These are often CAM releases, which are copies of the movie recorded in a theater. Despite their low quality, these CAMs are often downloaded hundreds of thousands if not millions of times.

To find out what effect these downloads have on box office revenues APAS Laboratory researcher Marc Milot conducted a thorough field study. By using download statistics from the torrent site Demonoid, in combination with movie ratings and pre-release buzz, the research estimates the effect of CAM piracy on box office sales.

The findings, published this week in a paper titled “Testing the lost sale concept in the context of unauthorized BitTorrent downloads of CAM copies of theatrical releases”, reveal an intriguing pattern.

Based on a sample of 32 widely released movies, the results show that box office revenue could be best predicted by pre-release buzz and to a lesser extent by the rating of the movies, which were both taken from Rotten Tomatoes. Interestingly, the amount of times a movie was pirated had no effect on its box office sales.

Instead of a link with sales, the amount of unauthorized downloads was affected by how visible these titles were on Demonoid. TorrentFreak contacted Milot, who believes that these results support the notion that many pirates download movies to discover new content.

“The research findings are the first to support with concrete behavioral evidence what BitTorrent file-sharers have been saying all along: that they don’t always download movies – in this case CAM versions of theatrical releases – they would have paid to view if they were not available on sites like Demonoid,” Milot told us.

This notion is supported by the fact that, unlike at the box office, the rating of a movie doesn’t affect the piracy volume. This finding is based on ratings by both Pirate Bay and Rotten Tomatoes users, to control for the possibility that pirates simply have a different movie taste.

According to the researcher, these results should caution the movie industry not to overestimate the effect of CAM piracy on box office sales.

“BitTorrent site users appear to be exploring and downloading the most visible movies, without caring how good or bad they are. It is in this way that BitTorrent sites and the box office are completely different systems in which people behave uniquely and with different motivations,” Milot explains.

“These findings should caution against the use of download statistics alone in calculations of losses – in this case lost ticket sales – to avoid overestimation,” he adds.

Whether the above will be a reassurance for Hollywood has yet to be seen. There have been several studies on the impact of movie piracy in recent years, often with conflicting results. The current research helps to add yet another piece to the puzzle.

Source: TorrentFreak, for the latest info on copyright, file-sharing and anonymous VPN services.

In recent years many of these sites have gotten a bad reputation as they are frequently used to share copyrighted files.

Today the Digital Citizens Alliance released a new report (pdf) that looks into the profitability of these sites and services. Titled “Behind The Cyberlocker Door: A Report How Shadowy Cyberlockers Use Credit Card Companies to Make Millions,” it offers insight into the money streams that end up at these alleged pirate sites.

The study, carried out by NetNames and backed by the entertainment industry, uses information from the busted Megaupload service to estimate the earnings of various other sites. Based on these and other assumptions it concludes that the top cyberlockers generate an average $3.2 million per site per year.

“Overall, total annual revenue across the thirty cyberlockers equated to $96.2 million or $3.2 million per site. One site gathered $17.6m per year in revenue,” the report notes, adding that it’s a conservative estimate.

The report brands these sites as piracy havens based on a sample of the files they host. All the sites that are listed are used predominantly for copyright infringement, they claim.

“The overwhelming use of cyberlockers is for content theft. Analysis of a sampling of the files on the thirty cyberlocker sites found that the vast majority of files were clearly infringing,” the report reads.

“At least 78.6 percent of files on direct download cyberlockers and 83.7 percent of files on streaming cyberlockers infringed copyright,” it adds.

Here’s where the researchers make a crucial mistake. The sample, where the percentage of allegedly infringing files is based on, is drawn from links that are posted publicly online. These are certainly not representative for the entire site, at least not in all cases.

For Mega the researchers looked at 500 files that were shared online. However, the overwhelming majority of Mega’s files, which number more than 500,000,000, are never shared in public.

Unlike some other sites in the report, Mega is a rather traditional cloud hosting provider that’s frequently used for personal backup, through its desktop client or mobile apps for example. The files that are shared in public are the exception here, probably less than one percent of the total.

There is no denying that there are shady and rogue sites that do profit heavily from piracy, but lumping all these sites together and branding them with a pirate label is flat-out wrong.

Aside from “exposing” the estimated profitability of the cyberlockers the report also has a secondary goal. It puts out a strong call to the credit card companies Visa and MasterCard, and hosting providers such as Cloudflare, urging them to cut their ties with these supposed pirate havens.

“They should take a hard look at the checkered history of their cyberlocker partners. Simply put, the businesses that simply exploit and expropriate the creative efforts of others do not occupy a legitimate place in the Internet ecosystem,” the report notes.

“Content theft is a cancer on the Internet. It introduces viruses and malware to computers, robs creators who rely on the Internet to sell their products, damages brands by associating them with illegal and inappropriate content and provides seed money for criminals to engage in other illegal activities,” it adds.

Hopefully future reports will have more nuance. At minimum they should make sure to have all the facts right, as that’s generally more convincing.

Source: TorrentFreak, for the latest info on copyright, file-sharing and anonymous VPN services.

One of the suggestions often made is to remove or demote pirate sites in search results. A lower ranking would lead fewer people to pirate sources and promoting legal sources will have a similar effect.

Google previously said it would lower the ranking of sites based on DMCA complaints, but thus far these changes have had a limited effect. A few weeks ago the company also began promoting legal options but this effort is in the testing phase for now.

The question that remains is whether these changes would indeed decrease piracy. According to new research from Carnegie Mellon University, they can.

In a paper titled “Do Search Engines Influence Media Piracy?” the researchers ran two experiments where they let participants use a custom search engine to find a movie they wanted to watch. The respondents could pick from a list of 50 titles and received a $20 prepaid virtual Visa card as compensation.

All search results were pulled from a popular search engine. In the control category the results were not manipulated, but in the “legal” and “infringing” conditions the first page only listed “legal” (e.g Amazon) and neutral (e.g IMDb) sites or “infringing” (e.g. Pirate Bay) and neutral sites respectively.

While it’s quite a simple manipulation, and even though users could still find legal and pirated content in all conditions, the results are rather strong.

Of all participants who saw the standard results, 80% chose to buy the movie via a legal option. This went up to 94% if the results were mostly legal, and dropped to 57% for the group who saw mostly infringing results on the first page.

TorrentFreak contacted Professor Rahul Telang who says that the findings suggest that Google and other search engines have a direct effect on people’s behavior, including the decision to pirate a movie.

“Prominence of legal versus infringing links in the search results seem to play a vital role in users decision to consume legal versus pirated content. In particular, demoting infringing links leads to lower rate of consumption of pirated movie content in our sample,” he notes.

In a second study the researchers carried out a slightly modified version of the experiment with college students, a group that tends to pirate more frequently. The second experiment also added two new conditions where only the first three results were altered, to see if “mild” manipulations would also have an effect.

The findings show that college students indeed pirate more as only 62% went for the legal option in the control condition. This percentage went up gradually to 76% with a “mild legal” manipulation, and to 92% in the legal condition. For the infringing manipulations the percentages dropped to 48% and 39% respectively.

According to Professor Telang their findings suggest that even small changes can have a significant impact and that altering search algorithms can be instrumental in the fight against online piracy.

“The results suggest that the search engines may play an important role in fight against intellectual property theft,” Telang says.

It has to be noted that Professor Telang and his colleagues received a generous donation from the MPAA for their research program. However, the researchers suggest that their work is carried out independently.

As a word of caution the researchers point out that meddling with search results in the real world may be much more challenging. False positives could lead to significant social costs and should be avoided, for example.

This and other caveats aside, the MPAA and RIAA will welcome the study as a new piece of research they can wave at Google and lawmakers. Whether that will help them to get what they want has yet to be seen though.

Source: TorrentFreak, for the latest info on copyright, file-sharing and anonymous VPN services.

This popularity makes BitTorrent an interesting target for attacks, which various anti-piracy companies have shown in the past. One of these possible attacks was recently unveiled by Florian Adamsky, researcher at the City University London.

In an article published in “Computers & Security” Adamsky and his colleagues reveal an exploit which allows attackers to get a higher download rate from seeders than other people.

In technical terms, the exploit misuses BitTorrent’s choking mechanism of clients that use the “Allowed Fast” extension. Attackers can use this to keep a permanent connection with seeders, requesting the same pieces over and over.

The vulnerability was extensively tested in swarms of various sizes and the researchers found that three malicious peers can already slow download times up to 414.99%. When the number of attackers is greater compared to the number of seeders, the worse the effect becomes.

The impact of the attack further depends on the download clients being used by the seeders in the swarm. The mainline BitTorrent clients and uTorrent are not vulnerable for example, while Vuze is partly affected.

“Vuze is only partly affected as it allows pieces to be downloaded 64 times and then all further requests are rejected. Nevertheless we have listed Vuze as vulnerable since it is possible to reconnect and restart the attack,” the paper reads.

Update: Vuze was quick to patch the vulnerability.

An older version of Libtorrent also had this vulnerability but that was fixed a year ago.

TorrentFreak spoke with Adamsky who predicts that similar results are possible in real swarms. Even very large swarms of more than 1,000 seeders could be affected through a botnet, although it’s hard to predict the precise impact.

“If an attacker uses a botnet to attack the swarm, I think it would be possible to increase the average download time of all peers [of swarms with 1,000 seeders] up to three times,” Adamsky tells us.

“If most of the clients would have a vulnerable client like Vuze or Transmission it would be possible to increase the average download time up ten times,” he adds.

In their paper the researchers suggest a relatively easy fix to the problem, through an update of the “Allowed Fast” extension. In addition, they also propose a new seeding algorithm that is less prone to these and other bandwidth attacks.

Update: The article was updated to clarify that only older version of Libtorrent were affected. According to the research Transmission currently has the “Allowed Fast” code commented out, but it could become vulnerable when it’s implemented.

Source: TorrentFreak, for the latest info on copyright, file-sharing and anonymous VPN services.

A new study conducted by economist Koleman Strumpf is one of the most comprehensive on the subject so far.

Drawing on data from a popular BitTorrent tracker and revenue projections from the Hollywood Stock Exchange he researches how the release of a pirated movie affects expected box office income.

The research covers 150 of the most popular films that were released over a period of seven years, and the findings reveal that the release of pirated films on file-sharing sites doesn’t directly hurt box office revenue.

“There is no evidence in my empirical results of file-sharing having a significant impact on theatrical revenue,” Strumpf tells TorrentFreak in a comment.

“My best guess estimate is that file sharing reduced the first month box office by $200 million over 2003-2009, which is only three tenths of a percent of what movies actually earned. I am unable to reject the hypothesis that there is no impact at all of file-sharing on revenues.”

So while there is a small negative effect, this is limited to three tenth of a percent and not statistically significant.

Interestingly, the data also reveals that movie leaks shortly before the premiere have a small positive impact on expected revenues. This suggests that file-sharing may serve as a form of promotion.

“One consistent result is that file-sharing arrivals shortly before the theatrical opening have a modest positive effect on box office revenue. One explanation is that such releases create greater awareness of the film. This is also the period of heaviest advertising,” Strumpf notes.

One of the advantages of this study compared to previous research is that it measures the direct effect of a movie leak on projected box office revenues. Previous studies mostly compared early versus late leaks, which is less accurate and may be influenced by other factors.

“For example, suppose studios added extra security to big budget movies which then have a delayed arrival to file-sharing networks. Then even if file-sharing has no impact at all, one would find that delayed arrival on file-sharing leads to higher revenues,” Strumpf tells us.

Another upside of the research lies in the statistical precision. The data includes thousands of daily observations and relatively precise estimates, something lacking in most previous studies.

The downside, on the other hand, is that the expected box office impact is estimated from the Hollywood Stock Exchange. While this has shown to be a good predictor for actual revenues, it’s not a direct measurement.

In any case, the paper suggests that file-sharing might not be the biggest threat the movie industry is facing.

Even if the negative effects were twice as big as the data suggests, it would still be less than the $500 million Hollywood spent on the MPAA’s anti-piracy efforts during the same period.

Source: TorrentFreak, for the latest info on copyright, file-sharing and anonymous VPN services.

Increasingly researchers are also examining the sociological links, causes and effects of copyright infringement. A new study conducted by Tennessee Tech University’s Jordana Navarro is a good example.

With a large survey Navarro and her colleagues investigated the link between piracy, internet addiction and deviant tendencies. The results were published in an article titled “Addicted to pillaging in cyberspace: Investigating the role of internet addiction in digital piracy,” which appears in the latest issue of the Computers and Human Behavior journal.

The researchers conducted a large-scale survey among 1,617 students from 9th through 12th grade. The participants were asked a wide range of questions, covering their piracy habits, as well as scales to measure Internet addiction and association with deviant friends.

The findings on the piracy side are comparable to many previous studies and show that movie piracy is most prevalent. Nearly 30% of the students admitted to pirating movies, and this percentage went down to 15% and 13% for music and software piracy respectively.

One of the more interesting findings is the link between piracy and Internet addiction. Here, the researchers found that students who have more internet addiction related issues are more likely to pirate software.

“Based on the results of the study, we can determine that high school students who have Internet-related problems due to addiction are more likely to commit a specific form of piracy involving the illegal downloading of software,” the researchers write.

The same group of software pirates were also more likely to hang out with deviant friends. This measure includes friends who pirate, those who threaten others with violence online, those who send nude pictures, and those who have used another person’s credit card or ID without permission.

“Not surprisingly, youth who committed this form of piracy were also more likely to have deviant peers. In other words, their behaviors were influenced by friends who committed similar or other deviant acts,” the researchers conclude.

Interestingly, the link between Internet addiction and copyright infringement was only found for software piracy. High school students who pirated movies and music were not more likely to have these type of problems. They were, however, more likely to associate with deviant or criminal friends.

“The remaining two forms of piracy for juveniles are not predicted by Internet addiction based on our findings. However, the results did support past findings that deviant peer association and piracy behaviors are significant related,” the researchers write.

According to the researchers the results are a good first step in identifying how various problems and deviant behaviors are linked, which could be helpful to shape future educational efforts.

Unfortunately, the paper doesn’t offer any explanations for the differences in the link between Internet addiction and various types of piracy. One likely explanation is that those who show more signs of Internet addiction simply spend more time on the computer, and are therefore more interested in software piracy and software in general.

For now, it appears that some more follow-up research is needed before it’s warranted to send the first batch of kids to piracy rehab.

Source: TorrentFreak, for the latest info on copyright, file-sharing and anonymous VPN services.

The law, which gives rights holders the authority to request the personal details of alleged copyright infringers, was met with fierce resistance from ISPs and the public at large.

At the same time, however, there were plenty of signs that the law stopped people from pirating. A day after it went into effect, Netnod Internet Exchange reported a significant drop in Swedish Internet traffic.

Inspired by the anecdote, the effectiveness of IPRED has become a topic of interest for economists at Uppsala University in Sweden. In a new paper they report their findings on the effect of the anti-piracy law on Internet traffic and music sales.

The main goal of the research is to examine whether the anti-piracy law did indeed have an effect, and to what extent. To make sure that the effect is unique to Sweden, both Norway and Finland were chosen as control groups.

The results, which will be published in the Journal of Economic Behavior and Organization, confirm that Internet traffic decreased quite a bit after IPRED went into effect, beginning abruptly the very same day.

Perhaps even more surprisingly, music sales also skyrocketed compared to the other two Scandinavian countries.

“We find that the reform decreased Internet traffic by 16% and increased music sales by 36% during the first six months. Pirated music therefore seems to be a strong substitute to legal music,” the researchers write, summarizing the results.

Interestingly, however, the overall effect on Internet traffic and music sales vanished after half a year. The only effect that remained was the increase in digital sales. Internet traffic and physical music sales returned to normal, in part because the chance of getting caught is quite low.

“The deterrent effect decreased quickly, possibly because of the few and slow legal processes. Law enforcement through convictions therefore seems to be a necessary ingredient for the long-run success of a copyright protection law,” the researchers note.

The researchers suggest that if more people are convicted, the effects may last longer. During the first few years only a handful of file-sharers were brought to justice, while hundreds of thousands took steps to circumvent the law.

“As the first court cases were only settled recently, it is still possible that further convictions would restore an effect that is more long-lasting,” they write in their conclusion.

The question remains, however, whether bankrupting people or throwing them in jail is the ideal strategy in the long run…

Source: TorrentFreak, for the latest info on copyright, file-sharing and anonymous VPN services.

The law heightened punishment for copyright infringement over the Internet and criminalized circumvention of DRM. In addition, it also introduced a safe harbor for Internet services, meaning that they can’t be held liable for their pirating users as long as they properly process takedown notices.

Initially these notices were mostly sent to consumer ISPs to alert them to pirating subscribers. At the time, rightsholders showed little interest in sending takedown notices to other online services, but this changed drastically in the years that followed.

New research by Stanford Law School’s Daniel Seng reveals that online services such as Google and Twitter have seen a surge in takedown requests in recent years. In fact, drawing on data from ChillingEffects.org, Seng finds that the number of DMCA notices processed by Google increased 711,887 percent in four years, from 62 in 2008 to 441,370 in 2012.

Titled “The State of the Discordant Union” and published in the Virginia Journal of Law and Technology, the paper discusses this upward trend. Two key changes Seng observes are that the average number of URLs in each notice is increasing, and that claims for multiple works are often included in a single notice.

Where copyright holders previously listed only one work per notice, there are now sometimes dozens of movies or tracks bundled in each. This is a worrying development according to Seng.

“It is disturbing to see the trend where more claims and more takedown requests are packed into each takedown notice. Up until 2010, each notice contained only one claim. But in 2011, the average number of claims per notice is 2.18, and in 2012, this average is 5.05,” Seng writes.

More copyrighted works per notice also means that the number of URLs per notice is increasing too. For example, between 2011 and 2012 the average number of URLs listed in each notice increased from 47.79 to 124.75.

According to Seng, these changes can be attributed to a small number of copyright holders. In fact, most copyright holders still submit only one notice.

“These increasing averages paint a slightly misleading picture. More than 65% of all reporters have only issued one notice, and almost 95% of all reporters have issued no more than 10 notices in 2012,” Seng writes.

The most active copyright holders up until 2012 were the RIAA, Froytal and Microsoft, each listing more than five million notices. Seng’s paper doesn’t include the most recent data, but Google’s Transparency Report shows that these numbers more than doubled again in 2013.

Interestingly, the surge in notices started right after the SOPA and PIPA bills failed to pass, suggesting that this is anti-piracy plan B for the entertainment industries.

The release of the paper documents an important change in the use of DMCA takedown notices and coincides with ongoing discussions between copyright holders and online service providers on how to improve the DMCA takedown process. What changes will be made, if any, remains to be seen.

—

_{*The number of notices sent by Yahoo in 2011 and 2012 are most likely 0 because they stopped reporting them to Chillingeffects.}

Source: TorrentFreak, for the latest info on copyright, file-sharing and anonymous VPN services.

The site’s status as one of the largest online media libraries has also piqued the interest of researchers, who are closely following what people are sharing.

Previously we documented how the number of files uploaded to The Pirate Bay increased 50% in just a year, and that more than one-third of the uploads are adult content.

A new study, conducted by researchers from the Institut Mines-Télécom in Paris and Madrid’s Universidad Carlos III, confirms the Pirate Bay’s expansion and adds several new insights. The researchers looked at large samples of torrents from various Pirate Bay categories, and polled the active sharers at various points in time.

“To the best of our knowledge it is the first study that weights several factors related to BitTorrent over a two years window,” researcher Reza Farahbakhsh tells TF.

From their data samples, which span from 2009 to 2012, the researchers draw three main conclusions.

• The number of high-definition video torrents on The Pirate Bay increased more than 500%.
• The median file-size of all torrents has doubled.
• Between 40-50% of all torrents point to video content and 80% of the total downloads come from these torrents.

During the latest measurement in 2012, high-definition video torrents accounted for 8.2% of the total, up from a mere 1.5% in 2009. As a result, the number of people downloading these files also surged, now making up nearly 10% of all downloads.

“The popularity of High-resolution PORN and VIDEO content follows the increasing availability of this type of content. While it only attracted 1.87% of the downloads in 2010, it has increased its popularity 5 times by receiving 9.62% of the downloads in 2012,” the researchers write.

Another result of the increased availability of high-definition videos is that the median size of all content indexed by The Pirate Bay has doubled over the years.

“The median value of the content size in 2009 was 223MB and increased by 53% (to 341MB) in the next five months, and it kept growing up to 370MB and 458MB in 2011 and 2012 respectively,” the researchers write.
The researchers’ data could act as a warning signal to Internet providers, who need to make sure they can handle further increases in their network usage. Not just with BitTorrent in mind either, but also other pirate sources such as cyberlockers where similar patterns may emerge.

“These findings are useful to those Internet players (i.e. ISPs, CDN operators) involved in the content distribution business in order to update their infrastructures, resources and algorithms to efficiently distribute and serve multimedia content,” the researchers conclude.

It will be interesting to see how the demand for high-definition content develops. With increasing broadband penetration and bigger screen sizes, the upward trend is expected to continue in the years to come.

Source: TorrentFreak, for the latest info on copyright, file-sharing and anonymous VPN services.

This conclusion was in part based on a working paper from researchers at the University of Amsterdam and Tilburg University. After the publication of the paper this week in the Telecommunications Policy journal, TF caught up with the authors to discuss the results as well as the wider topic of Internet censorship.

The researchers are happy with the peer-reviewed publication. It affirms that BREIN’s attempts to defame their work during the court hearing were unfounded. The court already concluded as much, as the findings were at the basis of the ruling to unblock The Pirate Bay.

“Our research, together with research by TNO showed quite indisputably that the measure has little to no effect,” says Joost Poort, lead author of the Baywatch paper.

“We do not want to take sides in the case by calling the decision the right one in a legal sense. However, we are very content with the fact that the effectiveness of blocking access to The Pirate Bay was taken into account in the decision,” he adds.

The Baywatch paper shows that censoring The Pirate Bay had no lasting net impact on the overall number of downloaders from illegal sources in the Netherlands. On the contrary, local piracy rates went up.

According to the researchers the Pirate Bay blockade could be easily circumvented. Thanks to the many readily available proxy sites online, users only had to update their bookmarks to gain access.

“Basically, the required knowledge to circumvent the blocking is no more advanced than the knowledge required to download from illegal sources, and the emergence of new mirror-websites and proxies is something that regulators can hardly keep up with,” Poort tells TF.

Researcher Jeroen van der Ham adds that people could simply move to other torrent sites, or move to other download platforms. They found that, in line with previous research, the initial effect of such anti-piracy measures wears off in about six months, as people turn to alternatives.”

While the researchers focused on the Dutch public they believe that their results should apply elsewhere too. This includes the UK, where numerous file-sharing sites have been blocked in recent years.

“We believe that the results can probably be generalized to any service that has a widespread user-base. Internet users have shown to be very inventive in circumventing blockades, or moving to similar services if that is not possible,” Van Der Ham says.

Censoring websites has no lasting effects on piracy. On the contrary, the researchers say they these measures may reflect negatively on the image of the entertainment industries. At the same time, the blockades threaten the open Internet.

“These measures bear a risk of alienating customers from the content industries and giving them incentives to adopt covert technologies such as dark nets, IP-spoofing and VPN,” Poort tells us.

“The interventions threaten the transparency of the Internet, effectively introducing censorship, Van Der Ham adds.

So how should copyright holders combat piracy? According to the researchers there is some evidence in the literature that harsher punishments and strong enforcement may be effective.

However, this also introduces the risk that customers will become further alienated from the entertainment industries. This could result in more customers going “underground” and consuming even less. Ultimately, the researchers believe that the piracy problem can be best solved by offering superior legal options.

In their Filesharing 2©12 report Poort and colleagues saw a decline in file-sharing for music over recent years, while sharing films and TV series increased. They are currently analyzing this data set to find out whether a change in legal offerings may explain these developments.

“All in all, the best strategy seems to be to arrange reasonably priced, up-to-date and easy-to-use legal supply,” Poort concludes.

Source: TorrentFreak, for the latest info on copyright, file-sharing and anonymous VPN services.