Which is still utterly absurd, because it implies that a harder working dev would be spending more time chatting to a bot.

I agree with the premise of “simple but hard”. However, I still want to underscore that large areas of the ocean will at any given time be covered in clouds or fog. Sure, once you find the ship the first time, you’ve narrowed your search radius significantly, but a ship that can move at 30 knots can move around 1500 nautical miles (2800 km) without being seen under just 48 hours of cloud cover. That means any intel on the position of a ship carrying weapons that can easily strike at ranges of 500-1000 km is fresh produce. Just a day after you spotted that ship, it can have moved almost 1500 km, and if you lose track of it under clouds during your next satellite pass, it can suddenly be 3000 km from where you last spotted it.

What this means is that the “hard” element here is significant. Even the “simple” element becomes complicated by stuff like night time and cloud cover. All this taken into account, there are very few countries in the world with enough surveillance satellites and processing capacity to actually keep a pin on a ship at sea over any significant period of time.

Wait, you didn’t have some git-based thing with an off-site upstream?

One of the worst things about this is that the person vibe coding just ends up shitting on the reviewers time. Like… you couldn’t even bother to write a real PR, and now you want me to spend time filtering your shit? Fuck off.

It’s 3.7x more lines of code that performs 2,000 times worse than the actual SQLite.

Pretty much my experience with LLM coding agents. They’ll write a bunch of stuff, and come with all kinds of arguments about why what they’re doing is in fact optimal and perfect. If you know what you’re doing, you’ll quickly find a bunch of over-complicating things and just plain pitfalls. I’ve never been able to understand the people that claim LLMs can build entire projects (the people that say stuff like “I never write my own code anymore”), since I’ve always found it to be pretty trash at anything beyond trivial tasks.

Of course, it makes sense that it’ll elaborate endlessly about how perfect its solution is, because it’s a glorified auto-complete, and there’s plenty of training data with people explaining why “solution X is better”.

but why countries?

For the same reason as personal property, just on a larger scale. You don’t want some rando to come and set up camp in your kitchen, so we draw lines on a map and say “this is your spot, you make the rules in your spot”. Similarly, we collectively don’t want some other group of randos to show up and tell us what clothes we’re allowed to wear, or that drunk-driving is suddenly allowed. So we draw lines on a map and say “this is our spot, where we make the rules”.

Of course, ideally, the whole world would agree on a reasonable set of laws and mode of administration, and countries wouldn’t be needed anymore. However, we’re pretty far off from that being possible. To put it on point: If we had no countries today, would you prefer the russian, Chinese, German or other set of laws and administration? How would you decide which to apply globally?

See, this is what I mean. I never said that. I never said anything close to that. At this stage you’re just making up opinions you want me to have so that you have something to attack.

It’s obvious that you’ve already decided that you “know” what the person you’re arguing with means and think, and then you’re attacking that instead of recognising that you’ve misunderstood them.

You can fine-grain nr. 2 even more: You can give access to e.g. modify files only in a certain sub-tree, or run only specific commands with only specific options.

A restrictive yet quite safe approach is to only permit e.g. git add, git commit, and only allow changes to files under the VC. That effectively prevents any irreversible damage, without requiring you to manually approve all the time.

You’re absolutely right. I mostly run a pretty simple local model though, so it’s not like it’s very expensive either.

Saying that it can serve the same purpose does not mean that I mean the two are equivalent in every aspect.

Just based on how you’ve responded so far it seems like you’re wilfully misinterpreting how I actually use an LLM for this purpose, especially with responses referring to LLMs causing people to commit suicide and offloading decision making or the thought process itself to an LLM.

It really seems like you’re wilfully misinterpreting what I’m writing.

That is correct. However, an LLM and a rubber duck have in common that they are inanimate objects that I can use as targets when formulating my thoughts and ideas. The LLM can also respond to things like “what part of that was unclear”, to help keep my thoughts flowing. NOTE: The point of asking an LLM “what part of that was unclear” is NOT that it has a qualified answer, but rather that it’s a completely unqualified prompt to explain a part of the process more thoroughly.

This is a very well established process: Whether you use an actual rubber duck, your dog, writing a blog post / personal memo (I do the last quite often) or explaining your problem to a friend that’s not at all in the field. The point is to have some kind of process that helps you keep your thoughts flowing and touching in on topics you might not think are crucial, thus helping you find a solution. The toddler that answers every explanation with “why?” can be ideal for this, and an LLM can emulate it quite well in a workplace environment.

Yes, absolutely, but there’s a huge span from completely removing the box to having “just” a chatbot.

For example, at my company, we’ve set up an agent that can work with certain design-files that engineers typically work with through a rather complex GUI. We’ve built a bunch of endpoints that ensures the agent can only make valid changes to the files, and that it can never delete or modify anything without approval. This saves people a bunch of time, because they can make the agent do “batch jobs” that take maybe 10 min in about 10 s. It’s not possible for this agent to mess up our database or anything like that, because all interactions it has with anything are through endpoints where we verify that files, access permissions, change logs, etc. are valid.

I think you’ve misunderstood the purpose of a rubber duck: The point is that by formulating your problems and ideas, either out loud or in writing, you can better activate your own problem solving skills. This is a very well established method for reflecting on and solving problems when you’re stuck, it’s a concept far older than chatbots, because the point isn’t the response you get, but the process of formulating your own thoughts in the first place.

Nah, you can run it in a box and limit its ability to interact with anything outside the box to certain white-listed endpoints. Depending on what you want to achieve, that can be more than safe enough.

Meh, they work well enough if you treat them as a rubber duck that responds. I’ve had an actual rubber duck on my desk for some years, but I’ve found LLM’s taking over its role lately.

I don’t use them to actually generate code. I use them as a place where I can write down my thoughts. When the LLM responds, it has likely “misunderstood” some aspect of my idea, and by reformulating myself and explaining how it works I can help myself think through what I’m doing. Previously I would argue with the rubber duck, but I have to admit that the LLM is actually slightly better for the same purpose.

I mean, there’s a good reason the first rules of firearm safety are to always treat a weapon as loaded, and to never direct the weapon at something you aren’t prepared to destroy. The key point being that you never know when some freak accident can happen with a loose pin, bad ammo, a broken spring, or just a person tripping and shaking the gun a bit too hard.

A gun should never go off by itself. You still treat it as if it can, because in the real world freak accidents happen.

You can’t effect the number of bit flips your users hardware has, but you can affect how often buggy code corrupts their memory or otherwise crashes your program.

Let’s say any app will crash about once a year on my machine due to a bit flip. If the app is crap and crashes hundreds of times for other reasons, the bit flip is irrelevant. If the app is robust enough that the bit flip accounts for 10 % of the crashes, that basically means the app is pretty much never crashing due to poor code.

But… why? Isn’t that just far more energy consuming and expensive to run? It sounds like replacing your car for a bus that sporadically stops working, even though you always drive alone.