Hacker Newsnew | past | comments | ask | show | jobs | submit | mcny's commentslogin

I reached point three and was nodding all along. I guess I am the NPC

I don't think this is true at all. I use opus 5 to build a small project. It bothers me that people simply handwave away concerns so easily.

The context size is still not nearly as big as it needs to be to store all the code an enterprise needs. And apparently as you increase the context size, there are more defects. So none of this is a solved problem. We are still in early days and there is a lot to be done.

I'm sure there are marketing people who will say "coding is solved" and other such snake oil but none of this is done, far from it!

That being said there is still enormous value in older models especially with tool calling which will let them access the latest data. I feel like we need to be a little more careful and the tools should cache in a smart way to avoid rework but clearly if we could have opus 4.8 level of work for like a one time payment of a system for local LLM it will have value for years into the future.

So I agree in a weird way that sol is good enough for certain tasks but really there is a long road ahead.


I don't think it is a fair comparison. The cost of raising a child can still be too high even in a fairly low gini country.

In fact I'd say gini has nothing to do with birth rate. Birth rate declines when parents have confidence that there is a good likelihood their children will survive to adulthood.

It is often easy to forget that just five hundred years ago almost one of two infants born didn't survive into adulthood. Low birth rates are good. Don't let anyone say otherwise.


I tried to find some information about how Japan manages to get away from criticisms around forced labor in prison. It seems like Japan does have very poorly paid labor, as little as usd 0.05 an hour, but the incarceration rate is low.

Once again, I don't know much about Japan but I doubt there is a large ethnic minority that is only about fifteen percent of the population but somehow accounts for almost half of all incarcerated people in Japan.

From Google:

> Low Incarceration Rates: Japan maintains one of the lowest incarceration rates in the developed world (roughly 35 to 40 prisoners per 100,000 residents, compared to over 500 per 100,000 in the US).


If I could wave a magic wand, I would implement this like some kind of token replenishment or quota system like every IP address gets to access old dot like a very low number and if you log in with the same IP address you get more quota or replenished a little faster if you also come in with no auth like from a private window. Of course, in an ideal world we wouldn't have to do anything like this but yes, regional is likely as well.

I know I worked with some programmers in India and apparently some ISP will do some kind of trickery so like a lot of customers share the same ipv4 address. So what I said above might not actually make sense in the real world.


Is that trickery NAT, network address translation? It is a pretty widespread technique I think, we’d have run out of ipv4 addresses a while ago without it.

Yes, it is called carrier grade nat but the problem is you can't assume that an IP address maps to a user or even a household neatly. That's the problem.

There are some defects like what does this mean?

> dir="auto" rather than inheriting the story's direction: quotes are reproduced verbatim, so a Hebrew story routinely carries an English one. Inheriting rtl left its closing punctuation stranded at the start of the line (KNEWS-453). -->

https://news.kagi.com/s/3gskhn


Kagi dev here, sorry about the bug; we had a malformed nested HTML comment that ended up partially rendering. This is now fixed.

Hey, I remember reading along before and around the time you were “officially” hired – happy Kagi could recognize your work, and seemingly continues to do so :)

Okay, strange, but that’s a bug, not a web app “feeling barebones” wrt. core functionality for most users, at least IMO.

Has that been reported on kagifeedback.org yet?


> You must have a valid mobile phone number to sign up. SMS charges may apply.

Reduces barrier of entry for mass AI spam, so I’m for it

As we all know spam text messages aren't a thing since it's very costly and hard to spoof cell phone numbers.

Those kind of SMS check usually ban voip providers. So you need a real line or a prepaid cellphone, and it can presumably only be used once or twice on this service before the number is "burned".

Such checks, although not perfect, do prevent a huge amount of spam and bots by having a very real cost for bots/spammers, and essentially no cost (other than their privacy) for the user.


Even a one cent toll is enough to change the economics of many bot behaviours.

Is this really where our law enforcement should be spending its resources? I would say rescuing kittens from trees is a higher priority than this nonsense.

Also worth noting wherever we get a chance — the Computer Fruad And Abuse Act itself is nonsensical and must be abolished. There is no salvaging it. It must be repealed with no replacement.

I don't think the FBI has enough agents to play tech support to anyone who doesn't understand how wifi works.


People allegedly broadcasting fake SSIDs on a plane to do credential harvesting on a plane shouldn't be prosecuted? Ok...


Have you ever seen a cat skeleton in a tree?

Still, I agree.


Why yes, yes I'd like to see someone's shit get pushed in for interfering with the wireless networks on the aircraft. It should have parked at the end of a taxiway and waited for LE to meet it on the pavement instead of even parking at the gate.

They didn't just jam the pax WiFi, their stupid stunt also causes havoc with the internal crew SSIDs available below 10k. Pilots and flight attendants were definitely not pleased having their electronic flight bags and some of the app-based comms drop off the face of the earth in flight.


> Most people dramatically under-size their air purifiers, or run them on a very low fan setting, and then they throw up their hands and just say that air purifiers don't work.

I believe something is better than nothing here. One of the biggest complaints against air filters is noise so maybe a good compromise would be to run them at full speed and full noise for a certain amount of time or something when nobody is in the room?


A false sense of security can be worse than nothing, because it prevents you from seeking out actually effective options.

I too would like such a "shy" air purifier, but manufacturers always seem to go the other way: when occupancy is detected they increase the fan speed.

Best option IMO is just to get an air purifier with a good CADR-to-decibel ratio and then (again) size it correctly. A surprisingly good option is something called the Airfanta 3 Pro, which is basically like those wildfire filter boxes except it uses PC fans.


The way to get air purifiers to really work well is to install them at the air intake, i.e. in the windows, or where the central air intake is, so all incoming air passes through them. I use indoor air purifiers too, but not as a substitute for ones at the intake. Note that tires and diesel fumes are prominent neighborhood sources of harmful particulates.

It is not expensive to run intake fans in the spring and fall seasons when active heating or cooling are not required.


That helps for pollution that comes from outside (traffic, pollen, wood smoke), but most of the microplastics are generated by moving/wearing synthetic textiles inside the home.

Positive pressure systems are great, love 'em, but there's a quantitative mismatch in this case. Above ~1 ACH your HVAC costs will go through the roof (even with heat/humidity recovery), but for effective filtration you need 6-8 ACH to catch the larger dust before your lungs do.


And for pm2.5, at least in Australia, it’s entirely generated from cooking. Outside air is very clean.


bigger filter = less noise to move more air


If I am hiring, I need these characteristics:

1. Person who cares 2. Person who is knowledgeable 3. These two must be the same person!


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: