A Proposal for the Creative Commons
The optimist says, "We live in the best of all worlds."
The pessimist says, "Yeah, I know."
When it comes to the Creative Commons and photography, it's no surprise that the reactions to my blog entries on the subject seems to come from extremists one either one side or the other. Their common misunderstanding is that I feel the Creative Commons itself is flawed. True, I've stated that because photography is a common, everyday thing that it is a poor fit for the CC. But this is the pessimistic viewpoint of a more neutral reality of my message: that it's the byproduct of how CC is assigned to photos, and how people generally acquire CC photos that creates most of its problems.
In this posting, I am going to present the "optimist" perspective of the exact same underlying premise, but with a twist: a proposal for a framework for how risk conditions for both sides of a CC-licensed work can be minimized. This is a framework, mind you--one that has its flaws like any other--but at least it presents an approach that plugs the most egregious holes in how CC is applied to photos, and how people pick up photos for use.
This simple scenario illustrates the problem at hand:
You're a judge. In front of you are two parties: a photographer and a company. The photographer claims that the company infringed upon his copyright by placing his photo on their web site without his permission. The company states that the photo was marked as CC-attribution on Flickr, and they complied with its terms precisely. The photographer counters that he never made the image available for CC-licensing.
As the judge, you have to determine three potential explanations for how the photo got on Flickr and was assigned the CC license:
In copyrights infringement claims, the legal process starts with the indisputable facts: (1) the photo is owned by the photographer, and (2) the defendant has used it. Normally, it's at this point that the onus is on the defendant to prove that they properly acquired the photo legally from the photographer (or his assigns, like a stock photo agency). Unless they can, they are liable for infringement. Though they can claim it had a CC-license, that part is irrelevant unless and until it is established that the photographer assigned that license to it. One of the problems with the current use of the CC is that, though the photographer may voluntarily assign the CC license to it, it's impossible to differentiate that from a fraudulent assignment, which is why all of this is being discussed. But unless and until such a case can be shown or volunteered, it remains the defense's responsibility to present proof of permission to use the photo.
In the above scenario, most judges would have a hard time finding a legal reason to rule in favor of the defense without such evidence or admission by the photographer, who may in fact be lying. Yet, even if a sympathetic judge were to see the possible ambiguity, there's the possibility that the company may be lying. And the worst part is that both sides could be telling the truth and end up paying the price together (because of the high costs of litigating) if it turned out that some third party caused all this.
While such conditions are always possible for any kind of copyrighted work, the CC worsens the situation because it provides all sides with more tools and opportunity to create these conditions than before the CC.
But again, the CC itself isn't the problem! It's that it is has no reliable infrastructure. Flickr allows anonymous users to attribute a CC license to any object at all, and that's the problem. And not just for CC and those on either side of the photo in the middle, but Flickr itself, as would be any other entity that also tried to enable CC-licensing.
Infringement claims are already going on, but if the number of such claims continues to rise, it isn't a stretch to find someone filing a claim against Flick for contributory infringement. For those not familiar, contributory infringement requires (1) knowledge of the infringing activity and (2) a material contribution (actual assistance or inducement) to the alleged piracy.
Note that the Digital Millenium Copyright Act (DMCA), which normally protects companies like Flickr (Yahoo) from direct copyright infringement because they merely host the photo on their site, it does not protect them from contributory infringement if the second part of its definition applies: that Flickr provides "actual assistance or inducement". My feeling is that Flickr's current presentation of offering users to apply CC licenses, while not fully disclosing its ramifications to either photographers or users of such photos, is both assistance and inducement, and thus, puts them at risk for contributory infringement.
Crafting a Solution
Back to optimism. Alleviating this problem is relatively simple in nature; it's the details of implementation that can get sticky. The solution is to have some kind of mechanism for identification. In the simplest case, Flickr could just act more responsibly by not allowing users to assign CC licenses to photos without at least one more step in identifying who the person is that's doing it. This very simple act can help to alleviate a chain reaction of bad things, thereby reducing some ambiguities around the photo. It also reduces some of the inadvertent CC-assignments, while also potentially giving some publishers a tad more confidence in where the image came from. Of course, none of this is foolproof, but this one step still helps quite a bit.
Though it would be a good stop-gap measure for Flickr to do this, it's not really a long-term solution, especially if more sites were to ever adopt CC-licensed photo hosting as Flickr does. That's where it becomes necessary to craft a more formalized solution. The trickiness comes in the balancing of the "effectiveness" dividend against the "implementation" costs.
In theory, the next step is to introduce a documented process of registering and authenticating users on both sides of the "transaction." While that sounds bureaucratic and scary, this is not hard, nor new to anyone that uses the internet. The question is really the degree in which the CC or its users wish to participate.
Why consider it? Because it gives publishers a little more assurance that the photo may belong to a particular party, just as it gives photographers access to information about who's using their images. And that's been the main problem that's caused all of this. Again, this isn't CC's fault, but it would certainly help if the CC itself provided tools for its affiliates to act more responsibly.
First, no one would ever be required to register themselves with the CC. It is used only as a vehicle for information that both content owners and licensees can use to give themselves an added level of assurance. A photo's user can have a step more confidence that the image is owned by a trackable person; and the photographer can potentially be notified when his image is used. When these come together, each party is more likely to act in compliance with the terms of any given CC license.
Note that when people register, they are issued a Creative Commons Identification Number, or CC-ID, which can be used wherever they post CC-licensed content. Users can always remain anonymous to the outside world, and even to the CC (except for an email address, which is as good as anonymous anyway), in which case, potential users of that person's content knows this ahead of time, which may give them reservations about using it due to risk.
Here are the basic levels of identity that may be associated with any given person:
Note that only people and companies register, not the works they submit. And it is only done once, after which an ID is issued, which is used to assign to CC-license to creative works.
Not every level of verification has to be implemented. (It may be too cost prohibitive in the early stages to deal with postal-mail submissions.)
The registration process has one of three states: "application submitted", "pending", and one of the above "confirmations."
So, is this solution merely a matter of having a registration process? Not quite--there is one more thing left to address: register with whom?
It could be the Flickr and/or other sites that wish to participate in the Creative Commons endeavor implement their own registration system for content contributors and acquirors, but this leaves some very big holes as to the credibility of any given registrar. Sure, people can trust Flickr (can't they?) but you can't necessarily trust any site that claims to collect personal information for the purpose of letting you have a CC-ID. And besides, someone's got to arbitrate the IDs.
My proposal for dealing with this is for the CC to be the ultimate gathering agent and host for CC-ID applications. It doesn't necessarily have to be the exclusive agent that everyone goes to (which would disrupt a streamlined user experience on other sites), in which case it would do the following:
Any company that wishes to give users the ability to register themselves and their works with the CC would implement the specifications and apply for certification from the CC to become a registrar. Once granted, it can accept user registrations; once accepted and confirmed, the user would get a CC-ID, which would be attached to any CC-license that the user may wish to assign to his creative works. The CC would list this new website as a credentialed CC-client, who in turn displays the CC logo and/or other certification identification. Potential users can check the validity of the owner and the work as a safety check.
It goes without saying that the CC is a trusted entity, and all software and specifications would be open-source software, the internals of which would be available for inspection on any site that has CC certification. This assures all participants there isn't funny business going on by a certified registrar. (One more caveat for CC certification is that the site must also provide a user interface for visitors to register to be a CC creator, or link to the CC's own registration forms.)
If the CC were to act as a registration hub in the manner I just described, sites like Flickr would hardly need to do anything differently than what we see today. The only major difference is that the checkbox to attribute CC-licensing to a photo would not be available unless the user has a CC-ID. (And of course, the user would be offered the opportunity to apply for one.)
Now publishers have choices about the level of protection they need, and may choose to avoid any content whose creator has not yet been "verified." To protect themselves from frivolous claims of infringement, they would also register as users of images, and for each image they use, they would be issues an ID with a timestamp, like any other receipt.
And, by the way, this process applies to anything that can be licensed under the CC, not just photography. For example, YouTube and other sites that host user-contributed copyrighted content could also employ such mechanisms.
Caveats
Clearly, there is no way to physically attach a CC-ID with specific works, but this is no different than how CC is attached to works in the first place. Once a photo is copied from one place to another (or, in print, etc.), it has completely lost the fact that it was ever a CC-licensed image. Currently, the only way to identify whether a photo has a CC-license (and its CC-ID) is the hosting site(s) where the photographer posted them originally.
(I had considered including as part of the proposal that the Creative Commons also act as a hosting site for CC-licensable content to address this problem, but that would be too onerous for a non-profit to implement. Hosting only registrant info is enough, especially if it needs to hold the master key to the CC-IDs that are handed out to certified registrars.)
As I mentioned, this is merely a framework for initial discussion. It is not intended to be a foolproof vehicle for eliminating fraud; it is only intended to provide the mechanisms by which implementations can be made more or less "secure" over time. However, I do believe that even this minimal set of guidelines provides sufficient infrastructure for licensees and users to have more confidence in the system.
One could argue that it's still possible for the photographer or a publisher to create a forged entity with the intent of entrapping licensees (by the photographer), or getting away with free licensing (by the publisher), or even maliciousness pranksterism (by an internet troll), but the registration process is more formal and involves several steps, which slows down the process enough to minimizes the likelihood of fraud to far more manageable levels.
Long-term Effects
Continuing with my optimistic view of the world... I believe that if the CC had some built-in protections that registration can provide, it would not only make the system better trusted, but it could stimulate broader adoption. Why would anyone want to get into the business of offering free images? Massive amounts of content. And as everyone knows, content translates to traffic. And the more eyeballs that go to a site, the higher its "value" in terms of potential advertising and other revenue streams.
The pessimist might say that selling paid-for content with CC content degrades price perception. This is a fallacy. Price acceptance is a direct result of perception of value, and that perception is not due to the mere existence of lower-priced items, or even free ones. (I wrote a long article about that here.) It's the chaotic and unstructured way in which various price points are managed that keeps value-perceptions low. A CC model that has a better managed infrastructure for how it is implemented (providing it has such an infrastructure) keeps the wild part of "free" from negatively affecting its paid-for cousins.
There's also evidence to suggest that when "free" content is mixed with paid-for content, there is a higher chance for those who currently opt to give away content to think, "Gee, I could be paid instead of getting nothing for this photo." This would be more likely than someone thinking, "Gee, I'd rather get nothing for this picture than a check."
Here are even stronger arguments in favor of a controlled environment for "free" content.
As the photo industry matures into a form that embraces the consumer (a phenomenon that I've always predicted to be an inevitability anyway), it begins a new era where there's now a justifiable impetus for participation and investment from a much broader community. CC is still currently too chaotic and risky for many people to seriously consider, not because of its set of licenses, but in how Flick--its only real advocate--supports it. By crafting a firm set of guidelines for how a CC-partner behaves and certifying those who are so compliant, the chaos turns into order, which attracts more participants. I've written before that the size of the photo licensing industry is closer to the $20B range, as opposed to the current $2B that most industry analysts think it is, and it could be that the CC acts as the catalyst that brings it out: it would pacify the wild-west aspect of the photo industry, which has been what's kept the smart money away.
The pessimist says, "Yeah, I know."
When it comes to the Creative Commons and photography, it's no surprise that the reactions to my blog entries on the subject seems to come from extremists one either one side or the other. Their common misunderstanding is that I feel the Creative Commons itself is flawed. True, I've stated that because photography is a common, everyday thing that it is a poor fit for the CC. But this is the pessimistic viewpoint of a more neutral reality of my message: that it's the byproduct of how CC is assigned to photos, and how people generally acquire CC photos that creates most of its problems.
In this posting, I am going to present the "optimist" perspective of the exact same underlying premise, but with a twist: a proposal for a framework for how risk conditions for both sides of a CC-licensed work can be minimized. This is a framework, mind you--one that has its flaws like any other--but at least it presents an approach that plugs the most egregious holes in how CC is applied to photos, and how people pick up photos for use.
This simple scenario illustrates the problem at hand:
You're a judge. In front of you are two parties: a photographer and a company. The photographer claims that the company infringed upon his copyright by placing his photo on their web site without his permission. The company states that the photo was marked as CC-attribution on Flickr, and they complied with its terms precisely. The photographer counters that he never made the image available for CC-licensing.
As the judge, you have to determine three potential explanations for how the photo got on Flickr and was assigned the CC license:
- Did the photographer lie so as to collect the statutory damages?
- Did the company lie so as to get free use of the image?
- Or are both parties telling the truth, and someone else did it?
In copyrights infringement claims, the legal process starts with the indisputable facts: (1) the photo is owned by the photographer, and (2) the defendant has used it. Normally, it's at this point that the onus is on the defendant to prove that they properly acquired the photo legally from the photographer (or his assigns, like a stock photo agency). Unless they can, they are liable for infringement. Though they can claim it had a CC-license, that part is irrelevant unless and until it is established that the photographer assigned that license to it. One of the problems with the current use of the CC is that, though the photographer may voluntarily assign the CC license to it, it's impossible to differentiate that from a fraudulent assignment, which is why all of this is being discussed. But unless and until such a case can be shown or volunteered, it remains the defense's responsibility to present proof of permission to use the photo.
In the above scenario, most judges would have a hard time finding a legal reason to rule in favor of the defense without such evidence or admission by the photographer, who may in fact be lying. Yet, even if a sympathetic judge were to see the possible ambiguity, there's the possibility that the company may be lying. And the worst part is that both sides could be telling the truth and end up paying the price together (because of the high costs of litigating) if it turned out that some third party caused all this.
While such conditions are always possible for any kind of copyrighted work, the CC worsens the situation because it provides all sides with more tools and opportunity to create these conditions than before the CC.
But again, the CC itself isn't the problem! It's that it is has no reliable infrastructure. Flickr allows anonymous users to attribute a CC license to any object at all, and that's the problem. And not just for CC and those on either side of the photo in the middle, but Flickr itself, as would be any other entity that also tried to enable CC-licensing.
Infringement claims are already going on, but if the number of such claims continues to rise, it isn't a stretch to find someone filing a claim against Flick for contributory infringement. For those not familiar, contributory infringement requires (1) knowledge of the infringing activity and (2) a material contribution (actual assistance or inducement) to the alleged piracy.
Note that the Digital Millenium Copyright Act (DMCA), which normally protects companies like Flickr (Yahoo) from direct copyright infringement because they merely host the photo on their site, it does not protect them from contributory infringement if the second part of its definition applies: that Flickr provides "actual assistance or inducement". My feeling is that Flickr's current presentation of offering users to apply CC licenses, while not fully disclosing its ramifications to either photographers or users of such photos, is both assistance and inducement, and thus, puts them at risk for contributory infringement.
Crafting a Solution
Back to optimism. Alleviating this problem is relatively simple in nature; it's the details of implementation that can get sticky. The solution is to have some kind of mechanism for identification. In the simplest case, Flickr could just act more responsibly by not allowing users to assign CC licenses to photos without at least one more step in identifying who the person is that's doing it. This very simple act can help to alleviate a chain reaction of bad things, thereby reducing some ambiguities around the photo. It also reduces some of the inadvertent CC-assignments, while also potentially giving some publishers a tad more confidence in where the image came from. Of course, none of this is foolproof, but this one step still helps quite a bit.
Though it would be a good stop-gap measure for Flickr to do this, it's not really a long-term solution, especially if more sites were to ever adopt CC-licensed photo hosting as Flickr does. That's where it becomes necessary to craft a more formalized solution. The trickiness comes in the balancing of the "effectiveness" dividend against the "implementation" costs.
In theory, the next step is to introduce a documented process of registering and authenticating users on both sides of the "transaction." While that sounds bureaucratic and scary, this is not hard, nor new to anyone that uses the internet. The question is really the degree in which the CC or its users wish to participate.
Why consider it? Because it gives publishers a little more assurance that the photo may belong to a particular party, just as it gives photographers access to information about who's using their images. And that's been the main problem that's caused all of this. Again, this isn't CC's fault, but it would certainly help if the CC itself provided tools for its affiliates to act more responsibly.
First, no one would ever be required to register themselves with the CC. It is used only as a vehicle for information that both content owners and licensees can use to give themselves an added level of assurance. A photo's user can have a step more confidence that the image is owned by a trackable person; and the photographer can potentially be notified when his image is used. When these come together, each party is more likely to act in compliance with the terms of any given CC license.
Note that when people register, they are issued a Creative Commons Identification Number, or CC-ID, which can be used wherever they post CC-licensed content. Users can always remain anonymous to the outside world, and even to the CC (except for an email address, which is as good as anonymous anyway), in which case, potential users of that person's content knows this ahead of time, which may give them reservations about using it due to risk.
Here are the basic levels of identity that may be associated with any given person:
- Name and Email Verified
Here, a registrant fills out the application and provides no more information than a name and an email address. This is the base case for each of the other registration forms. - Phone/Address Confirmation
This level is a higher form of confirmation, which includes name, email phone and address. It is identical to the requirements of copyright registration with the US government. - Signature Verified
This level is the highest form of confirmation because the registrant provides all the information noted above, but prints it out and physically sends it in for confirmation. (Fax submission is permitted for states whose laws recognized signatures on faxes.)
Note that only people and companies register, not the works they submit. And it is only done once, after which an ID is issued, which is used to assign to CC-license to creative works.
Not every level of verification has to be implemented. (It may be too cost prohibitive in the early stages to deal with postal-mail submissions.)
The registration process has one of three states: "application submitted", "pending", and one of the above "confirmations."
So, is this solution merely a matter of having a registration process? Not quite--there is one more thing left to address: register with whom?
It could be the Flickr and/or other sites that wish to participate in the Creative Commons endeavor implement their own registration system for content contributors and acquirors, but this leaves some very big holes as to the credibility of any given registrar. Sure, people can trust Flickr (can't they?) but you can't necessarily trust any site that claims to collect personal information for the purpose of letting you have a CC-ID. And besides, someone's got to arbitrate the IDs.
My proposal for dealing with this is for the CC to be the ultimate gathering agent and host for CC-ID applications. It doesn't necessarily have to be the exclusive agent that everyone goes to (which would disrupt a streamlined user experience on other sites), in which case it would do the following:
- Provide a set of specifications and requirements for what information is to be collected for user registration, how it needs to store and maintain the information, and a set of protocols for passing such information between the CC and its affiliates.
- Provide executable APIs in a variety of programming languages for reference site to implement the specifications.
- Offer "registrar certifications" in which third party sites (like Flickr) would apply to use the API and to comply with the terms and conditions.
- Provide a reference site in which users can register themselves if they don't want to use a third party site.
Any company that wishes to give users the ability to register themselves and their works with the CC would implement the specifications and apply for certification from the CC to become a registrar. Once granted, it can accept user registrations; once accepted and confirmed, the user would get a CC-ID, which would be attached to any CC-license that the user may wish to assign to his creative works. The CC would list this new website as a credentialed CC-client, who in turn displays the CC logo and/or other certification identification. Potential users can check the validity of the owner and the work as a safety check.
It goes without saying that the CC is a trusted entity, and all software and specifications would be open-source software, the internals of which would be available for inspection on any site that has CC certification. This assures all participants there isn't funny business going on by a certified registrar. (One more caveat for CC certification is that the site must also provide a user interface for visitors to register to be a CC creator, or link to the CC's own registration forms.)
If the CC were to act as a registration hub in the manner I just described, sites like Flickr would hardly need to do anything differently than what we see today. The only major difference is that the checkbox to attribute CC-licensing to a photo would not be available unless the user has a CC-ID. (And of course, the user would be offered the opportunity to apply for one.)
Now publishers have choices about the level of protection they need, and may choose to avoid any content whose creator has not yet been "verified." To protect themselves from frivolous claims of infringement, they would also register as users of images, and for each image they use, they would be issues an ID with a timestamp, like any other receipt.
And, by the way, this process applies to anything that can be licensed under the CC, not just photography. For example, YouTube and other sites that host user-contributed copyrighted content could also employ such mechanisms.
Caveats
Clearly, there is no way to physically attach a CC-ID with specific works, but this is no different than how CC is attached to works in the first place. Once a photo is copied from one place to another (or, in print, etc.), it has completely lost the fact that it was ever a CC-licensed image. Currently, the only way to identify whether a photo has a CC-license (and its CC-ID) is the hosting site(s) where the photographer posted them originally.
(I had considered including as part of the proposal that the Creative Commons also act as a hosting site for CC-licensable content to address this problem, but that would be too onerous for a non-profit to implement. Hosting only registrant info is enough, especially if it needs to hold the master key to the CC-IDs that are handed out to certified registrars.)
As I mentioned, this is merely a framework for initial discussion. It is not intended to be a foolproof vehicle for eliminating fraud; it is only intended to provide the mechanisms by which implementations can be made more or less "secure" over time. However, I do believe that even this minimal set of guidelines provides sufficient infrastructure for licensees and users to have more confidence in the system.
One could argue that it's still possible for the photographer or a publisher to create a forged entity with the intent of entrapping licensees (by the photographer), or getting away with free licensing (by the publisher), or even maliciousness pranksterism (by an internet troll), but the registration process is more formal and involves several steps, which slows down the process enough to minimizes the likelihood of fraud to far more manageable levels.
Long-term Effects
Continuing with my optimistic view of the world... I believe that if the CC had some built-in protections that registration can provide, it would not only make the system better trusted, but it could stimulate broader adoption. Why would anyone want to get into the business of offering free images? Massive amounts of content. And as everyone knows, content translates to traffic. And the more eyeballs that go to a site, the higher its "value" in terms of potential advertising and other revenue streams.
The pessimist might say that selling paid-for content with CC content degrades price perception. This is a fallacy. Price acceptance is a direct result of perception of value, and that perception is not due to the mere existence of lower-priced items, or even free ones. (I wrote a long article about that here.) It's the chaotic and unstructured way in which various price points are managed that keeps value-perceptions low. A CC model that has a better managed infrastructure for how it is implemented (providing it has such an infrastructure) keeps the wild part of "free" from negatively affecting its paid-for cousins.
There's also evidence to suggest that when "free" content is mixed with paid-for content, there is a higher chance for those who currently opt to give away content to think, "Gee, I could be paid instead of getting nothing for this photo." This would be more likely than someone thinking, "Gee, I'd rather get nothing for this picture than a check."
Here are even stronger arguments in favor of a controlled environment for "free" content.
- The increased awareness of copyright issues. Because "free" is an inevitable part of every industry that deals with creative works, the CC could offer an enforceable infrastructure that has teeth, thereby making the previously simple and common act of "lifting" subject to more serious financial ramifications.
- By institutionalizing free with a "mechanism," people will get used to the idea that there even exists a mechanism in the first place. This, in itself, is the first step towards getting many entities (content creators as well as users) to start to think about licensing.
- If "free" content is made available with the caveat that it must be used in accordance with certain terms, or there would be swift and immediate financial consequences, it reinforces the notion that all content is protected by copyright, regardless of the price paid.
As the photo industry matures into a form that embraces the consumer (a phenomenon that I've always predicted to be an inevitability anyway), it begins a new era where there's now a justifiable impetus for participation and investment from a much broader community. CC is still currently too chaotic and risky for many people to seriously consider, not because of its set of licenses, but in how Flick--its only real advocate--supports it. By crafting a firm set of guidelines for how a CC-partner behaves and certifying those who are so compliant, the chaos turns into order, which attracts more participants. I've written before that the size of the photo licensing industry is closer to the $20B range, as opposed to the current $2B that most industry analysts think it is, and it could be that the CC acts as the catalyst that brings it out: it would pacify the wild-west aspect of the photo industry, which has been what's kept the smart money away.
Labels: analysis, copyright, creative commons, dan heller, flickr, infringement, licensing, photo business, photography business, pricing, registration
posted by argv | 10:41 PM
<< Home