haolez

That was always my experience as well. I commend the effort, but I have never been able to use it as a daily driver. Gave up like 10 years ago or more.

Yeah, I thought it was just me - it hallucinates more than Flash, Flash-lite and Pro.

It's literally called _soft_ware (vs. hardware).

I don't want to bore my HackerNews fellows, so I'm sorry in advance, but wasn't there a lot of drama in the Nix community? I've read some time ago that a private company was buying parts of its infrastructure (or something like that) and forcing some changes upstream and people were not happy about it.

I'm interested in NixOS, but I'm at an age where I just want things to stay stable and change only when they really have to. I already have a lot of big mandatory changes from life itself, I don't need more :)

I'm writing this in good faith. I'm genuinely curious and interested in trying it again.

But the chain of thought as text can be elusive, no? It might not be related with the actual "reasoning" of the model. There was a paper about this.

I feel that a desktop like this would just occupy parts of my cognition for no good reason. Not sure.

If you use an LLM to write assembly, it is officially your (optimizing?) compiler :)

I'm not saying that Big Tech won't respect it. I'm saying that, if privacy and intellectual property is a major concern for your company, you can't trust a checkbox - you have to keep the data in-house.

Most companies won't have this level of risk aversion (maybe defence?).

It's naive to think that the check boxes in the other providers will protect your data.

Nice work!

Are you sure this isn't just a custom model that is API-compatible with OpenAI?

Better metric: your LLM enhanced version cannot be larger than the prompt. Otherwise, you are just reducing your signal to noise ratio.

I think the problem is the ease that you can create believable low effort articles and even software.

It's harder to filter for the good stuff when compared to previous hypes, like Blockchain and JavaScript frameworks.

That's an interesting viewpoint, but then, will the allocator's performance never matter for any use case that is not "wrong"? It doesn't feel right.

Yep. Early in my career I developed a simple driver for a CAN interface to be used in a medical device, which was my assigned project.

The same code was used in another project of this same company. The project was a missile. It felt weird.

Eventually solving for cost is a much easier problem than solving coding.

I couldn't adapt to the fact that, when I click, I have to be mindful of not moving the mouse sideways with the right amount of finger pressure.

This made me remember of a benchmark that I saw a few months ago about LLMs being unexpectedly _very good_ with Perl when compared to any other language. I couldn't find it right now. If someone knows what I'm talking about, please post it here :)

I've started using OpenSpec[0] recently to mitigate problems like that, but I'm still very early in this journey.

Can someone with more experience with it (or similar tools) chime in and confirm that this isn't just more AI snake oil? :)

[0] https://openspec.dev/

Ansible includes modules to handle cloud resources as well, such as AWS Lambda.

Fair point. We could classify AI generated articles in two categories:

1) articles generated with context data that's trivial to find (or even embedded into the model)

2) articles generated with context data that's hard to find or not publicly available

The problem is the signal/noise ratio in these articles. If the AI has written the article, then this same info could have been generated by my own AI, but tailored to my needs. So what, exactly, is the new info that this article is generating that I can use to consult with my AI? That's what I want to get out of this interaction.

Maybe my point is something on the lines of "Just send me the prompt"[0]

[0] https://blog.gpkb.org/posts/just-send-me-the-prompt/

This is pretty damn cool.

On a side note, _in the context of workstations_, I wonder if a hypothetical OS that reimplements the Windows APIs (like ReactOS, but with perfect modern hardware support) would be better for end users than a Linux distro with a modern DE.

In the past, this hypothetical OS would be a revolution. But I feel that, in recent years, this gap is not as big anymore and Linux supports way more apps than in the past. Such an OS might even not be relevant anymore, even if it exists.

Do I have a blind spot on this? Is there value in having a "working ReactOS" as of 2026 _for workstations_?

Option 1 is a PM.

Can anyone give some hints on what made Civ 1 special compared to other classic entries in the franchise? Despite the nostalgia factor, of course.

I don't have a stake on AI, but more and more I see the following patterns:

- people that give in to AI do so because the technical merits suddenly became too big to ignore (even for seasoned developers that were previously against it) - people who avoid AI center their arguments on principles and personal discomfort

Just from that, you can kind of see where this is going.

Kinda off-topic, but I wonder if a hypothetical OS that reimplements the Windows APIs (like ReactOS[0], but with perfect modern hardware support) would be better for end users than a Linux distro with a modern DE.

In the past, this hypothetical OS would be a revolution. But I feel that, in recent years, this gap is not as big anymore and Linux supports way more apps than in the past. Such an OS might even not be relevant anymore, even if it exists.

Do I have a blind spot on this? Is there value in having a "working ReactOS" as of 2026?

[0] https://reactos.org/

That's an interesting topic. Please, elaborate.

I was pretty surprised when I learned recently that the Java alternative for green threads doesn't use colored functions. It put Java in a higher place in my perception.