Could you please point me to legal definitions, in court or otherwise, that say it is not violating my copyright license to directly use my artwork in any shape or form for a non-fair use product? As in, a service you pay money for to create things based on the training data it has taken from me, is not fair use. Or point me to the legal definitions where I lose my copyright by posting things online? Allowing to scrape is not the same thing as giving derivative copyright license permissions. You aren’t disagreeing with me, you’re disagreeing with my legal rights.
hmm. That talks of data mining not of derivative work from the mined data. … I’ll leave the discussion to others, though.
What effect would a Creative Commons “ND” clause have? Is it reserving the right to make derivatives? And would machine generated stuff even count as a derivative?
DuckDuckGo translate:
Copyright and Related Rights Act (Copyright Act) § 44b Text and Data Mining
(1) Text and data mining is the automated analysis of one or more digital or digitized works in order to obtain information, in particular about patterns, trends and correlations.
(2) Reproduction of lawfully accessible works for text and data mining is permitted. The
reproductions must be deleted when they are no longer required for text and data mining.
(3) Uses pursuant to subsection (2) sentence 1 shall only be permitted if the rights holder has not reserved them. A reservation of use for works accessible online is only effective if it is made in machine-readable form.
Translation:
(1) Text and data mining is the automated analysis of one or more digital or digitized works in order to obtain information, in particular about patterns, trends and correlations.
(2) Reproduction of lawfully accessible works for text and data mining is permitted. The reproductions must be deleted when they are no longer required for text and data mining.
(3) Uses pursuant to subsection (2) sentence 1 shall only be permitted if the rights holder has not reserved them. A reservation of use for works accessible online is only effective if it is made in machine-readable form.
None of that says anything about creating profitable derivative work. In fact is specifies patterns, trends and correlations, which does not lead me to believe it is protecting visual works created from this data, those kind of things are only used to inform things, like information science.
Algorithms trained with that training data are considered to make new works, so your only change to stop your work getting into a Algorithm is by going against the ones that make the training data. And they will always pull the science card.
And respectively, the algorithm muches up the training data so extremely that less than 0,1% of a training data picture is inside the result, therefore its legally a new work, however that work itself doesn’t fall under copyright protection because that is only a thing for human art. And as already stated, the training data is harvested according to the law i mentioned.
Could you please point me to legal definitions, in court or otherwise, that say it is not violating my copyright license to directly use my artwork in any shape or form for a non-fair use product? As in, a service you pay money for to create things based on the training data it has taken from me, is not fair use. Or point me to the legal definitions where I lose my copyright by posting things online? Allowing to scrape is not the same thing as giving derivative copyright license permissions. You aren’t disagreeing with me, you’re disagreeing with my legal rights.
https://www.gesetze-im-internet.de/urhg/__44b.html
German law and many of the data mining companys are German.
hmm. That talks of data mining not of derivative work from the mined data. … I’ll leave the discussion to others, though.
What effect would a Creative Commons “ND” clause have? Is it reserving the right to make derivatives? And would machine generated stuff even count as a derivative?
DuckDuckGo translate:
The stuff the algorithm makes is considered a new thing because the impact of each individual pice of training data is basically unmeasurably small.
Where has that been proven legally?
Translation: (1) Text and data mining is the automated analysis of one or more digital or digitized works in order to obtain information, in particular about patterns, trends and correlations.
(2) Reproduction of lawfully accessible works for text and data mining is permitted. The reproductions must be deleted when they are no longer required for text and data mining.
(3) Uses pursuant to subsection (2) sentence 1 shall only be permitted if the rights holder has not reserved them. A reservation of use for works accessible online is only effective if it is made in machine-readable form.
None of that says anything about creating profitable derivative work. In fact is specifies patterns, trends and correlations, which does not lead me to believe it is protecting visual works created from this data, those kind of things are only used to inform things, like information science.
Algorithms trained with that training data are considered to make new works, so your only change to stop your work getting into a Algorithm is by going against the ones that make the training data. And they will always pull the science card.
Where are algorithms considered as being new and legally allowed derivative works in relation to visual works of art?
Germany.
And respectively, the algorithm muches up the training data so extremely that less than 0,1% of a training data picture is inside the result, therefore its legally a new work, however that work itself doesn’t fall under copyright protection because that is only a thing for human art. And as already stated, the training data is harvested according to the law i mentioned.