archontes

Outside of TOSs/agreements, NYT can't prohibit anything except distribution.

The Haber-Bosch process.

That isn't necessarily true. You're saying that the model weights would be a derivative work. Derivative work is insufficiently transformative to fully distinguish the output from the input.

In this instance, it's very likely that the process is sufficiently transformative. A set of model weights look nothing like the Mona Lisa, nor can they be directly transformed back into it. What it is NOT is the product of a creative process on the part of a human, and is thus ineligible for copyright.

It is as though we are able to distill meaning using an automatic process. Copyright doesn't protect meaning, only expression, and it only protects expressions that were generated by humans.

The network itself might be patentable if it isn't obvious.

If a network was copyrighted and it was found that the function of the network was inseparable from its expression, it would no longer be eligible for copyright on that basis.

I think people often misunderstand the application of copyright to GPL code in this way.

Now, will any of this stop people from CLAIMING copyright? No. It'll have to be fought out in courts.

No, we're talking about verifying superconductivity using a traditional magnet at an elevated temperature.

You could always use an electromagnet.

It's unlikely that model weights can be copyrighted, as they're the result of an automatic process.

You don't own data. You can sometimes copyright data.

https://www.americanbar.org/groups/science_technology/public...

You don't own data. You can copyright data, maybe.

https://www.americanbar.org/groups/science_technology/public...

The only thing that's copying is a copy. Copyright has never ever covered style.

More like, "Reading the internet".

I don't think anyone has claimed that model weights are a production of AI and are not copyrightable for that reason. It's that the model weights are not the product of human effort. They're the result of an automatic process.

Like... if I set up a Galton Board [1] and allowed the balls to fall into the bins, even though I set up the board and collected the balls, I wouldn't be able to copyright the sequence of the number of balls in bins 1 through 13.

https://en.wikipedia.org/wiki/Galton_board

I'm using Firefox right now. When I highlight in the bar and press tab, ONLY when the drop-down bar is showing, it moves the cursor down one to Amazon, which is not my default search engine, but the first one alphabetically.

So that doesn't work.

I also don't want to search my default search engine. I want tab to search for the panoply of websites that the omnibox will search. When I start typing ebay, I want "tab to search ebay".

I know I can set them up manually (a pain, but might be worth the investment to avoid the google overlord), however, I still don't get the "tab to search blah" function with a keyword. I have to type the keyword exactly -instead of having tab function as autocomplete- then space for my query.

What I want is omnibox tab to search.

All I want is tab to search. Actual, identical-to-omnibar tab to search.

Ikigai.

I'm afraid that until the Chinese government stops their influence campaigns in foreign countries, Chinese nationals living elsewhere will be justifiably suspect.

https://wisconsinwatch.org/2023/02/does-china-operate-police...

If you put your content on a billboard, what expectation should you have that you can control who reads it?

Why are you posting this without context? We've all heard of it before, making us all practically experts.

The policeman doesn't have to answer, and can literally lie to you.

I would like to humbly submit that we should address the beam before the mote.

https://i.imgur.com/UvyGJjq.jpeg

I'm sure that copying a work for your private consumption would be considered non-commercial in nature, but his assertion that it doesn't affect the potential market for the item wouldn't hold up.

It's clearly affecting the market for the original as you are removing yourself from the market for it. Creating a copy for your own consumption harms the copyright holder's ability to market a reproduction to you.

Probably not, as the L.H.O.O.Q [1] is the literal textbook example of a derivative work, which would fall under the explicit protection of copyright.

[1] https://en.wikipedia.org/wiki/L.H.O.O.Q.

US law as written does not permit this. Copyright restricts reproduction and the production of derivative work except for the reasons listed in fair use. Personal use isn't among those reasons (setting the US apart from like 20 other countries).

It's just de facto legal because of the difficulty of detecting those instances and the presumably nil return in litigating them.

Also, I believe this is irrelevant.

A model being able to produce an infringing work doesn't mean the model is an infringing work.

I can produce work myself that is infringing or not infringing as a result of learning.

What's incredibly impressive is that that phrase doesn't appear anywhere else according to google. It's ChatGPT being genuinely 'creative'.

That's not likely to be true. An AI can create a work that is infringing, a picture of a Marvel character, for example. But that doesn't make the AI or its weights or its training an infringement.

I understand that. And with the comments it's almost certainly infringing.

If you strip the comments out, though, it's questionable.

The thing is that it probably isn't. The thing you call "laundering" is probably actually fine. The thing that people misunderstand is that copyright is written such that it never protects function. If your code is open source, then I can write code that does the same thing, and if I can't because your code perfectly encapsulates the function, it's inseparable from that function and thus isn't eligible for copyright protection at all.

https://casetext.com/case/kern-river-gas-transmission-v-coas...

Your comments and creative expression are the only portions of your code that are protected by copyright. The function isn't.

Yeah, I'd be willing to wager Houston would be happy to snap up manufacturing and has plenty of skill in the talent pool. And proximity to Monterrey for cheap labor.

It almost seems to me that the stakes are higher for other people. Microsoft can afford to have a go, fail, and just move on, whereas other people might have more on the line and try harder.