There is a reason why countries in Europe have such laws. US did not have a major war for quite a while on their own soil which affects your thinking. We do not want to reignite national socialism or communism, we do not like to see news channels lying to us. We do not like Russian bot armies spreading propaganda in chats.
What truth is it that you cannot say in Europe? You can say pretty much anything and be critical and nothing will happen to you. And if something happens there are instruments like European court system which you can use to fight your case (there is no need to be rich for that).
Individuals get arrested for posting memes in the U.K. and fined and/or get phones taken away in Germany for the same because an unelected bureaucrat did not like it. 1984
At the end it mentions what the future engineers will do:
> Engineers spend more time writing specifications and models, designing systems at a higher level of abstraction, defining precisely what systems must do, what invariants they must maintain, what failures they must tolerate.
We do that already and the abstractions are very high. The other part is about knowing what the system is supposed to do way in advance, which is not how a lot of engineering is done because it is an exploratory problem. Very few of us write crypto or spend much time in a critical piece of code. And most importantly no user ever said if the software they buy is using proofs. Just like security these concerns are at the bottom of a barrel.
However you like it or not banning just one company is not a recipe for success. IMO the issue is in the procurement and how these tenders are worded. For instance, if the requirement is data residency backed by private keys and conf compute then put it in writing. The idea that some other vendor will come in and solve this problem without such a requirement upfront will not hold for long.
By and large MS problem is that our world gets fragmented and you need to have products that adapt, eg great firewall in China, strict data residency in Europe. It is difficult to achieve that without segmenting your products as well.
Irish infra is not great if you compare it to many advanced European countries. I hate they still do not have a train/tram connection from the airport to the city. Taxes also make you weep. Not to mention an immense risk of losing all those corp taxes and industry if US pushes ahead and creates barriers for companies to trade. It is great at many things but also has some downsides.
What limits? You can do pretty much what you want but make sure you can defend yourself in the court. I feel there is a bit of a disconnect in terms where people get the news where in US you kind of expect biggest news providers to be biassed, eg Fox, hence reliance on social media. In Europe gov media is quite strong and objective, and the idea that it restricts something is odd. A great example is the banning of RT, they lost licenses IMO in multiple countries, but the agency was spreading a lot of lies. IMO what we all want is objective news reporting.
Concrete examples - in Germany you are not allowed to insult politicians or the government in social media. In Italy, people have faced criminal charges for simply criticizing the prime minister.
When the government does not allow its population to freely speak against it, it's just waiting to be abused by one bad leader.
> Concrete examples - in Germany you are not allowed to insult politicians or the government in social media.
You're not allowed to insult anyone, https://www.gesetze-im-internet.de/stgb/__185.html , though the term "insult" is not nearly as broadly defined as in everyday speech. The law dates back to the 18th century, and has largely been unchanged for 150 years. I really don't understand the recent outrage over these and other laws. We have been fine.
> has largely been unchanged for 150 years. I really don't understand the recent outrage over these and other laws. We have been fine
The last 150 years of Germany have...ahem...not been what I would call "fine."
It would be interesting to have a replay of history without this law and similar ones related to it. Could be nothing different happens.
On the other hand, any law regulating speech is going to have a reverberating effect on the marketplace of ideas with 2nd and 3rd order outcomes that are impossible to disentangle after the fact.
almost all communication was oral 20 years ago, now-- especially since covid -- it's almost all, even casual comments, through text messages which can easily be used in evidence
That's a good point. Though I wouldn't say text as a medium is the critical factor, it's that more communication is taking place in the open (over social media) and being recorded for everyone to see.
However, I don't see how this would imply the law that's been in place for 150 years would suddenly be bad. In fact, one might argue that precisely because so much communication is happening in public now, more regulation is needed.
> Concrete examples - in Germany you are not allowed to insult politicians or the government in social media.
Germany restricts insulting individuals / your neighbour, police officer, a pastor or a minister. There’s no special law for politicians. Political criticism is protected under the Basic Law (constitution). Go ahead and be crucial about a politician’s actions but don’t insult their person’s honour or use a slur. That’s not your freedom of speech, that’s the dignity. In fact, you can even insult the government! You can say German government as the government is not a person.
>Your neighbor is still not allowed to defame you.
Anyone can defame anyone else on the US. The only time the libel or slander laws apply is when the defamed person can prove real harm in court. Not harm to dignity, but monetary loss, personal loss, or physical injury. These are very high bars to clear.
If people could sue and win just for proving willful or negligent defamation of character, a lot of extremist influencers would be in the poor house.
> A great example is the banning of RT, they lost licenses IMO in multiple countries, but the agency was spreading a lot of lies. IMO what we all want is objective news reporting.
You shouldn't need a "license" to publish a website.
They had TV licenses. Also they are the state media arm of a country that is in a proxy war with the EU and NATO. I don't think that situation would even pass muster in the US.
I have heard of RT lying but I have never actually seen examples of specific lies. Is there any list out there where they list any specific ones? If they do it a lot, it should be quite easy, no?
> The January 14, 2016, edition of Weekly Disinformation Review reported the reemergence of several previously debunked Russian propaganda stories, including that Polish President Andrzej Duda was insisting that Ukraine return former Polish territory, that Islamic State fighters were joining pro-Ukrainian forces, and that there was a Western-backed coup in Kiev, Ukraine’s capital.11
> Sometimes, Russian propaganda is picked up and rebroadcast by legitimate news outlets; more frequently, social media repeats the themes, messages, or falsehoods introduced by one of Russia’s many dissemination channels. For example, German news sources rebroadcast Russian disinformation about atrocities in Ukraine in early 2014, and Russian disinformation about EU plans to deny visas to young Ukrainian men was repeated with such frequency in Ukrainian media that the Ukrainian general staff felt compelled to post a rebuttal.12
> Sometimes, however, events reported in Russian propaganda are wholly manufactured, like the 2014 social media campaign to create panic about an explosion and chemical plume in St. Mary's Parish, Louisiana, that never happened.15 Russian propaganda has relied on manufactured evidence—often photographic. Some of these images are easily exposed as fake due to poor photo editing, such as discrepancies of scale, or the availability of the original (pre-altered) image.16 Russian propagandists have been caught hiring actors to portray victims of manufactured atrocities or crimes for news reports (as was the case when Viktoria Schmidt pretended to have been attacked by Syrian refugees in Germany for Russian's Zvezda TV network), or faking on-scene news reporting (as shown in a leaked video in which “reporter” Maria Katasonova is revealed to be in a darkened room with explosion sounds playing in the background rather than on a battlefield in Donetsk when a light is switched on during the recording).17
> RT stated that blogger Brown Moses (a staunch critic of Syria's Assad regime whose real name is Eliot Higgins) had provided analysis of footage suggesting that chemical weapon attacks on August 21, 2013, had been perpetrated by Syrian rebels. In fact, Higgins's analysis concluded that the Syrian government was responsible for the attacks and that the footage had been faked to shift the blame.18 Similarly, several scholars and journalists, including Edward Lucas, Luke Harding, and Don Jensen, have reported that books that they did not write—and containing views clearly contrary to their own—had been published in Russian under their names.
I found that source on the Wikipedia page for RT after a couple of minutes. You can find more pretty easily.
Thousands of people in the UK have been arrested for social media posts, some for speech recognized as protected by international organizations.
Germany is currently actively campaigning to force everyone to use their real names on all social media and force ID checks to do so, a clear chilling effect for free speech.
Macron has been railing against free speech specifically in recent months, calling it "bullshit".
Europe is against free speech, any argument to the contrary must contend with the above examples of them trampling on rights.
> Germany is currently actively campaigning to force everyone to use their real names on all social media and force ID checks to do so, a clear chilling effect for free speech.
Source? (Other than one derailed politician, which unfortunately we get to call our chancellor, having a moment? He's still not "Germany", though, not even "the German government".)
> Macron has been railing against free speech specifically in recent months, calling it "bullshit".
Huh? You're saying the German Chancellor does not represent the German government? [1] Large swathes of the CDU support it as well.
Macron was responding to criticism of the Digital Services Act, which contains censorship provisions for 'hate speech', which is repeatedly and routinely used by European nations to crack down on protected political speech. For example, it has been used as an excuse to censor political views leaning anti-immigration.
The UK in particular has used Ofcom as a weapon to target American companies that enable free speech communications, notably 4chan.
> Huh? You're saying the German Chancellor does not represent the German government?
I'm saying, there is a huge difference between a random utterance of the chancellor, which by next week he'll likely already have forgotten about, and "Germany actively campaigning" e.g. at the EU or federal level, both of which would require both ruling parties to get behind the chancellor's demands, which – based on how similar discourses have turned out in the past – is completely unlikely.
I'm not defending Merz's position, not by a long shot. I'm just saying that, based on previous experience, we're still quite far away from the "actively campaigning" stage and very, very, very far away from Merz's ideas being turned into law. I'm concerned about many things but this is not one of them. Civil rights organizations are already rallying and telling him how stupid he is¹ for suggesting that real name enforcement would be a good idea. :-) It's the usual political discourse.
¹) See how I am exercising my right to free speech and am not at all concerned about being charged for "insulting a politician"?
> the Digital Services Act […] The UK in particular
You do realize that the UK is not part of the EU? So I'm not sure how UK's supposed "weaponization" of Ofcom has anything to do with Macron's statement.
> which is repeatedly and routinely used by European nations to crack down on protected political speech.
I'm really looking forward to your sources here. The DSA does not contain any provisions that change anything about the legality of speech. It's mostly meant to harmonize procedural aspects across the member states.
But the Digital Services Act is EU-specific? Macron's statement referenced the DSA specifically, so I don't know what the UK has to do with that.
> As for the DSA censorship, I don't think you've read it.
I have. In fact, it seems you didn't read the links I shared, given that the second reference specifically addresses the – quite frankly – bullshit House Judiciary Comittee Republicans' report you linked to. (Again, to emphasize, this report was authored by the committee's Republican members only. In today's MAGA-controlled congress, I don't think such a report can count as authorative reference any longer.)
It cites specific sections of the DSA. Your previous claim was that DSA did not have hate speech provisions. Are you claiming DSA Article 22 does not exist, for example?
Just to be sure, by "it" you're referring to the committee report?
> Your previous claim was that DSA did not have hate speech provisions. Are you claiming DSA Article 22 does not exist, for example?
Please do quote the parts of DSA Article 22 that regulate hate speech or speech in general. It says absolutely nothing of the kind. It concerns itself with "illegal content" and defines procedures to handle it. What content is legal or illegal is defined by the laws already in place in the different member states. Also, procedures to handle illegal content already existed at a national level before DSA was enacted, so the only thing that DSA did was to harmonize them.
> some for speech recognized as protected by international organizations.
Can you share some concrete examples from reputable sources that show these? Every examples I've seen have been clear-cut calls for violence, or unambiguous harassment.
The only semi-concrete example that article gives:
> After the Southport stabbings, several people were questioned by police over false communications for spreading claims the attacker was a Muslim immigrant. In one instance, a man pleaded guilty to the offence for a livestreamed video on TikTok where he falsely claimed he was “running for his life” from rioters in Derby.
That very much seems like an attempt to harass or invite harassment against a group of people...
The propaganda take I keep seeing is that you can get arrested for misgendering people or something, but these are at least close to incitement to violence. Some clearly cross that line.
To be clear I’m closer to the American view. I think the bar should be very, very high for speech to be criminally actionable. Just pointing out that it doesn’t seem as nuts as some make it sound.
"Internet freedom declined in the United Kingdom during the coverage period due to a reported increase in criminal charges for online speech"
"A separate report from The Telegraph found that 292 people had been charged for spreading false information and “threatening communications” under the Online Safety Act between when it came into effect in 2023 and February 2025. Some civil liberties groups expressed concern that the laws were being applied broadly and in some cases punished speech protected by international human rights standards (C3)."
"Legal experts have also questioned the new rules. David Hardstaff, a serious crime expert at the law firm BCL Solicitors, said the fake news offence was “problematic both for its potential to stifle free speech if misused, but equally for its lack of clarity and consistency”."
> Isn’t one agent fast enough? Why lose accuracy over +- one week to write a compiler?
My thinking as well, IMO it is because you need to wait for results for longer. You basically want to shorten the loops to improve the system. It hints at a problem that most of what we see is a challenge to seed a good context for it to successfully do something in many iterations.
Very much the same experience. But it does not talk much about the project setup and the influence of it on the session success. In the narrow scoped projects it works really well, especially when tests are easy to execute. I found that this approach melts down when facing enterprise software with large repositories and unconventional layouts. Then you need to do a bunch of context management upfront, and verbose instructions for evaluations. But we know what it needs is a refactor thats all.
And the post touches on a next type of a problem, how to plan far ahead of time to utilise agents when you are away. It is a difficult problem but IMO we’re going in a direction of having some sort of shared “templated plans”/workflows and budgeted/throttled task execution to achieve that. It is like you want to give a little world to explore so that it does not stop early, like a little game to play, then you come back in the morning and check how far it went.
What truth is it that you cannot say in Europe? You can say pretty much anything and be critical and nothing will happen to you. And if something happens there are instruments like European court system which you can use to fight your case (there is no need to be rich for that).