It feels like you are making a computer program out to be more than it actually is right now. At the same time this all isn’t about what that program is doing. It’s about how it was built.

Well. When I copy and paste source code into my program and compile it it also doesn’t retain the actual code. It’s still not allowed.

If I on the other hand read source code, remember and reapply it in a sort of similar way later on then that’s totally fine. But that’s not what OpenAI did there. There wasn’t a human involved that read the articles and then used that knowledge to adjust the LLM.

There question i would have is where is the line there? Does that mean that as soon as there is some automated process that uses the data it’s fine?

E.g. could I have a script that reads all NYT articles, extracts interesting information and provides them in a different format to users?

But they aren’t forming take aways from it. They literally used that material to build this system. I also cannot just go around and take arbitrary data from anywhere and use it to build my own program. There are licenses attached to it and I have to be mindful of who’s work I can use to build my system and who’s I can’t use without explicit permission.

Building this system isn’t looking at other folks material and forming take aways from it. It’s literally using that material as input for building the system.

Might be a fundamental difference in opinion. I don’t see us anywhere near anything related to artificial life.

What they’ve built there is a product, a computer program and they used other folks data to build it without getting their permission. I also cannot go and just copy and paste source code from all over the internet to build my program. There are licenses attached to it that determine what you can or can’t do with it.

I feel like just because the term “learning” is involved people no longer view it as simply building or programming a system. Which it is.

But there is no one learning from it. It serves as a building block / source material to build these LLMs. I feel like the fact that it’s called learning gives folks the impression that it’s similar to what a human would do.

This comparison doesn’t make sense to me. If the person then makes money off it: yes.

Otherwise the question would be if copyright law should be abolished entirely. E.g. if I create a new news portal with content copied form other source, would that be okay then?

You are comparing a computer program to a human. Which… is weird.

Ah it will be at done point

And still it’s basically all Google.

Super easy. Especially since this is all under their control. So they could simply write those messages elsewhere if they wanted to. I’m not saying they do, but it’s technically possible and a walk in the park.

I would generally trust such a company to do it right. But that doesn’t save you when law enforcement and such get involved.

Yes you should. Because it’s not e2e encrypted then.

No they didn’t? My company just recently introduced it.

I’m unsure of many people know that StackOverflow also had enterprise offerings. Our company has their own StackOverflow instance with very specific content to our tech stacks.

I don’t care about stuff working OOTB - half the fun is messing around with things IMO.

I generally agree. Backups for me are just something I don’t want to tinker with. It’s important to me that they work OOTB, are easy to grasp and I have a good overview.

The web interface is important to me because it gives me that overview from any device I’m currently using without needing to type anything into a terminal. The OOTB is important to me since I want to be able to easily set this all up again even without access to my Ansible setup or previous configuration.

To each their own. I’m not saying your way of doing this is wrong. It’s just not for me. This is just my reasoning / preferences. It’s also the reason something like borg wasn’t my chosen solution, even though it’s generally considered great.

Features that are important to me are things like an easy overview of all backup jobs (ideal via a web UI), snapshots going back every day for a week and after that every month. Backup to providers like Backblaze or AWS and the ability to browse these backups and individual snapshots.

I’d assume that you can build all of this with git annex in some way. But I really want something that works out of the box. E.g. install the backup software give it some things to backup and an B2 bucket and then go.

What I’m curious about is that the git-annex site explicitly days that they aren’t a backup system, but you describe it as such.

Somehow “took me a while to wrap my head around it” doesn’t make me feel comfortable. Apart from git-annex themselves saying that they aren’t a backup system and just a building block to maybe create one, a backup system should imho be dead simple sind easy to understand.

Paid for the web interface as well. I really like that it’s super simple and just does it’s job. That would be the one I’d also recommend.

Look into Veeam. The free version should be enough for this workflow.

Nah. Why should I be against it? That’s a rather weird stance. I don’t really tend do deal in auch absolutes. But seeing that it’s a technical problem for you I don’t think there is merit in talking further about it. I don’t see it that way. Technology is just an enabler. It doesn’t much matter in these things.

What I meant by that is that I doubt that you can make your club so resilient. We are talking about a lot of social dynamics here. This isn’t a technical problem in any way. And the past has shown that network effects are a real thing. So inevitably if you give someone with a thousand times the resources and likely than the rest of the community the opportunity they will take it. It will become known as the main instance and everyone will join there. Smaller instances will become more irrelevant as they are already and at some point bow to what the largest instance dictates.

Take Lemmy for example. You can already see some of that happening with instances like beehaw. Do what they say or you get defederated. Naturally smaller instances will fall in line. What do you imagine happens if an instance joins that is as thousand times the size of the current entire network?

At some point it will be „do as we say or loose all your content“. Which will then lead to users switching instances where they have the access they want.

This is not a technical problem. The protocols can be nice and open. But that doesn’t help you if the network itself is fragile due to human nature.

What I meant is: It sounds nice in theory that you can build a social network in a federated way that is resilient to our social nature. I just have my concerns and going to watch with interest how it unfolds. It will likely take some years. But we‘ll see.

That sounds nice in theory. Going to be very interesting how it holds up in practice.