But you said "Any random movie, but also JPEG, sound, any type of data. .." - if he needed to generate a "key" first then he couldn't have demonstrated any random file could he? Or did he generate the key during the demonstration?
What is wrong with what Steorn is doing?
- Thread starter T'ai Chi
- Start date
-
- Tags
- free energy
But you said "Any random movie, but also JPEG, sound, any type of data. .." - if he needed to generate a "key" first then he couldn't have demonstrated any random file could he? Or did he generate the key during the demonstration?
Well, yes, he produced a key by reversing his algorithm. (This is a bit off-topic) Sloot had a kind of database with all possible sounds and colors, each marked with a key. The key of the encoded movie was used to fetch these sounds and colors and construct a movie out of it. His 'database' was max. 370 Mb in size and had to reside on the client. Given any key, the algorithm reconstructed the movie out of it. In fact... it was possible to construct a fully random movie by just entering a fully random key. Take a look at (English spoken) cs.unimaas.nl/p.spronck/Sloot (add 'www' in front and '.htm' behind).
Horatius
NWO Kitty Wrangler
- Joined
- May 9, 2006
- Messages
- 29,691
Do you know the number of this patent/application? I've searched on the name "Sloot", and haven't found anything that looks like a data compression algorithm. Was "Jan Sloot" his formal name, or would he have filed under a slightly different name?
If we can't locate this patent, it calls into question at least one aspect of this story.
So, reading that site gives me the impression that SDCS is about as useful as getting movies, pictures, sounds, heck even government secrets from alien races all over the universe from various strings in Pi. It can be done, and it will even be translated into English, it's just a matter of using a key that specifies the starting point and end point of where the data is.
Last edited:
Horatius
NWO Kitty Wrangler
- Joined
- May 9, 2006
- Messages
- 29,691
You do realize that article concludes that Sloot was a crank, right? I'm not sure what you're going for here.....
Pardon?
Colors, perhaps - if we go with a digital movie. But all possible sounds?
The voices of Humphrey Bogart, Darth Vader, Yoda and the Tarzan yell?
Blight
Scholar
- Joined
- Jan 16, 2007
- Messages
- 105
I call serious BS on this.
I write media playback software for a living, so i'm very experienced with this sort of material.
In fact, I read somewhere a very similar sounding hoax, where a guy claimed exactly the same thing, but what he was actually doing was transfering the massive amount of data through the power cord (he had an ethernet cable hidden within the power cord).
If steorn's device works, I doubt it breaks physics. Perhaps they've found some obscure way to convert gravity into energy, I don't really know. But I am very familar with computers, compression, encryption, hashing, etc to know this is complete BS.
Perhaps after he signed the contract he figured that he might be going to jail once they find his claims were all fraud and died of a heart attack due to all the stress.
I write media playback software for a living, so i'm very experienced with this sort of material.
In fact, I read somewhere a very similar sounding hoax, where a guy claimed exactly the same thing, but what he was actually doing was transfering the massive amount of data through the power cord (he had an ethernet cable hidden within the power cord).
If steorn's device works, I doubt it breaks physics. Perhaps they've found some obscure way to convert gravity into energy, I don't really know. But I am very familar with computers, compression, encryption, hashing, etc to know this is complete BS.
Perhaps after he signed the contract he figured that he might be going to jail once they find his claims were all fraud and died of a heart attack due to all the stress.
Horatius
NWO Kitty Wrangler
- Joined
- May 9, 2006
- Messages
- 29,691
Maybe you missed the last sentence in his original post?
Yeah, but now he's posting as if he thinks Sloot had something real. So does he think the same of Steorn?
I think I'm losing it here!
Foolmewunz
Grammar Resistance Leader, TLA Dictator
Ooh! Ooh! And then when you translate it, you have to get one clever guy to realize it's in three dimensions, not two, and then ya build this neat gyro thingy and Jody Foster gets to go see her daddy on a beach or some other heartbroken person gets to meet the original Martian and fly off to Neverneverland.
Go to the website of the European Patent Office and search on keywords 'Jan Sloot'. There you will find 2 filed patents.
The thing is, Jan Sloot demoed his invention in front of hard-to-fool technical guys from Oracle, Hewlett Packard, Philips. This is a recorded fact. There has been a lot of research in this case, but one thing remains: the demos where real and his device worked.
But I was just referring to this case because of the patent issue. Steorn can put his device or parts of its device under patent, without fearing that any nutcase can (re)build the machine. But I believe that Perpetuum Mobiles cannot get patented nowadays. So their patent does only cover their magnetic alternator.
Sounds like he didn't let anyone get near enough to conduct a proper verification so I'm not sure how you figure that out. Assertion of the matter under dispute as a bald statement of fact may not be the best way of persuading us!
All sounds like grade-A horse poo to me, and I find the analysis given in the site you linked to (which concluded he was a crank) very convincing.
I would also call into question how we can tell that:-
"(actually he died exactly one (1) day before releasing his algorithm)"
Since he died at all, he died before he could EVER release his algorithm. And even if you say he died exactly one day before he would otherwise have released his algorithm, there's nothing verifiable there at all.
Since I don't have a parallel universe machine, we can't say what would have happened if he hadn't died. All we can say is that we don't have his too-good-to-be-true algorithm (and there's no good evidence that anyone else has it). Choose to believe that he had found such a thing if you like, but don't pretend it's an indisputable "fact".
Matt
Horatius
NWO Kitty Wrangler
- Joined
- May 9, 2006
- Messages
- 29,691
Found them. They are Dutch Patents, which I don't know much about. And I can't read them, because they aren't in English. But if someone else can read it, they're at:
NL1005930C
NL1009908
There doesn't seem to be an online text version, so I can't translate it either. There are pdf's of the patents.
Well, to end this sub-thread: there is a movie in the make about the phenomenon Jan Sloot. He convinced Roel Pieper, he convinced several other people in several detailed demos. And you're right, Jan Sloot was a shy, paranoid man, hardly speaking any foreign languages, extremely anxious that people would discover his secret without his permission. Not a diehard hoaxer. But, he was obviously under great stress. The day he died was the day before he would sign the contract with the Roel Pieper Investment Company (it had an other name) and by doing this he agreed to release his algorithm. His initial share was $ 20.000.000,- plus a percentage of the licenses.
In Holland we have two camps: those who are convinced that Jan Sloot was a very, very clever conman, and those who think that Sloot stumpled accidently on something he could hardly comprehend. Let's put this to rest.
It doesn't matter who he convinced. It is mathematically impossible to do what he claimed. The same key will always generate the same movie/picture, correct? If you have a limited sized "key" of 1k, which I believe is what is claimed, then you are limited to 2^(8*1024) possible different outputs. I can create a program that can automatically make simple, unique 640x480 (32-bit colour) picture. I only need to create 2^(8*1024) + 1 pictures. This means that I have created at least one picture that his system can't encode.
For a 640x480x32 image, there are 2^(640*480*32) possible difference pictures. This means that his system could produce 2^8192 images, but there are actually 2^9830400 possible images. So his system can even come close to being able to reproduce all low res, still images, let alone full movies.
Con men throughout the ages have convinced some rather intelligent people. Without actually knowing what the demonstrations consisted of, we can't determine if the findings are valid. A full technical analysis has not been produced.
Why should we believe, without any evidence beyond simple stories, that someone has done something that is not mathematically possible?
Look, the Sloot case was a very big issue in Holland. All you bring up here and more against Sloot, has been done in Holland. The fact is, not only Sloot demoed his device, Pieper persuaded him to 'lend' him the device and took it with him on a tour in the USA, where several demos where given (by Pieper).
Ofcourse it is mathematically impossible. Therefore I and many other people are convinced that the explanation Sloot gave about his algorithm was purportedly invalid or simply wrong. But, it is also possible he found a kind of reverse hashing algorithm. I guess you are familiar with a i.e. MD5? MD5 gives you an unique 'fingerprint' of a file as a 16 byte message digest. Any modification in the sourcefile results in a completely different message digest. Although the algorithm of MD5 is wellknown, we can't reverse it and construct the original file out of the 16byte fingerprint. Maybe Sloot found a way to do this. I don't know. Nobody does at this moment.
But, the issue was that a Patent does not necessarily contain enough details to reconstruct a device or algorithm and still can be valid. Therefore I brought up the Sloot case. I think if we want to discuss Sloot and his invention we should do so in a separate thread.
Last edited:
Maybe. But why lie about how his contraption works? That alone raises a huge red flag.
From what I've read is that he repeatedly said that his invention was easy to replicate and he was very afraid that someone else would grab the money and run.
Uh, that is like saying that Sloot took a potato cut it in half and ended up with 42 oranges, but that's ok, since he may not be giving the REAL explanation. If his explanation is not actually what he did and the patents he has don't work, then we have 0 evidence of his demo being a type of compression, rather than just copying the movie to a laptop hard drive.
What is more likely, that Sloot's algorithm worked, thereby doing the impossible, or that he is a con man, and either conned Roel Pieper or hired him outright?
Hashes have collisions. That is my original point. There are far more (infinite in fact) different possible files, yet a finite number of MD5 hashes. That means one hash can apply to more than one file. It is NOT unique. So, even if you were to have a "reverse hash", which of the infinite files that can generate this fingerprint do we choose? The same goes for Sloot's algorithm.
FYI, MD5 is considered "broken".
By US patent law, a patent must describe the invention fully, or it is not considered valid. You can't partially describe an invention and still be granted a defensible patent. I suspect this is true for most other countries. The fact that the patent doesn't actually work is more evidence that Sloot was a scam artist.
Why do I continue with this side track in a thread about Steorn? Because it is so similar to what Steorn is doing. They are claiming to have done something that is impossible, yet other no evidence. Everything screams scam, right down to hiring "experts".