They take up so much space though.

The only problem I’ve had with Raspberry Pi is that some apps want to write a lot of stuff to “disk”, and the default “disk” on a Pi is a MicroSD card which dies if you keep writing things to it. Sure, you can always plug something into a USB slot, but that adds a bit of friction to the whole process.

Oh, also, I wish it were easy to power a whole bunch of Pi units. Each one needing its own wall wart is a bit annoying, and I’ve had iffy results using weaker, less steady power supplies with multiple ports intended for things like phones.

Well, you know those guys are extremists. Their slogan is literally “It’s Nerf or Nothin’!”

Average ebook size: 2.5 MB or so.

Meta downloaded 81 TB, or 81,000,000 MB.

81,000,000 / 2.5 = Approx 30 million books.

30,000,000 books * $250,000 per offence = $7.5 trillion

(are you sure you’re from programming.dev?)

Nice. I was being lazy, I was thinking of writing my own add-on or tampermonkey script, but I didn’t get around to it. I should have known there was already one out there.

There’s still a lot of old, useful and informative posts on Reddit that I find via a Google search. It annoys me that whenever I find one of those posts, I have to go and edit the URL to be old.reddit instead of www.reddit otherwise it’s so hard to use. Like, the useful thing is in the comments, but the comments are collapsed by default, so if you search for something you won’t find it until you expand those comments.

If they get rid of old.reddit I think I’m going to end up using the wayback machine to get that old post rather than trying to use that horrible new reddit interface.

For a while I was reading and posting on Lemmy, but still going over to check things out on Reddit (without an account) because Lemmy just wasn’t getting enough traffic to keep interesting posts popping up. I’m having to do that less and less these days.

Especially important to understand decentralization when it comes to topics like piracy. Unlike Reddit, a non-commercial Lemmy instance isn’t going to be shut down by a CEO who is worried about how it might affect their ability to sell ads. But, admins can get threatened, they can get overwhelmed, they can get hit by a bus. It’s good to know how how to find other communities on other servers, etc.

Is this really just the “backup site”? Boo, I say. Boooooo.

Also, wasn’t Reddit banning links to Lemmy or something? Maybe that was only in the run-up to the IPO?

Edit: elsewhere in this thread people are saying that trying to link to lemmy.world fails, and trying to link to lemmy.ml fails, but you can still link to other lemmy instances?

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.

True, and this is something that Bluesky actually seems to do better. Your posts are stored in a “PDS” (personal data store), so in theory they’re not tied to any particular instance.

I hope that a future version of the Fediverse design / ActivityPub considers how to handle this issue. Still, I’d much rather lose my past posts than lose my social graph. Past posts can probably be archived, but it’s much harder to track down people you used to be mutuals with on a different account and follow each-other.

The difference is that you can easily move to another Mastodon instance, and it’s designed so that when you do that your followers / followees come with you.

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?

He’s already gone. But, regardless, why sign up for yet another corporate social media site when every single one of them becomes enshittified after a few years. Are they just planning to abandon Bluesky eventually too? Or just hoping that this time it’s different?