Random Startup Generator
ykombinator.com
ykombinator.com
Ask HN: Does my startup need to code unicorn picture technology?If the implementing website did not actually copy the copytext from the generator and came up with the exact same language by chance, then the implementing party owns the copyright (or more accurately, the implementing party has not infringed upon the rights of whoever wrote the idea generator).
But let's assume there was actual copying. Then whoever write the idea generator owns the copyright on the copytext (assuming he or she didn't assign those rights to an employer, etc.). BUT that assumes the copytext is copyrightable to begin with.
Copyright law requires that a work be created by a human being in order for it be copyrighted. In particular, the U.S. Copyright Office refuses to register "works produced by a machine or mere mechanical process that operates randomly or automatically without any creative input or intervention from a human author."[1] Since the copytext here is computer-generated, you could argue that no one owns the copyright.
But that's a stretch. Although each individual idea is randomly generated by a computer, a fair amount of human creativity went into making sure the ideas it spits out aren't total gibberish. I don't know what the exact level of human input is required for something to be copyrightable, and as far as I'm aware, there's no court case directly on point here. But I think there's enough for whoever wrote the idea generator to make a good copyright claim.
One final aside though: If a start-up actually copies the copytext, they still might be protected under fair use. Fair use generally protects copying that is minimal and serves a different purpose than the original (e.g. quoting a passage from a book in order to critique it).
In this case, the work that is being copyrighted would have to be the set of all ideas generated as a whole, not the text of any particular idea. That is, assume the idea generator is capable of generating 100,000 possible random start-up ideas. Then the unit of "work" that is copyrightable is the set of all 100,000 ideas. Therefore, because you're only copying one of those 100,000 ideas, you might be able to argue that your copying is minimal. Moreover, your copy of the work serves a different purpose than the original (the purpose of the generator is parody, but in your case, you're actually being serious). There are other factors that go into fair use (economic effect and commercial vs. non-commercial use), so it's hard to say for sure whether you could get away with it, but it's certainly a plausible position to take.
This is quite false. A copyrighted work's content is protected by copyright, and how it is duplicated doesn't enter into the equation.
In principle, with a very powerful set of computers and a lot of time, I could randomly generate any work of art, visual, sound or print. The fact that the works were generated randomly doesn't make any difference.
There are many variations on the thousand monkeys at typewriters story going back decades, but the idea is if enough random sequences of words are generated, eventually a recognizable, copyrighted work will appear.
Many people don't realize that the "Birthday Song" is copyrighted. The first measure is a mere six notes long. Assuming I can render the tune within a range 12 semitones, I only have to generate 12^6 = 2,985,984 notes, to be absolutely certain to have randomly violated the copyright somewhere in the sequence -- and, of course, it's very likely that I will have succeeded well before generating all 2,985,984 notes.
But my method of generating the tune is irrelevant to the outcome, which is a potential copyright violation.
Under U.S. law, there must be actual copying in order to infringe upon copyright.[1] Note that is different from patent law, which does not protect independent "invention".
That doesn't mean you can get around the Happy Birthday copyright[2] by simply randomly generating six note sequences until you get the one you want. That's still considered copying for all intents and purposes.
On the other hand, let's say you somehow never heard the Happy Birthday song and somehow just came up with them on their own. That's fine -- no copying, no infringement. This is what I was referring to.
[1] http://en.wikipedia.org/wiki/Substantial_similarity
[2] There's actually a good argument that the song is no longer copyrighted. http://arstechnica.com/tech-policy/2013/06/filmmaker-picks-a...
No, not confused in the slightest.
> By "came up with the exact same language by chance", I mean "came up with the exact same language without reference to idea generator," i.e. your copy of the text was independently created.
Did you actually read my post? My "Birthday song" example shows that a randomly generated sequence of notes still constitutes a copyright violation.
> Under U.S. law, there must be actual copying in order to infringe upon copyright.
Yes, but you're confused about the meaning of "actual copying". Any method that substantially reproduces the original constitutes actual copying.
> That doesn't mean you can get around the Happy Birthday copyright[2] by simply randomly generating six note sequences until you get the one you want.
That is absolutely false, and I am shocked that you would post this nonsense in a public forum. The result is the only issue -- if it's a copy of a protected work, no one asks how the copy was generated.
> On the other hand, let's say you somehow never heard the Happy Birthday song and somehow just came up with them on their own. That's fine -- no copying, no infringement.
Also false, as much case law demonstrates. The difference in the case of simultaneous invention is the severity of the punishment, not the judgment that it's a violation.
Your reference to substantial similarity isn't relevant to what we're talking about. Substantial similarity addresses the issue of how alike two works are, not how a work was copied. Two works that are sufficiently different are judged not to be in conflict with respect to copyright issues. The method of copying is not a factor.
> There's actually a good argument that the song is no longer copyrighted.
Yes, and IMHO it shouldn't be copyrighted. It's a classic example of a perversion in copyright law.
Nope, you're flat out wrong and I'll stake my bar license on it. Actual copying requires that "the defendant actually used the plaintiff’s work in order to create his or her own work."[1] Courts typically require proof of "access to the copyrighted work" in addition to substantial similarity between the results.[2] Substantial similarity can help constitute circumstantial evidence of copying but does not indicate infringement in and of itself.[3]
---
>> That doesn't mean you can get around the Happy Birthday copyright[2] by simply randomly generating six note sequences until you get the one you want.
> That is absolutely false, and I am shocked that you would post this nonsense in a public forum. The result is the only issue -- if it's a copy of a protected work, no one asks how the copy was generated.
Read my original quote again. I think you missed the "doesn't" in my statement. We're in agreement that randomly generating content until it resembles a copyrighted work will get you in trouble, albeit for different reasons.
---
>> On the other hand, let's say you somehow never heard the Happy Birthday song and somehow just came up with them on their own. That's fine -- no copying, no infringement.
> Also false, as much case law demonstrates. The difference in the case of simultaneous invention is the severity of the punishment, not the judgment that it's a violation.
Can you point out a single case where that's true? There's at least one appellate level case saying the opposite (albeit with 3 notes, not 6):
'''[D]efendant submitted affidavits by experts analyzing the two compositions and showing that the only similarity between them is a recurring three note sequence found also in the works of Johann Sebastian Bach. The evidence presented clearly does not raise a question of fact as to whether the two compositions were so strikingly similar as to preclude the possibility of independent creation.'''[4]
You may be confusing patent law with copyright. Under patent law, independent invention is no defense (except as to damages). But copyright law != patent law (unfortunately IMHO -- some semblance of an independent invention requirement would make patent trolling less profitable).
---
Citations:
[1] http://www.americanbar.org/groups/young_lawyers/publications...
[2] See, e.g., Sid & Marty Krofft Television v. McDonald's Corp., 562 F. 2d 1157, 1162 (9th Cir. 1977), available at http://scholar.google.com/scholar_case?case=1674068343222286....
[3] See, e.g., Donald v. Zack Meyer's TV Sales and Service, 426 F. 2d 1027, (5th Cir. 1970), available at http://scholar.google.com/scholar_case?case=1791074878897978..., and Orgel v. Clark Boardman Co., 301 F. 2d 119 (2d Cir. 1962), available at http://scholar.google.com/scholar_case?case=9525184415609798....
[4] Ferguson v. National Broadcasting Co., Inc., 584 F. 2d 111, 114 (5th Cir. 1978), available at http://scholar.google.com/scholar_case?case=1791074878897978...
I look forward to future algorithms that will, on demand, create random but coherent music that closely resembles Mozart, or Beethoven (the string-quartet Beethoven, not the other one :) ) or another composer of the user's choosing. I know some pilot projects like this have been done, but they're not accessible yet.
Reasonable IANAL answer: the person who owns the copyright to (the strings in) the code.
Perhaps "Cloudulation" is what happens at the end of a successful project...