451–460 of 464 posts
Earlier quoted context omitted.
Careful there. My law firm in east Texas happens to own the patent rights to a system and method of concatenating a plurality of 1s and 0s for the purpose of generating an encoding of a creative work, and as such you are in peril of infringing on one of my claims with your concatenation of 1s and 0s!
General Mills owns the copyright on 0s, sir. (Cheerios)
Earlier quoted context omitted.
Well, the problem is that all this "vigilante[0] enforcement" was there to enable online services to work at all; and it has been further extended by private business agreements to give copyright owners powers above what the law requires. Any tort in law also has secondary liability attached to it, and it's generally accepted caselaw that online service providers get secondary liability if someone uses them to infrin…
May I ask if you are a lawyer? I've read many of your posts before and they're always extremely informative. You've taught me a lot of things about copyright. I'm not sure if you'd agree with what I say but I still wanted to say thank you.
I am, however, an American tech enthusiast born in the 90s, which means I got exposed to a lot of anticopyright, Creative Commons, Free Software, and Open Source propaganda at a very young age. Of course, my actual positions nowadays are a lot more nuanced than the ones I had decades ago.
Furthermore, I'm involved with software reimplementation[0]. And one constant of that whole field is that it's often unofficial, if not outright adversarial to the original hardware or software. If you aren't being very careful about copyright at every stage of the process, you are liable to get sued with the full force of the law. There's zero leniency with copyright these days, so you either have to be legally untouchable or go deep underground. I chose the former path, which means that I have to know a lot of this shit, even though I'm not a lawyer.
[0] Specifically I am one of the contributors to Ruffle, a Flash Player reimplementation that works in modern web browsers.
Earlier quoted context omitted.
People will think twice before using GCP. I know I'm.
But why didn't you do that all the previous times? - 13 years ago: https://news.ycombinator.com/item?id=350968 " On friday I woke up to find my google account had been disabled. No gmail, no adsense,adwords,webmaster tools,blog etc etc. No reason given. I've filled out the forms, got an automated response which is totally irrelevant " - 10 years ago: https://news.ycombinator.com/item?id=4013799 and it was happening o…
Earlier quoted context omitted.
Well, it has to happen at some point. Exceedingly unlikely, but never say never.
When the number of hash values (at least 2^128) exceeds the number of atoms in the observable universe (10^80), it's never going to happen. ... unless the hash function is broken, in which case it's not cryptographic anymore. Note that the birthday paradox doesn't apply when one of the inputs is trivial.
Consider a piece of data that is 128 bits long. There are 2^128 possible values. A 128 bit hash function will either yield a unique value for each possible data value (which will exhaust the entire range of unique hash values), or you'll have a collision. Now, lets add in pieces of data that are 129 bits (or any other length) as well - you'll have a collision immediately.
All software has bugs; I'm not mad at all that this silly test case was flagged incorrectly. The truly infuriating part is "A review cannot be requested for this restriction." Translation: "We have no idea if you actually own this content or not, but it would be _way too expensive_ for us to find out for sure! So you're out of luck, but don't worry — it's all worth it so we can make sure children can't stream Marvel…
It'd be nice if the law had a escrow appeal process. Alleged violator now posts a $100 escrow, now accuser has to do the same. Then Google reviews it, makes a decision, and loser has to pay for it.
I'm afraid they'll have incentives to automate that review, and then simply repeat that you can't appeal. Now you still can't access your file AND you're out of a $100 :-/
Earlier quoted context omitted.
When the number of hash values (at least 2^128) exceeds the number of atoms in the observable universe (10^80), it's never going to happen. ... unless the hash function is broken, in which case it's not cryptographic anymore. Note that the birthday paradox doesn't apply when one of the inputs is trivial.
Just because there are a lot of possible values doesn't matter. There are also a lot of possible pieces of data that could be fed into the hash function. In fact, an infinite number. Which is larger than 2^128. Unimaginably larger. Collisions will happen. Sure, intentional collisions might be next to impossible, but collisions will happen. Consider a piece of data that is 128 bits long. There are 2^128 possible value…
Earlier quoted context omitted.
Well, it has to happen at some point. Exceedingly unlikely, but never say never.
h(1) == h(some_real_data) does not have to happen at some point. Collisions are unavoidable, but the point of a cryptographically secure hash algorithm is to ensure that for all h( x ) == h(some_real_data), there are no values of x that can be found by brute force, or by cleverness either for that matter. Of course, it’s always possible that the particular hash function used could be hilariously broken…
Cryptographically secure hash functions state that it's unlikely to occur, but the fact that you can take data a greater size than the output will guarantee that you'll have collisions. If you have collisions and the hash function is generating output that is uniformly distributed across the possible output space, you will hit h(1)==h(some_real_data) at some point (edit: given that "some point" is a mathematical concept, not an actual point in time. Of course, there's no guarantee that we'll ever get collisions for hash functions with massive output spaces. But I'd argue that's a limit of computing, not of the math behind the functions themselves)