jeroenhd

Apple is a digital services company that happens to sell hardware. Their big money maker is their app store, and no Linux user is ever going to buy apps from the app store.

They still have the Darwin kernel open,but more and more of the open core is moving to closed components, a recipe for what Google started doing to Android. Now that they're no longer the hipster underdog, I don't think they care much about the brand marketing. You already believe they make the best laptops by far, what more marketing do they need?

From a privacy viewpoint: yes

From a practical viewpoint: the desktop tower in the corner of my room that acted as my NAS+router+home server+home assistant was running quite hot and the poor undervolted i3 CPU wasn't able to keep up.

> the EU is still a patchwork of entities that do not have a common vision of what the future should be

For many, that's a feature, not a bug. The EU follows a democratic system consisting of many different countries with different types of government and different ideologies. It's not a unified federal government, as much as some people would like it to be.

The whole 28th regime concept seems extremely flawed to me. I understand the desire from a business perspective, but as a citizen I do not want a company to opt out of national legal protections and obligations by operating under some fantasy government. Unless this concept will be subject to the strongest, best-enforced regulations and tax rates equivalent to the highest tax rates within the Union, I do not want this project to happen, and I predict I'm far from the only one.

It has a tendency of doing so, but in this case the body that was supposed to patch over the requirements for EU data transfer was flawed in its design.

The reason this house of cards was necessary in the first place is that the American government does not want to grant foreign citizens the rights necessary to ensure the privacy guarantees the EU requires.

American courts deciding that institutional independence is bad now is awful for American citizens, but it's not supposed to be very relevant to the EU like this.

If you think running a Linux desktop on an Ampere is bad, try running it on an M5 Mac!

The US supreme court is correcting the lies the American government made when they assured the EU and its citizens that they can be trusted with their data. It's not just the EU lying, both sides are awful at this.

I don't know why the EU wants to trust the USA so bad, it's clearly unwise. It makes sense, because banning EU companies from using AWS/GCP/etc. would bankrupt the EU into a recession, but the way they're going about these things is very annoying.

That said, if the USA would actually keep its promises and adopt legislation that solves the reasons why the EU cannot give out a decent competency decision, the problem would go away entirely.

The Biden administration set up a precarious body within the government to resolve the issue rather than go through the normal lawmaking process, probably because it wouldn't go through.

A small group of people from the EU parliament is going against the wishes of the EU commission in an attempt to force through a change that contains a subsection of the bill that tries to mandate E2EE scanning.

The way this is going is definitely worrying, but what you're saying is disingenous at best.

Furthermore, even if this passes somehow, that doesn't change the fact that the US remains an unreliable partner. Now we have two governments scouring through your data instead of one.

> I'd wager that that's the configuration of vast majority of the consumer systems out there...

I'm not so sure. Most people I know, including developers, only have one SSD, maybe a hard drive for photos or whatever that's not mounted inside /home.

The most popular mainstream Linux distributions (Android, ChromeOS, SteamOS) all default to a single disk setup, as do most virtual machines (which must exceed the number of home installs by at least a thousandfold).

For a particular brand of enthusiasts, running multi-disk is definitely more popular (I have 4 in my desktop), but I don't think those are the majority, and even among those splitting partitions across disks rather than just using a /media (or equivalent) to access the larger disk like you would an external drive is quite popular.

Multi-disk detection would be nice, but I can already imagine the absolute mess that would happen if you were to try to boot on a multi-distro PC, or even just a recovery flash drive on an existing Linux image. My recovery boot drive only has a root partition so auto-detecting the drive will work fine, but auto-mounting my home directory to /home would mess up the entire system.

The code would be so full of edge cases and weird guards that it'll inevitably just break without configuration, making the entire thing pretty damn useless.

A mac with a boatload of RAM can run models that will exceed the limits of any GPU not worth at least twice the Apple hardware itself.

You get fewer tokens per second, but at some point the balance between quality and quantity makes the large model size worth the spend.

When you're spending this kind of money, you may as well treat yourself to a pretty screen and some decent speakers. Nothing the competition doesn't offer these days, but you get them for free with the car-priced RAM upgrade so why go for less.

With the new Neo prices, they're more coming for the lower-end of the midrange market. The low-end market was half the price the Neo was before the price went up.

Google has gone after the low-end market through ChromeOS and has actually been very succesful at it. Sure, the $200 pieces of crap can't really do anything more than play a video and maybe type into a document at the same time, but that's better than what the Windows alternative offered for many years.

That just means your IT department isn't putting in the effort. Antivirus, remote management, MitM proxies, software permissions, macOS can do everything Windows can, it just takes three or four subscriptions and double the effort for the IT team.

macOS makes it kind of difficult to do these things, so when companies do deploy the same control they use on Windows, the problem actually becomes worse, because every management tool now needs to rely on hacks rather than officially supported APIs.

Exactly, we need Windows Lite! Just Win32, media encoders, graphics drivers, CoPilot, and enough to run movie maker and Office, the rest is useless anyway!

Or do you mean no Office support, just everything you need to run Steam?

Or maybe you mean no UI and just the admin interface so you don't need all that bloat when all you need is a server. Except for Powershell and the management interface, of course.

Actually, these days, we don't need applications, Copilot can do all that for us. Just Edge, Bing, and Copilot, the rest is bloat.

There's plenty of unnecessary crap in Windows but everyone who wants Windows Lite wants a different subset to be included in Windows Lite. Just imagine the tech support hell you'd be in when a client installs your software on a machine that doesn't have .NET or H.264 support or libraries for decoding JPEGs or a webview because their kid said they can make the PC less bloated. And you can bet that Windows' remote help support and all of the requirements for third party remote desktop are omitted from anyone's view of "minimal Windows"!

For what it's worth, various stripped-down Windows images exist online. They all use official Windows binaries with signed DLLs and everything so you could theoretically even verify that they're safe. Just don't expect your programs to work on them and be ready to reinstall when a system service inevitably corrupts.

> instead of having it as a layer which you can install and enable if you need so is a bit sinister

I don't really get it. If you don't want systemd, you can install your own OpenRC-based immutable OS. I don't know if that currently exists (Android does, I suppose? maybe ChromeOS?) but it's not like systemd is forbidding everyone from doing anything.

The biggest difference between systemd and alternatives is that systemd builds out the architecture that makes things like building immutable distros easy. If you have a thousand different components, all with slightly different roles, the amount of coordination necessary to get everything to play nice together makes it near impossible to achieve the same things a coordinated system can achieve.

I'm sure that if someone would build an alternative to systemd-sysext and prepare a prebuilt set of tooling and configuration that does what systemd does, there will be distros that will use them.

Secure boot can be overridden by just adding your own keys. At worst, Microsoft will refuse to boot Windows because you added a key yourself. The exception is extra expensive, specially labeled Enterprime (TM) hardware built for Windows sysadmins and Microsoft tablets built for consumers, but I don't know why you'd buy either if you don't want to give MS that much control.

systemd replaces the duct-tape-and-tie-wrap systems that many Linux distros used to run with a unified set of tools, which can be interchanged, but often aren't. You don't have to enable container support, DNS servers, or disk mounts in systemd, but support is there in case you need it. Some distros stopped shipping bespoke DNS servers, others still do, it depends on the distro; it's not like distros are forced to enable the whole set. I don't know any distro that actually does, actually.

If you're the kind of person that wants to deliberately choose what software handles your DNS and what software configures your TTY and which cron daemon you use, then systemd limits your options somewhat. If you want to go in and re-write the init scripts for a system daemon because you have specific requirements, you still can but it adds a level of indirection to the whole thing.

systemd works great for the 95% who can do with a reasonable set of defaults. The remaining 5% will run into annoyances about defaults and obscure configuration, and a small (but extremely loud) subset of those people hate the idea of one set of software taking over so many tasks.

In reality, the option not to use systemd has always been there, but many distros find it much easier to use and most software doesn't bother with the ancient init scripts anymore. You can install openrc, but you're going to need to write most of your service management configuration (often shell scripts) yourself

> converting Linux to a locked-down immutable OS

That approach is part of why Valve has been able to ship Linux to general consumers.

I don't know exactly why you'd go for amutable specifically, but not handing users the key is a huge benefit to any company-owned hardware out there.

> you can't spread your /home and / partitions on separate disks and use this

No, you'll need to add a line to your fstab manually if you want something autodiscovery cannot do, like multi-disk setups, LVM, and FDE with unsupported configurations. It's great if it works for you and a very minor change if it doesn't.

I don't recall ever installing a system with / and /home on separate partitions, though.

I do agree that systemd could use some competition, but until someone builds one, we're going to be dealing with some systemd defaults. Canonical's attempts seem to have been most complete, but outrage at change and different ideas have shut down most of Canonical's attempts to improve the situation.

It's far from janky, at least compared to what else is out there for Linux. If you're not writing systemd service files, you're almost surely back to writing unpredictable 80s style scripting jank.

dnscrypt supports ODoH so any device capable of running that will do. Other than that, you'll need individual app support (like browsers for instance).

If you want support on mobile devices without VPN-like apps, I think the best way to set it up would be to run something like a PiHole or equivalent, configure dnscrypt as your upstream, and then set up DoH/DoT from your phone to your DNS server. Mobile phones can do DoT/DoH natively already, but I don't think any commercial mobile OS has extended support to ODoH.

In theory you may be able to run dnscrypt in the background and point your DNS resolver to that, but I doubt it'll work reliably.

Facebook has been documented for sponsoring these bills in the USA.

But we also have years of research about how social media and other internet phenomena are ruining kids' lives.

I don't disagree with the principle of a lot of these laws, but many implementations are too flawed to be a mistake.

My DoH server ended up on a random list on Github at some point. I noticed when I saw what seemed like a small country suddenly use my DoH server.

Blacklisting the entire country worked, after that I moved my actual DoH resolver to a subpath. Because it's HTTPS, you can just run your DoH server at https://my-doh.example.com/066c591f-c976-4095-85fe-a49e62577.... Not as easy to remember, but you can send yourself and anyone you want to share the server with a link.

Other things to consider when setting up your own DoH server: setting up HTTP3 with HTTPS records and the like, 0-rtt TLS for the query server, ODoH support (upstream or as an endpoint directly), and of course DNSSEC validation (because you can't trust your clients to the validation themselves).

For DoT this is a lot harder. A random IPv6 address should work, but then you're stuck having to fall back to something else on networks with only legacy IP support.

The nitrogen problems has been ignored for so long and nitrogen compound deposition has been so intense for a long time that you can't really do anything anywhere without depositing nitrogen compounds in an area already suffering from the environmental effects.

The Netherlands is a very densely populated country compared to the most of Europe. Furthermore, there are industrial hubs bringing in polluted air through the wind from the west, south, and east, making up a significant source of the nitrogen compound deposition. All efforts to solve the problem have so far upset very powerful lobbyists whose income relies on being allowed to pollute more to stay ahead of the competition (not to mention caused violent protests).

There are many factors to the housing problem, it's not just nitrogen compounds. In my opinion, the entire construction sector coming to a standstill after the 2008 crisis was probably what kicked off a storm of seemingly unrelated issues, from population pyramids to the water table levels to investors using property to accumulate wealth to hyperintensive farming practices.

New measures to supposedly solve the nitrogen compound crisis have been announced. By the looks of it, I expect the agriculture lobby to riot again, and new elections by the start of next year when the government inevitably collapses itself again in an attempt to use populism to gather more votes.

I don't buy that anymore. The day America threatened to invade Canada and Denmark was the day America showed they cannot be trusted any more.

It's not like China can be trusted either, but China isn't planning any direct invasions to the west. Taiwan, perhaps, but they're playing a long-term tactical game rather than a "invade the country we don't like this week" game. They might get some info on you, but the data brokers in the west will sell a lot more details about you, pre-categorized and all.

If you're afraid of industrial espionage, Chinese companies may be a risk, but in that case you shouldn't be uploading your secrets to an AI company in the first place.

Companies selling these titles should know a minimum end date. Even if contracts don't get renewed, it's unlikely they will only have the rights for less than a year.

If that minimum drives customers away, these companies should put more work into ensuring their minimum availability is a good deal.

And if that one-time purchased software stops working at an arbitrary date, it should be subject to the same rules. Especially online software or software requiring servers to run.

You can still offer limited-time subscriptions, of course, and you can extend the minimum deadline for your server-dependent software to free as often as you want, just make sure people know what the deal is when they buy your software.

DVDs and other media also aren't yours to buy, they're just licenses and a physical container to use that license. You can buy software the same way you can buy a DVD, and you can rent software the same way you can rent a movie on a digital storefront.

The video mentions that drivers were needed to get the full speed on Windows, and that the Realtek Linux drivers didn't compile on a modern kernel. So it's probably software.

Realtek makes some pretty affordable networking chips but their Linux drivers can be a real gamble. Either it works out of the box or you're in for years of messing around.

People do it. I think the time and learning resources it takes to do it all yourself is probably not worth it for a company trying to make it big, especially if you stack your BV inside of a holding inside of a stichting to get something similar to the OP tax-avoidance-cum-risk-reduction strategy.

The risk of making a mistake and getting hit with a huge fine, as well as all the time wasted on crunching numbers and reading tax code, probably makes it very difficult to justify doing it yourself if you have a profitable business. If you do not make the money you need to make paying an accountant make sense, I think setting up a BV would probably not be the best approach.

Getting things like payslips right (you have to pay yourself a defined minimum wage and that means paying income tax and getting all the payslips correct if you set up a BV) may seem simple, but there's a reason companies pay good money to have that all taken care of. Just simple things like the possibility of a 53 week year can trip up someone who is just starting out.

But at the end of the day, plenty of people do their own paperwork, deliver the taxes themselves, and file the yearly required documents. It's neither impossible nor illegal to do it all yourself, just often unpopular.

> went to a lawyer

They studied AI and are building an AI company. I doubt the idea for the business structure come from a lawyer, to be honest. Especially such an overcomplicated setup with so many real-world issues that they're running into right now.

Sometimes criminals do put pressure on people to register companies in their name and show up to these meetings. Whether that's addicts, people with mental disabilities, or young people looking for a quick buck, the fraud mechanism is the same.

However, that does put the company in their name. On paper, they have full control over it. That's a risk to the criminals trying to use the company as a financial asset for laundering money.

Registering a private limited liability company in the Netherlands costs around 400 euros. If you can file all of the taxes and other legally required paperwork yourself, you can be set up in a week or two. You will be a salaried employee of your own company, though, with a minimal salary you will need to rake in.

The combination of "no personal risk whatsoever, minimal funds/risk coverage, maximal profit extraction" doesn't lend itself well to places with basic regulations.

Capital investments in Europe are definitely not as easy to obtain as in the US for various economic, cultural, and historic reasons, which all led to some pretty weird laws here and there, but the extra week it takes to set up a business isn't the cause.

The reason this all took so long and was so expensive is simple. As the author states:

> I wanted real limited liability

They wanted two different companies with different setups to get out of having to save up the funds or find investors while also paying the least amount of tax possible. They set up a two-company system with all the risk in one and all the earnings in the other. It's like one of those tax dodging schemes the multinationals like, except within a single country. That comes with overhead.

Funnily enough, they then end with:

> Which leaves the only real question. Why 25,000 at all? It is my company and my risk.

Weird to think it would be their own risk if they spend so much time, money, and effort setting up a system that explicitly removes all the risk from them.

All of this feels like it was based on a business plan generated by some over-eager AI that tried to optimize to tick as many boxes as possible, ignoring the real-world consequences of those choices.

> I don't understand why big tech and EC want to scan messages

I believe that's a form of corporate greenwashing. If you can prove your claims that you do everything in your power to prevent abusive materials, you're going to get less attention from annoying authorities next time a pedo network/terrorist cell hits the news.

Aside from that, there are a lot of well-meaning people who want to try every little thing to help stop horrific abuse. Police investigations are happening too slowly (if at all); if the police won't help solve the problem, going the civil route may help, even if just a little. I think it's an act of desperation rather than malice. Plus, just like there are plenty of people who say "they've got nothing to hide, I don't need encryption", there are plenty of people who feel like a tech panopticon is worth it if it catches some abuse cases. Besides, in cases like these, scientific evidence often doesn't matter as much as the emotion behind proposals, and most messaging providers couldn't appear more devoid of emotion if they tried.

I don't agree with the idea to scan every message for various reasons. If the police won't investigate criminals with the massive amount of power they already possess, overwhelming them with "abusive" material from an algorithm is only going to make it harder to filter out the real criminals. Plus, if a few large providers do it, that will put pressure on all the other providers that don't do it (see, for instance, that time they arrested the CEO of Telegram for not volunteering information without a warrant, like other messaging providers seem to do).

But, as much as I disagree, I do understand where those people are coming from. And then there are also the blatant comic-book villains that just want totalitarian government control over all information exchange, of course.