This question is inspired by the TED talk Copyrighting all the melodies to avoid accidental infringement and its Lobsters thread. Basically, the story is that Damien Riehl and Noah Rubin put every melody on a hard drive, thus automatically copyrighting them, in order to defend the independent creation of small musicians. In the following thought experiment, I am going to take one step further and try to copyright all creative works created hereafter. I'm not sure of its practicality, but I assure you it would be fun.

Let me first introduce the Infinity Hard Drive. It is simply a device that echoes back its input. For example, if you feed a document/painting/music into it, it would echo back the identical document/painting/music. Essentially, it is merely a repeater that is unable to store anything like a conventional one, but you will see why I call it a hard drive in a moment.

Say Alice publishes her creative work, and I would like to sue her for violating my copyright. I have the evidence that if someone feeds Alice's work into the Infinity Hard Drive, it will return the identical work. In fact, you can find anything in the Infinity Hard Drive, as long as you know what you are looking for.

You might say "Seriously? You are literally taking Alice's work to generate yours!" But think about it: what would Alice do if Bob sues her for violating his copyright law? She would read Bob's work, and try to find the relevant drafts on her drawer that proves her work is original and independent. Relevance is the key here: if Alice has thousands of drafts at hand, it would make no sense to submit all of them to a court in response to a lawsuit concerning a specific work. Alice can read Bob's work to find her relevant work, so why can't the Infinity Hard Drive read someone's work to find my original work? I am arguing that conceptually, the process is more like using Alice's work to query my infinite collection of creative works to compute the location where my work resides, rather than generating one from scratch. Maybe Alice doesn't need to read Bob's whole work, but this is because her collection is finite. The Infinity Hard Drive is infinite in capacity, so naturally, it takes more information to find a specific work. In fact, Damien Riehl and Noah Rubin also need to take a melody to locate its counterpart in their hard drive, so if their approach works, so will mine.

To reiterate, the Infinity Hard Drive serves as a "tangible medium of expression" of my infinite collection of creative works. From my understanding, the internal mechanism of such a media does not matter. It can be a piece of paper, an SSD, or the Infinity Hard Drive which conceptually contains everything. What really matters is the outcome, that "(works) can be perceived, reproduced, or otherwise communicated, either directly or with the aid of a machine or device."

Just to clarify, what I’m trying to (automatically) copyright is not the Infinity Hard Drive itself, or the process of "querying" stuff in it, or the idea of making one. I am copyrighting the conceptual contents within. Just like you don't copyright the ink on a paper or electronic signals on a hard drive; you copyright the conceptual content born.

I probably also need to prove that my work is "the result of creative decisions of a human being", but that is arguable. We know that a randomly-generated book is not a violation of copyright. Also, how could J.K. Rowling prove her work is creative instead of the accurate deception of an alternative universe? Anyway, if the contents of the Infinity Hard Drive is not copyrightable, then nothing is, because it contains everything.

Finally, how do I prove that Alice had access to my infinite collection of creative works prior to her creation? Well, you are reading this right now, and the Infinity Hard Drive is, in fact, ubiquitous, e.g. a mirror, or a computer that echoes back your keystrokes on the screen.

Now, can I sue you for copyright infringement with the Infinity Hard Drive? To extend the scope of this question a little, can I also claim its ownership? originality? If the answers to the above questions are all "No", then can I defend myself from a copyright infringement lawsuit with the Infinity Hard Drive?

  • I do not completely understand your idea but in all cases you are not the creator of anything, correct? And everything you feed in has already been in a fixed form by the actual creator? Feb 12, 2020 at 21:12
  • @GeorgeWhite Yeah, we all know I'm not the creator. The purpose of this question is just to highlight a loophole in the copyright law that we can probably exploit.
    – nalzok
    Feb 12, 2020 at 21:18
  • 2
    If you didn’t originate a document and it has been fixed by the author it is already copyrighted. The action of feeding it to your content addressable memory (there is a name for it) does not copyright something already copyrighted. If this did work, what was the loop hole? Feb 12, 2020 at 21:30
  • @GeorgeWhite That’s a really nice summary! The loophole is that everything is already in the addressable memory right now. I’m not feeding the document into the memory. Instead, I’m reading the document out from the memory by using the document from someone who “violated” by copyright as an address, just like Alice finding her own draft on a huge bookshelf which requires a long call number to navigate. The drafts are always there to be found.
    – nalzok
    Feb 12, 2020 at 21:38
  • 2
    Separately, you are doing nothing like the people in the video. They claim a finite number of melodies can be made with the notes, for example, on a piano. If they methodically generate all melodies of a specific length or shorter then an exhaustive body of melodies will have been fixed. For your thing to be analogous you would need to get those million monkeys typing for a million years to generate Hamlet and everything else that could ever be written. i am going to avoid an extended discussion. Feb 13, 2020 at 0:15

3 Answers 3


Copyright requires originality

Your infinity hard drive appears to be a machine designed to violate copyright by immediately copying anything presented to it.

Damien Riehl and Noah Rubin were creating original melodies. These do not have copyright until they are fixed in a tangible medium. That's what the hard drive is for. This is insurance against them being sued if they release a song and someone claims it violates that person's copyright - they can produce in court the melody with a date stamp of 2015 (or whatever). GitHub or similar would be even better evidence.

In this context there is a specific allegation that melody X infringes copyright. Riehl & Rubin can then go to their records (including metadata) and say no, here is melody X version 1 through n and they all predate your release so we didn’t violate your copyright.

  • This is exactly my confusion! Damien Riehl and Noah Rubin need a melody to locate the corresponding melody stored in their hard drive. The Infinity Hard Drive also needs a document/painting/music to find its counterpart. What's the difference then? A repeater is just a compact representation of a storage device where the content at each address is the address itself. In other words, the contents have already been created; they are just stored compactly. So.. does the representation of contents actually matter?
    – nalzok
    Feb 12, 2020 at 22:35

The Alice-Bob analogy has a gaping hole: Alice has actually produced copyrightable work, and did so before Bob. The Infinity Hard Drive, as you point out, doesn't produce or store anything - your question about having "the Infinity Hard Drive read someone's work to find my original work" makes no sense, as there is no original work that existed before Bob made it. You cannot ever show that you made this work before Bob, and Bob can clearly show a time period where his work existed and yours did not - the time before he showed it to you.

This isn't Alice reading Bob's work, and finding a draft among thousands that match and predate his work - it's Alice reading Bob's work and then copying it. Bob's original predates any copy that Alice might make, and therefore is deserving of the copyright. This seems rather like trying to copyright the "conceptual contents" of a pen, which you could also use to transcribe any text. Copyright protects creative works, not the potential to generate a work.

  • I see your point, but in the normal case, do I need to publish all my drafts before someone is suing me? I mean, Bob can steal Alice's work and publish it before her. Now, even if Alice has incomplete drafts to prove she is the original author, by your reasoning, she cannot have the copyright because "she cannot ever show that she made this work before Bob". In this case, there is original work that existed before Bob stole it, but it was not published before Bob's publication, so it can never be proven?
    – nalzok
    Feb 12, 2020 at 21:24
  • 2
    @nalzok You don't need to publish a work to have a copyright on it, copyright typically happens automatically in the US at the time of creation. But to prove it in a court of law, Alice will require some form of proof that her draft pre-dates Bob's - having published it previously is excellent evidence, but there are other ways. If Alice can't convincingly prove that her draft even existed before Bob's, then she'll have a very hard time proving that he copied it. Automatic copyright has its uses, but explicitly copyrighting something leaves no room for doubt here. Feb 12, 2020 at 21:30
  • I think the key is the definition of "creation". You can see the conversation between me and @George White above, especially this comment. The question is essentially arguing that the contents of all possible creative works have already been created right now. Using the Alice-Bob analogy: does Alice create her own work (sealed before Bob's publication) when it's unboxed in the court, or at a prior point? does the Infinity Hard Drive create its contents when it's read, or at a prior point?
    – nalzok
    Feb 12, 2020 at 22:23
  • 1
    @nalzok It seems pretty clear that the IHD creates its content when it reads Bob's work, and not a moment before. As you point out in the question, "it is merely a repeater that is unable to store anything" - it is literally impossible for the device to even remember anything it produced at any point in the past, so even if it had somehow produced Bob's work at some point, there is no record of it. A device with no memory is utterly useless in making claims about the past. It's like if I claimed I wrote Harry Potter, but forgot to save it. Feb 13, 2020 at 14:03

A lawyer would simply prove no such device could actually "fix" a representation of every copyrightable work (as we understand physics). If it actually stored every possible copyrightable work, you'd run into storage problems very quickly. Very, very quickly.

Let us consider every single possible tweet under the old character restriction, 140 characters. Considering only lower case English letters and spaces this amounts to a number that greatly exceeds the number of protons in the observable universe. Squared. (~10^200 vs. ~10^86)

And don't think you can cheat by doing something silly like claiming "well, I produced it earlier and I'm just backtracking to find when I produced it and wiped it to save space". If every proton and electron produced a different tweet of length 140 adhering to the above constraints for every Planck time for the entire age of the universe, you're still not even remotely close.

Given we've shown you can't possibly list out every tweet, the lawyer would feed your device a couple and after your device parroted back a couple of absolute nonsense strings that no sane person would ever copyright, you'd be exposed as a fraud with certainty far beyond the usual proof standard for such a case, which is only a bit better than 50%.

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .