For the best experience on desktop, install the Chrome extension to track your reading on news.ycombinator.com
Hacker Newsnew | past | comments | ask | show | jobs | submit | history | rbanffy's commentsregister

Weapons are designed with an opponent in mind, and guarded against the expected threat models from that opponent. Everything breaks down when the opponent does not what you want them to.

> US leadership descending the oceans of stupidity all the way to the Mariana trench.

And they voted for this not only once, but twice.


As the Soviet Union made us learn, you don’t need a big military victory to make your enemy spend themselves into defeat.

When you decapitate a well organised military, all you achieve is installing a new enemy you know little about you can’t predict their actions and that now know they are fighting for their own survival.

Not the best place to be.

Americans seem to underestimate everyone else.


If you throw away your principles because you are fighting an unprincipled enemy, you are no better than them.

That's a lovely thing to say, but if your existence is being threatened by an aggressor, I wouldn't blame you for throwing out the rulebook.

In my view, if someone invades your territory and starts attacking you, you have no obligation to follow any sort of "principles" or "rules" when it comes to how you fight back. Anything you need to do to the attackers in order to defend yourself and your people is, by definition, morally defensible.

(Do note that I said "need". Doing arbitrary messed-up things that don't actually further the goal of driving back the attackers is not ok.)


FWIW, during the Iran-Iraq war (where Iraq invaded Iran), Iran used a bunch of pretty questionable tactics like suicide squads of child soldiers.

It’s such a shock to the system to realise that “unprincipled enemy” referenced here is the US.

There is no if. We've already done that. So yes, we are no better than them. So answer the question. Why would Iran follow conventions it's enemy that started a war of aggression is not following?

Becaus two wrongs don't make right. If they are smart they will stick to the convention.

Hegseth explicitly ordered to give the enemy “no quarter”.

Hegseth is not in charge of the Iranian military.

Right but the reason we have rules against people declaring no quarter is to prevent a race to the bottom. It is absolutely reasonable to respond to a no quarter declaration in kind, which is... again... the entire reason we have prohibitions on it.

But he did publicly declare his intention to commit war crimes.

Actually even just declaring no quarter is itself a war crime.

Hes also liable for the death sentence, 18 U.S.C. § 2441 — War Crimes Act (1996) & 10 U.S.C. § 950t — Military Commissions Act (more relevent)

They won't face any US law. AIUI, they have been getting letters from the DOJ office of legal counsel that say it's legal. This effectively immunizes them (the DOJ can't turn around and charge you with a crime, if they advised you beforehand it was not a crime).

The best shot would be to turn them over to the ICC


> they have been getting letters from the DOJ office of legal counsel that say it's legal. This effectively immunizes them (the DOJ can't turn around and charge you with a crime, if they advised you beforehand it was not a crime).

This is not true.

OLC opinions are just that: opinions. They are non-binding and non-promissory. They are an important factor in any assessments as a norm, but definitely not dispositive and not legally binding.

The only real barrier is the pardon power, but I'm personally fine at this point with totally breaking the seal, trying and jailing every criminal in the administration(++), and consider the pardon power gone for good. Small price to pay.


> This effectively immunizes them (the DOJ can't turn around and charge you with a crime, if they advised you beforehand it was not a crime).

Where is the check or balance on this? The executive branch can apparently just launder itself wholesale of any crimes committed by its members.


> the DOJ can't turn around and charge you with a crime, if they advised you beforehand it was not a crime

this sounds like the kind of rules we, as a society, decided to dispense with, so the DOJ can absolutely turn around.


Alas, the USA isn't signed up to the ICC.

Sure, but, if somehow they fell into ICC custody overseas...

We've already committed several war crimes.

In case anyone else doubted this, I will save you the time to look it up. Yup, it's sadly true.

https://www.commondreams.org/news/hegseth-no-quarter-interna...


Yep. And war crime seems to have lost all meaning in the US.

But, even if you dismiss the idea of international standards, this is clearly very bad for US soldiers (and sailors, airmen, etc). I wonder if they see that.


> But, even if you dismiss the idea of international standards, this is clearly very bad for US soldiers (and sailors, airmen, etc). I wonder if they see that.

Even if you dismiss the idea of international standards, a no-quarter declaration is against _US law_, specifically subject to the penalty of death with no other lawful penalty defined: https://www.govregs.com/uscode/title18_partI_chapter118_sect....


Z/OS for ARM then? ;-)

I’ve been running VM/370 and MVS on my RPi cluster for a long time now.


Cool, can you share more about the setup?

A 4x RPi Zero Ws Docker Swarm cluster running the dockerised versions of Hercules with VM/370 Sixpack, VM/370 CE and MVS TK 4. All in an IKEA picture frame.

AIX for ARM? ;-)

Is modern ARM stuff done big-endian? because AIX is exclusively BE iirc

That, weirdly, should be fine; ARM is bi-endian in the sense of being perfectly happy to run either way. In fact, the easiest way I know of to test software on a big endian system is to run a perfectly ordinary Raspberry Pi with NetBSD's big endian port for it:)

Yeah, I know ARM is bi-endian (pretty much all non-x86 archs used nowadays are) but the question is if it's actually enough to have a software base for it. NetBSD having an ARM port in BE is great but most arm stuff is done for LE systems since MacOS, and NT, and most Linux stuff is LE. This isn't that much of a problem in the free software world because we like to test things on obscure architectures but for the kind of proprietary stuff that you'd want to run on arm might have problems (assuming it wasn't ported AIX already)

I never said it'd be an easy port, although there was an x86 (and s/390) port back when time itself was new.

edit: s/390 is big endian.


Not in the real world, but this is kind of how Asimov’s robots interpret their 3 laws - it’s about consequences much more than what the order is. Also, they weight consequences of inaction as well and might be driven to action when not acting could cause a violation.

Our AI is nowhere near the level of sophistication required to implement something like that, but it’s still an interesting idea.


That's a great connection.

You're right that current systems aren't close to that level of reasoning.

What I'm wondering is whether we can approximate some of it structurally — by defining when execution is allowed or not — even without that level of sophistication in the model itself.

Curious how far you think simple constraint systems can go before something like that kind of reasoning becomes necessary.


I’d say that kind of reasoning was needed since we invented the chatbot.

At least since we stopped using rules and embraced neural networks.


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

Search:

HN For You