Time for YouTube to get serious about cleaning up all the junk, spam and malware files on its site
YouTube is great for finding videos about pretty much everything. Need to learn how to fix a furnace or use the latest camera equipment? There’s bound to be a video shows you how. Unfortunately, amid the useful stuff, YouTube is also chock full of garbage. The question is, with its massive technical resources, why doesn’t the site do a better job keeping house?
I’ve written before about the epidemic of fake “full-movie” uploads that fill YouTube. That was in 2012. Now, six years later, the problem still exists. Apparently, YouTube isn’t concerned that its pages are full of spam files, many of them fake pirate movie uploads that lead users to sites rife with malware and money-making scams.
These fake uploads, promising full copies of hundreds of films, both indie and mainstream, are easy to find. Go to YouTube, search for a specific film title using the term “full movie,” and voilà, most results will lead to garbage. These bogus uploads fall into two categories. Some offer links to other dubious websites while others are merely dummy files uploaded to generate advertising income(for the user and YouTube). Some do both.
As for these offsite scam movie/gaming portals, it’s difficult to figure out who is actually behind them. The site URLs vary and include tzarmedia.com, gnomicfun.com, cnidaplay.com, jabirufun.com,flogame.com, among dozens of others. Curiously, a WHOIS search for these various domains indicates they are all registered via the same domain registrar, enom.com. One can’t help but suspect that this particular business model is being orchestrated by a few, linked operators. When I called their customer service number to ask questions I was given the proverbial run-around. Other contact information was essentially non-existent.
These scam movies sites share domain registrar and likely more
Why are such dummy files an issue? Not only do they pollute legit searches for content on YouTube, but they make the process of reviewing pirated content more difficult for rights holders. When I search for copies of my film using my Content ID account, I have to wade through dozens of these fake uploads.
Removing them is an incredibly time-consuming task as it seems YouTube has purposely chosen to make the Content ID dashboard as inconvenient as possible for users.
When I get page of results that is nothing but dummy uploads why can’t doesn’t YouTube offer a select all option so that I can remove them en masse? Instead–if I want to remove them–I’m forced to click and open each one and go through a 5 step process: select takedown, select title, acknowledge, fill in my signature and then click takedown. Instead, why not offer a select all option?
Another interesting twist is that many of these fake movie uploads also share links to legit social media sites like MTV’s Facebook or The Wrap’s Twitter account. I checked to see whether the operators of these sites knew about this and was assured they didn’t. It would appear these dummy uploads include such links drive more traffic to the bogus uploads and make them seem legit.
What can YouTube do to prevent this scheme? Why not utilize their own fingerprinting tech (Content ID) to detect and block these dummy files? If necessary, why not employ a team of actual humans to help with the task. I imagine if their engineers put their minds to it the task would be a relatively simple one. Certainly YouTube can afford to invest in keeping its house tidy?
Why is do advertisers allow themselves to be part of this junkyard?
Perhaps YouTube doesn’t take action against such uploads because it makes money off them? Here are just a few examples I found recently–bogus uploads with advertisements for New York Life, Walmart, Tide and Walgreens. These fake pirate full-movie uploads emblazoned with ads are a dime a dozen on YouTube. Do these advertisers know what they’re paying for? Do they care? Perhaps TAG, the Trustworthy Accountability Group, should take a look at this situation and pressure YouTube to take action.
Can you think of any other business that could get away charging money this for type of thing? Isn’t Internet commerce–and YouTube–mature enough at this point to operate a business where what you see is what you get? Apparently not… Imagine walking down the aisles of Target and finding half the merchandise to be knock-offs or empty boxes?
It’s not only the spam fake movie uploads and advertising scams that are problematic. As a study by the Digital Citizens Alliance found, YouTube is also rife with uploads that link to various types of malware including RATS (Remote Access Trojans), used by hackers to install malware that hijacks computers of unsuspecting internet users. Why is it OK for YouTube to continue to allow activity that scams–and possibly endangers–users?
As I mentioned, YouTube has the technical expertise and financial means to develop better algorithms and Content ID matching to weed out these garbage uploads if it chose to do so. Until then, the site will increasingly resemble a hoarders home with junk stuffed into every conceivable corner. Is that any way to run a business?
Why not make Content ID more accessible and transparent?
Much has been written about YouTube’s Content ID program, a fingerprinting technology that allows rights holders to find and claim their music or movies when uploaded to YouTube. The technology was introduced in 2008 in the wake of Viacom’s lawsuit against YouTube and since then has helped (some) creators mitigate the problem of piracy on the popular UGC (user-generated content) site.
Those who have access to the Content ID system can uploaded reference files and use a dashboard to choose how matches should be handled. They can be limited based on audio, video, and length. Matching content then can be blocked, removed, or monetized based on territorial rights.
I’ve written many pieces about what works and what doesn’t when it comes to Content ID so I won’t be redundant here, but this week I read some pieces which highlight some lingering issues that continue to limit the reach (and effectiveness) of this technology–most notably limited access and accountability.
Are audiobooks being ignored?
Author Ryan Holiday published a piece in The Observer asking, “When Will YouTube Deal With Its Audiobook and Podcast Piracy Problem?” He described how using search, he’d found the audio version of one of his books streaming, in full, on YouTube. The audio had been streamed 16,000 times. He observed, “It might not seem like a ton but the book had sold about 50,000 copies in audio—an additional 30% of that figure pirated it through a single video?”
Holiday goes on to repeat the oft-heard lament of filmmakers, musicians, et al who have found their pirated works streaming on YouTube. Like many of them, Holiday believes YouTube needs to make it easier for artists to protect their creations from this type of theft:
For its part, YouTube needs to get its act together and offer tools directly to publishers and authors. Audiobook piracy is real and clearly growing. The idea that songs and television and films all deserve protections from Content ID but authors don’t is absurd.
Now, to be fair, it’s not clear that Holiday himself has ever applied for Content ID access. It seems that the YouTube’s language dissuaded him.
I can continue to file these claims as the author but since I’m not a major publisher with a “substantial body of original material,” I can’t participate in YouTube’s Content ID personally.
If I were an author (or publisher) I would not hesitate to at least try to apply for a Content ID account. I had no difficulty being approved for a Content ID account in 2010 and only own the rights to two films, a feature and a documentary that I co-produced. The companies that distribute my film also have Content ID for their film catalogues.
…an evaluation of each applicant’s actual need for the tools. Applicants must be able to provide evidence of the copyrighted content for which they control exclusive rights…Content ID applicants may be rejected if other tools better suit their needs
The key here is the last sentence. There’s also Content Verification which seems to be another, higher tier of Content ID aimed at large companies. The other “tool” is simply sending a takedown via web form. That may be appropriate in very limited situations, but using it can be time-consuming and also requires a rights holder be proactive in searching for infringing content. Since Content ID does that work for you it’s the key reason it should be made available on a much wider scale. As Holiday noted, one copy of his book was accessed thousands of times….shouldn’t the onus be on YouTube to put a roadblock up to prevent infringement? Why should creators be required to be copyright cops too?
Lack of transparency has long been a flash point for musicians vs. YouTube
Frustration over YouTube’s Content ID and monetization scheme was also at the core of a blog post by 5-time Grammy winner, musician Maria Schneider published on Music Tech Policy. Schneider has become a strong advocate for artists rights and attacks Content ID on a number of levels including skewering YouTube over its lack of transparency (a common complaint among many musicians) and for its apparent refusal to grant access to artists–even those like her–who are Grammy recipients.
Content ID is reserved for big record companies with big catalogues, and probably selected independent artists whom YouTube believes will make YouTube a heap of money…In the press, YouTube has fought back against the recent flood of criticism, saying that all rights-holders can access Content ID – that they can get it through “third-party vendors.” These third party vendors often take between 20% to 50% of the revenue paid by YouTube—after YouTube takes its share.
I highly recommend reading Schneider’s post for her account of how Content ID has failed her and other musicians. As a filmmaker, in terms of using Content ID simply to combat piracy, my experiences have differed, but I do share many of her overall concerns. Unfortunately, the fact she (and others) are apparently being denied direct access to Content ID tools is an ongoing issue that inflames the lingering, legitimate mistrust creators have with Google.
Does it really need to be this way?
In spite of years of ongoing complaints like these, neither Google nor YouTube seem really to have made a genuine effort to work with artists–instead preferring to stonewall or sidestep debate. When YouTube officials do comment, they often find themselves tangled up in webs of their own making, as was the case when musician Zoe Keating took them to task in a very public exchange. Why not work with creators to solve some of these problems instead of demonizing them? There are ways to find common ground if only the powers that be at Google would care to (really) listen.
Content ID could make it easier for creators of all stripes to ask permission instead of simply taking content
Moving the Content ID machinery out of the shadows could pay dividends in other ways–perhaps by helping bridge the divide in disputes over copyright. Maarten Zeinstra, an “advisor on copyright and technology” recently penned a thoughtful blog post proposing that YouTube’s Content ID could become a useful tool for those who interested in utilizing content in legitimate ways. His piece, “YouTube should open up Content ID” was published this past May on the Dutch website, Kennisland. In it, Zeinstra noted:
YouTube should open up Content ID and make their register of rights holders publicly available...Let anyone be able to contact the rights owners of a certain clip or publicly contest that ownership. This creates an innovative new possibility in using content with permission or under copyright exceptions, and create legal certainty for copyright holders and remixers alike.
Now, I’m a tad skeptical as to what he means by “contest that ownership” but I’m open to the possibility that he’s merely describing a middle ground. If someone wants to make a mashup video using clips from my film or segments of archival footage from my documentary they could use Content ID to find that I am the rightful owner of the footage and can ask permission. Personally, I support the idea of fair use but also appreciate the fact that one should ask permission. I did so with footage used in my documentary and was always met with a positive response. Perhaps opening up Content ID in this way could foster a new respect for the work of creators and support the idea of asking, and not simply taking.
Improving Content ID would be in everyone’s best interests
Clearly, Google needs to do a much better job in providing access and accountability with its Content ID and monetization programs. Expand outreach to indie artists. Includethem in discussions about how to improve Content ID. Update the interface to make it more intuitive and user-friendly. Open the books so that creators can see exactly how much revenue is earned and where it goes. Be innovative and use Content ID to open new avenues to legitimate use of copyrighted content.
There’s little doubt that YouTube’s Content ID is a powerful tool that’s in dire need of an overhaul, both in terms of who uses it and how it’s used. This technology could provide so much more than it now does–but the ball is in Google’s court. I won’t hold my breath waiting for something to change–but there’s always hope isn’t there?
YouTube users claim Fair Use as a defense for uploading full copies of pirated movies
There was a lot of talk about fair use and takedown abuse at last week’s U.S. Copyright Office Section 512 roundtables in San Francisco. Many of those who spoke, bemoaned how poor, innocent uploaders were victimized, time after time, by malicious DMCA takedowns.
It’s a tried and true talking point, convenient, but disingenuous all the same. Some of us, myself included, tried to make the point that creators, whose work is routinely (and massively stolen), are often (doubly) victimized bymalicious fair use claims.
I thought I’d share an example of this that occurred just this week on YouTube. On Tuesday a full-copy of the Swedish indie film “Kyss Mig” (all 147 minutes of it) was uploaded to YouTube by a user aptly named “Free Movies.” As an added flourish, the user-name included the notation, “free movies bitches.”
In this instance YouTube’s Content ID system worked as intended. The Content ID user (an indie film distributor) had set the system to block uploads of a certain length in its territories. Even though the video was a full, pirated copy of the film, it wasn’t taken down, it was simply blocked. So far, so good right?
Wrong…This YouTube user didn’t seem to think the rights holder had the right to block the full, infringing copy and promptly disputed the block. S/he stated the reason as being:
Approval from copyright Holder is not required. It is fair use under copyright Law.
The user also added a note: “I don’t need to explain.” Clearly Free Movies didn’t bother to read YouTube’s information on disputing a claim or its explainer on fair use.
Despite all the testimony at last week’s roundtable about fair use–and how copyright holders seek out to punish those who claim it using malicious takedowns–it’s worth pointing out, yet again, that for every legit “fair use” claim, there are also false, and rather malicious, abuses of that defense. It’s a fact conveniently overlooked by the anti-copyright apologists.
Bogus “fair use” claim on YouTube
Take a gander below at the actual screen caps documenting this bogus “fair use” claim. Hopefully, officials considering DMCA reforms will acknowledge that creators can be twice victimized by abusive fair use claims.
I did in fact “reinstate” the claim (on behalf of the indie distributor I work for) so we’ll have to wait and see if this user goes on to file a counter-notice. If s/he does so, the film, in its entirety, will return to YouTube even though it’s CLEARLY infringing because we don’t have the financial resources to enforce the removal in federal court.
I’ve had the same thing happen after full pirated copies of our film were uploaded to YouTube. For creators trying to protect their work it’s a lose, lose…Perhaps YouTube should require it’s users to review “fair use” and “copyright” before they are allowed to uploaded content of a certain length? Why should creators be twice victimized while uploaders walk away unscathed?
139 DMCA NOTICES LATER, NOTHING CHANGES FOR THE GOOGLE DRIVE ACCOUNT
As I wrote previously, Google seems to ignore its own pledge to disable accounts of repeat (piracy) infringers. Today I sent another 31 DMCA takedown requests this week (170 over several months) reporting this same account for copyright infringement on behalf of indie film distributors I represent. So far, Google has removed 139 pirate links since last Aprilyet the account remains online sharing links to several hundred pirated films. I ask again, why is this account still active? After all, isn’t eligibility for protection under “safe harbor” dependent upon implementing a reasonable repeat infringer policy?
(i)Conditions for Eligibility.—
(1)Accommodation of technology.—The limitations on liability established by this section shall apply to a service provider only if the service provider—(A)has adopted and reasonably implemented, and informs subscribers and account holders of the service provider’s system or network of, a policy that provides for the termination in appropriate circumstances of subscribers and account holders of the service provider’s system or network who are repeat infringers; and(B)accommodates and does not interfere with standard technical measures.
All these DMCA notices were reporting links from one Google Drive account
Respect copyright laws. Do not share copyrighted content without authorization or provide links to sites where your readers can obtain unauthorized downloads of copyrighted content. It is our policy to respond to clear notices of alleged copyright infringement. Repeated infringement of intellectual property rights, including copyright, will result in account termination. [emphasis added] If you see a violation of Google’s copyright policies, report copyright infringement.
Of course Google no one at Google will respond to queries, so I guess creators are left, once again, to ask “How much is enough?”
MUSO’s Global Piracy Insights Report 2016 – Click for more
A report in today’s Torrent Freak noted that content protection firm (anti-piracy) firm Muso recently released its annual Global Piracy Insights Report for 2016 so I was prompted to take a look to see what what’s new on the piracy landscape. According to the report there’s been a, “massive shift towards direct downloads for music content – growing by 31% in 2015” In addition the report found that “28% of all visits to piracy sites in 2015 were through mobile devises, up 8% during the year.”
Viewing habits for pirated movie watchers also seems to have shifted over the past year as more and more users to viewing streamed content instead of downloading torrents. The study examined traffic from 14,000 pirate websites, encompassing 141 billion visits, and according to an analysis of the report on Advanced Television, discovered this trend:
Out of a total 78.49 billion film and television piracy site visits, 73.69 per cent (57.84 billion) were visits to streaming sites…the second most popular piracy delivery type was torrents, capturing 17.24 per cent of audience visits.
The report is available to subscribers only so I cannot delve deeper into the figures but I’m not surprised to see streaming gain a growing foothold as the favored viewing platform. Pirates, like the rest of us, have grown accustomed to watching shows streaming on Netflix, HULU and Amazon. It’s no wonder the same patterns persists when watching pirated fare.
For those who care about the impact of piracy on musicians, comes this unfortunate news:
2015 saw a 25% rise in the use of YouTube ripper sites, used primarily for downloading mp3’s from YouTube music videos. The ripper piracy from mobile devices overtook piracy from desktop devices, growing by 46% last year. The usage of these sites is far larger than many realise, in fact making up 17.7% of all visits to piracy sites for music content.
One piece of apparent good news from the report is that, according to MUSO researchers, “…piracy levels remained relatively throughout the year, with a 5% overall decline.”
Andy Chatterley, MUSO CEO ultimately focuses on what can be gained by studying these trends noting the report as helping creators develop a framework by which nudge consumers in better (legal) directions:
This report gives a complete picture of the piracy landscape and identifies key insights into piracy audience and behaviour. The Global Piracy Report is hugely valuable to right holders and for the first time looks at all forms of piracy traffic and not just p2p usage. In understanding the scale and mechanism of the audience we can be better informed to re-connect this audience to legal content.-MuSO press release
The report also found that streaming piracy in both the United States and UK was trending down, “likely to be due to legal music and video streaming services such as Spotify and Netflix.” But, before celebrate too much it also noted that in many countries, streaming piracy is actually increasing. What’s that old saying, two steps forward, one step back?
Facebook has been promising for some time to introduce tools that would allow rights holders to automatically detect and remove pirated content from its pages.
The company has endured a lot of bad publicity around the freebooting of viral YouTube videos on its pages, but Facebook’s also long been a place where pirated movies and music found a cozy habitat. That is–until now. I’ve recently begun to utilize this tool to manage Facebook DMCA takedowns and wanted to share my first impressions, but first a bit of background.
First of all, I’m thrilled that Facebook, with all its resources, has finally begun to take copyright infringement seriously. In introducing the new tool last month the Facebook development team explained why the company had finally stepped up:
Video has become an important part of the Facebook experience for people around the world, due in large part to the amazing creativity we’re seeing from all kinds of video publishers.
To provide the best experience for everyone who watches, creates and shares videos on Facebook, we work with our community to understand which tools they want us to build. Based on this feedback, on top of the measures we already have in place, we’ve been building new video matching technology to further help rights owners protect the content they own.
Signing up is easy and the interface straightforward and simple to use
I found signing up for the rights manager tool to be relatively straightforward. You must have a page to link the rights manager to and I initially applied for, and was accepted into the program, by using our film’s Facebook page. Once I received approval I was able to upload a reference copy of our film (and trailer) to the Facebook rights manager dashboard. A trailer I’d uploaded to our page previously was also listed. From there, Facebook’s automated digital matching tools went to work.
Facebook’s Rights Manager dashboard is pretty straightforward
Easily upload and maintain a reference library of the video content they want to monitor and protect. Publishers can upload content libraries and publish live video as references for Rights Manager to check against, including videos they are not sharing publicly on Facebook. Rights Manager then monitors for potential infringement of that content across Facebook.
Create rules about how individual videos may be used. Publishers can set specific match rules to either allow or report copies of their videos based on criteria of their choosing—for example, how much content has been reused, where the matching video is located or how many views the matching video has received.
Identify new matches against protected content. Rights Manager’s dashboard surfaces any new matches against a publisher’s uploaded reference files and live video. On the dashboard, publishers can filter matches by time, date or view count, and then either report potential copyright infringement or allow the matching content to remain published.
Whitelist specific Pages or profiles to allow them to use their copyrighted content. Publishers can specify Pages or profiles that have permission to publish their protected content without being monitored for potential infringement.
Protect their reference library at scale with the new Rights Manager API. We’re rolling out an API for Rights Manager to improve bulk uploading for publishers and to allow media management companies to support partners in managing, monitoring and protecting their content across Facebook. You can find out more about the Rights Manager API here.
Facebook’s tech support is responsive and proactive in working to improve the system
Facebook asks for feedback in an effort to improve its rights manager tools
I do believe this type of fingerprinting technology will be an increasingly crucial tool as we move forward in the battle against online piracy on sites like Facebook, but as with any new offering, there are glitches.
The good news is that so far, Facebook’s technical support team is quite responsive and the company seems to be making a concerted effort to sort through issues and improve the tool’s operation. Any time you remove an item from the dashboard a window pops up soliciting feedback. I’ve also had a fair amount of helpful email correspondence with the support team and have found Facebook’s prompt and open response to my queries offers a welcome contrast the less-than-stellar support offered by a (popular) site that shall remain nameless.
As with any new tech, there are some glitches
I also set up a Rights Manager account for an independent film distributor I work for and in the process of uploading dozens of reference files have found the “matching” to be rather hit and miss. At this point Rights Manager seems to do a great job detecting the company’s opening logo (and music) but little else. What makes it even stranger is that the tool detects the distributors opening logo and music and then matches it to the wrong reference file. Obviously ALL the titles I’ve been uploading share the same opening sequence from the distributor but when it comes time to actually issue the takedown to remove the infringing (matched) content, it auto-populates the form with the film’s title, which in these instances is the wrong one.
Lots of early glitches with Facebook’s Rights Manager tools
I’ve also come across situations where a single film title is simultaneously listed has having matched multiple reference files to different titles, but NEVER the actual reference file for that particular film. Consequently, rather than send a DMCA notice with incorrect information, which would be illegal, I have chosen to wait for Facebook to sort out this particular glitch. This is where their responsive tech support will, hopefully, come in handy.
I’ve also found that there’s a lot of uploaded content that doesn’t really match anything. Perhaps a song is playing in the background that matches the film’s soundtrack, but it’s difficult to tell? At this point the system’s matching capabilities clearly need to be dialed in order to better weed out innocent content.
As it stands, I have been manually removing this erroneous matches from the dashboard, but that takes precious time, and efficiency is one reason this system was developed in the first place. For larger entities there are API tools, but for independent, smaller entities, it seems that utilizing the dashboard will be best route.
Users can create “match rules” to fine tune content matching
Some of the hiccups I’ve encountered thus far are likely simple bugs in the system, while others may well be user error. Fortunately, Facebook has created tools that allow publishers/creators to fine tune the matches based on length of time, territory and content type.
I plan to spend some time working through the reference files I’ve uploaded to create appropriate match rules in the hope that it will result in fewer false positives.
Will creators be able to make money from their videos and music?
There’s also the question of monetization. Will rights holders be able to earn money from copies of their work uploaded to Facebook? It’s likely at some point in the future, but first Facebook will need to fine-tune Rights Manager. They can’t afford to complicate a system that’s still for all practical purposes in beta mode.
Overall I’m pleased with Facebook’s effort. Yes, it’s overdue and yes, it’s not (yet) perfect but it is a huge step in the right direction and hopefully can serve as a model for other social media and video sites across the web looking to do a better job thwarting piracy.
As I’ve written previously, I firmly believe UGC sites of a certain size (like Facebook, Vimeo, YouTube, et al) should be required to offer this type of tech in order to qualify for safe harbor. Of course that assumes the creaky old DMCA will be revised and the odds of that actually happening any time soon….well, I’ll leave that discussion for another day. In the meantime, I’m going to get busy on Facebook and upload some more reference files. So far I’ll give the new system a thumbs up!
As an indie film and broadcast journalism veteran, I'll share my perspectives on issues of interest to the creative community and beyond--Ellen Seidler
As for these offsite scam movie/gaming portals, it’s difficult to figure out who is actually behind them. The site URLs vary and include tzarmedia.com, gnomicfun.com, cnidaplay.com, jabirufun.com, flogame.com, among dozens of others. Curiously, a WHOIS search for these various domains indicates they are all registered via the same domain registrar, enom.com. One can’t help but suspect that this particular business model is being orchestrated by a few, linked operators. When I called their customer service number to ask questions I was given the proverbial run-around. Other contact information was essentially non-existent.
Removing them is an incredibly time-consuming task as it seems YouTube has purposely chosen to make the Content ID dashboard as inconvenient as possible for users.
Another interesting twist is that many of these fake movie uploads also share links to legit social media sites like MTV’s Facebook or The Wrap’s Twitter account. I checked to see whether the operators of these sites knew about this and was assured they didn’t. It would appear these dummy uploads include such links drive more traffic to the bogus uploads and make them seem legit.
Can you think of any other business that could get away charging money this for type of thing? Isn’t Internet commerce–and YouTube–mature enough at this point to operate a business where what you see is what you get? Apparently not… Imagine walking down the aisles of Target and finding half the merchandise to be knock-offs or empty boxes?
Why not make Content ID more accessible and transparent?
