So it is going to be a feeling. Is their smaller income going much farther now in how it benefits them, if so they feel better-off

Meanwhile the video fades more and more every few years, you can barely even make out faces and objects in the frames now. But it’s happened so slowly that he never really realized that it was happening all along

I did recently, specifically targeting lower capacities for the components that have been increasing (RAM and storage).

It didn’t seem like prices would be going down for a while and I didn’t have a desktop pc otherwise, so just went for it. We’ll see how it all plays out but I don’t think it was a terrible decision, as long as prices stay high for a couple years it still makes sense to just suffer through the increases

Will the developer of a new programming language be able to reach out to model companies to give a huge amount of training data, ensuring that the models are good at that new language? I don’t think a small team can write enough code, the models already struggle in medium-popularity languages that have years of history. They hallucinate lua functionality sometimes, for example, even though I’m sure there is lots of lua code out there.

So if most people use coding agents, we’re stuck with the current most popular languages because no new language will get past the barrier of having enough code that models can write it well, meaning nobody adopts the new language, etc.

Same thing with libraries and frameworks - technical decisions are already being made based on “is this popular enough that the agents can use it well?” Rather than a newer library that meets our needs perfectly but isn’t in the training data

I realized that they probably made that whole change, along with all of the ads, because they knew it would spark mild outrage and discussion from people who saw it — they’d discuss if assigned seating is actually better or worse than the previous fist-come-fist-serve system. I can understand either angle but I liked that they were different than other airlines, etc.

But really it’s because they removed the free checked bags that had been their policy forever, now you need to pay like any other airline. Which completely ruins their value prop. But by advertising the seating changes so heavily for months, they make you forget about that part that actually makes a much bigger difference in the experience

I think we’ve seen a wave of bad actors - either employees of LLM companies, or bots - pushing the idea hard of code quality not mattering and “the models will improve so fast that your code quality degrading doesn’t matter”.

I think the humans pushing that idea may even believe it, but I don’t think they’re usually employed as software engineers at regular non-AI companies, rather they have some incentive to believe it and convince others as well

1. It’s closer to 50 years, and even a partially degraded panel will work, just with less output

2. Even if we say 20 years, that means that you only need to buy panels once every 20 years! Not continuously. A complete and total interruption of solar panel production lasting 4 years will only mildly interrupt current output. How long can we last with a total disruption to oil supply chains?

They don’t know what Linux is, and know nothing about tech, they just know that we had a 30 minute lesson on “here’s Firefox, this icon means you need to install updates, here’s how you print”.

Oh and this was Linux Mint back in ~2016

Things have only gotten easier since then

Two cops in a car rolled up and jumped out of the car with guns drawn and screamed at us to put our hands up because we “looked suspicious”.

They then asked us what we were doing, we said “walking home”, and they put their guns away and said “be safe out there, we didn’t realize you were kids”.

Absolutely no idea why it warranted guns pointed at us

Same as if you use an image diffusion model. You can describe very clearly what you want, and iterate carefully until you get a picture that looks good. But nobody would say that they “drew a nice picture”, since they haven’t done any drawing.

(except maybe the mega-power-users who use the tool and have a warped view of their accomplishment)

Engineered ingredients may or may not be equivalent, but they often remove nutrients that existed in whole foods, then attempt to add nutrients back in through industrial processing. But we still don’t know the full affects of that delivery method, but we do know that it can negatively impact the gut microbiome.

There’s enough evidence out there to be highly skeptical of ultra processed ingredients

https://www.yalemedicine.org/news/ultraprocessed-foods-bad-f...

https://www.nature.com/articles/s41574-025-01218-5

I don’t think those links prove definitively that UPF is a direct cause of disease, but they show strong evidence that there are problems with UPF and we should probably eat more whole ingredients

If a person has a shitty idea that sounds good, they start writing about it. If they exercise some care in their writing, the act of writing itself is enough to make them realize that their idea is shitty.

By the way, it happens to me all the time! Even just on HN, I’ve bailed halfway through writing a comment because I realized that I didn’t know what I was talking about, lol.

But an LLM will gladly take that shitty idea and expand it into a very plausible article/message/post, that seems reasonable if you don’t think very critically about it. And it’ll be done with such a high-seeming level of care that any human author would’ve been fact checking themselves the whole time.

So it forces the reader to think even more critically, rather than letting our subconscious try to judge authenticity of the writer through the language they use.

For example, someone who says “my WiFi is broken” when referring to the fact that their computer is dead, we can quickly judge them as “not an expert at computers”. But if they say that “my M.2 drive has gone bad”, we inherently assume they have some understanding. —- when the first person uses LLMs to write, they sound as informed as the second person even if they are completely clueless and wrong

Also I mentioned criminal vs civil penalties being treated differently - I don’t believe the same scenarios apply to both. AND if you can prove that you lent a car to someone who got you a speeding ticket, then it’s on them - just that the owner of the car is responsible by default.

I hate these arguments that doing anything to restrict the deadliest machines in history is impossible because it discriminates against the poor. It doesn’t. But having a society where driving is the only option does, so I am all in favor of alternatives. But I also think that we should improve safety for cars when possible as well

It is commonplace to drive, but has high potential for danger and death. It seems ok to me to have a level of care required for owning a vehicle, and that includes being mindful of who you share your vehicle with.

Same thing with guns - if you blindly lend a gun to an acquaintance and they shoot a school, you will absolutely be charged with some crimes, either accessory to murder or manslaughter, where you have to prove that you weren’t being negligent by giving it to them. Guns are dangerous and owning them bears a higher level of responsibility to the owner.

Vehicles kill more people, they also deserve responsibility to own. If somebody breaks laws with your vehicle, it’s your responsibility by default unless you prove otherwise.

In the same way, if your car fails emissions tests, you can’t register it and it’s the responsibility of the owner to ensure that their car meets emissions standards.