Thanks for the recommendation, I’ll give it a try!

Sadly not, I’d also be interested in one!

I very much dislike Mozilla’s direction over the last decade. They’re introducing user-hostile features that subtly break normal browsing experience, even when disabled[0]. Not like Google is better, but I’m also trying to get away from Mozilla.

[0] On Firefox Mobile, there’s a “feature” which makes the address bar auto-complete domains of companies paying Mozilla. I noticed this with Netflix - I never visit, but when I start writing a URL with n, roughly every 10th time Netflix was suggested. You can disable this feature, but this doesn’t actually disable it. The address bar no longer auto-completes with Netflix, instead it just doesn’t autocomplete! So 9/10 times I can write n and press Enter, but 1/10 times I press n and search for the letter n.

Mozilla doesn’t care whether they break features, as long as they can make more money. I strongly dislike this approach by the supposedly “good” browser manufacturer.

Me too, very excited for the relaunch!

I don’t think that’s what we’re talking about? As an example, if I ask an LLM for good spots to get Bratwurst and it recommends places in New York, it’s fairly useless to me as a European.

How so? For pretty much any application in the EU, EU-centric results are obviously much better?

Can I freely download all the training data for any of those? I was under the impression they were all trained on non-licensed and copyrighted data.

I support FOSS LLMs, but which actually exist? Which LLMs have open-sourced all their training data?

I agree that we should incentivize open source work, but my worry is that by legitimizing partial open source as “open source”, you’re disincentivizing fully open source work. After all, why put in the effort if you’ll get the same result with way less work?

The incentive you’re asking for is a disincentive against full open source, and I can guarantee you that if the existing “open source” term wasn’t defended by hardliners, there’d be far less open source work in the wild than we have today.

Open source isn’t defined legally, only through the OSI. The benefit is only from a marketing perspective as far as I’m aware.

Which is also why it’s important that “open source” doesn’t get mixed up with “partially open source”, otherwise companies will get the benefits of “open source” without doing the actual work.

If they open source everything they legally can, then do they qualify as “open source” for legal purposes?

No, definitely not! Open source is a binary attribute. If your product is partially open source, it’s not open source, only the parts you open sourced.

So Llama is not open source, even if some parts are.

I appreciate the clarification, thank you!

Right, which means that you practically cannot give open source software for free to non-corporations while selling it to corporations while still being fully open source, as the corporations can simply get it for free from any non-corporation.

No, not in the way GP wrote. You’re not allowed to have your license discriminate between users, so you’d have to sell your software to everyone, not just big companies.

Either no one pays, or everyone pays.

No, because that would no longer be open in the open source sense.

It’s either open for everyone, or it isn’t open.

Edit: sorry to whoever doesn’t like it, but it’s literally how “open source” is defined

when the data used to train the AI is copyrighted, how do you make it open source?

When part of my code base belongs to someone else, how do I make it open source? By open sourcing the parts that belong to me, while clarifying that it’s only partially open source.