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.

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”.

Yeah look if we really are seeing diagnoses suddenly rise, and it’s not just “a better telescope”, maybe it is worth considering exploring environmental causes, diagnostic criteria, societal tolerance of certain traits etc. That’s fair.

Thanks for making an effort to understand what I’m trying to get at here. Honestly, this is like my primary point and I appreciate that you’re recognising it. It can get tiresome to discuss with people that never seem to understand that you have good intentions, thank you for seeing mine.

What concerns me about your responses is that “investigating why diagnoses are increasing” is used all the time to cast doubt on ADHD itself.

And I absolutely agree that that’s a big problem. My point is concerning when the same sentence is used in the positive, constructive, sense that this is clearly something we as a society need to look more closely at, because something is very obviously wrong on either an environmental or societal level if a large fraction of the population needs medication in order to function properly.

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.

I honestly have the impression that we agree on pretty much all points here but that we’re talking past each other. I agree to pretty much everything you’re saying, and I’m all for helping as many people as possible live as good lives as possible.

What I’m trying to say is basically that problematising the large volume of (and increase in) psychological diagnoses can be valid, and doesn’t have to be founded in trying to downplay those diagnoses. To take a very concrete example: Kids that are disposed to growing very short or tall can be offered growth (blocking) hormones, such that they grow to a “more normal” height. Today, very few kids are offered, or take, these hormones. Now, let’s say some area suddenly saw a rapid increase where 20 % of kids needed growth hormones to grow to “ordinary” height. I would say that we need to figure out what has happened: Is there something about the environment that has caused stunted growth to become ver common? Has the window for what is “normal” gotten narrower?

Of course, in this example, it’s very was to compare to historical records of human height. The same isn’t true for mental disorders. That doesn’t mean the same discussion isn’t worth having- at its core, this is a discussion about how we can make society as good as possible for as many as possible. That also involves discussing what should be treated as a disorder that disproportionately makes people’s life objectively worse, and what is within the “normal” range that we should rather build society around accepting.

You cannot equate ADHD and spectrum mental conditions with disease.

I agree, the only way I meant to compare them is that we diagnose and treat both with medication.

We as a society need to accept that and stop thinking these are diseases that need to be “fixed”.

I also agree 100 % with this, and it’s part of what I’m trying to get at with my “option a”. As of today, there are regions where over 20 % of the population are diagnosed with, and treated for, ADHD. At that point, I’m asking the question if we’re creating a problem by treating something that appears to be within the spectrum of how “normal people just are” as a problem that needs to be fixed. My point is exactly what you’re saying here: If a large fraction of the population has this “problem” that needs to be “fixed”, haven’t we just gotten to a point where we have a too narrow definition of what is “normal” and “healthy” human behaviour? Shouldn’t we in that case rather be looking at how we can structure our society in such a way that a larger span of the population is capable of functioning in it without medication, rather that trying to force everyone to conform to the same, ever narrowing, mould?

I’m in no way an anti-vaxxer or anti-science (I’m a researcher myself). I still think it can be justified to look closely at the large increase in, and volume of, various mental disorders. First of all: There’s no doubt that a lot more people are being diagnosed due to better diagnosis tools.

However, a major difference between psychological and somatic illness is that the divide between sick and healthy is (typically) a lot sharper in the latter case. Either you have an injury or infection, or you don’t, and we can measure that. In the case of e.g. depression or ADHD, there’s a much wider gray zone from e.g. “healthy person having a bad day” to “clinically depressed”.

The point I’m getting at is this: When a certain percentage of the population is diagnosed with a disorder, you have to ask whether we’ve started diagnosing ordinary human existence as a disease. Alternatively, you have to start looking at a systematic level for why an enormous portion of the population has a certain disorder. Where that limit should be is an open question, but I would argue that when something like 10-20 % of the population has a specific disorder, we’re no longer just looking at individual cases of disease but rather at (a) the possibility that the criteria for diagnosis are two wide, so we’re catching “healthy” people with it, or (b) we have a society-level problem (e.g. an epidemic).

I know of areas with ADHD-rates around 20 %. For a somatic illness, we would never let that kind of infection rate pass without taking a closer look at what’s going on at the societal level.

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.