It doesn't surprise me that there are a lot of misunderstandings over contracts and IP considering how many vtubers are young folk who got into it while struggling to find more traditional employment in the aftermath of the pandemic. Those of us who are more seasoned tend to assume certain things should be commonsense or even intuitive completely forgetting that gen-z and even late gen-y didn't get a lot of the early career experiences and mentoring we did.
I started having questions about Neurosama (that I didn't look up because I cba) Stuff like what parts of the neuroverse is copywritable etc, personally I think most of it since it's all made by humans the only part that isn't is the content so it's probably fine, it's just that listening to things Neuro says may not be copywritable probably and idk if that could even be made an issue so yea, probably all good
@@Cameron115 AI-generated content is not copyright protected, but products that contain AI-generated content are still completely protected. Vedal starting the stream is enough to make it his. Practically this means that everything about Neuro is copyright protected. The funny thing is that theoretically some things Neuro says would not be copyright, but it's not illegal to quote a real person either, so effectively it makes no difference.
Fun fact, this gets complicated depending on where the person doing the work is located. This can be resolved by indicating in the contract which jurisdiction prevails. Otherwise it could result in some epic bullshit. Additionally, Canada doesn't recognize work for hire regardless, in that case, 'conceptual rights' will secure full rights to the work. In the end, talk to a lawyer, all the time, every time.
It gets murky, if we talk internationally. E.g. the copyright in Germany simply does not allow for every right to be handed over. The original creator or interpreter always retains the rights of the conceptual idea of a body of work and cannot license this right away. That makes for very complicated cases, where e.g. a composition is in the public domain, but an individual interpretation/recording of said composition has a new set of copyright on top, yada yada. So always write in the agreement which jurisdiction takes precedence! Always.
I always put art I get from skeb in the BRB screen with QR codes to the skeb artist. Solves the credit problem since I have no end screen. If I use commissioned art anywhere else, it gets credited, every time. vtuber models are a special exception because Twitch has space to put those credits, but RUclips and other social media does not. Twitter is especially bad since it barely has room for the artist name.
I don't think that's true. From what I recall, people get hit with this all the time when it comes to weddings where the wedding photographer get the copyright by default and can do whatever they want with them. Copyright in the US goes to the person who makes it by default.
Nope. If you fail to have a contract in writing (as in HARD COPY and Signed) it's up to the court or arbitration to decide what the agreement was based on evidence in emails and works produced. For all intents, the assumption in the US, and ONLY the US with another US citizen is that it's work-for-hire. Any other country, (Japan, Canada in particular) you can not assign moral rights away. Only the copyright. Which means that even if the artist says it's a work for hire, it is not. They can just not enforce their moral rights. This is why Japanese artists basically own all artwork to characters they draw even if they do not own the character, and also why it seems like their doujin works market is more permissive. Because the IP holders would just waste time and money in court fighting their fans and it creates negative sentiment towards the IP holder. So they choose not to in most cases. This is why on Skeb for example I'll get like a dozen appeals at a time, and they often have one very common paragraph about "asking permission to make a fanart". Realistically, they don't have to ask westerners (fair use,) but it's just polite to do so. If you pay them for it, it's still not a work for hire. When you pay for a work through fiverr, vgen, skeb, etc there is some standard language of the art platform itself that assigns rights to the person paying or person producing the work, these are boilerplate things, and you can generally override that language in your request to the creator of the work. I am not a lawyer, and please do not take this as legal advice.
Because a thing like that could make things getting muddier. Firstly, she is not directly related to that case, she may or may not know the detail of that case which potentially could make thing goes harder for melody, etc. that is wise of her to not mention that except if it's needed.
What if the rigger uses some novel techniques to make it look better than what other artists could have done? Would that be considered transformative and would people be willing to accept compromises to get that? Or would they do better in industry jobs?
Rigging is more of a technical skill than an art(speaking as someone who rigs in 3D). You could theoretically patent a particular rigging process with new software, but that's assuming the patent was accepted, and it would need to be genuinely new and revolutionary. Re-rigging an existing model is unlikely to be considered transformative because all you're doing is pinning parts of the model to an armature and weighting them. The only way you make it "better" is by adding more bones and doing finer rigging with more reference points - like using 100 points to animate the mouth instead of 20.
@SigurdStormhand thanks for the insight! I was thinking of an Irish channel who did some clever rigging so the head of his 3d "sausage" had retractable "skin". All the people in the comments were commending him for the way he did it but I have no understanding of 3d modelling so I'm probably wrong
@ OK, so that would be a prime example of "I must rig this and I must scream", unless it was actually an animation and not part of the rig. I'm not exactly clear on how Live 2D rigging functions, but the basic premise is the same, the model can only "hinge" around certain fixed points. To get highly accurate movement - especially if you have two layers on top of each other, you must have MANY points of reference. This is why a lot of VTuber 3D models will have coats or capes that will clip through the arms. I order to prevent clipping you have to exponentially increase the complexity of the armature, and it has to have physics attached to it, and each individual vertex must be precisely weighted. At a certain point you give up and just choose which bits you're willing to let clip.
@@SigurdStormhandIm no lawyer so can really say for certain if it could be considered transformative but I must say that L2D rigging differs greatly from 3D rigging I agree completely on your explanation with the perspective of 3D rigging but in the case of L2D while still technical there is a lot more "art" involved As someone that also had a 3D rigging background I was quite surprised when I started working on L2D, what we would normally consider "rigging" in 3D, is only a small (even optional) tool for a L2D rigger and its a lot closer to a 3D animator who is creating a prebaked animation set (but in this case you are already also implementing it in engine) The only tool resembling 3D rigging, vertex weighted to a bone system, is the glue tool where you can weight vertex from one mesh object to another directly but it is incredibly imprecise and you cannot replicate anything close to a bone system as it can only link one vertex to another vertex, both must be on the original mesh and if the weight (or anything else) causes a child to exit the out of the limited bounding box of its parent you tank performance Most of the L2D "rigging" is basically creating an absurd ammount of morphers (not saying blend shapes because that another term misused in L2D) through the use of FFD deformers that can be linked to form the usual parent/children hierarchy you would find in a 3D bone system, with the biggest FFD deformer affecting all their children, etc In L2D you do have to mesh each individual art piece (they are basically 3D textured flat surfaces), and pre animate all of their movement , with those animations activating either by tracking or by physics you need to configure precisely, and these animations can also compound on each other or trigger certain animation states only possible through a combination of specific tracking information Also as it is now L2D is very similar to the early age of computer graphics or videogame creation where there are few standards and whatever solution "works" is good enough Also my pet peeve is that model optimization is absolutely dreadful but most riggers lack a 3D background so they dont notice and those who do are not paid enough to care until someone develops poper standards or L2D develops some basic tools that have been standard in 3D for decades (You cannot cut a mesh into smaller islands to optimize UVs... And there are worse things... Yeah... I know of models that have more vertex that many current gen AAA 3rd person characters with more, maybe even double the ammount of 8K textures that were used in the rendered Game of thrones dragons) There are breakthroughs pretty often that either introduce new tools, new techniques or standarize old ones and there are rigging styles that while later adopted by other riggers can be clearly traced to individual riggers that either discovered or popularized them Hope the info was helpful, and sorry if it is quite the messy wall of text, im sleep deprived and L2D is quite the shitshow to explain even in person... Let alone through text hahaha
Depends on the artist, but there’s like an entire month plus of work that goes into a great model. Between the art and rigging there’s over 200 hours of work into one model with their toggles. How much would you expect to get a month for nearly all your time for a month?
@ it’s a lot of vectors, how many outfits does said model come with? How many layers are they? How many toggles do you want them to make along with expressions. It’s a lot of questions that only the person commissioning can answer.
@@otaking3582You clearly don't know what you're talking about. You're not aware what it takes to make one so how about you don't tell professionals the amount of time that is spent is "doing it wrong".
It doesn't surprise me that there are a lot of misunderstandings over contracts and IP considering how many vtubers are young folk who got into it while struggling to find more traditional employment in the aftermath of the pandemic. Those of us who are more seasoned tend to assume certain things should be commonsense or even intuitive completely forgetting that gen-z and even late gen-y didn't get a lot of the early career experiences and mentoring we did.
Misunderstandings also can stem from people being from different countries with different laws or English not being their first language.
Had this exact question about vtuber models after watching Dougdoug's Tech Tuesday presentation.
I started having questions about Neurosama (that I didn't look up because I cba)
Stuff like what parts of the neuroverse is copywritable etc, personally I think most of it since it's all made by humans the only part that isn't is the content so it's probably fine, it's just that listening to things Neuro says may not be copywritable probably and idk if that could even be made an issue so yea, probably all good
@@Cameron115 AI-generated content is not copyright protected, but products that contain AI-generated content are still completely protected. Vedal starting the stream is enough to make it his. Practically this means that everything about Neuro is copyright protected. The funny thing is that theoretically some things Neuro says would not be copyright, but it's not illegal to quote a real person either, so effectively it makes no difference.
Fun fact, this gets complicated depending on where the person doing the work is located. This can be resolved by indicating in the contract which jurisdiction prevails. Otherwise it could result in some epic bullshit.
Additionally, Canada doesn't recognize work for hire regardless, in that case, 'conceptual rights' will secure full rights to the work. In the end, talk to a lawyer, all the time, every time.
It gets murky, if we talk internationally. E.g. the copyright in Germany simply does not allow for every right to be handed over. The original creator or interpreter always retains the rights of the conceptual idea of a body of work and cannot license this right away. That makes for very complicated cases, where e.g. a composition is in the public domain, but an individual interpretation/recording of said composition has a new set of copyright on top, yada yada.
So always write in the agreement which jurisdiction takes precedence! Always.
I always put art I get from skeb in the BRB screen with QR codes to the skeb artist. Solves the credit problem since I have no end screen. If I use commissioned art anywhere else, it gets credited, every time. vtuber models are a special exception because Twitch has space to put those credits, but RUclips and other social media does not. Twitter is especially bad since it barely has room for the artist name.
The *DEFAULT* in the US is that any commissioned work is a work-for-hire, absent some agreement stating otherwise.
I don't think that's true. From what I recall, people get hit with this all the time when it comes to weddings where the wedding photographer get the copyright by default and can do whatever they want with them. Copyright in the US goes to the person who makes it by default.
Nope. If you fail to have a contract in writing (as in HARD COPY and Signed) it's up to the court or arbitration to decide what the agreement was based on evidence in emails and works produced.
For all intents, the assumption in the US, and ONLY the US with another US citizen is that it's work-for-hire. Any other country, (Japan, Canada in particular) you can not assign moral rights away. Only the copyright. Which means that even if the artist says it's a work for hire, it is not. They can just not enforce their moral rights. This is why Japanese artists basically own all artwork to characters they draw even if they do not own the character, and also why it seems like their doujin works market is more permissive. Because the IP holders would just waste time and money in court fighting their fans and it creates negative sentiment towards the IP holder. So they choose not to in most cases.
This is why on Skeb for example I'll get like a dozen appeals at a time, and they often have one very common paragraph about "asking permission to make a fanart". Realistically, they don't have to ask westerners (fair use,) but it's just polite to do so. If you pay them for it, it's still not a work for hire.
When you pay for a work through fiverr, vgen, skeb, etc there is some standard language of the art platform itself that assigns rights to the person paying or person producing the work, these are boilerplate things, and you can generally override that language in your request to the creator of the work.
I am not a lawyer, and please do not take this as legal advice.
@@TenshiKisaifrom my quick google search, the exception would be an employee creating work within the scope of their employment.
Thank you for reminding me that I dislike society because it is full of people just trying to out-grift each other constantly.
Man I always love listening to geega.
Surprised Geega didn't bring up the whole Mel and Digitrevx situation as an example.
Because a thing like that could make things getting muddier. Firstly, she is not directly related to that case, she may or may not know the detail of that case which potentially could make thing goes harder for melody, etc. that is wise of her to not mention that except if it's needed.
Dredging that up would be very disrespectful however relevant it is
I had to slow that down to .75 cause I couldn't keep up.
In short: if someone makes something, that something belong to them. You need yo write contracts in a specific manner to show that they do not own it.
What if the rigger uses some novel techniques to make it look better than what other artists could have done?
Would that be considered transformative and would people be willing to accept compromises to get that? Or would they do better in industry jobs?
Rigging is more of a technical skill than an art(speaking as someone who rigs in 3D). You could theoretically patent a particular rigging process with new software, but that's assuming the patent was accepted, and it would need to be genuinely new and revolutionary. Re-rigging an existing model is unlikely to be considered transformative because all you're doing is pinning parts of the model to an armature and weighting them. The only way you make it "better" is by adding more bones and doing finer rigging with more reference points - like using 100 points to animate the mouth instead of 20.
@SigurdStormhand thanks for the insight!
I was thinking of an Irish channel who did some clever rigging so the head of his 3d "sausage" had retractable "skin".
All the people in the comments were commending him for the way he did it but I have no understanding of 3d modelling so I'm probably wrong
@ OK, so that would be a prime example of "I must rig this and I must scream", unless it was actually an animation and not part of the rig. I'm not exactly clear on how Live 2D rigging functions, but the basic premise is the same, the model can only "hinge" around certain fixed points. To get highly accurate movement - especially if you have two layers on top of each other, you must have MANY points of reference.
This is why a lot of VTuber 3D models will have coats or capes that will clip through the arms. I order to prevent clipping you have to exponentially increase the complexity of the armature, and it has to have physics attached to it, and each individual vertex must be precisely weighted.
At a certain point you give up and just choose which bits you're willing to let clip.
@@SigurdStormhandIm no lawyer so can really say for certain if it could be considered transformative but I must say that L2D rigging differs greatly from 3D rigging
I agree completely on your explanation with the perspective of 3D rigging but in the case of L2D while still technical there is a lot more "art" involved
As someone that also had a 3D rigging background I was quite surprised when I started working on L2D, what we would normally consider "rigging" in 3D, is only a small (even optional) tool for a L2D rigger and its a lot closer to a 3D animator who is creating a prebaked animation set (but in this case you are already also implementing it in engine)
The only tool resembling 3D rigging, vertex weighted to a bone system, is the glue tool where you can weight vertex from one mesh object to another directly but it is incredibly imprecise and you cannot replicate anything close to a bone system as it can only link one vertex to another vertex, both must be on the original mesh and if the weight (or anything else) causes a child to exit the out of the limited bounding box of its parent you tank performance
Most of the L2D "rigging" is basically creating an absurd ammount of morphers (not saying blend shapes because that another term misused in L2D) through the use of FFD deformers that can be linked to form the usual parent/children hierarchy you would find in a 3D bone system, with the biggest FFD deformer affecting all their children, etc
In L2D you do have to mesh each individual art piece (they are basically 3D textured flat surfaces), and pre animate all of their movement , with those animations activating either by tracking or by physics you need to configure precisely, and these animations can also compound on each other or trigger certain animation states only possible through a combination of specific tracking information
Also as it is now L2D is very similar to the early age of computer graphics or videogame creation where there are few standards and whatever solution "works" is good enough
Also my pet peeve is that model optimization is absolutely dreadful but most riggers lack a 3D background so they dont notice and those who do are not paid enough to care until someone develops poper standards or L2D develops some basic tools that have been standard in 3D for decades (You cannot cut a mesh into smaller islands to optimize UVs... And there are worse things... Yeah... I know of models that have more vertex that many current gen AAA 3rd person characters with more, maybe even double the ammount of 8K textures that were used in the rendered Game of thrones dragons)
There are breakthroughs pretty often that either introduce new tools, new techniques or standarize old ones and there are rigging styles that while later adopted by other riggers can be clearly traced to individual riggers that either discovered or popularized them
Hope the info was helpful, and sorry if it is quite the messy wall of text, im sleep deprived and L2D is quite the shitshow to explain even in person... Let alone through text hahaha
GEEGA!!!! FULL EXPLAIN VIDEO THIS???
This helps for usa people
Crys in North Mexico
Definitely interesting
All I know is that VTuber models are always outrageously expensive
Depends on the artist, but there’s like an entire month plus of work that goes into a great model. Between the art and rigging there’s over 200 hours of work into one model with their toggles. How much would you expect to get a month for nearly all your time for a month?
@odin5625 If they're spending 200 hours on a single model, they're probably doing something wrong.
@ it’s a lot of vectors, how many outfits does said model come with? How many layers are they? How many toggles do you want them to make along with expressions. It’s a lot of questions that only the person commissioning can answer.
@@otaking3582You clearly don't know what you're talking about. You're not aware what it takes to make one so how about you don't tell professionals the amount of time that is spent is "doing it wrong".