There's no reason to believe that weights are copyrightable. The only reason to pay attention to this "license" is because it's enforced by Apple, in that sense they can write whatever they want in it, "this model requires giving ownership of your first born son to Apple", etc. The content is irrelevant.
> The only reason to pay attention to this "license" is because it's enforced by Apple
Yes, but the most important reason to pay attention to ANY license for most people is because it is a signal for under what conditions the licensor is likely to sue you (especially in the US, which does not have a general “loser pays” rule for lawsuits), not because of the actual legality, because a lawsuit is a cost most people don’t want to bear while it is ongoing or cover the unrecoverable costs of once it is done, irrespective of winning and losing, and, on the other hand, few people care about being technically legal with their use of copyright protected material if there is no perceived risk of enforcement.
But even if that wasn’t true, and being sued was of no financial or other costs until the case is finally resolved, and only then if you lose, I wouldn't bet much, in the US, in the court system ultimately applying precedent in the most obvious way instead of twisting things in a way which serves the interest of the particular powerful corporate interests involved here.
> There's no reason to believe that weights are copyrightable.
I know this is a long, nuanced, ongoing discussion. I'm very interested in it, but haven't read up on it for years. Could you elaborate a bit on the latest?
I was always in the camp that opined that "weights" are too broad a term for any sensible discussions about conclusions like "are (not) copyrightable". Clearly a weight that's the average of its training data is not copyrightable. But also, surely, weights that are capable of verbatim reproduction of non-trivial copyrightable training data are, because they're just a strange storage medium for the copyright data.
> This. Tables of numbers are explicitly not subject to copyright; that’s a copyright 101 fact.
Ok, but there's clearly more nuance there. Otherwise I could claim that any mp3 file I wanted to distribute is just a table of 8-bit integers and therefore not subject to copyright.
I wanted to reply in this direction. Ultimately, literally everything and anything in SW is a sequence of numbers, that anybody could easily put in some kind of table form.
I don’t know where the catch is, but that sentence can not be true in general.
A table of numbers is copyrightable if it represents some creative expression by a human being. For example, a BMP representing a sketch is a table of numbers and clearly copyrightable.
Weights are numbers that come from an optimization process. To the extent that weights encode any creativity, they encode the creativity of the training data. But any company using AI models (including Apple) does not want that interpretation because they are using AI models that were trained on other people's copyrighted works. If weights could be copyrighted, we all of us would own them.
That is simply not true. The details might vary by jurisdiction and the protection might not be under the exact name of “copyright” but there most certainly are comparable legal protections for the contents of databases (“tables of numbers”). See for example: https://europa.eu/youreurope/business/running-business/intel...
Disney would like you have a word with you. Why would their pile of numbers that represent Avatar3.m4a be any more subject to copyright than Apple_2D_3D.bin. Or GPT52.mlx or Opus45.gguf?
It's probably just Apple layers avoiding getting involved in any copyright lawsuit over the copyrightability of weights, by avoiding licensing it except under what's clearly fair use anyway, making copyrightability moot.
Yes this seems more about protecting them from a lawsuit. I don’t think they actually give a shit about the weights or they wouldn’t release them at all. I suspect they just know they’re training dataset isn’t perfectly “clean” and don’t want to accept any more liability than they already have.
Not sure I've met one of those people in... a decade or so? Loving apple products has been an uphill road for a long time (and increasingly more so post-Jobs)
> Not sure I've met one of those people in... a decade or so? Loving apple products has been an uphill road for a long time (and increasingly more so post-Jobs)
You either a deliberately misrepresenting the facts or been livoning under a rock. I mean read any discussion about M laptops and you see apple fanboys noncritucally declaring them a revolution in computing.
> I mean read any discussion about M laptops and you see apple fanboys noncritucally declaring them a revolution in computing.
I see a lot of people extolling the battery life, displays, and trackpads. And probably an equal amount of complaining about the increasingly locked-down and un-customisable nature of macOS. We all like the hardware, and fight the software more by the day.
The blind zealots of the "I'm a Mac, and I'm a PC" era just aren't very common anymore.
Quantum theory says that quantum computers are mathematically plausible. It doesn't say anything about whether it's possible to construct a quantum computer in the real world of a given configuration. It's entirely possible that there's a physical limit that makes useful quantum computers impossible to construct.
Quantum theory says that quantum computers are physically plausible. Quantum theory lies in the realm of physics, not mathematics. As a physical theory, it makes predictions about what is plausible in the real world. One of those predictions is that it's possible to build a large-scale fault tolerant quantum computer.
The way to test out this theory is to try out an experiment to see if this is so. If this experiment fails, we'll have to figure out why theory predicted it but the experiment didn't deliver.
> One of those predictions is that it's possible to build a large-scale fault tolerant quantum computer.
Quantum theory doesn't predict that it's possible to build a large scale quantum computer. It merely says that a large scale quantum computer is consistent with theory.
Dyson spheres and space elevators are also consistent with quantum theory, but that doesn't mean that it's possible to build one.
Physical theories are
subtractive, something that is consistent with the lowest levels of theory can still be ruled out by higher levels.
Good point. I didn't sufficiently delineate what counts as a scientific problem and what counts as an engineering problem in QC.
Quantum theory, like all physical theories, makes predictions. In this case, quantum theory predicts that if the physical error rate of qubits is below a threshold, then error correction can be used to increase the quality of a logical at arbitrarily high levels. This prediction can be false. We currently don't know all of the potential noise sources that will prevent us from building a quantum logic gate that is of similar quality as a classical logic gate.
Building thousands of these logical qubits is an engineering problem similar to Dyson spheres and space elevators. You're right that the lower levels of building 1 really good logical qubit doesn't mean that we can build thousands of them.
If our case, even the lower-levels haven't been validated. This is what I meant when I implied that the project of building a large-scale QC might teach us something new about physics.
> The way to test out this theory is to try out an experiment to see if this is so. If this experiment fails, we'll have to figure out why theory predicted it but the experiment didn't deliver.
If "this experiment" is trying to build a machine, then failure doesn't give much evidence against the theory. Most machine-building failures are caused by insufficient hardware/engineering.
Quantum theory predicts this: https://en.wikipedia.org/wiki/Threshold_theorem. An experiment can show that this prediction is false. This is a scientific problem not an engineering one. Physical theories have to be verified with experiments. If the results of the experiment don't match what the theory predicts then you have to do things like re-examine data, revise the theory e.t.c.
But that theorem being true doesn't mean "they will work given enough time". That's my objection. If a setup is physically possible but sufficiently thorny to actually build, there's a good chance it won't be built ever.
In the specific spot I commented, I guess you were just talking about the physics part? But the GP was talking about both physics and physical realization, so I thought you were also talking about the combination too.
Yes we can probably test the quantum theory. But verifying the physics isn't what this comment chain is really about. It's about working machines. With enough reliable qubits to do useful work.
You're right. I didn't sufficiently separate experimental physics QC from engineering QC.
On the engineering end, the question on if a large-scale quantum computer can be built is leaning to be "yes" so far. DARPA QBI https://www.darpa.mil/research/programs/quantum-benchmarking... was made to answer this question and 11 teams have made it to Stage B. Of course, only people who believe DARPA will trust this evidence, but that's all I have to go on.
On the application front, the jury is still out for applications that are not related to simulation or cryptography: https://arxiv.org/abs/2511.09124
If he's anything like me (doubtful but roll with it), the commit history when prototyping is probably something like "commit", "commit", "fixed a bug", etc.
Wouldn't you want the opposite? Once domestic production ramps up you gradually lift import restrictions to create more competition. I guess that's if the intention is to improve the domestic market in the national interest, rather than to just make people rich.
That is exactly what you never want to do under protectionist policies.
Domestic producers are shielded from Chinese competitors.
This means they are under less pressure to reduce prices and innovate.
I wouldn't read too much into the national security justification.
It's a political argument to an economic policy.
If this is about military capability, why ban all foreign manufacturers, including proven innovators like Helsing and Baykar?
Instead of blanket bans, targeted contracts could leverage Ukraine tested designs while building domestic capacity.
Innovation happens under competitive pressure. The US just created a domestic vacuum.
The national security justification is that we need expertise building/designing drones. We won't get that if we allow China to out-compete domestic manufacturers.
Essentially migrating codebases, implementing features, as well as all of the referencing of existing code and writing tests and various automation scripts that are needed to ensure that the code changes are okay. Over 95% of those tokens are reads, since often there’s a need for a lot of consistency and iteration.
It works pretty well if you’re not limited by a tight budget.
It's all very well being more public, until a government decides to make 5 years of social media history an entry condition[0], and moreover imprisons those people who are denied entry instead of simply sending them home on the next flight[1].
I have no problem with this per se, as I have no plans to go to the US this decade, but I do worry about contagion. Perhaps being a public person on the internet is an idea whose time has come and gone.
My father lived through the Uruguay dictatorship in the 1970s and avoids all social media whenever possible.
I was an infant at the beginning of democracy, so I haven't taken that much care.
Now, it seems he'll be vindicated once again, I do plan on visiting the USA and I'm hoping my social media won't be an obstacle (fortunately I don't think I have anything, but who knows, maybe I liked a meme or something).
Was Obama funding Aaron Swartz's efforts to scrape JSTOR?
Some people have the personality trait of loving to build collections or archives. Either for idealistic reasons (knowledge deserves to be free) or just because it's fun.
When that personality trait intersects with technical ability, we get projects such as the Internet Archive, Archive Team, Library Genesis, etc. There is no reason to assume state sponsorship, and 2/3 of those definitely aren't state sponsored.
To be snarky, do they? The average user doesn't even know what JS is.
Users want websites that are fast and solve their problems, with a good UI. They don't care how it's made.
Make websites that people enjoy using. A good developer can do that with any set of tools, though a no-JS approach is limited in scope.
reply