The Infinite File Download
infinite-download.ma.ttias.be
infinite-download.ma.ttias.be
I didn't expect this project to make it to the frontpage. Right now, it's pulling close to 500Mbps in terms of bandwidth. It's cool, I like to see my graphs explode.
However, at one point, I may have to pull the plug on this. Bandwidth isn't entirely free and there are limits to what my little uplink and server can provide.
But in the meanwhile, enjoy the Silicon Valley startup parody: https://ma.ttias.be/infinite-file-downloader/
(Yeah, I know this is impractical and won't do anything against >/dev/null )
It'd be amusing to see people fill their disks at a far faster rate than they expect their network connection to allow.
Though, note that what you're suggesting is basically impossible for a work of any meaningful size, and the act of searching for a specific work in a random stream probably would prevent it from being considered an independent creation.
http://www.techpatents.com/Blog/independent-creation-is-a-de...
The maths tend to work against you as well.
π is known to never repeat (that is, it's irrational); it is not, however, known to be normal.
That means the digits in its decimal expansion (to pick a base) are not known to be normally distributed. This means that it is not necessarily the case that every possible sequence of decimal digits is present in π.
For example, for all we know it's the case that beyond a certain (massively huge) number of decimal places, π never contains another '7'. That would render some sequences impossible, while still preserving the proven-to-be-true property that π never repeats itself in its entirety (that is, that it's irrational).
You mean uniformly distributed. Being a normal number states more than that: any finite sequence of digits are uniformly distributed.
While you are right that Pi isn't proven to be normal, constructing a normal number isn't exactly hard. I think the following number is proven to be normal (binary representation, I added spaces for clarity):
0.0 1 00 01 10 11 001 010 011 100 101 110 111 ...
It's clear to see that this number contains all possible finite binary sequences.
In what theory? We can very well theoritically conceive that you just ask for those "sample words" from someone else who has the book either legitimately or illegitmately, and he gives them to you.
Even if the set of sample words is different for each metadata file, the book would be the same, so you just need a person willing to share the words with someone that doesn't own the book.
Now, an actual lawyer might obliterate this through some logic I don't know, and/or appeal to "bit color" as others are pointing out.
P.S. I do have to admit my idea reminds me of warrant canaries, a "clever hack" around the law that I ultimately believe can't really be lawful.
P.P.S. CleanFlicks et al may be relevant case law, as from what I understand they made varying efforts to "ensure" their customers held a license, but were basically still distributing copies of works they didn't own.
Yeah, but this way you get the electronic version that you want + the key (from some other user). In the end it comes to those distributing they keys, and it's not different from site offering trial-versions and others giving out serials and cracks.
>The point is to make it so that the distributors can reasonably argue they are not actually distributing copyrighted material
Just because it's encrypted it doesn't mean it's not copyrighted material. If I encrypt "Star Wars" and give it for download (letting others give the key), they'll still be all over my ass.
Note that retrieving any "useful information" about the original is believed to be similarly impossible from sha256(a_copyrighted_work) and encrypt_{sha_256(a_copyrighted_work)}(a_copyrighted_work).
There is not any such thing as random bytes. Here's a byte: 0xca. Is it random?
Well, if I got it from a fair die, perhaps. But if I got it by printing the hex of a Java program, of course not. Bytes cannot be random, they can only be created in a random way (or not).
In your hypothetical you are not creating the bytes in a random way, you are creating them from a book. So the bytes have nothing to do with so-called "random bytes", but are in fact a derivative work of the book, at the very least.
> Any one who considers arithmetical methods of producing random digits is, of course, in a state of sin. For, as has been pointed out several times, there is no such thing as a random number — there are only methods to produce random numbers, and a strict arithmetic procedure of course is not such a method. - John Von Neumann
But to your broader point, no, this is not a question of semantics. There is an actual syntactical difference being asserted here.
That difference is this: randomness has nothing to do with numbers. 0xca--the same number--can be either "random" or "non-random" depending on how you got ahold of it.
When we talk about numbers--prime numbers, even numbers, rational numbers, etc., we are talking about properties of the number; whether they are divisible by 2 and so on. But "random numbers" have no property by which we could recognize them. It is entirely a question of where the numbers were made, not what they are or what they look like.
To take it in a slightly different direction, since parody has been upheld, what about a deterministic, reversible, mechanical parodic transformation? AFAIK machine-created works can't be given a copyright, but I'm not asking for that, merely that a machine can create a sufficiently transformative derivative work.
In fact, maybe it doesn't even need to be parody. In https://en.wikipedia.org/wiki/Perfect_10,_Inc._v._Amazon.com.... search engines were permitted to serve thumbnails of copyrighted images because the copies "served a different function than [the original] use – improving access to information on the Internet versus artistic expression." In the same way, an e-book could be "improving access" rather than offering "artistic expression."
I guess establishing "transformation" is a bit tricky -- presumably ASCII-encoding an English book's text is not transformative. But it seems like somehow you could come up with something interesting and legal, even if it's not quite like my original formulation. Google Books' excerpts are probably like this.
Oh, and after re-reading the Colour article, I see I am not even all that innovative: http://monolith.sourceforge.net/
In the same way torrents are lists of hashes, which are used as keys to find byte blocks in the "cloud" of torrent clients.
Most ISPs/hosting providers have some bandwidth to spare, to be able to cover the spikes in traffic and support their growth. As long as this Infinite Downloader doesn't consume _too_ much traffic, we'll be happy to sponsor it. Bandwidth/peering agreements are made in terms of commitments, and as long as you stay beneath your committed data traffic, additional bandwidth isn't charged.
Once it starts to saturate uplinks or pose in any way a problem to other clients, it'll get shot down. But I don't see that being the case any time soon.
HBO's Silicon Valley, anyone?
Obviously I'm not going to try it, but it can serve as a good reminder about how much data you can go thru on a modern (or in my case simi-modern) connection in hardly no time at all.
Gotta love caps I suppose...
( I accidentally left it to download. The resulting file is 1.991.671.687 bytes. )
With a compression algorythm we could safe 100% of the bandwith (trending to infinite, xD).
By the way, the donwload speed always drops, even when using axel.