the thermal paint idea is cute though

Thermal paint is very much a new development.

but pedantry is alive and well

Exactly, when you dig into all the complaints people have about this tech, they’re ultimately just symptoms of the underlying capitalist relations.

cause it’s a mechanical turk

The talk is reality check for anyone who thinks open source is still in its honeymoon phase. He basically argues that we have been living through a Gilded Age of open source from about 2000 to 2020 where everything looked like rapid growth and success on the surface while the foundation was actually rotting. Just like the original Gilded Age had its robber barons and railroad monopolies he points out that we have traded genuine freedom for the convenience of proprietary platforms like GitHub and Slack. He is pretty blunt about the fact that the industry has shifted from community driven passion projects to venture capital backed rug pulls where companies like Redis or HashiCorp just swap licenses the moment they need to squeeze more profit out of users.

He highlights how the XZ backdoor and the Log4Shell mess exposed that the entire internet is basically held together by three tired volunteers in a trench coat and how new regulations might actually make those people legally liable for bugs. He also goes off on how AI is being shoved into everything not because it helps developers but because VCs want to replace them, and he is clearly not a fan of how companies like Red Hat and Fedora are tying everything to AI tools now. It is a really sobering look at how we stopped caring about the principles of free software and just became pragmatic consumers who are okay with locked down ecosystems like macOS or Android as long as they are shiny.

He thinks we can still fix this but it requires us to stop being spectators and actually start mentoring the next generation on why these values mattered in the first place. He basically says that if we just treat open source as a way to get free labor for corporations it is going to end up as a dead hobby like ham radio. The main takeaway is that the era of easy growth is over and if we actually want a future where we control our own computers we have to stop picking the convenient path and start fighting for the principled one again.

reverse engineering this stuff is pretty challenging unfortunately

For most use cases though, you don’t really have much of a benefit of running Linux over Android on a phone though. There’s enough Linux compatibility on Android already to make it work seamlessly with your Linux devices. In my opinion, as long as the stack is open source and well supported, it doesn’t really matter whether it’s Android or Linux based.

It would’ve been a cool world if we got Linux that could work seamlessly between desktop and mobile. Imagine if you had architecture where apps were built as services with an API, and then you could connect either desktop or mobile UI to them. Heck, at that point you could even make custom UIs across apps, or pipe them together the way you do with shell scripts. And then you could also have a device like a phone which has all your apps and data, and you could plug it into a dock with more memory, GPU, etc. So, you wouldn’t have to juggle a bunch of devices and sync data between them.

Is it oversampling or just the fact there are a lot of users from China?

You can install Google app store in a container, and all the apps I’ve used work fine on it out of the box. It absolutely works fine as a daily driver.

I’d argue that Graphene is a better thing since it’s based on an OS that’s been designed for mobile from the ground up. I expect it’s going to be a while before Linux UX on mobile catches up to desktop, but Graphene works great already.

I just got a Pixel 9 last week and put GrapheneOS on it. Couldn’t be happier with it so far. The install was completely painless using web installer. All my apps worked out of the box. Google Store works fine in the sandbox. UX is good, and you don’t have any of the crap Google normally loads like all the adaptive services, and all the other junk that runs in the background.

that’s the hope

Yeah, that’s fair. If you’re already stuck on Signal, then it’s difficult to make a move to something else. I’m mostly talking about people who are using something like WhatsApp, and it’s better to make a fresh move to a platform that doesn’t have the issues Signal has.

The thing is that there’s nothing special about Signal that makes it better than alternatives like SimpleX. I just don’t see why it should be promoted instead of them. Yes, it’s better than WhatsApp where meta has a master key and can read your messages, but why settle when you can use a platform without compromises?

Yeah, there are network effects at play here. Getting people to move off a platform is very difficult because they need their contacts to move to, and their contacts need theirs in turn. Some people are willing to use multiple messaging apps, but most don’t. I’d argue that’s why it’s important to promote alternatives to Signal. The more popular they become the easier it is to get people to move to them.

The question here is why not get people to switch to a better platform like SimpleX or even matrix with something like Element. I don’t find that Signal does anything better in practice.

Pretty much yeah, and they’ve had a really good marketing campaign too. They got a whole bunch of prominent tech influencers incessantly pushing it, and it just feels like a massive astroturf campaign to me. Like you said, if a random person pitched this idea, they’d be laughed at, but you get some people with clout to do it, and it sticks because everybody respects them and trusts them.

I don’t think we’re saying anything new here. I’ve explained my point and the problem with Signal collecting phone numbers. You can make your own decisions on whether you think that’s acceptable practice or not.