Basically, at that time, RSS scrapers were arguing that, by putting content into an RSS feed, one was giving permission to use it on other sites, essentially creating an implied license to republish it. However, as I talked about in the previous article, the legal realities are much different and RSS scraping without per mission is, almost certainly, a copyright infringement.

However, while the legal realities haven’t changed much in the past five years, the people doing the scraping have. Spammers and sploggers, now wary of duplicate content issues, have largely abandoned RSS scraping in favor of other techniques. Today, the scrapers are fewer but place themselves as editors, curators and collectors, people building moderated lists of great content.

This shift hasn’t done much to alter the legal realities of scraping nor has it done much to placate creators who still see this as one of the most common issues they face.

The truth is that, even with this new veneer, RSS scraping is still not legally or ethically acceptable. Whether it’s curators or spammers, those who scrape from RSS feeds are in a dubious position and one that seems to be getting worse every day.

The Past Five Years Of Law and Scraping

The past five years of legal history have been strangely quiet on the issue of RSS scraping. Despite how common the behavior is, very few suits have dealt with the issue.

The best known of those cases was Gatehouse Media vs. The New York Times. Which saw Gatehouse Media, the owners of “Wicked Local” brand sites as well as hundreds of smaller papers, sue the New York for the Times’ scraping of their RSS feeds for inclusion on Boston.com’s “Your Town” section.

The suit only centered around the headlines and excerpts from the stories involved but the Times felt their position was weak enough to warrant settling the matter publicly and quickly. In the end, the New York Times agreed to stop scraping Gatehouse feeds and respect restrictions placed by Gatehouse Media on the content.

Related cases on the issue of data scraping, sometimes called data mining, have largely been equally negative for the scrapers. Though only at the summary judgment phase at last report, the Snap-on Business Solutions Inc. v. O’Neil & Assocs., Inc, highlights the other legal perils of scraping.

In that case, Snap-on produced and maintained a database of auto parts for Mitsubishi. After two years, Mitsubishi began to look at other vendors for the contract but Snap-on would not give up control over the data. Mitsubishi eventually hired an outside contractor, O’Neil, to scrape the content out of the database and bring it into a new system. When Snap-on learned of the scraping, they filed suit.

In the summary judgement phase of the case, the judge ruled that Snap-on likely had arguments regarding the Computer Fraud and Abuse Act (CFAA), Trespass to Chattels and Breach of Contract. The court rejected a copyright infringement argument, but only because the content copied did not qualify for copyright protection, unlike with RSS feeds.

The case shows, as I pointed out years ago, that scraping is a legal minefield. Even cases that seem to go the way of the scraper, such as the Cvent, Inc. v. Eventbrite, Inc. case, are highly fact-specific and seem to hinge more on poor case preparation than the law itself. (Note: Even in that “victory” the copyright claims and the unjust enrichment claims survived dismissal.)

Instead, most seem to follow the route of the Ticketmaster L.L.C. v. RMG Technologies, Inc. case, a 2007 win for Tickemaster against a sniping service that was snatching up popular tickets using an automated process. In that case, the court ruled RMG was infringing copyright by merely browsing the relevant pages since they were doing so in violation of Tickemtaster’s “browserwrap” license.

In short, the legal realities for scraper are even more bleak than they were five years ago. The implied license argument that’s so popular among scrapers has been eroded and, all in all, it’s almost impossible to scrape legally, RSS or otherwise.

Yet, what’s changed in the last five years isn’t so much the law, but the scrapers themselves and that’s where things have truly gotten interesting.

The Death of the Spammer Scraper

Back in 2006, your “typical” RSS scraper was probably a spammer, someone seeking a quick, hands off way of filling a large number of sites with search engine friendly content to rise in the rankings and, eventually, usurp the original work for certain keywords.

Those days, however, are gone. Though scraping spammers still exist, most spammers moved on from this method as Google and the other search engines improved their duplicate content detection, making it a less effective technique. Methods such as content spinning, content generation and even cheap outsourcing have proved to be more effective and equally reliable.

This decline has largely mirrored the overall decline in traditional (reader-based) RSS usage. RSS is falling out of vogue, at least as a tool for scraping and reading, but not as a tool for “curating”.

The reason is that tools for integrating RSS into existing websites have grown much more common and easier to use in the past five years. Though some were developed for the use of spamming, other tools were meant to allow authors to integrate all of their sites in one place. However, some authors have latched onto these tools as a way of bringing in the work of others without permission.

This has created a situation where the people doing the scraping are fewer in number, but likely much more dangerous. Where search engines were relatively effective at filtering out spammers, these sites tend to appear to be much more legitimate, increasing the likelihood they could be mistaken as originals.

Fortunately, the law doesn’t make a great distinction between spammers and those who scrape with less nefarious intentions, but many who engage in this practice have, according to emails I’ve seen, have claimed to have an ethical or even legal right to engage in the scraping, calling themselves “editors”.

This has set the stage for some ugly battles that, while they haven’t reached the courtroom yet, have certainly been heated on the Web.

Indeed, this argument seems to be one that’s moving out of the courtroom and into the court of public opinion, a place where it’s likely to stay given how straightforward the legal issues seem.

Bottom Line

In the end, consider that the New York Times Company, one of the most powerful media institutions on the planet, couldn’t or didn’t want to defend scraping of just headlines and summaries, there’s little hope for a successful defense of full RSS scraping. This is especially true in the light of other, related scraping cases.

However, those who want to scrape and those who are willing to allow their feeds to be scraped do have options. Creative Commons, for example, has modules for RSS feeds that enable applications to detect what they are allowed to do with a feed.

To those who don’t wish to allow it, I encourage you to put in your feed itself a notice stating that you do not wish to allow republishing and that the feed is for private personal use only. Though it shouldn’t be necessary under the law, it’s a wise move that blocks many of the potential arguments a scraper might raise. Furthermore, such footers can greatly help with the detection of scrapers.

All in all, RSS scraping has definitely changed in terms of who is using it and why, but the threat isn’t all that different and the legal realities have hardly changed at all.

This means that RSS scraping can be easily fought, just that the people you’re moving against may be a bit more vocal in their views.

In 2006 I wrote an article entitled “Why RSS Scraping Isn’t OK” that laid out many of the arguments provided in favor of RSS scraping and republishing and laid out counter-arguments to them.

By in large, in the years that followed that article it seems that Internet has agreed. RSS scraping is no longer a hot topic on the Web but it is also not an accepted practice by any stretch. When sites are accused of scraping they are usually shouted down and are viewed as spam bloggers.

However, a ruling from earlier in 2006 may be coming back to raise new questions about how implied licenses affect the Web. The Field v. Google case, which dealt with Google’s cache, took a look at the implied license question from a different perspective and came up with some interesting answers, namely that Field, a writer, had granted Google an implied license to cache and display the cache of his content by posting it on the Web and not taking additional steps to block it.

That ruling is now being revisited and reapplied in another case, this one dealing with the controversial copyright litigator Righthaven and the new application may impact all creators on the Web and make nearly all content copying legal.

Though I am no fan of Righthaven and their practices, this particular theory against them is very dangerous and well worth analyzing closely.

The Righthaven Case

Righthaven is a company set up by the shell corporation for the Las Vegas Reivew-Journal (LVRJ) to enforce the copyrights of the paper. The group purchases rights from the paper and then sues bloggers and other sites that have republished stories from the paper, demanding upwards of $75,000 for a settlement (though the cases usually settle for much less).

Their practice is controversial for several reasons including that they do not file takedown notices or cease and desist notices first, meaning that the suits usually come without warning, Also, the company has targeted sites with many mitigating factors including ones that only reposted small portions of the original content, were the subjects for some of the articles and, in some cases, sites who did not upload the work themselves but had them uploaded by users. The latter raises significant safe harbor issues.

One of the publishers sued was Jan Klerks, who runs the forum SkyScraperCity.com. A user at the forum posted a LVRJ article and Righthaven has sued Klerks for copyright infringement.

Initially, the court issued a default judgment against Klerks when she failed to respond to the suit but Klerks is now requesting the suit be vacated (PDF) on multiple grounds, including the following:

1. Fair Use: Claiming that the copying was a fair use.
2. De Minimis Copying: Meaning the copying was so small and technical as to be non-infringing.
3. Implied License: Claiming that the LVRJ, by their actions, granted an implied license for the reuse of the content.

It is the last argument that has gotten the most interest and is the one I’m focusing on today In the filing, Klerks argued that several actions the LVRJ took essentially gave her user (or anyone else) permission to copy. Those actions included:

1. Using Sharing Services: The LVRJ uses social media buttons and other tools to enable users to share links to stories.
2. Not Disabling Copy/Right Click: The site enabled users to right-click and otherwise copy content, not taking any steps to block it.
3. Posting the Work For Free and Making Available in Google: The article in question was still available for free on the LVRJ site and is searchable by Google.

Interestingly, the motion does NOT mention several very clear potential arguments that likely apply including DMCA Safe Harbor protection, which Klerks likely qualifies for as the host of the forum, or jurisdictional issues as Klerks is from Illinois and was sued in Nevada.

Still, the implied license argument has drawn a lot of attention from all over the Web and rightfully so. To support that argument, Klerks cites the Field v. Google case, revisiting the case in a very different light.

The Field v. Google Case

In the Field v. Google case, the court ruled that Google, among other things, had an implied license to store and display caches of Field’s content for several reasons, the biggest being that Fields was aware of the robots.txt standard for blocking archiving and the potential use of meta tags but failed to do either, therefore, according to the ruling:

Thus, with knowledge of how Google would use the copyrighted works he placed on those pages, and with knowledge that he could prevent such use, Field instead made a conscious decision to permit it. His conduct is reasonably interpreted as the grant of a license to Google for that use.

This ruling was somewhat unusual in that copyright law normally does not place any requirements on copyright holders to try to stop copying before a copy can be considered an infringement. However, the court also found in this case that the cached copy was transformative enough to be considered a fair use and that, even without this license, Google’s use would not be an infringement.

(Note: More about the verdict from the EFF)

Still, it is easy to see why Klerks would cite this case in her argument on implied license but the implications if this argument is successful could be very severe for anyone who publishes content online.

Analysis of the Argument

Personally, I don’t think Klerks’ implied license argument is going to take flight. Even though there are some similarities between her case and the Field case, there are several critical differences when it comes to the issue of direct infringement:

1. No Industry Standard: First, there is no industry standard akin to robots.txt for blocking human copying. No right-click scripts and other anti-copy protections are nowhere near commonplace and they disable other, useful features a webmaster may wish to enable. Comparing the two is a tremendous stretch.
2. Copier is not Passive: In the Fields case, the plaintiff did not object to the initial copying and the cached version is an automated process initiated by a visitor at Google, Google is passive in the process. Reposting a work on the Web is not a passive process that the person doing the copying has to initiate.
3. Weaker Fair Use Argument: Though the fair use argument should be weighed separately from the implied license/direct infringement issue, such copying and pasting is far less transformative than Google cache and, generally, is capable of replacing the function of the original. This, inevitably, is weighed into the ruling to at least some degree.

Also, Klerks’ argument regarding the sharing buttons seems a bit of a stretch as well. These buttons enable the sharing of links and not the whole content and other buttons, such as the email button are not for posting the content on another site, but rather, sending via a private means of communication.

In short, many legal scholars feel that the Field case was a bit of a stretch on the implied license issue and this argument seems to take the Field ruling and stretch it even farther, likely past the breaking point.

However, the honest answer is that one does not know what the court will decide and, as with fair use issues, courts are notoriously erratic. Attorney Eric Goldman said it best when he indicated that there is no way of knowing if an implied license exists here and, if one does, what it covers.

In short, even if an implied license does exist, it may not go as far as Klerks wants or needs it to. Instead of covering republishing on the Web, it may only cover sending to one or two people or posting to a some other limited use.

Until the court rules, we simply will not know.

In the meantime, we are faced with the possibility, no matter how remote, that courts could rule that posting a work on the Web is the same as giving an implied license to copy and republish it freely, something that could all but eliminate copyright protection for many Web publishers.

Take Action Now

Even if the argument is not likely to succeed, it has opened up a Pandora’s box and created uncertainty that has to be addressed. In 2006, when dealing with RSS scraping, I encouraged people to not take any chance and license their feed using a Creative Commons or other applicable license.

This advice now stands, more than ever, for content on the Web too.

The reason is quite simple, any implied license is trumped by an actual license. This is because implied licenses, by their very nature, are only for cases where an actual license has not been granted or doesn’t exist. So, even if the court does rule that publishing grants a broad implied license, any actual license would have to be followed first.

This is a big part of why it is so important to license your works correctly. Not only does it prevent misunderstandings, but it also prevents you from accidentally giving away more rights than you intended via an implied license.

In short it is best to decide your license for yourself and not let the court choose and hope it picks the right one.

Bottom Line

If you publish content on the Web, this affects you. It is that simple. As I said before, I am no fan of Righthaven or their tactics but this is an argument against them that could have very far-reaching and very negative consequences for almost everyone who posts content online.

Granted, the odds of the argument succeeding (at least to the extent Klerks needs it to) seems very slim but it does add another reason for webmasters to pay close attention to their licensing practices and work to make them clear.

In short, if you’ve been relaxed about licensing your content, now is the time to change that. It’s better to take such action now than possibly get caught off guard by this, or any other, ruling that affects your rights.

Ryan McGinnis of Backing Winds posted an article detailing copyright policy of news aggregator Fark, which just underwent a major redesign.

According to McGinnis, when the site was redesigned a new legal policy was introduced one that lays claim to the copyright of all works posted to Fark.

Though it appears that the policy is nothing new, the Web Archive shows the text being on the page as far back as December 2005, his analysis of the current policy is correct. Fark does claim the copyright of any and all material posted to the site.

This is a stark contract to most other sites that only take what the implied license grants them, the right to republish the posted material on their own site. It’s a very powerful rights grab that may be caused by poor wording but, nonetheless, could have very strong impact on copyright holders posting their work on the site.

What the Policy Says

If you read the Fark legal policy, under the header of “Copyright Notice” it says the following:

Fark.com is the legal owner of all copyright interests of Fark.com content. Each and every submission to Fark.com carries with it an implied assignment of the entire copyright interest in the submission. In exchange for the content and publication of that submission on Fark.com, Fark.com grants back to the submitter a non-exclusive, non-transferable and royalty-free license to republish that submission in any and all forms.

Compare this, for example, to the Terms of Use for bulletin Board service ezboard (see item 5).

ezboard.com does not claim any ownership rights in the text, files, images, photos, video, sounds, musical works, works of authorship, or any other materials (collectively, “Content”) that you post to the ezboard Services. After posting your Content to the ezboard Services, you continue to retain all ownership rights in such Content, and you continue to have the right to use your Content in any way you choose. By displaying or publishing (“posting”) any Content on or through the ezboard Services, you hereby grant to ezboard.com a limited license to use, modify, publicly perform, publicly display, reproduce, and distribute such Content solely on and through the ezboard Services.

It becomes clear that Fark’s copyright license, as it is written, is actually reversed. Instead of giving itself the limited license to reproduce the work, which it needs in order to display the work, it takes the entire copyright and then gives the user the limited license.

However, that isn’t the only problem with Fark’s policy. It makes several missteps when it comes to copyright law at large.

Problems with Fark’s Policy

Fark’s policy, as strong as it may appear to be, is most likely invalid. U.S. Copyright Law makes a policy such as this completely impossible.

See Title 17, Chapter 2 § 204(a) – Execution of transfers of copyright ownership

A transfer of copyright ownership, other than by operation of law, is not valid unless an instrument of conveyance, or a note or memorandum of the transfer, is in writing and signed by the owner of the rights conveyed or such owner’s duly authorized agent.

In short, a copyright can not be transferred without a written and signed contract between the two parties. Simply having a policy that states “You are signing your copyright over to us” does not meet that requirement. You can have people give up certain rights, and many are given up by the implied license, but you can not force someone to give up their actual copyright without a signature.

However, there are also issues with the notion that there is an “implied assignment of the entire copyright interest in the submission”. Though the exact extent of the implied license of comments, forum posts and other submitted works, is up for debate, most agree that it does not extend to full copyright ownership.

It is clear that, at the very least, the text needs to be rewritten. Fortunately, Fark has already expressed an interest in doing exactly that.

What it Would Mean

The question remains, if Fark’s policy were found to be valid and remained unchanged, what would it mean?

First, you would not be able to sell your work or give anyone else the right to post it. If you posted a work to Fark, you could not legally sell it to a publisher, to a record label or to a movie studio. You could include it in a book that you yourself made, but you could not sell the copyright or even your right to republish to anyone else. You and Fark would be the only people allowed to publish the work.

Second, Fark can do pretty much anything they wish with the work. They can compile it into a book, CD or video and sell it, they can license it to other sites and redistribute it as they see fit under whatever licenses they see fit.

Third, any Creative Commons License you originally placed the work under would be invalid. Since you are no longer the copyright holder, you no longer have the right to set the licensing terms. This could lead to trouble for anyone that reuses your content under the terms of the license but later runs afoul of Fark’s new licensing scheme, whatever it might be.

Finally, though you have the right to republish your own work, since that right is not perpetual, it can be ended at any time. This can mean that either Fark can terminate the license or they can sell the rights to someone else who will do it instead.

This makes it fortunate that the copyright license Fark is presenting is, most likely, completely invalid. If it were valid, it would create a lot of headaches for copyright holders posting their works to the site. It makes it even more fortunate that Fark is looking to change the license and will, hopefully, have a new one up soon.

Fark Responds

In an email from Drew Curtis, the owner of Fark, he reassured me that of the following:

We’re not, we’re asking for a non-exclusive right to republish. Submitters still own their submissions, we’re asking for reprint rights in case we can use it. We have no intention of acquiring ownership of submissions.

He went on to say that:

“We don’t really have any plans for anything. In the event we did use any content we’d attribute it, in fact I’ve gone out of my way in the past to make sure that whenever media outlets have requested use of Fark photoshops that they contact the owner directly and not republish unless they receive permission from the owner him/herself.”

This certainly is very reassuring and it makes it seem as if it is a case of badly-written license copy, not one of malicious intent. Curtis has asked for input from myself and other in rewriting the copy and I have already provided my input.

We will have to see how and if the terms of use change.

What To Do

Even though the license, most likely, would not be considered valid and it seems as if Fark does have genuinely good intentions, it is still worthwhile to take a few steps to ensure that there are no copyright mix ups on Fark or any other site.

1. Post Things in Links: A license only applies to what actually appears on the site. Rather than post your long essay or your photos, post links to them. No license can be claimed to anything not hosted on the site.
2. Post Only What You Can Lose: Only post what you are ready to give up at least some rights on. I doubt that there will be much interest in reusing general conversation but essays, photos, videos and other works are more likely to become targets for repurposing. Be wary of what you post and realize that, even under the best of circumstances, you will surrender some rights to them unless they are posted on your site.
3. Monitor Your Works: As always, monitor your works and be prepared to act on any infringement that you see. Be mindful of your rights and be prepared to defend them if needed.

All in all, if you take reasonable steps and realize that the copyright license is there and what it says, you’ll probably be safe. The important thing is to simply be aware and keep monitoring the situation.

Conclusions

All in all, I don’t think that Fark is likely to do anything abusive with this copyright policy. It has been in effect for well over a year and nothing has happened to date. By all appearances, it just seems that it was a case of poorly-written license copy that ended up biting both Fark and some of its users in the end.

Furthermore, it appears that it will be rectified very soon.Still, the entire episode highlights the importance of looking at a site’s policy pages before posting anything of value to it. You need to be aware of the rights that you are giving up by participating on any kind of community or forum so that you can be aware of what the rest of your rights regarding the work are.

It makes good sense. We read contracts before we sign them, why not take a few moments and read the terms of use on a site before posting? I admit I am as bad as most about this, but it is a habit that I am working to change.

At the end of the day, this is most likely a learning experience. Nothing bad has happened and, almost certainly, nothing bad will happen, but it shows why everyone should be more careful with where they post. The danger is clearly there.

Update: Just before this story went up, Curtis emailed me with a new draft of the text that addresses all of these issues. Once the new text is looked at by an attorney, he plans on using it.

Update 2: This story has now appeared on Fark. So, with that in mind, welcome Fark visitors! Also, a big congrats goes out to Media Temple for keeping this site alive. Under this kind of load, my old host would be dead and buried.