However reverse engineering, even for interop, doesn’t absolve you of EULA things you agreed to, which can include reverse engineering. And any such dispute can drag you into court for years and easily 10s of thousands of dollars, just to defend.
A EULA also is not zero legal weight. It is a contract for any legally enforceable term, for the most part. And one can sign away a lot of what you’d think cannot be signed away, such as the right to trial if you agree, even by EULA, to arbitration.
And in any case, if you want to fight a corp, you’ll easily burn through $50k min through hundreds of thousands to get through the hassle.
So simply trying to waive it away through unfounded belief is not a good strategy.
You’re assuming the patent is the only thing in the code. Theres likely also trade secrets, as many pieces of secret sauce are not patentable. No one files patents on things they want as trade secrets after all.
Trade secret law doesn't work that way. You can use it against an employee or business partner who signed a contract with you, or a competitor who paid off your employees to give them information. But someone finding your secrets independently is fair game.
Third party reverse engineering is never illegal for this reason. You can only violate copyright by publishing decompiled code, which is not the case with this driver. They might claim breach of contract if their EULA bans reverse engineering but that only applies to the person who entered into the agreement. If someone else decides to republish it then it's out of their hands.
Copyright never covers facts. Only patents can cover facts and they are supposed to be useful to understand. In my opinion engineers should be penalized for not reading relevant patents before building something, but in practice lawyers tell us not to because reading means we are tainted.
So if I need to write code in a specific way for the driver to work then that way is a fact not covered by copyright. There are sometimes false facts put into maps because those false facts are copyright and if found prove copyright was violated - with various levels of success that would not fit into a comment.
They most certainly did not claim in any threats or filings that they’d only be limited to patent claims. It’s nearly grade school legal boilerplate to state such in any filing. Having been on both sides of such things, you simply state in one sentence in any such filing something like “X, but not limited to,..”.
I think OP is both right, and there was people still complaining about it. Atom was seriously slow. Everyone said you couldn’t build a fast text editor that could compare to native ones. VS Code came and it was significantly faster. Atom was slow to the point of not being usable to some, while VS Code was perfectly usable. The people wanting vim or Sublime performance were still upset and wrote about it, but vscode was perfectly usable and fast enough for the majority of people. It was indeed held as an example of how to make a performant electron app. Still an electron app, but not absolute slog. Tbh, if you disable all extensions in vscode, it’s still fairly fast, and doesn’t even take that much memory. It’s not as usable without extensions though.
More parameters without significantly more data fails; it loses the ability to generalize and simply memorizes.
And we’ve consumed pretty much all the data, and that part is getting harder as data is now expensive, Balkanized by owners, litigated and charged for. Couple that with model collapse as models eat their own output (not solved for any models, despite more than a decade of the problem), and add that lots of previously human curated datasets (stack overflow style), and it very well may be the models are not getting much better.
Very large models trained with sgd variants don't tend to over fit, especially when using few epochs. See double descent.
The coding/math breakthroughs were caused by training models on the subset of their own outputs which pass a verifier, see reinforcement learning from verified rewards.
That's the historical wisdom, but there are reasons to believe the opposite: that past a certain point more parameters are always better. Basically, that overfitting don't matter, because a large enough model will memorize and build the general solution simultaneously. Then weight decay and SGD noise encourage forgetting of the memorized solution so the general solution takes over.
Alternatively, one can claim sqrt is a single symbol, and there are extensions requiring a digit. Log, ln, also have hidden bases, but they’re also valid symbols. Cos/sin may be in radians or degrees, both in common use, and no real place to denote it.
One could even claim division is raising to the -1 power, thereby disqualifying it in exactly the same manner you disqualify sqrt. I’ve never seen sqrt with a 2 index; power -1 is seen quite often for division.
OK, fine. But they are also an ad network, which absolutely acts as a publisher. I think they should be liable for the revenue they bring in from knowingly fraudulent ads.
Legal definitions can should change in order to keep up with society and technology.
No where does it say they know which ads are fraudulent and continue that particular ad.
It’s nuts how many people simply don’t read or think.
Walmart can say, for example, they know they’ll lose 3B to theft. That doesn’t mean they are wanting it, or allowing it, or know exactly which item will be stolen when and where. It means, given their scale, and from evidence they gather after the fact, this is an estimate of what happens when they have so many locations.
Now go read carefully, and don’t make stuff up about what you read, and think about the facts presented.
You're just responding here by saying "by law they are not thing X" in response to people who are saying "maybe the law should be different because they sure act like thing X".
No. Did you read the law? You wanting them to be a publisher because you don’t understand the difference between a legal use of a word and a popular use of a word does not make them a publisher any more than the right to bear arms means I can go shirtless.
Publisher, in the legal sense, has a long history, and content providers are not that any more the the word arm in the legal sense means human arms.
This sub thread literally starts with “how are they not a publisher?”, not with “we should change the legal definition of publisher.” As such, it would be reasonable for people to know the colloquial vague idea of “publisher” and the legal term are not identical.
When social media chooses what to display in a user’s feed—even if the content is 3P generated—
then it starts to look an awful lot like a publisher. Would you consider a newspaper not to be a publisher of its own op-Ed page? How about the old Reader’s Digest?
So a comic book store, by choosing what to put in the front of a store, is a publisher. Me, by choosing what Christmas decorations to put out front, is a publisher.
I wish people would simply learn to read a little law instead of routinely not understanding simple concepts.
Your analogies don’t hold. There’s no act of content reproduction (the essence of publication), as there is with the examples I provided. DMCA is about copyrightable subject matter.
You make it sound like there's a cut and dry answer to this. There isn't. Moreover, not all kinds of content are alike. There's content that Meta itself produces, and there is content that 3Ps produce (UGC, or user-generated content), which is what Section 230(c)(1) of the CDA pertains to. (Apologies for mentioning the DMCA in my previous response, I erred there.)
It's an interesting question as to whether Meta's act of selective republication of others' content constitutes its own content. This matter hasn't been adjudicated; I expect it will be in the fullness of time.
Parties and their attorneys disagree on legal and factual questions all the time. Meta has a position; others may differ. These disagreements are why we have a dispute resolution system (i.e., the courts) in the first place.
No other business is liable in this manner. Run a classified ad in a newspaper, use it to scam someone, newspaper not liable. Hack into router at Starbucks, use it for fraud, Starbucks not liable. Put ad in your front lawn, for fraud, you’re not liable.
Making each of these liable really makes it easy to destroy your competition or enemies - commit crime via such methods, they suffer.
This is not true in the case where the newspaper knows it is a scam, right? So as soon as Facebook says "we've reviewed this scam ad you notified us of, but found it complies with our terms and conditions", they're pretty clearly knowingly participating.
Point me to the federal law that makes newspapers not liable for the content they publish. No other businesses have safe harbor protections codified into law.
No other platform has the scale that Meta does. Any company making as much money as they do can't claim they can't spend an extra $20 million/year to even pretend to fight the issue.
The larger the company, the larger the obligation they have to be a good citizen.
Also productivity costs more in tooling than ever before per worker, so it's not a surprise that a smaller amount of capital goes to humans when the gains are majority from capital outlays, not labor improvements.
You know your claim simply doesn’t follow. Costs and profits can both go up. Why make such a silly straw man, since it makes the rest of anything you want to show much less well received?
It’s perfectly reasonable for gains in productivity to reward those taking risks for the gains.
And your graph also makes clear both corporate profits AND remuneration to employees have gone up significantly. And that profits are far more volatile than remuneration: most people willingly take stability over risk and choose salary over entrepreneurship, which is a reasonable choice.
Corporate profits are what has been growing retirement accounts, which has huge benefits to the majority of people nearing retirement. It’s not the us-vs-them binary option commonly sold.
You claim a huge increase over wage costs, I point out profits offset that increase. The topic is the fact that U.S. workers took home their smallest share of capital. You said 'yeah because costs went up'. I asked 'but profits went up'. You then said 'huh?' then jumped to something other than the topic we were discussing, 'U.S. workers took home their smallest share' and went to 'boomers retirement accounts'. So... point to me.
reply