should there be efficient trains between London and Manchester?
And preferably at prices lower than what an equivalent journey would cost by airplane.
HN user
meet.hn/city/jp-Suginami
Interests: Cybersecurity, DevOps, Gaming, Open Source, Programming, Startups, Web Development
---
should there be efficient trains between London and Manchester?
And preferably at prices lower than what an equivalent journey would cost by airplane.
Yeah, but you only pay the switch on the much cheaper model, the first bit is done with smaller context on the expensive model, so you still save money in the end.
Kimi k3 is noticeably more censored than k2.5
It’s starting to sound like Claude.
I think my main problem with the current 'aligned' version of models is that they're aligned with the very worst of the California social justice warriors.
Yeah, when I saw this in the Claude Code CLI I was completely baffled. I've been using the thing through their agent SDK for several months now so I wouldn't have to deal with any of the wonky shit they change every second week in the CLI.
That feels like moving the goalpost. First they would "never not respect the enterprises IP" then the next message "Oh, but it's fine as long as they introduce new terms that you can reject"
Either they respect IP, or they don't. Clearly they don't.
Explicitly signing a contract with enterprises to respect their IP and then proceeding to break that contract with your own customers.
You mean all the conditions that are attached to Fable use? My enterprise is deliberately holding off because those are unacceptable.
Ack! It looks so… actionscript. Why does a UI look actionscript? I can’t even begin to imagine why it feels like that.
My music piracy has been buying a casette recorder and recording anything I want from the aux port of my PC. It’s been fantastic.
Like, at some point we have to start considering teens natural disasters, and put it on the company to prevent something as banal as a password reset that can be requested on a phone from compromising their _entire_ fucking system. These kids aren’t most at fault here.
You can absolutely blow your lid, you just have to apologize afterwards and admit that you were wrong. This is very hard for some adults to do to a 5 year old.
2.5x the scaling efficiency, so 4 times the price? What is happening here? Did the subsidies dry up with the discrepancy between chinese and US models?
We’re a long time from the time Stripe represented freedom from the establishment.
Civil rights organisations are staffed by citizens. What is the problem here?
Wait? They previously allowed everybody and their mother to request the information? Limiting it to ‘citizens of the european union’ seems eminently reasonable.
Games are also very resilient to sloppy AI coding
Depends on what kind of game you are building. I can tell you that even Claude Opus absolutely hates desync bugs and has a rather hard tracing them. Maybe Fable or chatgpt-5.6 are better?
I think the big difference is that the only thing anyone ever needs a pot to do is hold fluids (or soil). It's a lot easier to spec out than a modern software product.
Do you know any adult that uses their phone less than two hours a day? Hell, do you know any child above the age of 12?
This tracks with my son’s observations on my wife’s phone use. She’ll tell him to stop watching youtube, then go right back to doing so herself.
It doesn’t really seem to compute how hypocritical that is.
Yeah, so you attach it to a person, and then you don’t have to wait at all.
I do not understand what sane person attaches taxes to a property instead of to a person.
are these sort of things just examples of selective prosecution
Maybe, but they hit the rich. All the selective prosecution in the US hits those least able to resist.
If I use electricity to do something stupid, I still have to pay for the electricity. Intelligence is just another utility.
Agentic coding is absolutely not a failure. It's just not the 10x that CEOs really wanted it to be.
It’s absolutely 10x faster for coding. But coding is only 10% of my job. The other 90% is figuring out what to code.
You can easily figure this out though.
A 20x plan gives you between 8-10B tokens per week (99+% of which are cache_read), and has been for the past ~3 months (since I started tracking it). I have not seen any quota movement despite their claims to be raising or lowering usage limits.
I'm fairly certain which side of the conversation I'd have trouble understanding. I have trouble understanding it all the time.
I think that increasingly, the US will have to be passed by for these things. Clearly we’ll have to start looking to China for world leadership, to be the land of the free.
It’s really weird when you go to one of the open models and suddenly the same context window stretches nearly 3-4 times as long.
Can’t you just hand it in?
That’s only the case if the zines are evidence of anything. What were they evidence of?