It is simply not feasible for humans to keep the entirety of large projects in their heads all at once. When working on a project, one is constantly forgetting and relearning code along the way. The situation is even worse for projects where multiple people work together and no one knows what everyone else was thinking. We look at the code and try to build up sufficient context in our heads to make the necessary changes. When we misjudge that context or misunderstand existing code, we end up making mistakes.
It's a made up excuse because no one has ever tried doing that. Instead the core activity of software engineering (and engineering in general) is to break down things into modular components, ensure that they work well, produce specs, then assemble the result. Which is why no one read the source code for open(2) system call before using it. Instead they read the manpage and only when there's an issue there's a need to compare the specs with the implementation.So it's a matter of communication between humans. If person A is responsible for module 1 and person B's module 2 depends on the former. It's up to A to provide a proper spec to B. It's not up to B to learn everything about module 1's implementation and doing so is kinda foolish.
So something like
Any new code that is added is therefore always done with limited understanding.
Is a red flag. The codebase and workflow should be that new additions should deal with a limited context (as in one PR shouldn't touch unrelated things) at at time and the contributors should be able to come with a full understanding of what they are doing. If they cannot, it's either a failure of the codebase, the docs, the workflows, or the contributors.Meanwhile, anybody who's maintained a large scale software project knows that even with the best practices, problems can still be introduced into the codebase. Nobody is arguing against limiting context or doing as much as possible to ensure the contribution is doing the right thing. Yet, the idea that you can have full understanding of how the feature will interact with every edge case is frankly laughable.
The notion that you can architect a perfect codebase where context and scope of the effect of any change is always completely clear is idealistic in the extreme. Of course, we should strive to keep the code quality as high as possible, nobody is arguing against that. The reality is that large projects are complex, and nobody has come up with a way to write code in the way where it's possible to perfectly control the scope of the effects of any one change. Every single large scale project out there has bugs, and regressions no matter how much the authors strive to make it perfect.
That's not what I'm saying. What I'm saying is that the contributor should make a reasonable attempt at producing a bug free patch. And it's up to the project's leaders to make it easy to do so. And one of the easiest way for the latter to do so is producing design docs and isolating modules as much as possible so that it's efficient to get the needed understanding.
So both parties each have its own responsibilities into making quality contributions happen. I'm not against LLM usage. What I'm against is making the project difficult for human understanding because human understanding is limited or whatever. That's why I point out the README and the Makefile.
Personal attacks for someone pointing out some issues with the LLM apologetics. Nice.
disclaimer: 27 year professional in this field and I mostly agree with OP.
Also, I have no idea what any of this has to do with LLMs. Nothing discussed here was LLM related in any way. Nor am I doing any apologetics for anything here. I'm stating basic facts that anybody with even a modicum of intellectual integrity would acknowledge.
But if you have any actual counterpoints to what I said then feel free to share them.
https://github.com/jrouwe/joltphysics
https://news.ycombinator.com/item?id=50202762
https://jank-lang.org/
I believe what you're trying to say there, with "drop in replacement", is that jolt emulates common Java interop forms in Clojure JVM such that code written for Clojure JVM doesn't need to be ported to run on jolt. This a compelling feature which other dialects are tackling as well.
What I would mention as a key difference between jank and jolt, aside from the C++ story and native project management story, is that jolt is very much AI coded and jank is not. jolt doesn't disclose this on its website or Github readme, but it's something which clearly matters to a lot of people.
Jolt aims to faithfully implement the totality of features that current JVM Clojure implements along with shim for Java surface to facilitate existing code to run.
Meanwhile, I don't see the tooling used to develop the language as the key difference in the language itself. I understand this is important for some people, and that's fine. However, that has little to do with the language itself , its features, or its overall functionality. So, for me personally, that is not an interesting distinction. What matters to me is what I can use the language for, and what code it is able to run.
Protocols, records, and Clojure structs will definitely be a part of jank. They have been lower priority than improving some of the more commonly used Clojure features, smoothing out the C++ interop edges, and providing excellent tooling. Still, I have been working toward supporting them already. jank will have them before its production release.
> Jolt aims to faithfully implement the totality of features that current JVM Clojure implements along with shim for Java surface to facilitate existing code to run.
Yes, this is a neat approach. Making dialects maximally portable across existing Clojure JVM code is compelling and I think jolt's momentum, due to your usage of AI, has allowed you to pioneer an approach here faster than other dialects have done. In the coming couple of years, I can see us arriving at a more wide-spread approach of registering these shims so that the large corpus of Clojure JVM code can be used in various dialects without requiring library authors to add reader conditionals everywhere.
However, "Clojure" these days doesn't only mean Clojure JVM. We don't yet have a specification for Clojure, but reasonably the intersection of Clojure JVM, ClojureScript, and ClojureCLR is the target compatibility point for most dialects. For example, ClojureScript doesn't have JVM interop, emulated or otherwise. But nobody would ever say that ClojureScript is not Clojure. I discussed this exact point in my Conj talk this year, since there seems to be some outdated reasoning that "Clojure" means Clojure JVM. It does not.
> Meanwhile, I don't see the tooling used to develop the language as the key difference in the language itself. I understand this is important for some people, and that's fine. However, that has little to do with the language itself , its features, or its overall functionality. So, for me personally, that is not an interesting distinction. What matters to me is what I can use the language for, and what code it is able to run.
Yep, I understand this philosophy. I can appreciate it. The reason why I bring up disclosure is because you and I both know that this is divisive. There are people who will see jolt as a front-runner, due to its AI usage, and there are people who would rather abstain or at least wait until the dust settles. Providing people the necessary information on the label so that they can make this decision themselves is the respectful, transparent thing to do. I need to do better here as well and I have already publicly stated I will update the jank repo and website with clear disclosures for how I have used AI, which is mainly for troubleshooting and bug fixes. This will be coming with the beta release.
Jeaye, you’ve been doing amazing work on jank over the years, please keep going!
I've been occasionally checking up on jank for a few years now, and thanks for resisting the temptation to go slop and doing things the proper way.