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

If you like this kind of puzzle with a Sci-Fi theme then you might also like this:

http://www.boundvariable.org/task.shtml

It was the 2006 edition of the ICFP conference programming contest and it makes you write a small VM which given the provided binary blog leads you to multiple challenges/games.


I didnt know this one, awesome!!

Isn't WhisperX the canonical choice for running Whisper?

While whisper and whisperx is python implementations, the whisper.cpp wins the benchmarks.

Maybe for running locally? whisper.cpp is nice because you can embed it pretty easily in apps for various targets like iOS, OSX, Android, wasm, etc.

Yet prisons are overfull everywhere in America

Maybe we need more prisons

See, here's the problem. You've revealed yourself as a conservative, at least in social issues, and that usually comes with all the conservative beliefs. Why should we believe you, why should we believe you're not just taking some position today that makes you money at the expense of the rest of us? That's why 180 degree policy reversals are weird. I just don't know what to believe about what you're writing - sincere position or throwaway trolling.

The German economy is indeed heavily influenced by its manufacturing sector and its reliance on large trade surplusses. Consequently the German workers can't be granted huge wage increases because it would reduce their competitiveness internationally.

In contrast the US economy is service-based and thus isn't constrained to scale their prices up so much. Of course manufacturing becomes very hard to maintain when skilled workers can earn >100k.

All stats show that top US tech companies earn incredible amounts of money but their share of GDP growth isn't as big as the Mag7 share of the stock markets makes you think. (<15% of GDP growth in 2024)


How does Australia fit into that model? Their economy is based more on resource extraction than on technology or manufacturing, and wages are high relative to most other developed countries.

Well solely is dramatizing but to a large extend it is true that the clima is changing because of human actions.

Certainly the rise in Co2 from 280 ppm to 420 ppm since the onset of industrialization is rather very likely due to human actions.


> A culture evaporating

Well if you being your sentence like this, then you already made it clear that you aren't serious about discussing, because that is not what is happening.

Any discussion that lazyly tries to skirt what Germans have done 80 years ago and tries to move into similar directions is just plain ignorant.

We must get back to the real issues: 10m Ukrainian refugees because we let Putin murder in Ukraine. 10m Syrians being displaced due to the world's inaction to stop a civil war. The list could go on.


Fear mongering.

Current net immigration inflows into Germany are below 0.5% of population.

The big immigration waves of the last 20 years can be directly linked to devastating wars: Afghanistan, Syria, Ukraine.

How many generations did it take for the Germans to become Americans in the US? Did it make Americans disappear?


> How many generations did it take for the Germans to become Americans in the US?

It's not a question of generations. There were a lot of German-speaking towns in the US, but World War I-driven xenophobia pushed them to Americanize.


Germans mostly assimilated and culturally were similar to existing Americans, sharing the same religion and similar values. Can’t say the same for Muslim immigrants at all.

What about the Irish in America then? Culturally, they were not Protestant and were regarded as very different for a long time. How come we don't hear about this as an issue in the US today?

They had a tough time integrating but ultimately they were Christian and European and had western values. Muslims don’t have any of that and their ideals of freedom amd rights aren’t compatible with western values.

Let's spell the mistakes out:

You get 60% of last salary, not full salary.

You get it for up to 12 months not 24.

You lose 3 months of unemployment money if you quit rather than being fired.


Even a birth rate as South Korea's does not mean the population will disappear over night. It will shrink. It will mean things will change. Infrastructure will be overprovisioned and housing will be cheap. It will mean other things will be prioritized by politics (such as kindergardens and work life balance).

In any case it won't be a catastrophy as life in North Korea.


A good analogy might be the Black Death: it didn’t destroy Europe, it changed priorities, freed the serfs, started valuing labor more, and ultimately led to a stronger Europe in the future.

If we consider that the world had 1bn people a bit over 100 years ago, it puts many things in perspective. We have generations of time to turn things around if we think it is necessary.

No, the Black Death is not a good analogy, because it killed all ages indiscriminately.

Our "plague" "kills" the young and productive - by their never being born in the first place. We are headed for something we have never seen, ever: a society dominated by old people in pure numbers terms.

The worry is that it is a design for stagnation and decay rather than greater strength. (There won't be money or people to maintain infrastructure because the elders will demand healthcare.)

I don't know what will happen and nor does anyone else really.


It killed all but made people more valuable and, more importantly, valued in general. If kids become more rare, we will probably start treating and nurturing them better. We complain about childcare costs now but if kids are few enough society might just happily take up the bill.

If only it were that easy. Austria and Switzerland have nerly identical birth rates, but in one, childcare is almost free and in the other, it is very expensive.

South Korea and Hungary (among many others) have tried paying people to have children, without success. It pulls a few births forward in time, but then the birth rate declines again.

There is something more fundamental going on.


One question I was wondering about regarding the open models released by big labs is how much more the could improve with additional training. GPT-OSS has 2.1m hours of training, how much score improvements could we see at double that?

I think GPT-4.5 was potentially the original GPT-5 model that was larger and pre-trained on more data. Too bad it was too expensive to deploy at scale so that we never saw the RL-ed version

As we saw with GPT-5 the RL technique of training doesn't scale forever

Unless GPT-5 is 30% cheaper to run than o3. Then it's scaling brilliantly given the small gap between release dates. People are really drawing too many conclusions from too little information.

I meant scaling the base training before RL.

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

Search: