What a strange selection of censored words… “fail to comply”, “The billionaire owner…”, “would strenghten the Australian”, “has yet to be passed”

Or you get a message at 3 am the next day, where they apologize that they didn’t made it, but that they will deliver sometime between 8 and 22 that day.

E2E is just one part of the puzzle, you got to have a open source P2P or federated architecture as well, otherwise you have to trust a nebulous company or person intrinsically. People change and companies can be bought, but you will be stuck with their platform in order to contact your acquaintances, and changing that means loosing your contacts.

That is why the DMA is important. But you will be even better off just directly choosing a chat platform, where the users are in control.

Yeah, the whole article is a bit fishy:

In addition to generating clean electricity, the new ITO-silver window coating creates a cooling effect by allowing only the visible part of the light spectrum to pass inside. Other parts of the spectrum are reflected outside.

So how would a room actively cool down, when you let only the visible light spectrum inside? Sure it might not get as hot as if you let all light inside, but it will also not get colder.

I started using Fedora Silverblue on a tablet, seems to work fine so far, but requiring a reboot in order to install new system packages is a bit cumbersome and the process itself takes a while, but ordinary Fedora also doesn’t win any races when asked to install a new package

I think switching to FCOS or Flatcar on servers that just use containers makes sense. Since it lessens the burden of administrating the base system itself. Using butan/ignition might be unusual at first, but it also allows to put the base system configuration into a git repo, and makes initial provisioning using ansible or similar unnecessary. The rest of the system and services can be managed via portainer or similar software.

I also do not have long term experience with FCOS, but the advertised features of auto-update, rolling-release, focus on security and stability makes it a good fit for container servers, IMO.

An alternative to Debian on servers might also be Apline Linux. Which also has more a focus on network devices, but some people use it on a desktop as well.

If you have many different systems, and just want to learn to operate them all, maybe NixOS might be interesting. Using flakes, you can configure multiple machines from just one repo, and share configurations between them. But getting up to speed on NixOS might not be so easy, it has a steep learning curve.

What’s ACC?

ACC - Advanced Charging Controller, which allows to set charge limits, thus extending the battery life, which should have been part of Android from the beginning,

Anyway I would strongly discourage using root under Android as it breaks the security model.

Security isn’t a binary, security works like an onion, you have multiple layers of security and multiple decisions to make on every level. Currently you might be right, that having root access to a device might compromise it in some ways, but that isn’t necessarily so and depends on how it is done.

You should find ways around using root and if you can’t you probably shouldn’t be doing on your phone anyway.

This kind of thinking is the ‘I know better than you’ mentality, that I sometimes see around people advertising GrapheneOS. Having ‘root’ permissions to the device is owing it, I want to decide what to do with it, not the vendor of the ROM, or who ever else. They aren’t me, they don’t know what I want to do with it.

The goal of security models is allowing me, the owner, to do what ever I want with my device, while preventing others, non-owners, un-trusted applications or the internet from doing what they want with my device. If the security model doesn’t allow me, the owner, to do what I want, then it failed its job at least partially.

Root is very dangerous as it can survive a factory reset.

Why is that dangerous? The first thing I do, when I get a new phone is boot into the boot loader, and overwrite the whole partition, then the system is trusted again, at least if I trust the vendor of the boot loader. When I want to do a factory reset, I do the same, overwrite the flash with a fresh OS image.

IMO, there are other reasons why the current implementation of root are dangerous: They currently considered binary and I think they could be implemented more gradually. Like one application having root over individual other applications, e.g. accessing their files. Allowing/Disallowing individual privileged system calls, or access to specific system files, etc. All of this could be hidden behind a switch in the developers menu. Maybe only allow applications to gain root access when using a registered hardware token, etc.

As for MicroG, it is sandboxed but it does require device admin for full functionality. It isn’t running as root but it requires a lot of device permissions. You can turn off the permissions you don’t need but that could break things.

In order for MicroG to work full, you need to fake the signature, which requires a patch to the system, or root privileges.

Like others already said, you can still root your GrapheneOS, there are two ways to do this:

1. Just unlock your bootloader, flash Magisk or whatever, done. Disadvantages, you cannot lock your bootloader again, thus creating a huge security gap where an attacker, when gained physical access to your phone, overwrites your boot partition and you boot your compromised system without noticing. Which is bad, IMO.

2. Recompile GrapheneOS with Magisk installed, signed it with that key and use this key in your bootloader to lock it. You essentially created a GrapheneOS fork, can no longer use their OTA update server and use the security updates, etc. You need to create this yourself.

Yeah, they don’t prevent you from doing it, the same as original ROMs don’t prevent you from doing it.

Exactly right. However all the downsides you have when doing that sort of defeats the purpose. So a GrapheneOS native way to control your device would be nice.

Well, I never really missed being able to pay via NFC on a phone, but I also never done it. My NFC chip in my card works fine.

When my baking app started detecting my rooted phone, I just switched to using their web-app via Firefox, which allows you to create a direct link to it as an “App”. Which is probably better anyway, than installing random proprietary apps on a phone. And logging into it every time is also easy with a password manager.

So I guess, as long as the banks still offer a website, I am good.

I am currently using a rooted LOS with MicroG. It certainly is not as secure as GrapheneOS in terms of app sandboxing, encryption, regular security updates, etc., but I have control of the system, in case I need it, for instance ACC, F-droid privilege extension (F-Droid auto updates), ReVanced Manager (not using it currently) etc.

I trust GrapheneOS much more than Apple, but both go into a similar direction with their understanding of security. IMO taking control away from the user might be a good option, if you are dealing with just regular consumers, but I don’t really like the “one-size-fits-all” approach of it. And it is my device, I should be allowed to decide what I want to do with it.

BTW, this is just a personal annoyance of mine. The GrapheneOS devs do a very good job.

I would like to switch, but there are a couple of points that are still holding me back right now:

• Charge limits, on LOS I can root the phone, install ACC and still use the OTA updates, if I apply the patch afterwards. (Will be resolved in A15)

• Option for sandboxed MicroG, IMO privacy is also very important for security, and people should be able to decide if they like more privacy or more security.

• Option for rooting sandboxed apps from outside. IMO I, and a person, like to have full control over my phone. Trust often comes with control. If I choose to trust one app to have root access to another app in order to inspect it, then this should be possible. Sandboxing could allow one app to have root access to individually chosen other apps, thus limiting the impact compared to system-wide root access. Maybe offer rooting gated behind a separate hardware token authentication. (sudo like) A lot there can be improved IMO, while still providing it and making it more secure in general.

I know that my understanding of security and privacy might be different from what GrapheneOS understands, but as a long time Linux Admin, I don’t like black boxes, I like to peek into them, modify or patch them, when they do something I don’t want them to do, etc. So that when I enter personal information into them, I am still in control what happens to them, at least that is my desire.

Taking control away from the user in order to “improve security” might be a valid approach to some, but it is not something I have much trust in.

“Some firefox users like firefox” vs “many chrome users enjoy what they have” sounds to me like something that could have a source. Many sound to me more than some, so this is a comparison, which can be given a better foundation by supplying some numbers.

• Just as some Firefox users like Firefox, many Chrome users enjoy what they have too. They don’t want to lose that.

Do you have some source for that? IIUC, you mean that more Chrome users like Chrome than Firefox users like Firefox, right?

IT guys will stop using it…

No, they will not, if they didn’t already. Because convenience it key.

The browser war is over, and humans lost, corporations won. Google and other huge corporations control the biggest websites and most of the access to content on the internet.

They just need to make it inconvenient to use ad-blocking browsers.

They built their business on advertiser gambling, which seem to be flawed concept, because they keep on squeezing that tube for every penny more and more, in a race to the bottom.

But they are still in control of both browers and content so they have options to keep squeezing more.

So you want to use a ad blocker? Well, the browser that supports them might not be white listed (anymore) by the bot detector, and you have to solve captchas on every site you visit, until you come to your senses and use a browser, where ad blocking is no longer possible.

Oh, and all that is ok, because of “security”. Because letting the users be in control of their devices and applications is “in-secure”. They are just doing that to protect you from spam and scams, just trust them! Trust them, because they don’t trust you!

That is how capitalism works, privatize earning and nationalize losses.

Capitalism needs the deep pockets of the government to not collapse into itself.

Well… cults with an exit are just groups of people.

Apple looks more like a cult, similar to Scientology. Linux user as well.

But to install from local storage, you first download or fetch a storage medium from a remote location with the file on it. There isn’t that much of a difference IMO.

I would not call it side-loading when I download a file and then install it on the same device. Because that is how it has always worked. I never before heard people describing downloading and executing a setup.exe as “side-loading”.

Until some time ago, I always though that “side-loading” is something different. Since I first saw “side-loading” used in ADB, so I thought that it means using another system on the side to load and install software onto a target system.

So to me that seems fitting, but now it seems to be used differently. How is installing software using just one device “side-loading”. What side do they mean?

I don’t agree with the generalization here. Sure, it is generally advisable not to rely on security through obscurity, but depending on the use-cases and purpose it can be effective.

I dislike DRM systems with a passion, but they, especially those for video games like denuvo, can be quite effective, if the purpose is to protect against copying something for a short time until it gets cracked.

Otherwise I agree that software developed in the open is intrinsically more secure, because it can be verified by everyone.

However, many business and governments like to have support contracts so want to be able to sue and blame someone else than themselves if something goes wrong. This is in most cases easier with closed source products with a specific legal entity behind it, not a vague and loose developer community or even just a single developer.