6 ms·

Wolfram has been claiming copyright to the output of Alpha for a long time. http://www.groklaw.net/article.php?story=20090518204959409 http://www.groklaw.net/a

Wolfram has been claiming copyright to the output of Alpha for a long time. http://www.groklaw.net/article.php?story=20090518204959409 http://www.groklaw.net/article.php?story=20090518204959409 It's a bit weird because Wolfram is not the one solely responsible for Alpha's output; the user has to give Alpha the right input to produce the right output, so I don't really know how this could work in court.

rahuldottech 7y ago
Think of it this way: I have a black box. You give it some input, it does some magic, and gives you an output (a piece of "art"). Can I copyright such "art"? Sure, you triggered the event. But my black box did all the actual work, right? Similarly, if you say something to your friend and that inspires them to create a painting. Can you claim ownership to that? (Food for thought: can you really "copyright" or "own" a mathematical expression or graph, though? Because that's most of Wolfram Alpha)
- jordigh 7y ago
  The trigger can be really complicated, though, complicated enough to merit the nontriviality threshold of copyrightability. If anything, Wolfram and the Alpha user should be coauthors, but Wolfram doesn't seem to think anyone else has thoughts complex enough to meet the coauthorship threshold.
  - umvi 7y ago
    I mean, he was the youngest student ever to get a PhD from Caltech.
    - icebraining 7y ago
      Yeah, but did he ever win the Putnam?
  - rahuldottech 7y ago
    Fair enough, I get your point. That said, I'm not sure "copyrighted" mathematics/graphs would ever really hold up in courts. I don't know how to really phrase this, but isn't maths mostly (completely?) dealing with "universal truths"? Related article (I've only skimmed through it): http://www.mondaq.com/india/x/784302/Copyright/Mathmatical+Equation+Per+Se+Cannot+Be+Copyrightable http://www.mondaq.com/india/x/784302/Copyright/Mathmatical+E...
  - gwern 7y ago
    The Alpha user and Wolfram certainly could share a copyright for nontrivial inputs which represent a _de minimis_ creative contribution, but Wolfram can simply put a copyright assignment in the Terms of Service license agreement the Alpha user agrees to, and I bet if you read carefully through enough legalese, you'll find that somewhere (just in case).
- umvi 7y ago
  > can you really "copyright" or "own" a mathematical expression or graph, though? Yes, you can. Graphs, numbers, it doesn't matter, it's all information. Can I copyright my book? What if someone converts my entire book from letters to numbers and them modulates the numbers as a mathematical graph using, say amplitude. Can he freely distribute that graph? Or better yet, can he convert that graph from pixels on a screen to radio waves and broadcast it?
  - jordigh 7y ago
    Raw data generally does not meet the copyrightability threshold in the US (datasets are often copyrighted in the EU, though). The lack of copyrightability of data (say, phone numbers or maps) is why there are paper towns on maps, one kind of copyright traps: https://en.wikipedia.org/wiki/Fictitious_entry https://en.wikipedia.org/wiki/Fictitious_entry Mathematical facts are usually also not copyrightable for the same reason: they're just considered raw data. However, we can obviously copyright a mathematics book or article because there's more than just raw facts in it: there's style, presentation, and many other signs of authorship.
  - mikecsh 7y ago
    Your comment reminds me of illegal numbers [1] where a number is illegal because of the information it represents. [1] https://en.m.wikipedia.org/wiki/Illegal_number https://en.m.wikipedia.org/wiki/Illegal_number
  - rahuldottech 7y ago
    Unsure about other places, but how it works in my country: http://www.mondaq.com/india/x/784302/Copyright/Mathmatical+Equation+Per+Se+Cannot+Be+Copyrightable http://www.mondaq.com/india/x/784302/Copyright/Mathmatical+E... > In a legal sense also, you can copyright the content of a math book (such as diagrams, word problems, and illustrations) but the formula, problems, proofs, theorems themselves cannot be copyrighted. You cannot copyright a fact which is established and natural i.e., 1+1=2.
- otakucode 7y ago
  Your black box is a camera. And there is a mountain of legal precedent saying that I own the copyright, and you can not. Also, copyright applies to anything fixed in a medium of expression. You can not copyright a formula, but you can copyright its expression as soon as it is fixed in a medium. You cannot copyright the abstract idea of a graph, but you can copyright its expression or rendering once fixed in a medium.
baroffoos 7y ago
With that logic, the camera manufacturer owns copyright to all photos taken with it.

rahuldottech 7y ago

Think of it this way: I have a black box. You give it some input, it does some magic, and gives you an output (a piece of "art"). Can I copyright such "art"? Sure, you triggered the event. But my black box did all the actual work, right? Similarly, if you say something to your friend and that inspires them to create a painting. Can you claim ownership to that? (Food for thought: can you really "copyright" or "own" a mathematical expression or graph, though? Because that's most of Wolfram Alpha)

jordigh 7y ago
The trigger can be really complicated, though, complicated enough to merit the nontriviality threshold of copyrightability. If anything, Wolfram and the Alpha user should be coauthors, but Wolfram doesn't seem to think anyone else has thoughts complex enough to meet the coauthorship threshold.
- umvi 7y ago
  I mean, he was the youngest student ever to get a PhD from Caltech.
  - icebraining 7y ago
    Yeah, but did he ever win the Putnam?
- rahuldottech 7y ago
  Fair enough, I get your point. That said, I'm not sure "copyrighted" mathematics/graphs would ever really hold up in courts. I don't know how to really phrase this, but isn't maths mostly (completely?) dealing with "universal truths"? Related article (I've only skimmed through it): http://www.mondaq.com/india/x/784302/Copyright/Mathmatical+Equation+Per+Se+Cannot+Be+Copyrightable http://www.mondaq.com/india/x/784302/Copyright/Mathmatical+E...
- gwern 7y ago
  The Alpha user and Wolfram certainly could share a copyright for nontrivial inputs which represent a _de minimis_ creative contribution, but Wolfram can simply put a copyright assignment in the Terms of Service license agreement the Alpha user agrees to, and I bet if you read carefully through enough legalese, you'll find that somewhere (just in case).
umvi 7y ago
> can you really "copyright" or "own" a mathematical expression or graph, though? Yes, you can. Graphs, numbers, it doesn't matter, it's all information. Can I copyright my book? What if someone converts my entire book from letters to numbers and them modulates the numbers as a mathematical graph using, say amplitude. Can he freely distribute that graph? Or better yet, can he convert that graph from pixels on a screen to radio waves and broadcast it?
- jordigh 7y ago
  Raw data generally does not meet the copyrightability threshold in the US (datasets are often copyrighted in the EU, though). The lack of copyrightability of data (say, phone numbers or maps) is why there are paper towns on maps, one kind of copyright traps: https://en.wikipedia.org/wiki/Fictitious_entry https://en.wikipedia.org/wiki/Fictitious_entry Mathematical facts are usually also not copyrightable for the same reason: they're just considered raw data. However, we can obviously copyright a mathematics book or article because there's more than just raw facts in it: there's style, presentation, and many other signs of authorship.
- mikecsh 7y ago
  Your comment reminds me of illegal numbers [1] where a number is illegal because of the information it represents. [1] https://en.m.wikipedia.org/wiki/Illegal_number https://en.m.wikipedia.org/wiki/Illegal_number
- rahuldottech 7y ago
  Unsure about other places, but how it works in my country: http://www.mondaq.com/india/x/784302/Copyright/Mathmatical+Equation+Per+Se+Cannot+Be+Copyrightable http://www.mondaq.com/india/x/784302/Copyright/Mathmatical+E... > In a legal sense also, you can copyright the content of a math book (such as diagrams, word problems, and illustrations) but the formula, problems, proofs, theorems themselves cannot be copyrighted. You cannot copyright a fact which is established and natural i.e., 1+1=2.
otakucode 7y ago
Your black box is a camera. And there is a mountain of legal precedent saying that I own the copyright, and you can not. Also, copyright applies to anything fixed in a medium of expression. You can not copyright a formula, but you can copyright its expression as soon as it is fixed in a medium. You cannot copyright the abstract idea of a graph, but you can copyright its expression or rendering once fixed in a medium.

baroffoos 7y ago

With that logic, the camera manufacturer owns copyright to all photos taken with it.