balder1991

It doesn’t really solve it as a slight shift in the prompt can have totally unpredictable results anyway. And if your prompt is always exactly the same, you’d just cache it and bypass the LLM anyway.

What would really be useful is a very similar prompt should always give a very very similar result.

At least it used to be true.

Even assuming it’s not malicious, the script can mess up your environment configuration.

Not sure that’s even possible with ChatGPT embedding your chat history in the prompts to try to give more personal answers.

I think it would be better to just use Docker/Podman at this point?

Exactly, it’s really weird to see all this people claiming these wonderful things about LLMs. Maybe it’s really just different levels of amazement, but I understand how LLMs work, I actually use ChatGPT quite a bit for certain things (searching, asking some stuff I know it can find online, discuss ideas or questions I have etc.).

But all the times I tried using LLMs to help me coding, the best it performs is when I give it a sample code (more or less isolated) and ask it for a certain modification that I want.

More often than not, it does make seemingly random mistakes and I have to be looking at the details to see if there’s something I didn’t catch, so the smallest scope there better.

If I ask for something more complex or more broad, it’s almost certain it will make many things completely wrong.

At some point, it’s such a hard work to detail exactly what you want with all context that it’s better to just do it yourself, cause you’re writing a wall of text to have a one time thing.

But anyway, I guess I remain waiting. Waiting until FreeBSD catches up with Linux, because it should be easy, right? The code is there in the Linux kernel, just tell an agent to port it to FreeBSD.

I’m waiting for the explosion of open source software that aren’t bloated and that can run optimized, because I guess agents should be able to optimize code? I’m waiting for my operating system to get better over time instead of worse.

Instead I noticed the last move from WhatsApp was to kill the desktop app to keep a single web wrapper. I guess maintaining different codebases didn’t get cheaper with the rise of LLMs? Who knows. Now Windows releases updates that break localhost. Ever since the rise of LLMs I haven’t seen software release features any faster, or any Cambrian explosion of open source software copying old commercial leaders.

> I don’t even see usage going down

Depends on the price to use these cloud LLMs.

You’re over complicating something that is very simple. The stock market reflects people’s sentiments: greed, excitement, FOMO, despair…

A bubble doesn’t need a grand catalyst to collapse. It only needs prices to slip below the level where investors collectively decide the downside risk outweighs the upside hope. Once that threshold is crossed, selling accelerates, confidence unravels, and the fall feeds on itself.

Brazil had many times a judge punished WhatsApp by blocking it in Brazil, and all the times that happened, Telegram gained hundreds of thousands of new users.

I have this impression too, I open Mastodon and the timeline usually doesn’t have many interesting things.

Yeah we have open source models too that we can use, and it’s actually more fun than using cloud providers in my opinion.

They could, but that’s a recipe for going out of business in the current environment.

> This is a bit related to external dependencies vs build yourself (AKA reinvent the wheel). Quite often the external library, long term, causes more issues than building it yourself (assuming you _can_ build a competent implementation).

I feel like this happens mostly because simpler is better, and most of these dependencies don’t follow a good “UNIX” philosophy of modularity, being generic etc. something that you’d notice the standard libraries try to achieve.

Most of these third party dependencies are just a very specific feature that starts to add more use cases until it becomes bloated to support multiple users with slightly different needs.

Actually I’m just thinking that knowledge about how to crash Safari is valuable.

Because when you simply “read” you’re not necessarily learning. The illusion of knowledge is real, where you nod in agreement that you understood something but when it’s your turn to do it, you have no idea how to. You need to do something yourself to actually learn it, and it involves struggling, frustration, eventually insights etc.

Ironically, I don’t think the comments on this website have a good contrast, as explained in this old critique: https://www.nngroup.com/articles/low-contrast/

No, because an agent doesn’t learn, it’s just continuing a story. A kid will learn from the experience and at the end will be a different person.

And if it is true that the language is just the last step after the answer is already conceptualized, why do models perform differently in different languages? If it was just a matter of language, they’d have the same answer but just with a broken grammar, no?

The only relevant definition for them: https://techcrunch.com/2024/12/26/microsoft-and-openai-have-...

Yes, like stated in this video: https://youtu.be/COOAssGkF6I

So you went straight https://motherfuckingwebsite.com/ ?

And the new Xcode’s settings just adopted the new design, it’s awful.

Google has always been much better than the competition. Even today with their enshittification, competitors still aren’t as good.

The only thing that has changed that status quo is the rise of audiovisual media and sites closing up so that Google can’t index them, which means web search lost a lot of relevance.

I watched the video and it’s basically a “Jupyter notebook” type of app where you can type code and chat with the AI.

It’s nice cause it makes the interaction more dynamic and iterative. Honestly the “changing the answer” thing is something I always did on LM Studio when I wanted to change course. Definitely better than the limited interfaces of chatbots today, but I’m not sure it’s “revolutionary” by any means.

Still, it’s something I’d prefer until someone finds a better way to interact with LLMs. The ability to add stuff, remove stuff, move things around etc. probably help a lot when you’re creating something. It better matches the state of our minds. Also I appreciate the “using AI for learning instead of producing slop or — like many posts I’ve seen before — optimized spam.

I’m not sure what’s up with the course though. Seems more like a way to try to monetize something that wouldn’t be monetizable in any other way.

In fact I’m sure someone more knowledgeable than me could just create a Jupyter Notebook plugin that would replicate most of what this is?

I suppose that’s because people already lost faith in the current system and where things are going as it is.

This is true when there’s plenty of competition that matter. When everybody and every business is in one app, the network effect forces everyone to be there or be invisible. So the “essential infrastructure” label is kinda debatable. I suppose it’s essential for many businesses.

So your analogy should be more like there’s one big shopping mall network in the city that basically everyone has to go because certain stores are only there — and the owners bought any competitor that seemed to start becoming popular in the past so there’s no perspective of competition either.

That might be a recent phenomenon caused by the inevitable heat of the CPU getting closer and closer to its limit? Like explained in this video: https://youtu.be/AOlXmv9EiPo

My Mac Mini M2 never does any noise, even when I run FFMpeg the fans don’t spike. It just gets slightly warmer. Still, unless I’m doing these high CPU bound activities, every time I touch it it’s cold as if it was turned off, which is very different than my previous Intel one that was always either warm or super hot.

And if you don’t want to pay a subscription, VMWare also breaks no sweat since a long time ago, and it’s very polished at this point.

I recently tried VirtualBox and it’s finally catching up, seems to work without any problems but I didn’t test it enough to find out the quirks.

> They are all (AFAIK) based around android 9 or 10.

That’s not what I see when I lookup tablets. My old cheap Samsung Tab S6 Lite (2022 version) has Android 14 with security patch from May 2025.

My SO bought a cheap Xiaomi tablet this month with Android 15 and patch from August 2025.

Yeah, I think in proportions, tablets have always been a small niche — Not small in absolute numbers, but in comparison to phones. And since companies that make tablets also make phones, the phones always get all the priority.