, cpu memory, etc, and reuse that memory over and over. Ie: Reinitialize it.
They might say this, but there isn't a good technical rationalization here since anyone can create a global data structure just as easily in C++.
> Casey Muratori and Jon Blow have pushed this concept frequently
Ah... The school of what I like to call "maximum opinions and minimal evidence". Aggressive arrogant dismissal of anything except their exact view (and for Muratori, you're also "woke" for good measure), coupled with a complete lack of _hard evidence_ to back up their views. In that regard, they're not unlike "investment advice" instagram influencers.
Games ? Many have talked about it, but many also make their games work inside of Unity and so on, so, depends on the project.
My angle is systems programming, and there, it absolutely matter. If you are performance sensitive, then you try to avoid crossing the user-space -> kernel boundary more than you have to.
Eg, ask for lots of memory, manage with arenas.
Interestingly Odin and Zig both lean into this heavily. Rust went a different route but has tried later to bolt on pluggable allocators.
> My angle is systems programming, and there, it absolutely matter. If you are performance sensitive, then you try to avoid crossing the user-space -> kernel boundary more than you have to. Eg, ask for lots of memory, manage with arenas.
This gives the misleading impression that ordinary memory allocators are materially different from arena allocators. They aren't. Both types of allocators first ask for a big block of memory from the kernel, then dole that memory out in userspace. There's no need to cross the userspace/kernel boundary more often than you need to, especially when you consider that you can replace the standard platform allocator with whatever you want.
To wit, C doesn't emphasize arena allocation anywhere near as much as Zig et al do, and yet nobody alleges that C is somehow less suitable for systems programming than these languages. Have you considered why that is? Because, for the most part, arena allocation doesn't make a difference, and in the places where it actually does make a difference, you can trivially build an arena allocator on top of the standard allocator.
I've been using Odin for about 6 months now, and to be honest, it's hard to find fault with it. I've used it for STM32 microcontroller firmware, web and desktop applications, and all are performant and compile quickly.
My one issue is (and I'm fully aware it will never happen) I do wish there was some sort of first-class solution to inheritance. I've grown to love procedural programming, but some problems really are just better solved with a more OOP approach. Just because classes exist does not mean they need to be used.
But as far as a language to "get stuff done" with as few tradeoffs as possible, Odin is about as good as I can imagine a language being.
I'd like to hear more about Odin + STM32 MCU firmware, do you have any good resources? I'm also curious how difficult it may be to using it with ESP32 (ESP-IDF) / RP2350.
Like Go and Rust, the coolest thing (and maybe only cool thing) about this language is the name.
Now that LLMs are out of the bag, I expect to see a lot of new programming languages. In such an environment, one is better off promoting what you can do with it, rather than the language itself and its quirks.
Show me what you made with this language - it will help me better understand the use case(s) and trade-offs.
I say it because I see Odin following Rust in that it is relying a lot on its name in marketing. And on these microscopic language quirks that people don't really care about en masse, things people easily work around in other languages.
Doesn't seem very useful (sales-wise) to spend so much time talking about garbage collection or whether or not a language has native tuple support - and also the branding just doesn't matter that much. Go does a better job of this (what is Go's logo? Idk either. I know Swift's though. Go is 4x more popular - both "released" around the same time by similar companies).
JavaScript has the worst name of all time yet it's the most popular language because of how heavily relied upon it is by all kinds of consumer technology.
Python also - not a great name really, still wildly popular because of its use cases (and having a high quality library ecosystem).
Turbopack was a great example of when, why, where, to use a new, novel language (Rust) in an area I'd never consider it (web dev). It also opens my mind up to Rust doing other things like that in the ecosystem where I'd never consider it before. Now I kinda know what Rust is for.
It opens up questions like "is it designed only for graphics and GUIs or is it more backend in general, is it low level systems or network" etc. then after that I'd ask whether or not it's strongly typed, compiled, and get into all those language quirks (which don't matter that much).
After seeing the game demo, I read that it wraps basically every major GL - more questions related to graphics programming but then the site keeps saying:
"The Data Oriented Language"
...on every page.
Up until now I had began to think it was a graphics oriented language, with all the GPU talk and game dev demos. But now it's a data language? And what does that even mean?
Then there's a book for sale that I keep running into. This isn't helping!
I love the name "Odin", and I like the idea of a new graphics-oriented programming language that somehow does something useful. Push me over the edge! Help me fall for this thing, it's so much like Rust in this way - seems amazing! But why? Why do I want this?
If it were me: I'd narrowly focus on 1 thing at first, something controversial and powerful - private MMO servers with good server-side physics and anti-cheat, or torrents, or running local AI models.
Build a beautiful IDE designed around your language ecosystem (that works for other languages too) and give it away for free. Optionally include a 7b LLM in the IDE to autocomplete Odin syntax to teach people the language. Get private servers for games like WoW and FF14 (illegal in most countries) online and teach people how to deploy their own. Tout the fact that no backend language in all of gaming can simulate physics to such a degree (fluid dynamics, etc.) and that no front-end language in gaming wraps all major GL providers so seamlessly in a single, easy-to-learn syntax.
Data-oriented is a programming paradigm, just like object-oriented and procedural and functional. It has nothing to do with Big Data. It's about the way things are done. It prefers something like an ECS (data-oriented) rather than a class hierarchy (object-oriented). "Graphics oriented" isn't a thing.
Also, i disagree with your point about "promoting what you can do with it, rather than the language itself and its quirks". Like, what? Every language can do everything another language can. As long as it's turing-complete and has some interface for FFI or something similar, it can do anything. You can make a full modern SaaS in C if you really wanted to, from backend to frontend. The language itself and its quirks are what would make you maybe consider not doing that (as much as I love C, that would just be stupid if your goal is anything other than fun and experimentation).
I can see all the great software and games that were made with C++. Doesn't make me wanna use it though, the language sucks.
Your paragraph about IDE and the whole name thing just seems very out of touch to me. Are you a marketing / HR / sales person perchance?
Interesting, I am thinking/expecting we will see a massive decrease in new languages. Or people might make new languages but the will not get any adoption.
A new language now has to clear the ever growing hurdle of not being in the LLM training data.
Unless the language provides an absolutely incredible technical or runtime advantage over every other language that LLMs “speak” well I think it will really struggle to gain adoption.
Additionally, a language’s qualitative benefits to human writers arguably matters less and less.
I used to live in my IDE. Now I use it maybe an hour each day even in a JVM based language. IDEs dont really matter as much anymore.
Not with a new language but novel approaches within a framework or paradigm feel outside the LLM wheelhouse.
Even if the LLM is adept at novel languages the developer still has to learn it and learning new languages now when most programming is done via a prompt-review loop feels like it has lower ROI.
I think it would be better to approach it from the other side, the priority is not to design a language for LLMs but a language more suitable for humans to think with. And not a natural language like English, which is inconsistent and allows illogical formulations, but something like Esperanto or Interlingua (Latino sine flexione). Something that is based on mathematics and logic at the bottom, like Lean, with enough abstraction layers for a person to be able to "speak" with the machine intuitively.
I don't know but I would imagine there are a lot of inefficiencies in modern languages from an LLM perspective that it could strip out, reduce token costs, improve speed etc.
An LLM-only oriented language doesn't make sense, because without human generated training data there is nothing for the model to learn from.
But if a human-oriented language were to be designed to also be better for LLMs, I think it would involve deeply expressive syntax that can succinctly but distinctly represent a very broad set of common operations. Succinct so context can be managed well, distinct so completions don't confuse one thing for another, broad so as much "reasoning" can be taken away from the LLM as possible. An anti-C. A new take on the goals of Java and Go to be languages that protect the application from the Junior Developers You're Likely To Hire.
I somewhat think it would also involve application state images ala Smalltalk. LLMs seem okay at generating small deltas. Many deltas sequenced together invites compounding error. LLM generated apps are unlikely to lead to common libraries being compentized and extracted out of the application to share with other applications; it seems like LLM code generation is already a "married to a specific project" act already. So, having a living state image might reveal some benefits by leaning into incrementally developing the application in situ, as a whole.
Having fun with this.
Never bought into rust (have studied, have a (mostly AI-generated app in rust).
Wrote some Zig but Odin is even less overhead for me. I first loved Zigs built-in build system but having tried to wrap/use C libraries from both, I must say I prefer Odin. Wrapping some sqlite 3 API’s for my first little Odin program - just because I need so little of the API that it seems easier this way - and speaking to C from Odin is a pleasure.
That is, imho, where Rust fails the most - the second part is the C++’ish approach to memory management (RAII) - that’s not how systems programming or games (I’m told) tend to work.
To each their own. I had some fun with Rust too, but for me, Odin seems the most appealing :)
Did you try Swift? Its interoperability with C (and even C++) is great IMHO.
That is, imho, where Rust fails the most - the second part is the C++’ish approach to memory management (RAII) - that’s not how systems programming or games (I’m told) tend to work.
What does this mean? Who told you that?
Casey Muratori and Jon Blow have pushed this concept frequently.
They largely don't deeply elaborate, which is sad because I am a professional game developer who is interested in precisely presented knowledge so I can apply it to my work.
My interpretation is that games want to allocate large pools of resources, like GPU buffers, cpu memory, etc, and reuse that memory over and over. Ie: Reinitialize it.
In my experience, RAII is my preferred pattern for certain things (std::lock_guard), and you can almost certainly express the majority of these "uber game dev patterns" using RAII/smart pointers/etc, but the c++ implementation of these "uber game dev patterns" tends to be more complicated (and imo esthetically ugly) compared to really well written C.
These new languages (zig, odin, jai) appear to be attempts to improve C to allow an alternative to C++ that doesn't have the ugly baggage that C++ has.
From what I understood, their critique of RAII is twofold: coupling of allocation and initialisation, and enforcement of deallocation. The ease of use of smart pointers makes it tempting to allocate/free of temporary structures even within one single function. Given enough number of such occurrences, it kills performance by a thousand cuts. Also I remember they mentioned it’s not necessary to free memory if you’re about to close your program, because the OS will take the memory back. Obviously you need to gracefully deinitialise some things, like audio or other devices, but that’s beyond the discussion.
As on some references, Ryan Fleury did an episode on Wookash podcast on RAD debugger showing ECS like approach.
IMO, their RAII critique is a but nuanced, but because of their personality the discourse often gets polarising.
Edit: the sibling comment just proved my last point.
Casey Muratori and Jon Blow are both hyper-dogmatic "my way or the highway" types, and, crucially, neither of them has built any of the super high fidelity types of game that would require that level of optimisation. They're basically influencer types.
I think those guys both hate C++ so much that they want to dismiss everything about it instead of using all the features that work for them like most people.
My interpretation is that games want to allocate large pools of resources, like GPU buffers, cpu memory, etc, and reuse that memory over and over. Ie: Reinitialize it.
They might say this, but there isn't a good technical rationalization here since anyone can create a global data structure just as easily in C++.
> Casey Muratori and Jon Blow have pushed this concept frequently
Ah... The school of what I like to call "maximum opinions and minimal evidence". Aggressive arrogant dismissal of anything except their exact view (and for Muratori, you're also "woke" for good measure), coupled with a complete lack of _hard evidence_ to back up their views. In that regard, they're not unlike "investment advice" instagram influencers.
Games ? Many have talked about it, but many also make their games work inside of Unity and so on, so, depends on the project.
My angle is systems programming, and there, it absolutely matter. If you are performance sensitive, then you try to avoid crossing the user-space -> kernel boundary more than you have to.
Eg, ask for lots of memory, manage with arenas.
Interestingly Odin and Zig both lean into this heavily. Rust went a different route but has tried later to bolt on pluggable allocators.
> My angle is systems programming, and there, it absolutely matter. If you are performance sensitive, then you try to avoid crossing the user-space -> kernel boundary more than you have to. Eg, ask for lots of memory, manage with arenas.
This gives the misleading impression that ordinary memory allocators are materially different from arena allocators. They aren't. Both types of allocators first ask for a big block of memory from the kernel, then dole that memory out in userspace. There's no need to cross the userspace/kernel boundary more often than you need to, especially when you consider that you can replace the standard platform allocator with whatever you want.
To wit, C doesn't emphasize arena allocation anywhere near as much as Zig et al do, and yet nobody alleges that C is somehow less suitable for systems programming than these languages. Have you considered why that is? Because, for the most part, arena allocation doesn't make a difference, and in the places where it actually does make a difference, you can trivially build an arena allocator on top of the standard allocator.
Yeah, that's weird. I first learned about RAII from professional game developers at gamedev.net.
I've been using Odin for about 6 months now, and to be honest, it's hard to find fault with it. I've used it for STM32 microcontroller firmware, web and desktop applications, and all are performant and compile quickly.
My one issue is (and I'm fully aware it will never happen) I do wish there was some sort of first-class solution to inheritance. I've grown to love procedural programming, but some problems really are just better solved with a more OOP approach. Just because classes exist does not mean they need to be used.
But as far as a language to "get stuff done" with as few tradeoffs as possible, Odin is about as good as I can imagine a language being.
I'd like to hear more about Odin + STM32 MCU firmware, do you have any good resources? I'm also curious how difficult it may be to using it with ESP32 (ESP-IDF) / RP2350.
This was a pretty funny video for a language launch:
https://www.youtube.com/watch?v=dLPAqXi9In0
Like Go and Rust, the coolest thing (and maybe only cool thing) about this language is the name.
Now that LLMs are out of the bag, I expect to see a lot of new programming languages. In such an environment, one is better off promoting what you can do with it, rather than the language itself and its quirks.
Show me what you made with this language - it will help me better understand the use case(s) and trade-offs.
I say it because I see Odin following Rust in that it is relying a lot on its name in marketing. And on these microscopic language quirks that people don't really care about en masse, things people easily work around in other languages.
Doesn't seem very useful (sales-wise) to spend so much time talking about garbage collection or whether or not a language has native tuple support - and also the branding just doesn't matter that much. Go does a better job of this (what is Go's logo? Idk either. I know Swift's though. Go is 4x more popular - both "released" around the same time by similar companies).
JavaScript has the worst name of all time yet it's the most popular language because of how heavily relied upon it is by all kinds of consumer technology.
Python also - not a great name really, still wildly popular because of its use cases (and having a high quality library ecosystem).
Turbopack was a great example of when, why, where, to use a new, novel language (Rust) in an area I'd never consider it (web dev). It also opens my mind up to Rust doing other things like that in the ecosystem where I'd never consider it before. Now I kinda know what Rust is for.
This is maybe the best example I've found for Odin: https://odin-lang.org/showcase/solar_storm/
It opens up questions like "is it designed only for graphics and GUIs or is it more backend in general, is it low level systems or network" etc. then after that I'd ask whether or not it's strongly typed, compiled, and get into all those language quirks (which don't matter that much).
After seeing the game demo, I read that it wraps basically every major GL - more questions related to graphics programming but then the site keeps saying:
"The Data Oriented Language"
...on every page.
Up until now I had began to think it was a graphics oriented language, with all the GPU talk and game dev demos. But now it's a data language? And what does that even mean?
Then there's a book for sale that I keep running into. This isn't helping!
I love the name "Odin", and I like the idea of a new graphics-oriented programming language that somehow does something useful. Push me over the edge! Help me fall for this thing, it's so much like Rust in this way - seems amazing! But why? Why do I want this?
If it were me: I'd narrowly focus on 1 thing at first, something controversial and powerful - private MMO servers with good server-side physics and anti-cheat, or torrents, or running local AI models.
Build a beautiful IDE designed around your language ecosystem (that works for other languages too) and give it away for free. Optionally include a 7b LLM in the IDE to autocomplete Odin syntax to teach people the language. Get private servers for games like WoW and FF14 (illegal in most countries) online and teach people how to deploy their own. Tout the fact that no backend language in all of gaming can simulate physics to such a degree (fluid dynamics, etc.) and that no front-end language in gaming wraps all major GL providers so seamlessly in a single, easy-to-learn syntax.
Make the IDE icy af
Data-oriented is a programming paradigm, just like object-oriented and procedural and functional. It has nothing to do with Big Data. It's about the way things are done. It prefers something like an ECS (data-oriented) rather than a class hierarchy (object-oriented). "Graphics oriented" isn't a thing.
Also, i disagree with your point about "promoting what you can do with it, rather than the language itself and its quirks". Like, what? Every language can do everything another language can. As long as it's turing-complete and has some interface for FFI or something similar, it can do anything. You can make a full modern SaaS in C if you really wanted to, from backend to frontend. The language itself and its quirks are what would make you maybe consider not doing that (as much as I love C, that would just be stupid if your goal is anything other than fun and experimentation).
I can see all the great software and games that were made with C++. Doesn't make me wanna use it though, the language sucks.
Your paragraph about IDE and the whole name thing just seems very out of touch to me. Are you a marketing / HR / sales person perchance?
Interesting, I am thinking/expecting we will see a massive decrease in new languages. Or people might make new languages but the will not get any adoption.
A new language now has to clear the ever growing hurdle of not being in the LLM training data.
Unless the language provides an absolutely incredible technical or runtime advantage over every other language that LLMs “speak” well I think it will really struggle to gain adoption.
Additionally, a language’s qualitative benefits to human writers arguably matters less and less.
I used to live in my IDE. Now I use it maybe an hour each day even in a JVM based language. IDEs dont really matter as much anymore.
> A new language now has to clear the ever growing hurdle of not being in the LLM training data.
I found this to be far less of a prblen than I thought it would be.
Do you have practical experience?
Not with a new language but novel approaches within a framework or paradigm feel outside the LLM wheelhouse.
Even if the LLM is adept at novel languages the developer still has to learn it and learning new languages now when most programming is done via a prompt-review loop feels like it has lower ROI.
I wonder when we'll see new languages created specifically with LLM's in mind.
I think it would be better to approach it from the other side, the priority is not to design a language for LLMs but a language more suitable for humans to think with. And not a natural language like English, which is inconsistent and allows illogical formulations, but something like Esperanto or Interlingua (Latino sine flexione). Something that is based on mathematics and logic at the bottom, like Lean, with enough abstraction layers for a person to be able to "speak" with the machine intuitively.
What would that look like?
I don't know but I would imagine there are a lot of inefficiencies in modern languages from an LLM perspective that it could strip out, reduce token costs, improve speed etc.
An LLM-only oriented language doesn't make sense, because without human generated training data there is nothing for the model to learn from.
But if a human-oriented language were to be designed to also be better for LLMs, I think it would involve deeply expressive syntax that can succinctly but distinctly represent a very broad set of common operations. Succinct so context can be managed well, distinct so completions don't confuse one thing for another, broad so as much "reasoning" can be taken away from the LLM as possible. An anti-C. A new take on the goals of Java and Go to be languages that protect the application from the Junior Developers You're Likely To Hire.
I somewhat think it would also involve application state images ala Smalltalk. LLMs seem okay at generating small deltas. Many deltas sequenced together invites compounding error. LLM generated apps are unlikely to lead to common libraries being compentized and extracted out of the application to share with other applications; it seems like LLM code generation is already a "married to a specific project" act already. So, having a living state image might reveal some benefits by leaning into incrementally developing the application in situ, as a whole.