Sure, but I’m just playing around with small quantized models on my laptop with integrated graphics and the RAM was insanely cheap. It just interests me what LLMs are capable of that can be run on such hardware. For example, llama 3.2 3B only needs about 3.5 GB of RAM, runs at about 10 tokens per second and while it’s in no way comparable to the LLMs that I use for my day to day tasks, it doesn’t seem to be that bad. Llama 3.1 8B runs at about half that speed, which is a bit slow, but still bearable. Anything bigger than that is too slow to be useful, but still interesting to try for comparison.

I’ve got an old desktop with a pretty decent GPU in it with 24 GB of VRAM, but it’s collecting dust. It’s noisy and power hungry (older generation dual socket Intel Xeon) and still incapable of running large LLMs without additional GPUs. Even if it were capable, I wouldn’t want it to be turned on all the time due to the noise and heat in my home office, so I’ve not even tried running anything on it yet.

The only time I can remember 16 GB not being sufficient for me is when I tried to run an LLM that required a tad more than 11 GB and I had just under 11 GB of memory available due to the other applications that were running.

I guess my usage is relatively lightweight. A browser with a maximum of about 100 open tabs, a terminal, a couple of other applications (some of them electron based) and sometimes a VM that I allocate maybe 4 GB to or something. And the occasional Age of Empires II DE, which even runs fine on my other laptop from 2016 with 16 GB of RAM in it. I still ordered 32 GB so I can play around with local LLMs a bit more.

I’m not going to defend Apple’s profit maximization strategy here, but I disagree. Most people won’t end up buying a cable and adaptare because they already have one, and in contrast to those pieces made of plastic and metal, the packaging is mostly made of paper. I’m pretty confident that the reduction in plastic and metal makes up for the extra packaging that’s produced for the minority that does buy a cable and/or adapter.

I would look into how Matrix handles this, for example. It involves unique device keys, device verification from a trusted device, and cross-signing. It’s not just some private key that’s spread around to random new devices where you lose track of.

They’ve implemented it in such a way that you only have access to an encrypted chat on a single device, so no syncing between devices. Syncing E2EE chats across devices is more difficult to pull off, but it’s definitely possible and other services do that by default.

“Uncensored”: https://x.com/KarlMaxxer/status/1823753493783699901. I don’t know if this is really true, but if it is, it’s something that they should’ve called out in their article.

In many countries the age of consent depends on the context. By the looks of it, the Czech Republic is one of those countries:

The age of sexual consent in the Czech Republic is 15.

Additionally, the section of the Czech penal code 40/2009 Sb. covering “crimes against family and children” contains § 202 which criminalizes a “seduction to sexual intercourse” of any persons under 18 years by any promise or provision of payment, benefit, privilege or profit, for sexual intercourse, masturbation, exposure or similar behavior.

https://en.wikipedia.org/wiki/Ages_of_consent_in_Europe

Right, I see what you mean now. I misread your comment as explaining something that was already clear.

A false positive is when it incorrectly determines that a human written text is written by AI. While a detection rate of 99.9% sounds impressive, it’s not very reliable if it comes with a false positive rate of 20%.

The kernel is written in C and a bit of assembly. When support for a new language is added, that’s big news and worthy of headlines. The other languages are just there for various tools and helper scripts and are not used for kernel code, so not newsworthy by any means.

The black one is the 1x and is the tastiest of the three in my opinion. The 3x seems to be tastier than the 2x (both red), but that’s based on memory, as I haven’t tasted them side by side. We buy the 1x quite often and add some extra ingredients to it (egg, spring onions, crispy chili oil, cheese). It’s spicy, but not extremely so according to our taste, especially not with the extra ingredients.

“Our analysis shows that the two voices are similar but likely not identical,” Berisha said.

They also point out the main differences between the two voices in the paragraphs below this quote. I do believe that they hired a voice actress and that they didn’t train on SJ’s voice, or at least not entirely. However, I wouldn’t be surprised if there was big push for finding a voice similar to SJ’s voice in Her, no matter how much they deny this.

You’re probably thinking of Finland. Often rated as the happiest country in the world, yet also scoring high on depression rates, which gets pointed out often on the internet. See https://worldpopulationreview.com/country-rankings/depression-rates-by-country. Norway scores pretty low.

For me they only work in relatively quiet environments, or with earplugs. As soon as a car drives by it completely drowns out the sound. With music that might not be an issue, but with podcasts or calls it’s very annoying. I’ve bought earplugs especially for this, as my other earbuds have issues with wind while running, but it does feel like it’s defeating the purpose a bit. I guess turning them all the way up would also work, but that doesn’t feel healthy. Other than that I like them and the mic quality is also good according to people I’ve spoken with over the phone.

WSL 1 is a compatibility layer that lets Linux programs run on the Windows kernel by translating Linux system calls to Windows system calls, so in that sense I understand the name: it’s a Windows subsystem for Linux [compatibility]. It doesn’t use the Linux kernel at all. With WSL 2 they’re using a real Linux kernel in a virtual machine, so there the name doesn’t make much sense anymore.

And it does proper split DNS by default, using the search domains of each interface. That way you can configure a global DNS resolver while still being able to resolve local hostnames and without leaking other queries. I just hope they’ll also add DoH support, which is less likely to be blocked on a corporate network.

Note, however, that the mere fact that all those apps exist for iOS adds a lot of value for Apple too. Apple wouldn’t sell nearly as many iPhones if the most important apps weren’t available on their platform. They spin it as if they are only creating value for the app developers without asking for much in return, while the App Store is an enormous cash cow, which they’ve been able to build due to the lack of restrictions (pre DMA). A good API is not just a service for app developers, it’s a way to enhance the user experience and sell more phones, because of all the work that app developers do to turn it into useful and exciting features.

I don’t doubt the fact that they take some margin to extend the lifetime of the battery, but if we take iPhones as an example, they:

• charge at a slower rate when nearing 100%
• try to postpone charging the final 20% until the last moment before disconnecting from the wall outlet
• can be software capped at 80% by the user (in newer models)

This makes me suspect that that the margin between what’s reported in software as 100% and the actual capacity of the battery is less than 20%. This also makes sense from the standpoint of the consumer expecting a long battery life on their expensive high-end device, putting pressure on the companies to make the margin smaller and the charging algorithms smarter. Just my observations, of course.

I’m pretty sure that Chrome’s alternative is designed by Google to track you in a way that’s harder to block and gives them more control over the advertising market by forcing advertisers to play along and use their method instead of collecting your data directly. Sure, it’s more private, but it’s still tracking you.

Firefox, on the other hand, is focusing on completely blocking cross-site tracking. They have no incentive to completely block 3rd party cookies as long as there is also a legitimate use case for them, but I guess they will eventually also block them if Chrome is successful in forcing websites to stop relying on them for core functionality.

But then when you’re talking about 10:00 hours without specifying anything else, it actually means something completely different in the local context, apart from it being the exact same time globally. It doesn’t tell you whether it’s night or day at the other persons location. Your default point of reference in that system is the world, while even today, time is mostly used in a local context for most people. When I’m talking to someone abroad and I say “my cat woke me up at 5:00 in the morning”, I expect the other person to get the meaning of that, because the other person understands my local context.

When planning meetings you’d have to now the offset either way, because I’m not going to meet at idiotic times if there is an overlap in working hours between the two countries, which is something that you’d have to look up regardless of the time system. And if I send out a digital invite to someone abroad, the time zone information is already encoded inside it, and it shows up correctly in the other person’s agenda without the need to use a global time. In that sense UTC already is the global time and the local context is already an offset to that in the current system. We just don’t use UTC in our daily language.

But if it helps: I do agree that in an alternative universe the time system could’ve worked like that and it would have functioned. I just don’t see it as a better alternative. It’s the same complexity repackaged and with its own unique downsides.