> A diff can show what changed in the artifact, but it cannot explain which requirement demanded the change, which constraint shaped it, or which tradeoff caused one structure to be chosen over another.

That's not true... diffs would be traceable to commits and PRs, which in turn are traceable to the tickets. And then there would be tests. With all that, it would be trivial to understand the whys.

You need both the business requirements and the code. One can't replace the other. If you attempt to describe technical requirements precisely, you'll inevitably end up writing the code, at very least, a pseudocode.

As for regenerating the deleted code out of business requirements alone, that won't work cleanly most of the time. Because there are technical constraints and technical debt.

So what did you say about version contol?

> By regenerable, I mean: if you delete a component, you can recreate it from stored intent (requirements, constraints, and decisions) with the same behavior and integration guarantees.

That statement just isn't true. And, as such, you need to keep track of the end result... _what_ was generated. The why is also important, but not sufficient.

Also, and unrelated, the "reject whitespace" part bothered me. It's perfectly acceptable to have whitespace in an email address.

What the author actually wants is ADRs: https://github.com/joelparkerhenderson/architecture-decision...

That’s a way of being able to version control requirements.

I’m sorry but it feels like I got hit in the head when I read this, it’s so bad. For decades, people have been dreaming of making software where you can just write the specification and don’t have to actually get your hands dirty with implementation.

1. AI doesn’t solve that problem.

2. If it did, then the specification would be the code.

Diffs of pure code never really represented decisions and reasoning of humans very well in the first place. We always had human programmers who would check code in that just did stuff without really explaining what the code was supposed to do, what properties it was supposed to have, why the author chose to write it that way, etc.

AI doesn’t change that. It just introduces new systems which can, like humans, write unexplained, shitty code. Your review process is supposed to catch this. You just need more review now, compared to previously.

You capture decisions and specifications in the comments, test cases, documentation, etc. Yeah, it can be a bit messy because your specifications aren’t captured nice and neat as the only thing in your code base. But this is because that futuristic, Star Trek dream of just giving the computer broad, high-level directives is still a dream. The AI does not reliably reimplement specifications, so we check in the output.

The compiler does reliably reimplement functionally identical assembly, so that’s why we don’t check in the assembly output of compilers. Compilers are getting higher and higher level, and we’re getting a broader range of compiler tools to work with, but AI are just a different category of tool and we work with them differently.

The lossy aspect mentioned in the article just sounds like you forgot to write comments or a README. simple fix

The only way to do this is with a mathematically precise and unambiguous stored intent, isn't it? And then aren't we just taking source code?

The way we solve the why/what separation (at minfx.ai) is by having a top-level PLAN.md document for why the commit was built, as well as regenerating README.md files on the paths to every touched file in the commit. Admittedly, this still leans more into the "what" rather than "why". I will need to think about this more, hmm.

This helps us to keep it well-documented and LLM-token efficient at the same time. What also helps is Rust forces you into a reasonable code structure with its pub/private modules, so things are naturally more encapsulated, which helps the documentation as well.

The only practical obstacle is:

> Non-deterministic generators may produce different code from identical intent graphs.

This would not be an obstacle if you restrict to using a single version of a local LLM, turn off all nondeterminism and record the initial seed. But for now, the kinds of frontier LLMs that are useful as coding agents run on Someone Else's box, meaning they can produce different outcomes each time you run them -- and even if they promise not to change them, I can see no way to verify this promise.

People need to remember how good it feels to do precise work when the time comes!

Looking at individual line changes produced by AI is definitely difficult. And going one step higher to version control makes sense.

We're not really there yet though, as the generated code currently still needs a lot of human checks.

Side thoughts: this requires the code to be modularized really well. It makes me think that when designing a system, you could imagine a world where multiple agents discuss changes. Each agent would be responsible for a sub system (component, service, module, function), and they would chat about the format of the api that works best for all agents, etc. It would be like SmallTalk at the agent level.

Would love people's thoughts on this: https://0xmmo.notion.site/Preventing-agent-doom-loops-with-p...

I think commenters here identified many of the issues we would face with it today, but thinking of a future where LLMs are indeed writing virtually all code and very fast, ideas like these are interesting. Our current tooling (version control, testing, etc.) will certainly need to adapt if this future comes to pass.

would not be unfamiliar to mechanical engineers who work with CAD. The ‘Histories’ (successive line-by-line drawing operations - align to spline of such-and-such dimensions, put a bevel here, put a hole there) in many CAD tools are known to be a reflection of design intent moreso than the final 3D model that the operations ultimately produce.

If you use an LLM and agents to regenerate code, a minor change in the "specification" may result in huge changes to the code. Even if it's just due to forcing regeneration. OK, got that.

But there may be no "specification", just an ongoing discussion with an agentic system. "We don't write code any more, we just yell at the agents." Even if the entire sequence of events has been captured, it might not be very useful. It's like having a transcript of a design meeting.

There's a real question as to what the static reference of the design should be. Or what it should look like. This is going to be difficult.

Not sure what stops you from doing that just right now.

Deltas are just an implementation detail, and thinking of Git as diffing is specifically shunned in introductions to Git versioning.

1. https://dspy.ai/

CONGRATULATIONS: you have just 'invented' documentation, specifically a CHANGE_LOG.