royal__

I'm gonna make the claim that all of these are results of how someone uses an agent (poorly), not a result of the agents themselves.

This was my gut reaction as well but I think this is just a poor abstract. What they actually did appears to be much deeper than reinventing agents.md

Thx claude

Abstracts are not meant to be entertaining or memorable; they're meant to be highly functional. When you're reading a billion papers, a good abstract is important because it lets you get the gist of the paper and whether or not it's relevant to what you're researching in about 10 seconds. Make your abstract extremely clear and well structured about what you did, why, and your biggest finding, and the worry about giving it a good hook.

How much scope do you give it? I find it's easier to stay in control when you give it scope down the chunks of work you give it.

These kinds of updates always have this romantic scenario of someone having Claude develop a presentation or something on their way to work between multiple devices, which actually feels a little sad and does not align with what happens in my life at all.

Advertisers want this.

It's hard to take this seriously when it has wonky Claude website design. Rewriting emacs in Rust also feels like a nonsensical task.

I've been seeing lots of demos like this, especially with games. Yes, they are impressive, but I'm surprised nobody is talking about some of the things I've noticed with GPT-6: the code it writes, by default, is surprisingly messy and obviously unmaintainable. It's also very...not human. No human would write code like it does. I get the sense that the optimization of these models on benchmarks is causing their behavior to morph into a very brute-force type of approach. It makes you think: when code is suddenly very cheap, does good style and organization even matter as much? Or were those just important for humans? (I think they still do, for the record).

I agree that this is an example of interesting and good AI usage. The person who made it obviously put thought into it. Unfortunately a lot of vibe coded sites do not have that same level of care

No Mans Sky has sold over 10 million copies. I'll let you do the math. They are still basically an indie company as well.

This is the answer.

Why does it have to be black and white. You can use AI without vibe coding.

Woah hold on buddy, that take is way too reasonable for an internet forum!

I legitimately thought this was real for about half of it and it made it very engaging, well done.

See, I think that's cool. It's like a hobby or craft that you're invested in. It feels different from taking lame photos with your smart phone just for the cred.

The simple reason that a lot of the games made without a custom engine don't sell is because they're simply turds. Games don't sell based on what engine they were made in. Games sell because they're good games.

The biggest thing I've learned from doing stuff like this is that there are no shortcuts. At some point or another, to truly learn something deeply, you've got to dig in to the boring details and do things the hard way. LLMs can help with this...but I find it's usually tempting to try and just offload the boring stuff to them, which doesn't work.

Bleh. My eyes glaze over reading slop like this. It's a neat idea to have LLMs argue but something about the writing style just feels gross now...

OP seems to be wrestling with this feeling that LLMs are simultaneously incredibly useful and also leave a very hollow feeling inside, like it's cheating. The redefinition is how we choose to move forward with hobbies and art when we know these tools are available.

We continue to grapple with AI and how it's impacting our lives by trying to find comfort in comparisons to the past, or to other things that we do. But the reality is, there's never been anything quite like this before, and it's forcing us to redefine how we value work, art, craftsmanship, and output. Sometimes, those redefinitions are extremely uncomfortable.

I don't even know what Zig is but I've seen this topic come up so many times on this site that I'm starting to think the people who are actually doing this are unsure themselves whether it's a good idea or not.

To be fair they do acknowledge this directly in the abstract. Kinda hard to have a good heuristic for this stuff, what would you propose?

Thanks mate, I think you hit it on the head: the hardest problems in tech were never really tech problems. They're often people and market problems, and throwing more agents at the wall isn't going to magically solve them.

> I prefer CLI over GUI but I also prefer to have a life over making my life harder.

I think this is actually a really profound statement. Many people seem to get caught up in trying to take the most efficient path when you could take a somewhat less efficient path that requires much less time. It can become a form of yak-shaving.

Now, don't get me wrong: there's absolutely a benefit in investing time to learn efficient methods. But life is also short, and you've gotta choose where you invest.

The problem I have with this kind of approach is 1) it emphasizes scaling up a much as possible, which I don't believe is necessarily the most valuable thing, and 2) I really don't want my job to be band aiding agent problems, because it's like herding cats and there will never be an end to it. I'd rather just...get hands on and be involved in the code I am working to create.

A lot of folks are defending McMansions here, but let me provide one reason why they might be objectively bad architecture: they seek to cram all of the bullet points that home buyers want into one thing, without regard for good quality design. For example, high vaulted ceilings that throw off the proportions of everything else. And then what happens is they're built quickly and poorly with poor materials. McMansions are kind of a form of enshittification. Also, when you build them all over the place, like in Utah, you hurt a whole class of people who want live in a place but don't want to have to deal with living in houses like that.

Since you asked...I've gone to the effort to pull out the parts of the article that I think show it:

"That’s the part most benchmarks can’t see, and it’s what this post is about." Classic "it's not x, it's x", shows up in various forms throughout the article.

"To me, this is the most fascinating finding from this entire experiment - we saw very clear alignment tax being paid by certain models, which directly impacted their performance in this zero-sum game." - Usage of em dash. Now, yes, there's nothing wrong with using em dashes. But this feels like a weird place to use one. Also I counted at least 6 other emdashes in this article. Most people do not use em dashes that often.

"and a memory system that kept doubling down on what worked without second-guessing or doubting itself." - Doubling down is a classic Claudism.

"I want to be careful here..." - "wanting to be careful here" is another classic Claudism.

"The same game world, completely different results when in a different “task”." - "same X, completely different X" is another common one from Claude, as proofed by the repeated pattern later down: "These models were all given the same rules, same game world, and same tools, but each of them approached the game on a personality-level that is completely different from each other."

"It begs the question" - author used this twice in the article.

I'm guessing the author wrote a draft and then had Claude spruce it up a lot. I could be wrong and I'd be happy to be proven otherwise.

That all makes sense. I wasn't trying to attack Emacs or defend Neovim, for the record. I liked Emacs and didn't have any problems with it (except some window jankiness). I was mostly just curious about the ecosystem.

The big reason I switched is because a lot of the big features of Emacs (org mode, magit, "living" in Emacs, advanced text manipulation, the extreme extensibility of the software) were things that sound really amazing on paper, but in practice I just don't really need/use/care about, and that's just my preferences. But once again, Emacs is cool and I totally respect what it can do.

For the record, btw, id DID download your app and I am going to spend some time using it. Other commenters here are talking about how spaced repetition isn't what's important, which is true in the same sense that you can't just use Duolingo to learn a language. But that doesn't make it useless.