>French media reports that an OFMIN (small office) warrant was issued against him by the French National Directorate of the Judicial Police based on a preliminary investigation on Durov's supposed "failure to control content on Telegram".
You can make the same argument about Minecraft since servers download you texture packs, data packs or skins. I don't think that Apple should stop these apps, that's their game model.
I think that the point is to be easy to be changed, since Mermaid's first use case is to be used in building diagrams/graphs with data fetched at runtime like telemetry or finances data. I guess you can probably use Mermaid and take the resulting html and add it to the resulting document if you really want.
A few years ago, if you said you buy AMD, people would think you are hallucinating, but now it looks like it's the only reliable vendor for x64. Intel was once the king of reliability, but in the last years, it looks like the king of bugs.
If you complain the the current generation of LLMs isn't as impressive as some people try to prove it is, the shills complain that you used the wrong prompts and maybe you should fine-tune the model. I don't think people should spend much time trying to write a prompt to find a working solution because they could just do that. In the author's case, if he used the summary provided by ChatGPT, I doubt that anyone would have realized it's wrong, but that doesn't mean it's good either.
The problem with LLMs is the lack of reliability and coherence. If it gives you a wrong answer and you ask it if it's sure, in most cases it will show another wrong answer and you need to go through multiple of these hops to get something fine.
The real thing I've learned from this is that most malware doesn't get ring 0 access, but these "antiviruses" solutions can shutdown the whole critical infrastructure everywhere. I've never seen malware that infected millions of devices in seconds and made them unusable for days and maybe weeks to come.
You can't since man pages present the libc implementation, that's more useful if you work with syscalls that use structures since it shows you what to search for in libc to copy it in other language.
Let's be honest, he will resign in 2 weeks and move to the next company to be a glorified salesman. I wonder if the companies affected by this will sue Crowdstrike and the CEO or he'll just leave in peace pretending he did a good job there.
Most LTS strategies are like this, enterprises run the LTS version on the server while the consumers run the latest version. In a way, it is beta testing, but the consumer isn't really mad about it since he gets new features or performance boosts. LTS users usually update once every 3-6 months or if a serious CVE comes out, while normal users update daily or weekly. To be honest, I know servers running whatever the latest version of nodejs is, instead of LTS, mostly because they don't know that node has a LTS policy.
Fixing it isn't hard and it could probably be automated with WinPE, but the real problem is BitLocker. This makes the process tedious since you can't automate it totally and you also need the key which might require a login on microsoft's site and then writing it by hand or finding the stick that had the key. I was reading that sysadmins on reddit were complaining that they lost the passwords to the microsoft accounts so they are doomed.
I've talked with a guy from Bangladesh over the last 3 days and he said that he was at a party and he was about to be killed, so he stopped going and started filming how other students are being beaten by local police. The fact that the government blocked the internet is pretty sad and I wish that no more people die.
The antivirus did its job, now you can't get viruses. Jokes aside, I've checked their website and it was full of AI buzzwords so I guess that happens when you focus on nonsense instead of what your customers actually need (I know that all antiviruses have a machine learning component, but usually you don't advertise it as some sort of AI to get better stocks).
Poetry is so slow, please use uv, for some of my projects, poetry takes 20-30 seconds to update/add new packages while uv completes it in seconds. But I can agree that's better to use a package manager than plain old `requirements.txt` file since you can pin the version you want to use for development or production.
I opened your website and found it downloads 1.8mb of js files. It looks like it's too large for a website that could as well just be a static site made with hugo.
I don't use Vivaldi, but I think it's fine on their side to avoid falling into this AI trap. You have to consider that there were 2 possibilities, they used their API key to send requests to openai/claude or they did what Mozilla did and integrated gemini/chatgpt at the expense of the user data, both of which are pretty scummy and barely help the end user. The only way to integrate AI fine into browser is to ship a model to run local, but that would require the user to have the computing power to run it. Google is trying this with chrome by integrating gemini mini in the browser as a 100mb blob.
I don't really get the appeal of this, I'd just use vscode with Jupyter if I really wanted "ai" integration since I can then access the whole ecosystem of extensions. The idea isn't that bad, but it lacks purpose.
Refrigators now have "software" so you can order Uber and doordash, everything used to build it is cheaper and overal less reliable and the whole structure isn't meant to last. After all, how will manufacturers make money if your refrigerator works for 10-15 years? The whole industry is built around constant (3-5 years) replacement of electronics.
The old logo looked cute, the current one looks like a meta copycat. I wish they went for something with more personality, not this type of corporate logo.
As usual, Twitter is impressed by this, but I'm very skeptical, the chance of it breaking your program is pretty high. The thing that makes optimizations so hard to make is that they have to match the behavior without optimizations (unless you have UBs), which is something that LLMs probably will struggle with since they can't exactly understand the code and execution tree.
I kinda agree with the article, very few people go out of their way to review your application, usually it's people who want to report issues or are just there to give you 1 star and call it a day. Personally, I don't review the application when I'm prompted, but I know a lot of people that do this and it's probably better for the marketplace to get ratings from more people to rank the applications.
Also, I read in this article that Google forbids that review gating tactic, but I know 10-20 apps that asked me for feedback, I refused and then sent me to a random website to tell them what's wrong with this app. After that, of course, I went out of my way to drop 1 star since this is the worst tactic to gather engagement and turns people away from your app.
There is a bigger chance that a toddler smashing a keyboard finds a bug than gpt5. LLMs can't understand intent, so they literally work like `grep` with little to no understanding of the context, so most of the time it will false flag good code.
There are already a lot of tools already to find bugs, like fuzzers, but I am sure that LLMs won't be one of them.
Raspberry Pi is mostly an expensive toy at this point, it costs 60-80$ and it's pretty hard to buy one to begin with. You can't use it for most electronics projects because it's overkill and also it's not as powerful as a n100 or a second hand pc. If you are lucky, you can snipe a bunch of ryzen 3 hps or dells (if you don't need a lot of performance) for 60-80$ or ryzen 5s at 120$ (if you need a bit more).
>This is a step backwards for European innovation, competition in AI development and further delays bringing the benefits of AI to people in Europe
How would be using European users' data to train an AI a step forward for European innovation in AI when we're talking about Meta, which is an american company? Also I don't think that europeans are mad that they don't get these AI "features" that Meta tries to push, they only push to the degrading quality of facebook/instagram that were already pretty much dead.
Unless the contract made with Nuvia states explicitly that a successor has to resign a new agreement, I doubt that Arm has any chance to win in the court.
Investors don't even know what NVIDIA is selling, I was listening to a random investor podcast and they were talking about Intel, AMD and NVIDIA, but no one knew what exactly they are selling, they only knew they are part of this AI bubble so that's why you should invest in them
I know a guy who invested 60-70k in hosting servers and started a startup in cloud computing. He makes around 6-7k a month so around 4-5k after cutting all the costs (basically a salary since he maintains that all day long). I think it's a fine business even in the modern times, but the market is probably oversaturated and it's kinda hard to get consistent revenue since if you raise the price, most people will leave you and you kinda need to be trusted to have a company buy a lot of computing power from you.
It's pretty embarrassing for Apple because it's a pretty common thing for Apple to close random developer accounts with pretty respectable applications. At least, Google is a little bit more transparent from my experience, I could easily publish my application there, but on Apple, you have to go through tons of mental gymnastics and their cerficates are $100/year compared to one $20 certificate from Google.
While I can see the advantages of Python over VBA, I feel like running the code in the cloud instead of locally will be a big no from a lot of people, especially enterprises running offline Office installations.