DanHulton

I’ve caught bugs in code reviews before, but I would agree that it’s probably not the most crucial thing you should strive to do with code reviews.

That said, I think I’ve done a tremendous amount of PREVENTING bugs from being introduced via code review. Caught all kinds of weird code patterns, duplication, wrong levels of abstraction, etc. Essentially, the kinds of places that bugs like to hide and grow in.

I still think that’s valuable, and it’s the kind of thing that models generally aren’t great at, to this day. It generally relies on having a large amount of context about the project and about what’s being solved, so if you wanted to do this via automated review, you maybe COULD, but it would probably get pretty expensive, especially on larger projects.

I’m having a hard time understanding why you’re going so hard for the Laffer curve for your defence here. Like, a quick look at the Wikipedia page says that something along the lines of 70% is probably optimal for revenue generation, and I GUARANTEE you these billionaires aren’t paying anywhere close to that. So why not let’s move them significantly closer? If anything, the Laffer curve would tell us we SHOULD be significantly raising taxes on billionaires.

> "People need to understand that there's serious consequences for these types of things."

Why? Seriously, what is the point of demonstrating "serious consequences" for the crime of "military personnel mass-shooting other military personnel"? Is this a regular problem that needs to be corrected for? Do you think anyone in their right mind _doesn't_ already understand that there will be serious consequences for such an action? And if they're not in their right mind, exactly how do you think this will improve deterrence?

What an utterly boneheaded statement.

Additionally, if there’s more “knowledge” of the code and best practices and such baked into your model, it has to spend fewer tokens reasoning about it or even potentially looking it up. I think even one web search to retrieve additional information not also in the model will probably blow up any supposed benefit you’d get by a slightly more token-efficient language.

Oh wow, I had the same misread and the same gut reaction.

That is just one thing an intelligent actor can do. That is not intelligence.

I think we're both somewhat right, somewhat wrong. _Some_ peasants -- serfs -- were considered property of their lords, or well, of the land that the lord oversaw. If they ran away, they were rounded up and brought back, usually with a beating or worse. But there were also free peasants that rented said land. I didn't really mention them, because they deliberately chose to rent said land - they _could_ have tried their luck anywhere, but they _decided_ to try their luck on this lord's lands, and so, I thought, weren't really relevant to the conversation.

But it's not exactly irrelevant, either, right? It does add to the confusion to leave it out, so, my bad.

As far as the black plague, what gave even serfs so much leverage was the fact that there were so few people in all categories around anymore. Fewer peasants to work the land, so the prices of the free peasants went up commensurately. But even serfs who wants to run far away and set themselves up as free peasants elsewhere could frequently get away with it because a) fewer men on horseback to spare to chase them down, and b) the other lord they went and rented land from didn't have enough peasants to work his land either, and had a real vested interest in making sure they weren't caught and brought back to their original lord.

It might be eventually doable, as an experiment or as a flex, sure. But it's never going to come close to being cost-reasonable versus the equivalent infrastructure here on earth.

> Those peasants could have tried their luck planting a field in the middle of nowhere, far from any village.

Except not really. Peasants were considered the property of their lords. If they decided to go and “try their luck” at a neutral location and that was reported to their lord, armed men on horseback would be very likely to show up not long after, to return them back to the lord’s land, but not without a solid beating and/or maiming first.

Sometimes, certain forms of civilization just are oppressive, rather than collaborative. They tend to be out-competed by forms that are less oppressive, true, but it’s not guaranteed and not always fast and not always permanent.

I’ll even go one step further - I don’t even like saying “Artificial Intelligence”. I think even that anthropomorphizes the machine too much. I prefer “Simulated Intelligence”, and I feel like that describes what is going on much better.

We are, through this process, simulating intelligence. These models aren’t intelligent, but they can simulate it. Every simulation has a degree of fidelity, and we’re not at 100%, not even with the top models. When you think about it in those terms, I find it becomes a lot easier to keep their limitations in mind. Additionally, it becomes easier to remember that this is an algorithm that you are running, and are responsible for, not another being that you can ascribe blame to.

Every time LLM-defenders get upset that people apparently don't understand how LLMs work, why is it they _immediately_ pivot into examples and statements that demonstrate that they don't understand how _people_ work?

"you would be autocomplete too" "thoughts are just tokens" etc

You're not helping your case the way you think you are.

15% more price for 25% more battery is worth it alone. Add in that every time I adjust the volume without needing to dig out my phone, and I’ll feel even better.

$20 is a single meal at a fast food restaurant, which I’ll have forgotten about the very next day. This is a much better long term investment.

Not saying this is what’s going on, but you can ABSOLUTELY have nostalgia for a time that wasn’t your own, too.

>Starting September 14, we're permanently raising standard weekly limits in Claude Code by 25% for Pro, Max, Team, and seat-based Enterprise plans. Until then, the current 50% increase will be in place.

> Compared to today, this works out to a 17% reduction in weekly limits on Claude Code. We’re working on exciting changes that will make it feel like you’re getting more from Claude, while having more visibility and control of your usage. Can’t wait to share them.

They are both raising the limits by 25% and apparently reducing them by 17%. I think they mean to say you can do more, but what a terrible press release.

Exactly this. You still have to understand the abstraction layer the LLM is working at in order to be able to form an opinion about the quality of the code. (If you DON’T care about the quality of the code, there’s always been ways to get an extra 1000% short-term speed improvement by just not giving a shit.)

AI isn’t “a code abstraction layer” any more than Jira is, or the engineer themselves is.