Deebster

I’d assumed it was servers running on renewable power, although I’m not sure how they measure that. I know some hosting companies and CDNs have that as an option, but I don’t see how you’d know if each server chose that option so I guess it’s more like “servers with green hosting companies”.

Don’t downvote stuff just because you’re not interested in it! There’s no algorithm you’re training, you’re just being rude to people.

Downvotes should be for worthless content and people being dicks.

I just now noticed it was gone. Did it just vanish one day, or did its users at least have some hint?

edit: looks like it was a surprise: https://lemmyverse.link/mander.xyz/post/12163154

Mbin seems like a healthy project, and the only sensible move from kbin.

So, they’re just going to add a QR code? Of course, you could already do that, but having it built in and be the default process would probably help.

I think of a lurker as someone who doesn’t post - I guess your definition is someone who doesn’t interact at all (besides making an account and subscribing, I assume). But yes, I mean users who only vote are now counted (it’s not using views afaik).

I assume this latest bump is due to lemmy.world updating and now counting lurkers when assessing active users.

It’s weird how well making it roleplay works. A lot of the “breaks” of the system have been just by telling it to act in a different way, and the newest, best versions have various experts simulated that combine to give the best answer.

No, you don’t need to train it, it’s just about the prompt you feed it. You can (and should) add quite a lot of instructions and context to your questions (prompts) to get the best out of it.

“Prompt engineer” is a job/skill for this reason.

That would be good - I often want to see what an unfamiliar instance is about by checking out their homepage sidebar and local communities and currently it’s several clicks and involves going directly to the site, effectively logging me out. At least the new versions have the new feature that sends you back to your home instance to subscribe.

The only plus is that sometimes the remote instance has interesting styles or other customisations that I wouldn’t see unless i visited properly.

IPs and access logs, plus email addresses aren’t and are the kind of thing law enforcement wants.

It’s not the same expression. I mean “a bug” as in a software error and “on my side” as in it’s not Lemmy’s fault.

Now I see that OP is the creator of the tumblr clone Wafrn I’m sure it’s just a typo and that this is the intended meaning.

Or “a bug on my side”, depending on what they mean. Prepositions are hard.

It’s been downvoted because it was posted multiple times (misunderstanding/technical error) but I agree. Each of the posts have different replies so it’s a shame we can’t merge them into one excellent thread.

Hmm, getting origin servers to expose themselves this way is a clever hack. As noted, any bad actors probably already know this trick to bypass Cloudflare/whatever anti-DDOS layer.

As a fix, I guess you can either send your server’s outgoing connections through a proxy/VPN or use your hosting company’s firewall to block all non-Cloudflare inbound traffic.

Good to hear something about Solid again. I was aware of it already (having Sir Tim Berners-Lee behind it makes it occasionally news-worthy) but it’s really not made any ripples that I’ve noticed recently. It seems to be ticking slowly along, but I think until we can get a good look at ActivityPods 2.0 it’s hard to get too excited.

Unsolicited advice: Mbin a fork of kbin which fixes a lot of kbin’s problems. fedia.io was the second largest kbin server until it switched to mbin.

(I only found your post because I was searching for why kbin.social was down)

Yeah, it’s bizarre since kbin even also calls them magazines, so any half-hearted search will find the concept.

You’re 100% right, I think we do a lot more to stop content rotting like this. Now it’s half gone, with any contributions from other instances unable to federate, fracturing any more conversation.

Maybe we need to do two things:

• ensure that popular instances aren’t single-admin shows (so there needs to be a team with the skills, finances and motivation to run if one or more disappear, or at least the credentials in escrow so someone can step in)
• have some system to backup/migrate content

The migration one is tricky, because ActivityPub is very tied to urls. I guess that you could make dummy account for the conversation you brought over, so you’d at least have a read-only record.

reddthat is a great instance, and I’m glad to see that it’s not just a single-admin setup. If you vanished (I hope not!), does the team have the access, knowledge, finance and motivation to continue?