Making money from your stolen images
Ever had a photo stolen from your website? Did it make you angry? Did you feel violated? Did you feel powerless? Are you apprehensive about putting your photos on photo-sharing or stock-photo sites because you fear the problem would only be worse? I know it'll sound like a shocker, but you may be sitting on a small pot of gold. To explain why, you need some background.
Think about this scenario:
Most people would say the $1 would be stolen. Why? According to studies repeated in various contexts, a similar theme underlies most acts of theft: the decision to steal is driven primarily on the perceived "consequences." An article in the Journal of Applied Social Psychology showed that the lower the perception of consequences, the likelihood of stealing is pretty equal among genders and racial backgrounds. Innocently taking a small candy at the checkout counter is something most everyone's done at one point in their lives simply because the consequences of the act are very low. It is only when the perception of the consequences go up that the true risk-takers (poor single men under 25) are more likely to steal.
Consequences
What does this have to do with photography? In my hypothetical question about the web designer, most people would assume the $1 image would be stolen. Whether you're aware of it or not, your "intuition" guided you to that assumption because the perception of consequences is lower because the value of the image is lower. Which is also the same rationale as the thief. Odd, though it may sound, this is totally wrong--people attribute stealing images to stealing other things, or for other kinds of crimes or mistakes. Consider one of these two scenarios:
The common theme here is "compensation based on the value of the item." In law, these are called "compensatory damages" because you, the evil-doer, "compensates" the victim for whatever you did--broke a window, or stole a candy bar. Hence, stealing a $1 image is perceived to have lower consequences than the $349 image because people don't know of other kinds of violations.
Thing is, stealing an image is not really against the law, so much as using the image is. It's like saying that falling off of a tall building doesn't kill you--it's the sudden stop that does. So, stealing the image isn't what gets you into trouble, it's putting it into use that does. In this case, "copyright infringement." And here's the part most people don't know: the law on infringements does not specify compensatory damages, but statutory damages, which is a very big difference indeed.
To illustrate how big, there was an article PDN reported on November 20, 2006: Corbis Settles With TemplateMonster; Wins $20 Million Judgment From Mystery Companies . In short, TemplateMonster stole a bunch of images from Corbis and used them as background images for "templates" that were sold to web designers who would use them to create web pages. The reference to the "Mystery Companies" in the article title is that other people and companies were involved that were either difficult to track down, or impossible. (Turns out that the companies were empty shell organizations.) Still, Corbis had a nice windfall from those they did identify.
When the article came out, photo forums were abuzz with scorn for the notion that photos were stolen, and fear began to circulate that their images would be stolen too. Many even proclaimed they were going to remove images from social photo sites and microstock sites because they can be easily stolen.
My immediate response was "Huh? Didn't you read those headlines? $20M! that's an 'M'! You're kidding me, right? Remove your images from stock sites? Hey, sign me up! Steal my images! Please!"
And that's when I started putting the pieces together and realized something may be going on in this industry that I hadn't considered before, which started me on a long-term hunt for stories, tidbits, interviews, and of course, legal research into the dark world of copyright infringement. Or, should I say, very lucrative world... for the copyright holders.
Statutory Windfall
In "statutory damages," the fine is based on an entirely different formula, irrespective (usually) of the actual "cost" of the item (or, in this case, the license fee). In fact, in the case cited above, the judge awarded Corbis the maximum amount allowed by law: $30,000 per image, plus $2,500 for each act. What's more, the court may (and usually does, according to statistics of such violations in court records) also award attorney's fees. In other words, you not only get the money, but the infringer pays your legal bills.
How much would the images have cost the company if they licensed them legitimately? It's hard to say given the company and the number of photos used. But, suffice to say, one could license the same kinds of images on microstock sites for $1 to $10 each.
And that's where you start hearing the click of light switches going off in some people's heads: if you knew you could gets tens of thousands of dollars if a kid stole a $1 candy bar from your front counter, there may be a lot of money if you put a lot of candy bars out for hungry children to take, especially if you make it very easy for those kids to take them, and especially if those kids have no idea of what the penalty is, and even more especially if we're talking about adults working at businesses that all bear super-size responsibilities.
So what does that mean for photographers and for photo agencies?
Opportunities for Everyday Photographers
I recently got this email, which is similar to a constant and frequent stream I get on an ongoing basis:
On Jun 25, 4:24pm, jrus424@msn.com wrote:
I get so much of this, that it's staggering for reasons that will become clear soon.
First, let me point you to an invaluable resource: The website for the US Copyright Office. This link is a well-written, easily understood list of the most common topics about copyrights that can be understood by most anyone, including photographers and even those who work for our own Attorney General.
Statutory awards can range from $750 to $30,000 per infringement, or even be as high as $150,000 if the infringement was willful or intentional. (Sound effect of Vegas-style slot machine bells going off.) To determine your situation, consider these two steps:
If you have done one or the other of the above, you should find a lawyer specializing in Intellectual Property and start letter-writing. Don't contact the hotels first--you don't want to say anything that may make negotiations harder for you later. Make a screenshot of the website that's using your photos for archival and legal reasons, then shop for lawyers. (If it's printed matter, just make copies.)
If you haven't yet registered your photos, or put watermarks on them, do it now. While it won't have an affect on this case, do it for future cases. And always do this for new photos you put online. Soon, the copyright office's new online system for filing your copyright registration will be available, which will make the whole process easier. Details on their website at http://www.copyright.gov/eco/index.html.
Even without having registered your photos, you could have a lawyer file a claim, but the likelihood of a recovery is less certain, and will be considerably lower. According to the law, you still own the copyright, but unregistered images that have been infringed are only subject to "compensatory damages." That is, a $1 photo can recover $1. The hotel's lawyers are going to look at the two points noted above and realize you've got a harder case to prove. Still, if your lawyer is good, you might eek out several hundred, especially because most larger companies don't want stuff like this hanging over them. Your mileage may vary, but it does mean that your lawyer is going to have to do the cost-benefit analysis of time vs. money recovered (since he's the one doing all the work).
If you do not plan on filing a claim, you could try to contact the hotels and state that you "will" a file a claim unless they want to settle early. Here, you risk having your bluff called.
In fact, this is what I did in my early days, and I found that some paid, others didn't. Big companies were more likely to do so, and to pay better; little companies usually said, "Go ahead, sue us." In the end, the time and energy were draining, not to mention the constant, distasteful "conflict," which made the whole practice... well, not to my liking. Perhaps it was because I was doing this without a lawyer, but also because I was making far more money dealing with honest licensees, that I didn't really want to pursue the thieves. And let's be honest: you grow your business by working with the paying clients because they turn into better paying clients. Catching thieves may be morally rewarding, but it doesn't "grow" your business. Of course, it's a great side-business for an existing business, if your sources are such that you can have these lawyers working for you on the sidelines.
Recently, due to finding just one-too-many infringements of my own images, and inspired by Corbis' windfall, I finally got around to using a lawyer to do this work for me. And, because I have registered my images with the copyright office, and because I have my copyright notice on all my photos, I have a "slam dunk" shot at every single case I bring forward. (It's more slam dunk than the last time that phrase was used back in 2003, but we won't get into that now.)
Although I'd implied there are pretty substantial awards to be had for infringements, "common" violations by smaller companies and less serious uses probably yield much smaller payoffs that may range from $2000 to $5000. Given my own back-of-the-napkin calculations of the bona fide infringements of my images that I currently have documented, I could add another $70-100K to my bottom line.
In determining my bigger opportunities, I used Yahoo's site-search feature that lists about 90,000 sites that link to mine from around the web, 5000 of which use my images and link back to my site. Some of them are personal web pages and such that I wouldn't pursue, but imagine the number of image uses that do not link back to me and that use my images. Simple math would suggest that if I got 1000 violations at $2000 each, that's $2M. Not a bad kitty for a single photographer. (Suddenly, my company's valuation is higher.)
The Hurdles
So, if it were that easy, why doesn't everyone do it? One stumbling block is that it's hard to find the violators. Sure, I've done brute force Google searches and found many violations that could be worthwhile, but it's very time-consuming. The real business opportunity is if you can automate the process so you don't spend your time doing it.
In fact, there are some companies that currently do just this: http://www.ideeinc.com, and http://www.picscout.com both have image recognition software (that they have each respectively patented) that can find identical matches for photos, even if they've been rotated, shifted, colored, cropped, or mangled in many ways. The process is simple: they first analyze all your photos by doing an analysis of the pixels and coming up with a kind of "fingerprint" ID. Then they crawl the web, or look at printed material (that they scan) and generate reports for where your images are found.
Sounds great, right? Here's where the next hurdle lies: it's really expensive and time-consuming to crawl the web. So much so, that the resources necessary to do so are so prohibitive that the fees these companies charge eat away at your bottom line. This, almost to the point where, again, it's still more cost-effective to let the thieves go and just continue working with the honest people. (Actually, there's a cross-over point where the number and nature of the violations is high enough that it becomes worthwhile.)
With their limited resources, it takes months and months before actual matches are found, largely because the searching algorithms are rather dumb--they just search randomly. As a human, I can make more intelligent searching decisions, like going to images.google.com and searching on keywords that I know will match my specific images. And true to form, I can usually find several infringements if I put the time into it. But again, there's that time thing again. Too bad those companies don't use more intelligent search methods to come up with more likely matches.
Why don't they? To do so, you have to have already crawled and index the web before searching with keywords makes sense. This begs the question, why not just crawl google's already-indexed search results? Primarily because they are not allowed to. (Google once had an API and an agreement mechanism available to do this, but they've since terminated the program.) So, these visual-search companies crawl the web like "ships passing in the night." A very foggy night with no compass or GPS.
So, although the small image-recognition companies have really great technologies to find the matches, that's not the barrier to entry. It's the mere indexing of the web. Once you do that, the image pairing is simple. So much so, they could probably find all copyright infringements by all photographers in a matter of a few days if they were only allowed to churn through Google's warehouse of data.
Hold onto that thought for a moment.
Opportunities for Agencies
What about agencies and/or photo-sharing sites? Flickr hosts hundreds of millions photos--let's round it off to 1 Billion images. If .1% (one tenth of one percent) of these photos had infringements, that's 1 million violations. Even a low estimate of $500 per violation yields $500M. That's a half a billion dollars. And think of what that would do to sign-ups at Flickr, not to mention Yahoo!'s stock price. People would be Flocking to Flickr! (I've been dying to use that expression.)
Best of all, that barrier that plagues me and everyone else doesn't apply to Flickr because they are owned by Yahoo!, who's indexed the entire web. They could very easily scour the web at the blink of an eye.
Suddenly, the business prospect of Flickr is more interesting. I've said before that Flickr should enter into the licensing business directly, and while I still believe that to be a rosy picture, it's even rosier if you add their natural ability to police infringements. Best of all, NO ONE ELSE COULD COMPETE WITH THAT! (Sorry, I hate using all-caps, but in this case, it's warranted.) They would literally corner the market in stock photos by simple virtue of their back-end infrastructure and existing photo assets.
Nevertheless, would Flickr ever do it? Sigh. Not anytime soon. The New York Times has had daily articles about how Yahoo! is imploding faster and faster: they're losing their vision, having serious upper management problems, and all business units suffering for money, resources, and personnel. Rumor has it that are executives waiting for stock and options to vest (so they can leave). Each business unit within Yahoo is choking for air, and Flickr is not anywhere near the top of the pecking order. So, no, I don't expect to see anything "new" from Flickr, let alone anything innovative, until the bigger company gets it house in order.
This then leads to Google.
Would they get into photo licensing? Not without acquiring a pretty big player, and that isn't quite likely, mostly because the photo industry isn't perceived to be important enough because even the biggest players are having problems making money. (This is a topic for another blog article, I promise.) Given Google's size, this article alone wouldn't even get past a receptionist's desk.
On the other hand, Google does have interest in the social side of photography through its Picasa products, so while we might not see much in the way of photo licensing, it's not totally inconceivable they could be interested in the photo matching game. Why not: it would be a natural fit for their general offerings of search tools, too.
Yet another potential benefit might be its own public relations boost. Google has been the target of many copyright violation claims, ranging from very public ones, such as the YouTube case involving Viacom, all the way to publishers and universities whining about Google's project of scanning books and documents found in libraries to help make the world more comprehensively indexed. Though none of these claims are sticking, simply introducing a mechanism that helps people find "exact matches" of photos on the net could be precisely what the Public Relations doctor ordered.
But, again, there's been nothing even hinted at about Google blinking in this direction, so for the moment, the aforementioned image-recognition companies (Idee and Picscout) have a long time to try to build their businesses without being usurped by major players who could take them out with a flick of a switch. (Oh, and add cognisign.com to the mix: they have great visual-recognition technology as well, but have already expressed no interest at all in pursuing a business in tracking copyright violations.)
Returning to the viability of image-recognition companies, one might argue that image matching is not that hard--or rather, not hard enough to command a secure role as the "leader" of the industry, because the real barrier is still the searching/indexing of the web. Whoever figures that out will be the clear winner.
The other side of the coin
For the sake of argument, let's say Google or Yahoo unleashed a simple mechanism in their image search that says, "find all instances of this image on the net." Click it, and you instantly see every page that uses that image, or portions of it. If you're a photographer, you've just found all (electronic) infringements. What would such a world look like for photographers and agencies, and how would it affect the stock photo industry?
The cynical view would be that the major agencies are really playing a baiting game, and this would pull the rug right out from under them. If they are selling cheap candy bars to kids who they know will steal them, the game will be up. Well, not immediately, but quickly thereafter. The availability of instant search results for infringements would generate a huge influx of money for copyright holders, but quickly evaporate as it became so ubiquitously known what was going on. It'd be like having a police cruiser sitting at every stop sign in your city. There would be a huge inflow of revenue from traffic violations at the start, but within a day or two, there would never be another violation again--people would just know the cops are sitting there.
The less cynical view is one that reflects more of what we're seeing in the licensing world now. For example, Getty's images are so easy to "copy" because the licensing is based on the honor system, but it may be for reasons other than to "entrap" users into infringements. I have personally spoken with many of their clients who tell me they use Getty images all the time for short little pieces without paying for them because they feel justified. This was how the candy bar analogy came up: a client said to me, "Getty will look the other way if I steal a candy bar now and then so long as I do my major shopping at their grocery store. If you did the same thing, you'd get more big buyers like us, too." I wouldn't do it, but not because I didn't believe in the strategy, it's because such a strategy only works for a large company like Getty. It doesn't work on small scale operators.
If the "search switch" were turned on, what would that do to Getty's business model of "give a little away so you can collect the big clients." And therein lies the real question: do you pursue your violators for large sums of money? What if they are also spending large sums in groceries? If they aren't, can you convert them? Do you take the short-term win (of pursuing infringements) at the expense of potentially alienating your clients and long-term growth? And what about the photographers the agencies represent? Whether Getty or a microstock, if they don't pursue infringements, then they can be a contributory infringer. (I alluded to this in an earlier blog about Flickr's liabilities in this area.)
Making it even more complicated is the inherent value of having your images stolen in the first place. In fact, Viacom's claim against Google's YouTube has generated a stir by television industry analysts, as well as within the walls of Viacom itself: the fact that TV shows are available online has actually helped viewership, and thus, increased advertising rates for those shows. Speaking of my own experience, the fact that so many sites use my images and link to me is one of the main core reasons why my site ranks so highly, which is what brings me the traffic that converts to buyers. So, a little copyright violation can actually do your business good.
Converting a criminal into an honest buyer is warm and fuzzy, but what about the tactic that made them a criminal in the first place? It'd be interesting if a defendant actually tried to use the legal excuse of the attractive nuisance doctrine. "They don't prevent me from downloading it without paying for it! It's too easy! I didn't know! The rules are cryptic." That's hard to say, and perhaps somewhat of a stretch, but if that magic switch were ever pulled by a search company, and thousands upon thousands of copyright infringement cases suddenly came up, the courts would have a lot on their hands.
Suddenly, the business aspect of copyright infringements isn't so cut and dry.
On the darker side again, there's this:
in an article by Bug Shell, Getty purchased the negatives to the National Archives, and is now on a litigation spree against those who use those images, regardless of the fact that owning the film does not grant ownership of copyright, and the fact that the photos are in the public domain. Getty claims that their "scanning" of the negatives (dusting and other corrections) creates a "derivative work," but many legal observers are taking great exception to this claim.
And the industry pendulum continues to swing back and forth.
Summary
The net-net is there's big money in copyright violations, especially if photos are registered with the copyright office and they are watermarked. What keeps this from becoming a more prominent figure in the industry itself is two-fold: the lack of awareness of how lucrative it can be, and the "resources" available for tracking the infringements in the first place. But, like anything that has the potential for money, there will be those who start down those paths to see where they lead. For those playing the home game, keep in mind that the issues faced by agencies are not shared by individual photographers. That is, while an agency may have business risk by being too aggressive in pursuing violators, individuals certainly do not. Even as I type this, yet another email has just trickled in from another concerned photographer, "I'm going to take my images off Flickr because I don't want them stolen." My response to him will be to read this article.
Think about this scenario:
A web designer is making a web page for the corner hair salon, and there are two images to choose from: one costs $1 and the other cost $349. If the web designer was going to steal one of the images, which image do you think that person would choose?
Most people would say the $1 would be stolen. Why? According to studies repeated in various contexts, a similar theme underlies most acts of theft: the decision to steal is driven primarily on the perceived "consequences." An article in the Journal of Applied Social Psychology showed that the lower the perception of consequences, the likelihood of stealing is pretty equal among genders and racial backgrounds. Innocently taking a small candy at the checkout counter is something most everyone's done at one point in their lives simply because the consequences of the act are very low. It is only when the perception of the consequences go up that the true risk-takers (poor single men under 25) are more likely to steal.
Consequences
What does this have to do with photography? In my hypothetical question about the web designer, most people would assume the $1 image would be stolen. Whether you're aware of it or not, your "intuition" guided you to that assumption because the perception of consequences is lower because the value of the image is lower. Which is also the same rationale as the thief. Odd, though it may sound, this is totally wrong--people attribute stealing images to stealing other things, or for other kinds of crimes or mistakes. Consider one of these two scenarios:
- If you break someone's window with an errant golf club that mysteriously flew out of your hands only seconds after the ball itself landed in the sand pit, then the owner of the window is entitled to "compensation" to cover the cost of fixing it.
- If you get caught stealing a candy bar from a store, you have to pay for it.
The common theme here is "compensation based on the value of the item." In law, these are called "compensatory damages" because you, the evil-doer, "compensates" the victim for whatever you did--broke a window, or stole a candy bar. Hence, stealing a $1 image is perceived to have lower consequences than the $349 image because people don't know of other kinds of violations.
Thing is, stealing an image is not really against the law, so much as using the image is. It's like saying that falling off of a tall building doesn't kill you--it's the sudden stop that does. So, stealing the image isn't what gets you into trouble, it's putting it into use that does. In this case, "copyright infringement." And here's the part most people don't know: the law on infringements does not specify compensatory damages, but statutory damages, which is a very big difference indeed.
To illustrate how big, there was an article PDN reported on November 20, 2006: Corbis Settles With TemplateMonster; Wins $20 Million Judgment From Mystery Companies . In short, TemplateMonster stole a bunch of images from Corbis and used them as background images for "templates" that were sold to web designers who would use them to create web pages. The reference to the "Mystery Companies" in the article title is that other people and companies were involved that were either difficult to track down, or impossible. (Turns out that the companies were empty shell organizations.) Still, Corbis had a nice windfall from those they did identify.
When the article came out, photo forums were abuzz with scorn for the notion that photos were stolen, and fear began to circulate that their images would be stolen too. Many even proclaimed they were going to remove images from social photo sites and microstock sites because they can be easily stolen.
My immediate response was "Huh? Didn't you read those headlines? $20M! that's an 'M'! You're kidding me, right? Remove your images from stock sites? Hey, sign me up! Steal my images! Please!"
And that's when I started putting the pieces together and realized something may be going on in this industry that I hadn't considered before, which started me on a long-term hunt for stories, tidbits, interviews, and of course, legal research into the dark world of copyright infringement. Or, should I say, very lucrative world... for the copyright holders.
Statutory Windfall
In "statutory damages," the fine is based on an entirely different formula, irrespective (usually) of the actual "cost" of the item (or, in this case, the license fee). In fact, in the case cited above, the judge awarded Corbis the maximum amount allowed by law: $30,000 per image, plus $2,500 for each act. What's more, the court may (and usually does, according to statistics of such violations in court records) also award attorney's fees. In other words, you not only get the money, but the infringer pays your legal bills.
How much would the images have cost the company if they licensed them legitimately? It's hard to say given the company and the number of photos used. But, suffice to say, one could license the same kinds of images on microstock sites for $1 to $10 each.
And that's where you start hearing the click of light switches going off in some people's heads: if you knew you could gets tens of thousands of dollars if a kid stole a $1 candy bar from your front counter, there may be a lot of money if you put a lot of candy bars out for hungry children to take, especially if you make it very easy for those kids to take them, and especially if those kids have no idea of what the penalty is, and even more especially if we're talking about adults working at businesses that all bear super-size responsibilities.
So what does that mean for photographers and for photo agencies?
Opportunities for Everyday Photographers
I recently got this email, which is similar to a constant and frequent stream I get on an ongoing basis:
On Jun 25, 4:24pm, jrus424@msn.com wrote:
Two of my photos were lifted from Flickr and are currently being used on a hotels' website (w/o authorization). Should I take any action? My photos on Flickr are all rights reserved but others have told me to just forget it.
I get so much of this, that it's staggering for reasons that will become clear soon.
First, let me point you to an invaluable resource: The website for the US Copyright Office. This link is a well-written, easily understood list of the most common topics about copyrights that can be understood by most anyone, including photographers and even those who work for our own Attorney General.
Statutory awards can range from $750 to $30,000 per infringement, or even be as high as $150,000 if the infringement was willful or intentional. (Sound effect of Vegas-style slot machine bells going off.) To determine your situation, consider these two steps:
- If you have registered your photos with the copyright office, you can sue for statutory damages, and even recover your legal expenses.
- If your photos have had a visible watermark on them with your name and/or copyright notice, then the act is considered willful and intentional.
If you have done one or the other of the above, you should find a lawyer specializing in Intellectual Property and start letter-writing. Don't contact the hotels first--you don't want to say anything that may make negotiations harder for you later. Make a screenshot of the website that's using your photos for archival and legal reasons, then shop for lawyers. (If it's printed matter, just make copies.)
If you haven't yet registered your photos, or put watermarks on them, do it now. While it won't have an affect on this case, do it for future cases. And always do this for new photos you put online. Soon, the copyright office's new online system for filing your copyright registration will be available, which will make the whole process easier. Details on their website at http://www.copyright.gov/eco/index.html.
Even without having registered your photos, you could have a lawyer file a claim, but the likelihood of a recovery is less certain, and will be considerably lower. According to the law, you still own the copyright, but unregistered images that have been infringed are only subject to "compensatory damages." That is, a $1 photo can recover $1. The hotel's lawyers are going to look at the two points noted above and realize you've got a harder case to prove. Still, if your lawyer is good, you might eek out several hundred, especially because most larger companies don't want stuff like this hanging over them. Your mileage may vary, but it does mean that your lawyer is going to have to do the cost-benefit analysis of time vs. money recovered (since he's the one doing all the work).
If you do not plan on filing a claim, you could try to contact the hotels and state that you "will" a file a claim unless they want to settle early. Here, you risk having your bluff called.
In fact, this is what I did in my early days, and I found that some paid, others didn't. Big companies were more likely to do so, and to pay better; little companies usually said, "Go ahead, sue us." In the end, the time and energy were draining, not to mention the constant, distasteful "conflict," which made the whole practice... well, not to my liking. Perhaps it was because I was doing this without a lawyer, but also because I was making far more money dealing with honest licensees, that I didn't really want to pursue the thieves. And let's be honest: you grow your business by working with the paying clients because they turn into better paying clients. Catching thieves may be morally rewarding, but it doesn't "grow" your business. Of course, it's a great side-business for an existing business, if your sources are such that you can have these lawyers working for you on the sidelines.
Recently, due to finding just one-too-many infringements of my own images, and inspired by Corbis' windfall, I finally got around to using a lawyer to do this work for me. And, because I have registered my images with the copyright office, and because I have my copyright notice on all my photos, I have a "slam dunk" shot at every single case I bring forward. (It's more slam dunk than the last time that phrase was used back in 2003, but we won't get into that now.)
Although I'd implied there are pretty substantial awards to be had for infringements, "common" violations by smaller companies and less serious uses probably yield much smaller payoffs that may range from $2000 to $5000. Given my own back-of-the-napkin calculations of the bona fide infringements of my images that I currently have documented, I could add another $70-100K to my bottom line.
In determining my bigger opportunities, I used Yahoo's site-search feature that lists about 90,000 sites that link to mine from around the web, 5000 of which use my images and link back to my site. Some of them are personal web pages and such that I wouldn't pursue, but imagine the number of image uses that do not link back to me and that use my images. Simple math would suggest that if I got 1000 violations at $2000 each, that's $2M. Not a bad kitty for a single photographer. (Suddenly, my company's valuation is higher.)
The Hurdles
So, if it were that easy, why doesn't everyone do it? One stumbling block is that it's hard to find the violators. Sure, I've done brute force Google searches and found many violations that could be worthwhile, but it's very time-consuming. The real business opportunity is if you can automate the process so you don't spend your time doing it.
In fact, there are some companies that currently do just this: http://www.ideeinc.com, and http://www.picscout.com both have image recognition software (that they have each respectively patented) that can find identical matches for photos, even if they've been rotated, shifted, colored, cropped, or mangled in many ways. The process is simple: they first analyze all your photos by doing an analysis of the pixels and coming up with a kind of "fingerprint" ID. Then they crawl the web, or look at printed material (that they scan) and generate reports for where your images are found.
Sounds great, right? Here's where the next hurdle lies: it's really expensive and time-consuming to crawl the web. So much so, that the resources necessary to do so are so prohibitive that the fees these companies charge eat away at your bottom line. This, almost to the point where, again, it's still more cost-effective to let the thieves go and just continue working with the honest people. (Actually, there's a cross-over point where the number and nature of the violations is high enough that it becomes worthwhile.)
With their limited resources, it takes months and months before actual matches are found, largely because the searching algorithms are rather dumb--they just search randomly. As a human, I can make more intelligent searching decisions, like going to images.google.com and searching on keywords that I know will match my specific images. And true to form, I can usually find several infringements if I put the time into it. But again, there's that time thing again. Too bad those companies don't use more intelligent search methods to come up with more likely matches.
Why don't they? To do so, you have to have already crawled and index the web before searching with keywords makes sense. This begs the question, why not just crawl google's already-indexed search results? Primarily because they are not allowed to. (Google once had an API and an agreement mechanism available to do this, but they've since terminated the program.) So, these visual-search companies crawl the web like "ships passing in the night." A very foggy night with no compass or GPS.
So, although the small image-recognition companies have really great technologies to find the matches, that's not the barrier to entry. It's the mere indexing of the web. Once you do that, the image pairing is simple. So much so, they could probably find all copyright infringements by all photographers in a matter of a few days if they were only allowed to churn through Google's warehouse of data.
Hold onto that thought for a moment.
Opportunities for Agencies
What about agencies and/or photo-sharing sites? Flickr hosts hundreds of millions photos--let's round it off to 1 Billion images. If .1% (one tenth of one percent) of these photos had infringements, that's 1 million violations. Even a low estimate of $500 per violation yields $500M. That's a half a billion dollars. And think of what that would do to sign-ups at Flickr, not to mention Yahoo!'s stock price. People would be Flocking to Flickr! (I've been dying to use that expression.)
Best of all, that barrier that plagues me and everyone else doesn't apply to Flickr because they are owned by Yahoo!, who's indexed the entire web. They could very easily scour the web at the blink of an eye.
Suddenly, the business prospect of Flickr is more interesting. I've said before that Flickr should enter into the licensing business directly, and while I still believe that to be a rosy picture, it's even rosier if you add their natural ability to police infringements. Best of all, NO ONE ELSE COULD COMPETE WITH THAT! (Sorry, I hate using all-caps, but in this case, it's warranted.) They would literally corner the market in stock photos by simple virtue of their back-end infrastructure and existing photo assets.
Nevertheless, would Flickr ever do it? Sigh. Not anytime soon. The New York Times has had daily articles about how Yahoo! is imploding faster and faster: they're losing their vision, having serious upper management problems, and all business units suffering for money, resources, and personnel. Rumor has it that are executives waiting for stock and options to vest (so they can leave). Each business unit within Yahoo is choking for air, and Flickr is not anywhere near the top of the pecking order. So, no, I don't expect to see anything "new" from Flickr, let alone anything innovative, until the bigger company gets it house in order.
This then leads to Google.
Would they get into photo licensing? Not without acquiring a pretty big player, and that isn't quite likely, mostly because the photo industry isn't perceived to be important enough because even the biggest players are having problems making money. (This is a topic for another blog article, I promise.) Given Google's size, this article alone wouldn't even get past a receptionist's desk.
On the other hand, Google does have interest in the social side of photography through its Picasa products, so while we might not see much in the way of photo licensing, it's not totally inconceivable they could be interested in the photo matching game. Why not: it would be a natural fit for their general offerings of search tools, too.
Yet another potential benefit might be its own public relations boost. Google has been the target of many copyright violation claims, ranging from very public ones, such as the YouTube case involving Viacom, all the way to publishers and universities whining about Google's project of scanning books and documents found in libraries to help make the world more comprehensively indexed. Though none of these claims are sticking, simply introducing a mechanism that helps people find "exact matches" of photos on the net could be precisely what the Public Relations doctor ordered.
But, again, there's been nothing even hinted at about Google blinking in this direction, so for the moment, the aforementioned image-recognition companies (Idee and Picscout) have a long time to try to build their businesses without being usurped by major players who could take them out with a flick of a switch. (Oh, and add cognisign.com to the mix: they have great visual-recognition technology as well, but have already expressed no interest at all in pursuing a business in tracking copyright violations.)
Returning to the viability of image-recognition companies, one might argue that image matching is not that hard--or rather, not hard enough to command a secure role as the "leader" of the industry, because the real barrier is still the searching/indexing of the web. Whoever figures that out will be the clear winner.
The other side of the coin
For the sake of argument, let's say Google or Yahoo unleashed a simple mechanism in their image search that says, "find all instances of this image on the net." Click it, and you instantly see every page that uses that image, or portions of it. If you're a photographer, you've just found all (electronic) infringements. What would such a world look like for photographers and agencies, and how would it affect the stock photo industry?
The cynical view would be that the major agencies are really playing a baiting game, and this would pull the rug right out from under them. If they are selling cheap candy bars to kids who they know will steal them, the game will be up. Well, not immediately, but quickly thereafter. The availability of instant search results for infringements would generate a huge influx of money for copyright holders, but quickly evaporate as it became so ubiquitously known what was going on. It'd be like having a police cruiser sitting at every stop sign in your city. There would be a huge inflow of revenue from traffic violations at the start, but within a day or two, there would never be another violation again--people would just know the cops are sitting there.
The less cynical view is one that reflects more of what we're seeing in the licensing world now. For example, Getty's images are so easy to "copy" because the licensing is based on the honor system, but it may be for reasons other than to "entrap" users into infringements. I have personally spoken with many of their clients who tell me they use Getty images all the time for short little pieces without paying for them because they feel justified. This was how the candy bar analogy came up: a client said to me, "Getty will look the other way if I steal a candy bar now and then so long as I do my major shopping at their grocery store. If you did the same thing, you'd get more big buyers like us, too." I wouldn't do it, but not because I didn't believe in the strategy, it's because such a strategy only works for a large company like Getty. It doesn't work on small scale operators.
If the "search switch" were turned on, what would that do to Getty's business model of "give a little away so you can collect the big clients." And therein lies the real question: do you pursue your violators for large sums of money? What if they are also spending large sums in groceries? If they aren't, can you convert them? Do you take the short-term win (of pursuing infringements) at the expense of potentially alienating your clients and long-term growth? And what about the photographers the agencies represent? Whether Getty or a microstock, if they don't pursue infringements, then they can be a contributory infringer. (I alluded to this in an earlier blog about Flickr's liabilities in this area.)
Making it even more complicated is the inherent value of having your images stolen in the first place. In fact, Viacom's claim against Google's YouTube has generated a stir by television industry analysts, as well as within the walls of Viacom itself: the fact that TV shows are available online has actually helped viewership, and thus, increased advertising rates for those shows. Speaking of my own experience, the fact that so many sites use my images and link to me is one of the main core reasons why my site ranks so highly, which is what brings me the traffic that converts to buyers. So, a little copyright violation can actually do your business good.
Converting a criminal into an honest buyer is warm and fuzzy, but what about the tactic that made them a criminal in the first place? It'd be interesting if a defendant actually tried to use the legal excuse of the attractive nuisance doctrine. "They don't prevent me from downloading it without paying for it! It's too easy! I didn't know! The rules are cryptic." That's hard to say, and perhaps somewhat of a stretch, but if that magic switch were ever pulled by a search company, and thousands upon thousands of copyright infringement cases suddenly came up, the courts would have a lot on their hands.
Suddenly, the business aspect of copyright infringements isn't so cut and dry.
On the darker side again, there's this:
in an article by Bug Shell, Getty purchased the negatives to the National Archives, and is now on a litigation spree against those who use those images, regardless of the fact that owning the film does not grant ownership of copyright, and the fact that the photos are in the public domain. Getty claims that their "scanning" of the negatives (dusting and other corrections) creates a "derivative work," but many legal observers are taking great exception to this claim.
And the industry pendulum continues to swing back and forth.
Summary
The net-net is there's big money in copyright violations, especially if photos are registered with the copyright office and they are watermarked. What keeps this from becoming a more prominent figure in the industry itself is two-fold: the lack of awareness of how lucrative it can be, and the "resources" available for tracking the infringements in the first place. But, like anything that has the potential for money, there will be those who start down those paths to see where they lead. For those playing the home game, keep in mind that the issues faced by agencies are not shared by individual photographers. That is, while an agency may have business risk by being too aggressive in pursuing violators, individuals certainly do not. Even as I type this, yet another email has just trickled in from another concerned photographer, "I'm going to take my images off Flickr because I don't want them stolen." My response to him will be to read this article.
Labels: business model, copyright, corbis, economics, getty, infringement, licensing, microstock, registration
posted by argv | 11:17 PM
<< Home