the concept of file folders and directories, essential to previous generations’ understanding of computers, is gibberish to many modern students.

This is so weird to me. Aren’t people at all curious? Like, I would never try to fix a car’s engine, but I have a basic understanding of how one works. I wouldn’t install a toilet, but I know about J-traps. I wouldn’t write my own 3D engine, but I know the basics of how they work.

Files and folder is such a fundamental and basic thing. Where’s the basic curiosity?

Most music files may be MP3s, but music files are rare these days. I wouldn’t be surprised if most people under 30 have never interacted with a music file at all, they just use streaming services.

Canadians were clever enough not to enshrine gun ownership as a right in the constitution. As a result, Canadians have limited gun rights, and almost no handgun rights. Sure, having guns for hunting is pretty common, especially in rural areas. But, the idea of a gun for home defense or a gun for taking on a tyrannical government is something that never took hold in Canada.

So, instead of Canada becoming the 51st state, the suggestion is that Canada just loses BC? How about we dismiss all the options where Canada loses territory?

Also, if BC seceded, it would only be a matter of time before Quebec did too.

I’m not fundamentally against the idea of states splitting off or joining up. There’s no reason that the configuration of countries should always stay as they are in 2025. But, the reasons should be good. If there truly is a “Cascadian” culture, then yeah, maybe a nice separation agreement could be negotiated that’s fair to everyone. But, having spent time in Vancouver, Seattle, San Fransisco and LA, I sure don’t see it. The cultural difference even between SF and LA is pretty huge. And, I can’t imagine that most people in BC would be keen to accept the guns of America, and the lack of free health care. Or, going the other way, would Americans be willing to give up their guns to join Cascadia?

It just seems to me that every time the world adds borders or moves borders, the result is conflict. I hope that over time there are fewer borders, and that the borders matter less. But, the only way to do it while avoiding war is really to do it slowly.

Worth noting that when Google was founded, Microsoft was in the middle of a long antitrust investigation, which was documenting every illegal thing they had done to maintain their monopoly and hurt every company that challenged it.

The “evil” in the Don’t Be Evil motto was widely seen as a reference to that company and that behaviour. From early on, Google saw Microsoft as a threat. They ran Linux servers, and tried to make sure as few employees as possible were running Microsoft on their desktops and laptops. A lot of internal tools were developed to try to avoid any kind of dependency on Microsoft, including ones that eventually became available externally like Google Docs etc.

Now, 25ish years later, it’s Google who are being investigated for leveraging their monopoly in a way that hurts consumers. IMO, they still never stooped as low as Microsoft did. Google paid Apple and Mozilla billions to be the default search engine. Microsoft used lawsuits and patents to try to drive their competition out of business. But, it’s still a monopoly that harms the world.

Anyhow, I’m glad that Google originally had the “Don’t be evil” motto, and also had this bit about AI principles that avoid the risk of harm. They act like useful warrant canaries because when they’re removed you know something’s up.

What annoys me is that people are buying the idea that BlueSky is federated.

Not only is it not federated, the very architecture they designed means that it’s probably not federateable, at least not by normal users.

The way they designed it, a relay is required to collect and forward every single BlueSky post. That means, as the service grows, it becomes more and more impossible for anybody but a company to run a relay. Someone did some calculations back in November when it was a significantly smaller network, and they calculated that at a minimum it costs a few hundred dollars, possibly as much as 1000 bucks a month just to handle the disk storage needs for a relay on a leased server. The more the network grows, the more those costs skyrocket.

What good does it do to have a network that theoretically can be federated, but practically costs so much to run a single node that nobody except a for-profit company can manage it?