I had just heard the expression: “It helps the general effort.” It was in a Discworld audiobook. It made me think: “Yeah, I could use some help right now.” That was still rattling around in my head when I had to come up with a username. I don’t remember what I was having trouble with, so I guess that worked out.

There’s that episode of The Orville where social media is the law.

https://orville.fandom.com/wiki/Majority_Rule

What’s AmeBlog?

I had a look at what tumblr says and it’s probably a good option. It’s not likely that they will try to find sneaky ways around the settings. The liability risk is in no relation to the potential gain from selling that data. Under EU law, such opt-outs must be respected when training AI. For now, the major US companies can be expected to abide by that. In the future, we may see special models for the EU. A few open source models by Chinese companies already exclude use in the EU.

Reducing scraping takes skill and a major effort, which tumblr can bring. A catch is that there is a conflict between serving images to lots of people but not to scrapers. Sufficiently determined large scale scraping operations will still succeed, but maybe no one will feel that it’s worth the effort anymore. It’s impossible to prevent individuals from saving images. So AI hobbyists or small artists could still use your images for training and share the products. When fans re-upload your images, they may become part of large scale datasets after all.

It depends on what exactly you want to achieve. If you want money, then upload it to Adobe, Shutterstock, and such. That’s the best offer you can expect. No one will offer you big bucks for an AI license just because you have protected your drawings.

If you want to have your own site, then you could rely on Cloudflare for handling the technical and legal side of preventing scraping/AI use.

But I guess the main worry for any artist is other artists who use AI. That’s where Glaze and Nightshade come in. It’s already been suggested but you should know how much you can expect.

These tools target the original Stable Diffusion 1.5. IIRC they also work on SD 2.0 because they reused some components. I am not sure what other versions, if any, could be affected. Certainly not the newer ones.

It goes without saying that the major companies were never affected, could not be affected. Since no one mentions it, I guess it’s self-evident but I want to repeat it for the uninitiated.

I think these early models are still used partly because they have lower hardware demands and partly because they are less professionally censored (ie more suitable for porn).

Anyway, the effectiveness against hobbyists, your competitors, or other small scale AI users, is also limited. They may not use a susceptible model, especially if they make SFW images. If their model is susceptible, then these tools may waste a few hours of their time and maybe a bit of money. But it won’t get rid of the competition or even significantly harm them.

I agree with everything. The thing is, I’ve been thinking about the psychology behind this lately.

When Fedi-Fans complain about Bluesky, it is usually based on the misunderstanding that it also is instance based. It really doesn’t seem to occur to many that things might be done differently. But I think it may go a little deeper.

A common complaint is that it’s too expensive to run a full relay. People want to self-host it all. They want to feel that they are in control and don’t need anyone. It’s not particularly rational but people do lots of silly things chasing that feeling. The rational start would be to move somewhere remote and grow your own food. Instead, people buy a pick-up truck or degoogle their phone.

That architecture also appeals to a more tribal mindset. An instance is “our” place. We just pull up the drawbridge when bad people come and we are safe here in “our” castle.

I think to some people that is more appealing than the more open design of atproto.

On Bluesky, there is all this waffle about some people trying to get someone banned. They might find such tribal architecture more appealing.

I’d argue that the modular design is a more radical approach to decentralization.

Yes, but only if you can do without instances. Instances seem to be important to many in the Fediverse.

I realize my title has little to do with the post

I was going to say. Red dwarf is the name of a ship and Constellation is a class of ship. In any case, you can’t have a proper ship without a bridge.

This is a rare useful (and frugal/practical) niche for blockchain. Immutable verification is a core principle.

You’re right on everything else, but this is just no. You never need blockchain.

One just need someone who makes it credible that the hash and timestamp were not tampered with. Even posting the hash on Reddit would do it for most people. Reddit isn’t going to commit fraud for some random person. And that random person is probably not able to hack the database undetected.

Recomputing lots of hashes isn’t difficult. A blockchain doesn’t add any trustworthiness on its own.

Forgeries are nothing new. Society has always worked on trust and best guesses. Nothing has changed.

Relevant xkcd: https://xkcd.com/2650/

Recomputing the hashes is not a problem at all.

Yes, but that doesn’t seem sufficient for some. Conservatives certainly would like to remove trans people from the public completely. Aside: It’s foolish for trans people to copy these tactics, assuming this comes organically from the trans community. These people are certainly acting like the heels in some right-wing propaganda play.

Bluesky offers several ways in which users can remove unwanted content from their experience. Easiest is for users to block Singal; banning him from their personal part of the network. Blocklists can be shared easily. Users can also spin up their own moderation service.

I probably shouldn’t go into the details of what Bluesky can do on a technical level. Incidentally, that blog post contains errors.

In short: On a technical level, the Bluesky company can greatly reduce the visibility of someone. But they would likely run into legal problems if they used that on Singal. The EU regulates what can be done quite strictly. Maybe they could benefit from some industry friendly “loopholes”. I’d have to look that up.

That needs a longer explanation.

An instance does not interact with all other instances. It only syncs with other instances when users follow someone there, join a community, …

But that’s also a problem. It means you can’t search the entire Fediverse from a particular instance and find new and interesting discussions and people. There is no discovery feed. For that, you need something like Bluesky’s relay. That relay actually does keep up with what everyone is posting and archives it.

But that’s one aspect of Bluesky that draws a lot of criticism by Fedi people. A full relay is expensive to run and not something anyone can self-host. Pruned down versions are doable, though. If everyone actually did run their own relay, then one would get you the combinatorial problem.

In practice, large instances are the Fediverse solution to the discovery problem. You can see what the many users on that instance post. Also, the many users subscribe to many things and so a large instance will cache much content from elsewhere. That architecture encourages centralization.

There’s other difficult issues. So you have a little server that serves your content to a few followers. Some celebrity with millions of followers would have to rent an entire server rack. But what if little old you interacts with a celeb and now all their followers try to fetch your content from your little server? Common problem. You just need caching. EG the celebrity rack also serves your content to their followers and takes the load off your server. But now whoever is doing the caching can also filter replies. There’s no simply solution there.

Yes. On Bluesky, they could be individually muted or blocked. You can make and share blocklists, make your own custom feeds that exclude such posters, or even create your own moderation service that removes (or blurs, …) posts for your subscribers. Obviously, that is not satisfactory for some people.

Yes. It’s only a problem if you expect or want the Fediverse to be the future of social media, which it isn’t.

This does raise a question relevant to the Fediverse. Some Bluesky users are lobbying to have Jesse Singal banned, whoever that is. Of course, a hallmark of a decentralized network is that there is no central authority that could actually do that. Implicitly, this demand is a rejection of the very concept of decentralization.

Once people find out what decentralization means, are they even willing to tolerate it?

Bad rant. Wrong on technical aspects and self-contradictory. Anyway, off-topic here.

Probably, depending on the context. It is possible that all 6 models were trained on the same misleading data, but not very likely in general.

Number crunching isn’t an obvious LLM use case, though. Depending on the task, having it create code to crunch the numbers, or a step-by-step tutorial on how to derive the formula, would be my preference.

The Fediverse is fine as it is. But some people hope that it is the future of social media which isn’t likely at all.

The Fediverse is called decentralized because no one instance has a majority of users. But the only way to make sure that this state continues is if the most popular instances close registration for new users. That’s an easy choice to make for the volunteers running instances. More users mean more expense and more work. That’s what I mean by choices. A growth-oriented company would not do that.

I don’t see what’s so special about being able to make new accounts. You can do that anywhere. What you can’t do is take your followers, communities, or posting history with you. For many people, that is not acceptable, and not just influencers, for whom their accounts are their business capital. That’s a big part of what enables social media monopolies.

I don’t see how the Fediverse solves any of the problems with social media. The hobbyist driven Fediverse isn’t going to take off. Any professional, growth-oriented instance would have all the familiar problems.

AP is a standard for letting servers communicate, while ATP is that and more. You could build what ATP does on top of AP, or make both compatible. What matters is really the communities and ecosystems behind these protocols.

AP is behind the Fediverse. The basic building block of the Fediverse is the instance. Every instance is its own self-contained, centralized social media service, that optionally federates with other instances via ActivityPub. There is nothing about AP that encourages decentralization. To the contrary, the way things work rn encourages centralization (but that’s pretty technical).

Case in point, Trump’s Truth Social is a Mastodon instance that choses not to federate. If it was open for federation, the Fediverse would look quite different. Or perhaps more likely, most other instances would choose to defederate.

I explain this because a few weeks ago, there were some posts pitching the Fediverse as decentralized social media. But the Fediverse is what it is because the people running the servers choose to do things a certain way. This is not a result of technical or legal features.

@Proto is the result of a project to make Twitter decentralized. That is, not a decentralized alternative, but actual Twitter with all its users. We might never have heard much about it if Musk had not taken the wrecking ball to Twitter. The team created Bluesky as a proof of concept.

Current social media companies have monopoly power over their users. @Proto seeks to structure social media in such a way that that is impossible. It is quite sophisticated. Improvements may be possible, but it certainly is good enough to solve the technical aspect of social media monopolies. Of course, the technical part was never the hard part. We will see if the economics work out. But the real challenge is the legal angle.