Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Show HN: The Mog Programming Language

moglang.org

Post reply on HN
Print view View original

81–90 of 91 posts

Re: Show HN: The Mog Programming Language

#81

Post by pedrovhb »

Very interesting, thanks for sharing.

I'm curious about the focus on the language design and features with regards to agent orchestration vs. being a general purpose/ML architecture oriented language. The headline examples go "Agent hook", "Async HTTP with retry", and then "FFT on tensors", and that last one seems different from the others. It's easy to imagine Mog being the backbone of agent coordination in a project using more standard languages, so I imagined that would be its role; but then I'd expect primitives/abstractions to be more geared towards this role specifically. For instance, a rich subprocess interface with special handling of stdin/stderr and maybe process interaction and lifecycle is something I'd expect to see before tensors and math-y stuff. Is the goal for Mog to ultimately be a general purpose language designed for LLMs to write, or one meant for agentic harnesses and orchestration/integration?

Re: Show HN: The Mog Programming Language

#82

Post by codethief »

Hi Ted, congratulations on shipping this! You clearly put a lot of thought into this!

> If the host wants the Mog program to be able to do I/O, it has to supply the functions that the Mog program will call.

Have you considered implementing this in terms of algebraic effects? After all, controlling side effects is not only useful at the boundary between host and agent-generated code but also within any given code base to separate pure from non-pure code. I suspect that functional code will be much easier to understand for LLMs at scale because of state locality. (I actually suspect the same for humans but it's hard to obtain good data to answer this.)

Re: Show HN: The Mog Programming Language

#83

Post by gnabgib »

zeendo wrote: post #80
dang wrote: post #5

Earlier quoted context omitted.

If you keep running these bots we will ban your main account as well. Please stop now. https://news.ycombinator.com/newsguidelines.html#generated

I'm asking this with probably half curiosity and half rhetorical intent but....why even give a warning?

They've been a member for 16 years, better for them to change their ways than be lost forever?

If you stepped over a line you didn't know.. wouldn't you prefer a kind reminder rather than a sudden loss of access?

Re: Show HN: The Mog Programming Language

#84

Post by andreybaskov »

reitzensteinm wrote: post #53
andreybaskov wrote: post #7

Earlier quoted context omitted.

It's a legitimate question to ask about any new language post AI - given there is no training dataset, any other language would work better with AI. The bigger problem is maintainability over the long term, Deno is built by Node.js creator and is maintained for half a decade now, that's hard to compete with. In a way it's much more about social trust rather than particular syntax.

Coding is a verifiable domain, so I think you actually have it backwards on that first point. We can now synthesize Stack Overflow sized datasets for an arbitrary new language, and use those to train LLMs to understand it. It's expensive of course, but if a new language is genuinely better for LLMs to write and understand, that would not be an issue.

Synthetic codebases are certainly an option, but if top models remain closed I don’t see them building datasets for every new language.

Re: Show HN: The Mog Programming Language

#85

Post by andreybaskov »

embedding-shape wrote: post #19
andreybaskov wrote: post #7

Earlier quoted context omitted.

It's a legitimate question to ask about any new language post AI - given there is no training dataset, any other language would work better with AI. The bigger problem is maintainability over the long term, Deno is built by Node.js creator and is maintained for half a decade now, that's hard to compete with. In a way it's much more about social trust rather than particular syntax.

> given there is no training dataset, any other language would work better with AI. I guess it depends on what "would work better" really means, but I don't think it's always a given. I've made my own languages, there is no available training set on exactly those, but AI with a prompt can figure out how to effectively use them as much as any other language, it seems to me. I guess it helps that most languages are mor…

In basic scenarios agree it’s possible. I’ve similarly toyed with building a small language and LLM picks it up. But that’s an additional context and it’s never felt like it worked “out of the box” vs say Typescript. Just saying that it’s an adoption barrier established languages don’t have.

Re: Show HN: The Mog Programming Language

#86

Post by reitzensteinm »

andreybaskov wrote: post #84
reitzensteinm wrote: post #53

Earlier quoted context omitted.

Coding is a verifiable domain, so I think you actually have it backwards on that first point. We can now synthesize Stack Overflow sized datasets for an arbitrary new language, and use those to train LLMs to understand it. It's expensive of course, but if a new language is genuinely better for LLMs to write and understand, that would not be an issue.

Synthetic codebases are certainly an option, but if top models remain closed I don’t see them building datasets for every new language.

It's all about relative difficulty. It's not trivial to convince LLM vendors to include your pet new language in their internal synthetic datasets, and you can build your own and publish it but it'll be fiddly and expensive.

But compared to the immense amount of effort that goes into convincing a critical mass of humans to learn and write about your new language, and using _that_ material to train an LLM, I think it's fair to say things have gotten easier, not harder.

Re: Show HN: The Mog Programming Language

#88

Post by 1flei »

I think the general idea is cool, and a lot designs make sense to me. However, due to the lack of the training set, I may wonder whether the AI-spec language design is that helpful for an agent to use the language compared with those languages with extensive training data set (e.g. Python, Javascripts etc.)

Re: Show HN: The Mog Programming Language

#89

Post by robertwer »

Besides the aspect of self improvement I’m impressed with the ability to put a language specification into 3,200 token. This is like a micro-kernel which extends itself.

Now I wonder, shall we as humans still be able to read and review this code or is the sandbox + permission model as last line of defense sufficient?

The 3,200 token specification is not a formal one, I wonder whether having it formalized into a real grammar would help LLMs to generate code?

Re: Show HN: The Mog Programming Language

#90

Post by toniantunovi »

This is a fascinating approach, solving the problem at the language level. The capability-based permission model is elegant.

A complementary angle we've been exploring with LucidShark (lucidshark.com) is attacking the same problem from the workflow layer rather than the language layer: instead of constraining what the LLM can write, you run SAST, SCA, and linting automatically after every generation step, before anything touches CI or production.

The nice thing about that approach is it works with existing languages today — Python, TypeScript, Go, etc. and plugs directly into Claude Code or Cursor as a pre-commit gate. The downside is it's catching issues after generation vs. preventing them structurally like Mog aims to.

I suspect the long-term solution is both layers: safer languages for greenfield AI-native projects + robust static analysis for the 99% of existing codebases where you can't change the language.

Post reply on HN