impossiblefork

I think this is a ruling that's hard to justify. There are many frameworks under which the judge could have chosen to rule, it could be treated as assault, or some type of aggravated assault/battery, if the lead can be measured in the people who consumed it, it could be treated as fraud, since the water is presumably sold, etc. and reasoning about that it would ordinarily be a punishable tort would make it clear that the government can't impose that sort of thing either, just as it can't send anyone out to punch people.

Yes, I agree, and of course I sort of try to qualify it a bit. I think cheap, easy transportation has its own value. Creating a sort of fluidity, competition and so on, and it's not like one can't imagine a situation where somebody is the guy who controls all the rail deliveries to a town, or where he has bought all the rail transport demand in the past, so now he gets to be first. Roads, meanwhile, are open to everybody.

At the same time, I really do think that it could be beneficial to shrink trucks down, automate them, or create something which has so many bogies that the road wear is not greater than on a car, so I sort of argued for what I saw as theoretically fair without actually being sure that the outcome would be what I hoped for.

Yes, but if you don't tax it correctly you will misallocate resources.

Diesel will get used excessively and you will get more road wear, as the user proposed. If you tax it, you will pay more directly for goods, but as costs rise people will be incentivized to find more efficient transport solutions.

It might still be worth it if road wear is cheap or the predictable misuse is predictable and that saves money somehow, but in all likelihood little harm would come from a fair treatment of diesel vs. gasoline etc. taking into account road wear of each vehicle class. It would be more natural to tax vehicle weight, hours driven etc. directly based on a sound model of societal cost though.

Yes, I'm specifically talking about the song.

If propaganda is to be used, it should be to inform the public without deception as to the sender or the purpose, not used to push a political position with public money with deception about the sender or the purpose.

I didn't initially understand your comment, but you mean basically something like that the origin of feminism was in socialism and the CIA saw that it was going to be unavoidable or at least popular and wanting to avoid this leading to socialism becoming popular they decided to make a bastardized version that cuts off any socialist aspect and just turn the whole thing into the modern liberal feminism?

Yes, and people understood it as it happened.

There's a Swedish song, which sees it like this "daghemmets utbygnadstakt regleras av konjunkturen, så tillmäts kvinnan sin plats i den ekonomiska strukturen, en billig arbetskraftsreserv till hands när fabrikerna visslarn, tillbakapressad av daghemsbrist, om lönsamhetens drivhjul plötsligt gnisslar" "The extension of day care is controlled by the economic demand, so woman is given her place in the economic structure, a cheap work-force reserve at hand when the factories want it; pushed back by too few daycares in case demand slows".

The interpretation isn't anti-feminist, and they see the game rather as a cynical game having little to do with what is in the interests of the women themselves.

Yes, but a driver's license can be accepted as ID.

Just because a driver's license isn't proof of citizenship doesn't mean that it isn't sufficient, since it will have some identifier connected to the identity of the voter. You would show the driver's license, then the registrars would find your name on the list, strike it out, give your voting envelopes, and then you go wait for a private booth to become free and then you go into that booth and put your votes in those voting envelopes.

A Swedish driver's license doesn't mean Swedish citizenship. It can still be used to check somebody's identity. Citizenship is checked by seeing whether the person is on the voter rolls, not by their identity card.

That isn't how the Statute of Rome works. The Statute of Rome applies to things done on the territory of signatories.

So if a US person has operated in the territory of a signatory, and committed something suspected to be a covered crime, and there is no US attempt to prosecute them, they can be prosecuted.

The US militarily can't intervene. They don't have the capacity to fight the Dutch, or French or the like, and the political cost of attacking other NATO countries is of course enormous-- it would be the end of the US as a superpower and lead to an actual EU-US war which would probably last for decades. There's a reason the US Servicemen Protection Act has an out, that it doesn't require any action. Think of it, not as a real policy, but as a dog making sure there's a fence between it and another dog before it starts barking.

I don't see how they could legally intervene. Most EU countries have very independent court systems. Once the warrant is sent to the states, it will be enforced.

The ICC already has the authority to punish people who interfere with its operations.

It's on them to prosecute people who cooperate with sanctions and the like to retaliate against the judges or lawyers or others working for the court, and they can do that any time they like.

Literally, they can write a warrant for the president of their bank, or the highest Microsoft rep in Europe, or anybody involved in any way, and the ICC-signatory country they're in must hand them over.

They have the authority.

Well, that is not the court's position, or the position of any of the signatories the Rome Statute.

"Article 70 - Offences against the administration of justice

Offences against the administration of justice

[...]

(c) Corruptly influencing a witness, obstructing or interfering with the attendance or testimony of a witness, retaliating against a witness for giving testimony or destroying, tampering with or interfering with the collection of evidence;

(d) Impeding, intimidating or corruptly influencing an official of the Court for the purpose of forcing or persuading the official not to perform, or to perform improperly, his or her duties;

(e) Retaliating against an official of the Court on account of duties performed by that or another official;"

So I'd say 70 (e) covers the sanctions imposed on those court officials. They are clearly officials of the court, and the sanctions are clearly part of retaliation against them. Consequently it stands to reason that anyone who participates in their enforcement or drafting or really in any way can in principle be prosecuted.

I think what's needed here is for A/I to go the ECHR.

A state failing to prevent foreign sanctions from restricting the freedom of association of its citizens is certainly a breach of article 11. It doesn't matter whether it's caused by a foreign state, the duty of Italy is clearly to ensure freedom of association in practice.

The ICC has tools for countering this kind of thing though.

They can order anyone who impedes them seized and handed over for judgement to the ICC, including bank CEOs and the like participating in retaliatory measures against them. They haven't done this, presumably for political reasons, but they can.

Yes, my system is intended to be used on proper computers. I am not interested in phones at all.

Yes, definitely in the 80s.

No, of course there isn't a visible receiver address.

You have a list of peers, P1,...,PN, every second, at a specific pre-decided time, slightly fuzzed to avoid timing stuff, you send a message to each of them. If you don't put one in, it's a dummy message chosen at random and encrypted. But if you have a message, you send that at the appointed time.

Consequently, from the outside of the system, from the PoV of someone who does not have the keys, its behaviour is indistinguishable from a system where no one is sending any messagess at all.

No, you always send a dummy message the appointed time.

If you have a real message, or a network maintenance message, that substitutes for a dummy message. You never send messages other than at the appointed time.

Delayed or dropped messages won't matter, since you only send a message at the appointed time. So the network is literally indistinguishable from one where you have sent no messages to all participants and to even someone with complete control of the network.

Ah, blockchain-based. Maybe it's fine. I prefer my idea with dummy messages and no ledgers or anything.

I don't understand your comment.

The issue is complicated.

Many EU governments care, because Turkey cares, but the problem is that Turkey's position that they're terrorists is legally iffy, so a lot of European courts disagree, and this creates a schizophrenic situation where politicians hold that they're terrorist and will claim that it's the government position, while the actual government position in the courts is that they mostly aren't and that it's a case-by-case question.

No. With dummy messages you are guaranteed that traffic analysis will not be possible.

Because you connect to the network, then start sending messages to everybody continuously, a message every second, or every ten seconds, or every 100 seconds, you will, every second, or every ten seconds, or 100 seconds, be able to send a message to whoever you want, and you will not be distinguished in any way from the other network participants, even if an adversary controls all nodes but that of the person you want to communicate with. Without dummy messages, traffic analysis will always be possible.

If they knowingly and deliberately assisted the PKK in a forbidden way, there would likely have been an Italian prosecution. Since this hasn't happened, it probably isn't true.

The illegality of the PKK in the Europe is also something quite complicated. The commission have declared it a terrorist organization, but here in Sweden such declaration have no judicial meaning and we define terrorism by whether the organization satisfies objective criteria defined in our law, which the courts determine from case to case (which in Sweden means that if you send money the PKK and prosecutor brings the case and argues that they're terrorist organization, and you argue badly, and he wins, then you might be convicted, but when another guy does the same thing, and argues well, he wins); and the situation in Belgium is similarly complicated, with a legal dispute about whether PKK was a terrorist organization, which PKK won, and thus the PKK was, in Belgium, declared by a court to not be a terrorist organization.

Signal will allow traffic analysis. You send the message to signal, if they wish they can write down to whom they send it on.

It is impossible to make a secure system without the continuous transmission of dummy messages to large numbers of users.

Yes, but it's fine.

If you receive 10,000 messages per second, that's nothing. That means that you can have 10,000 people online simultaneously with 1 message per second and 100,000 people online simultaneously with 1 message every ten seconds even in the version where there is no retransmission.

Yes, but then you have to accept that they can't be part of a group that needs security.

If they aren't willing to do things correctly, not be lazy etc., and you tell them that this means that adversaries will be able to pinpoint them, and pinpoint you and everybody else, and they don't express that they're going to take the effort to be reasonable, then maybe you shouldn't talk to them.

Yeah, but it is possible to make distributed systems.

I've written up a little toy in Erlang where you have lots of connections to peers and send dummy messages, all of the same size, to all your peers continuously and then, because of this, you can obviously send a message to whoever you like without anyone being able to know to whom you sent it. With a bit of extra delay you can add retransmission, and then you can have huge networks. Complete resistance to traffic analysis.

The only thing you need to accept to have this kind of security is to accept that message timing might be 1/s, or 0.1/s, but for things like e-mail that isn't a problem at all.

My system is not done. I had intended to finish it if ChatControl 2.0 passed, but since it didn't, and there's upon that a war on I didn't put it out, and I thought it was a bad idea to popularize this kind of thing since there was a war on, but it's quite straightforward. Many here could probably write something like it.

Very strange to decide to only supply a communications system to people ideologically aligned with one. That seems like it would make one stand out to an undesirable degree, and it makes me wonder whether these groups actually do this, or whether they're setting others up. If this is correct I almost whether they're going after something set up by the Italian government.

Still, it seems like this would likely to be protected under the US first amendment, and it would probably be reasonable to bring legal action in the US to avoid silly stuff happening in Europe, when this in inevitably used to send goons and annoying infiltrators across the sea to Italy and Europe.

One could hope so, or it would be best for me if it were. I don't think I can count on that though. I don't think I can hope for anything better than 30-70 in my lifetime, and I think 10-90 almost requires you to buy the design firm.

I don't agree. At the moment companies like NVIDIA take several times what it costs to make a chip. I think the fair split for the technology contribution is more like 50-50, maybe even 30-70 in favour of the manufacturer.

With competition we will actually have the fair split, whatever that is, and thus much lower prices.

At the moment, to have a big AI firm, or really AI firm at all, you need to be blessed by NVIDIA, in the form of receiving circular financing for your compute. They know that their prices aren't fair, or competitive.

Commoditization of inference is the end of that. The end of the mega-premium on inference hardware, and it's good not only for people who like running their LLMs, but it's the first step towards commoditization of training.

Sorry, I seem to have messed up this comment, I didn't mean prompt injections, I meant leakage of human thought into the model.

Basically the idea is that if I train a model that speaks like a human, it will follow human biases. If I train a model that talks like Claude, it will follow the bias of my Claude-style examples.