Judge Orders Google To Turn Over All YouTube User Data To Viacom
Wired's Threat Level blog says that the judge in the Viacom/Google lawsuit has made a ruling forcing Google to turn over "every record of every video watched by YouTube users, including users' names and IP addresses," to Viacom.
Viacom is arguing that it needs the data to prove that its copyrighted material is more popular than user created videos.
Wired says:
Although Google argued that turning over the data would invade its users' privacy, the judge's ruling (.pdf) described that argument as "speculative" and ordered Google to turn over the logs on a set of four tera-byte hard drives.
The judge also turned Google's own defense of its data retention policies -- that IP addresses of computers aren't personally revealing in and of themselves, against it to justify the log dump.
The EFF has responded to the ruling, calling it "a set-back to privacy rights," that "will allow Viacom to see what you are watching on YouTube. "
Judge Orders YouTube to Give All User Histories to Viacom [Wired] (Thanks, Everyone!)
Court ruling will expose viewing habits [YouTube]
This is a test using rich text formatting and html links. It's the generic "company" ad that should appear on all posts with the Company category if they don't have an ad attached to a specific company.
Post a comment
Comments:
Does this mean Viacom is going to start following the music industry's oh-so-successful model of filing lawsuits against those that consumed their content via the internet?
Google, what do you have to say for yourself? Tsk tsk.
Anybody wonder whether or not this will prompt more RIAA lawsuits for distribution of copyrighted material on YouTube?
If they intend to prove that their content is more watched than other content, why on earth do they need people's IP addresses and user names? That seems a little over-reaching.
And why didn't Ben enable commenting on his "How to comment" article?
@tedyc03: And I guess it would be the MPAA.
Clearly I'm a tool who can't read or think and I'll go hide now. :-P
Google must monitor the content, at some point? After all, people cannot upload, say, porn to YouTube, can they?
Point being, if there are mechanisms in place, why are they not applied to copyrighted materials?
The only way this should be allowed to happen is if all personally identifiable information is anonymized before the records are turned over. If Viacom is truly using this information for statistical research, there should be no need for users actual logins/IPs etc. Shame on Viacom for trying to get us to buy their explanation of "need", and shame on Google for keeping the information in the first place.
The court truly blew it on this one.
This is actually a much bigger deal. In the past, as far as I am concerned, you really couldn't get in any trouble for visiting a site and looking at its content. Except for maybe child porn. In the past, you weren't a criminal unless you hit the download button.
Now, just being on a website, you can get in a lot of trouble. The internet is heading down a dangerous road, and I hope google stands up to this. Hell, pay off some senators, do something, we can't let this happen.
I understand technically, your browser caches the youtube video, so it's similar to downloading, but come on. That is no where as criminal as going after a .torrent file of a full movie. Shame on them.
so viacom has asked for literally petabytes of data? or if google wants to a few dozen trucks of paper. Otehr have mentioned that Youtube stores all it's records for teh vdieos under video ids meaning even when viacom gests tehdata they're going to have to look up every video id to ever know what it was for and the thing is that if viacom filed for a DMCA take down all they'll get is a standard "this video has been removed due to a DMCA takedown request" messege.
I completely agree that this is a setback for privacy rights. To what end does having a user's name/IP address serve Viacom in proving whether Viacom Video X is more/less popular than User Video X? Anonymous data could serve the very same purpose.
What prevents Viacom from looking at this very same data and using it as a mailing list of sorts to attack those it believes infringed on its copyrights? In no way am I suggesting infringement is appropriate, but finding them in a manner giving them access to data of this sort is akin to saying their assuming all are guilty until proven innocent.
I hope Google/YouTube takes this to appeals (duh). I'm curious as to how they will interpret the rationale behind this ruling on privacy.
This has been the danger of Google all along. Google's motto (cute as it was) was "Do no evil", but by storing this data, the ABILITY to use it for evil was there.
For example, if you built a world-destroying weapon but vowed to never use it, the fact that you created the weapon in the first place makes you somewhat culpable when some "evil doer" (turning Google's phrase, not coining a Bushism) takes your weapon and blows everything up.
Google could CHOOSE not to store this. They could CHOOSE to store all data anonymously. But then it wouldn't have the marketability as not all the data can be linked.
But the hope was "It's Google. They do no evil. So my data is safe with them. I know they have all this on me, but I trust them with it."
And now what I KNEW would happen has happened--the government gave an order that said "gimme". I expected it, honestly, to happen due to some federal investigation, not because Viacom are a bunch of whiny bitches, but nonetheless it has happened.
Worst: it's a drop in the bucket compared to all the data Google has on us, but it now sets a precedent for the next company to pull all gmail records, and the next to pull all Google Calendar records, etc, etc, etc. Finally our search history etc. will be open. And not just to Google, and not just to the government, but to a corporation like Viacom.
This should be appealed, resisted, and even go to the Supreme Court if need be...but if Google gives in, bye bye privacy.
I wonder if turning over the Judge's personal data to a private corporation or publicly publish, it would also be a merely "speculative" invasion of privacy, in his opinion?
They should be limited to the data of videos only covered under copyrights they hold. This judge is crazy. They easily could have proved the exact same thing by providing a list of infringing videos and asking for specific information on those, but aggregate totals and statistics on all other videos.
Now they have a ton of data which according to the RIAA is more than enough to sue someone over infringement. I cannot wait till we get the first lawsuits based on video titles with the word "daily" and "show" in them that of course had nothing to do with the "daily show". Every YouTube user now has to worry about a lawsuit from Viacom weather they did anything illegal or not.
The law keeps Viacom from inappropriately using users' wacky screen names and whatever "personal data" they may have entered into the YouTube website. What that information could be, I don't know. I certainly didn't tell them my underwear size or home address.
The biggest fear from this should be that Viacom will finally be able to prove what everybody knows but pretends they don't ever see or understand: users spend a lot of time on YouTube watching and illegally uploading copyrighted music videos, TV shows and movie clips. Surprise.
I wish they'd do the same for P2P, so they can have numbers. Then all those people so hell bent on damning Comcast et al. for infringing on their "right" to transfer gigabyte-size files over the network 24 hours a day will no longer be able to keep a straight face when they say they don't understand why copyright holders should be suspicious of them and would want more accountability.
Do the logs differentiate between 5 seconds of playback and me going "This sucks, close window" and watching the entire thing?
@Nyses: In another article about this on ArsTechnica ([arstechnica.com]), it's stated that Google's code "scans incoming videos against a database of video fingerprints; matches are flagged for human follow-up". Google would need a video fingerprint of the TV show (for example) from Viacom, then they could scan their db for possibly infringing content.
As far as porn goes, I think there have been numerous cases where porn or illegal material (eg Girls Gone Wild-style stuff with minors) has been taken down after the YouTube community flags it as such. There's far too much incoming data for Google to review every video.
Sounds like a good time to organize an internet boycott of all things Viacom.
Although I am not sure I really watch any Viacom stuff anyway.
@logicalnoise: I'm never really a grammer nazi, but I had to read your post 4 times to decipher your words.
Anywho, congrats to the judge for catching that data retention policy/ IP address slip up on google's part. It would appear Google's legal department didn't debate in high school...
@Nyses: the community at large monitors what is uploaded, and then flags it if its inappropriate. I've seen several 10 minute long full on porn videos posted that took months to remove, because people were uploading them and then setting them to "friends only."
@InThrees: depending on how they log, sorta... while it won't give you an exact length that you watched, if you browsed to another page, you could compare date/times with video lengths...
And it was my understanding that Viacom is requesting access to ALL videos that have been removed, not just the ones that they specifically own. Why should viacom have access to content it doesn't own, and has no claim to? How would that prove anything?
In my experience, the most popular videos on youtube are the ones of girls shaking their half naked asses to hip hop... Sorry viacom.
@sleze69: Probably b/c of the fallout from when Gizmodo did it and what that devolved into, namely repeating everything the article said not to repeat, and people apologizing for breaking the rules once or twice, just in case.
Does anyone know how MUCH info they will be handing over? By how much, how many billions of records and how much data in file size. It's almost the def of information overload. I for one have been known to watch 10 videos in a row. Hell, I wonder how many pages "Battle at Krueger" is going to be. And what about embed videos. Is there a record for every page load?
Ah! That explains the sudden absence on YouTube of a whole slew of videos...
Just to clarify both the consumerist and wired post - they had to turn over "user names"...not "user's names" - if that makes any difference.
I could see how IP addresses / user names could be needed, instead of pure anonymous data. There's some strength to "our video was watched 10,000 times by 10,000 different users" instead of "10,000 times by 10 users".
Still, yes - you could still sterilize the data and keep the same "multiple views" information...but that would take time and effort (money).
@Git Em SteveDave wants a Lego build buddy: For user name, ip address, video ID, date and time - let's conservatively assume 100 characters.
4,000,000,000 bytes / 2 bytes per character / 100 characters per log entry = 20 million log entries
Now if they compressed the log data instead of a straight text file...oh boy...
@Gilbert: Because it's harder to fake numbers when there's real data backing them up? Google could just say "two people viewed your infringing content, omg swaerz!". If Google has to supply the IP addresses and usernames (if available), lying about it becomes a lot more difficult. I don't agree that giving users' IP addresses out is A Good Thing(tm), but I can understand where Viacom is coming from.
@InThrees: They can probably determine that based on how much data was sent before the connection closed.
@Concerned_Citizen: Yes ok, but how do you identify videos on YouTube owned (intellectual property-wise) by a given entity? Google has their own fingerprinting technique, but an h.264 encode will fingerprint differently than an xvid encode, and resizing both will result in different fingerprints... so how many permutations do you check against?
@XianZomby: "The law keeps Viacom from inappropriately using users' wacky screen names and whatever "personal data" they may have entered into the YouTube website. What that information could be, I don't know."
What law is this that you speak of?
O, and you should leave the internet immediately. Do not pass GO.
Viacom's about to find how much porn gets deleted (and that most users are looking for this verboten porn, not 10 minutes of some shitty TV show).
This looked more like a motion by Viacom to gleam every bit of source code, and operating procedures of YouTube, then an attempt to thwart piracy.
Viacom wanted source code of the Video ID code, their search, user search keywords to their videos, database schemas, and advertising.
Google was their own undoing for the ruling on user ID and IP addresses (see pages 11 - 14 of the judges ruling).
I totally agree this is a very dangerous step... But aside from the (im)morality of this, the supposed utility is equally questionable. One of the claimed objectives for requiring this information is to look at the relative popularity of copyrighted videos compared to user-generated videos. But how could this possibly be done without knowing the copyright status of each video? Anyone who has looked for a music video or concert knows most of the returns are user-generated knock-offs. Titles aren't useful in determining copy-right status, and even watching the video itself it can be very difficult to know if the content is copyright infringing. This is a data-grab, plain and simple.
@NikkiSweet: They'll need Nedry to get in there to debug all the lines of code. That's one big excel file.
@howie_in_az: I understand the implications, but again, not only do we have to protect our right to privacy, but we have to operate under the notion that Google and YouTube are innocent until proven guilty and that they will abide by their sworn submissions.
It's up to Viacom to prove any attempts to alter the numbers on the defendants' part, not up to them to unnecessarily view sensitive data under the guise of preventing otherwise.
@sleze69: The other guy pretty much nailed it on the head. The first instinct when an article says "don't respond with 'lol' is to respond with 'lol'".
We're all a little childish like that sometimes. Maybe just me.
@Git Em SteveDave wants a Lego build buddy: "Yeah, so I'm going to need you to come in on Saturday..."
At the risk of being "agist", Judge Louis Stanton is over 80 years old (Reagan appointee, 1985). How come I detect another believer that the internet is "a series of pipes". As someone else suggested I hope Google hands over the data on trucks full of the data printed out from old dot matrix printers... Good luck digging through that...
@greatgoogly: Ahhh, I can just imagine the sounds of a thousand dot matrix printers running with a box of paper in the front, a box in the back, and a mile long sheet of that old connected paper with the tear off holed sides. Then someone over at Viacom has to pay someone to rip each page apart, and also the edges off. I see some guys in hard hats and hearing protection, carrying little oil cans, and walking in between the printers giving them some oil now an then. Ahhh, the deafening clacking!!!!
So if I am surfing through Ultrasurf (or TOR for that matter) I'm aok right?
@Andronicus1717: What law? This ruling by a federal judge, which I consider to be "law":
"Google has been ordered to turn over YouTube user data to Viacom. But Viacom will be guilty of contempt of court if it uses that data for anything other than specifically proving the prevalence of piracy on YouTube, a source close to Viacom told CNET News.com on Thursday.
That's serious business. Contempt of court is the sort of thing that can get a lawyer's license taken away.
On Wednesday night, a federal judge ruled that Google must turn over YouTube user activity--videos watched, IP addresses, usernames--to legal foe Viacom as part of a long-running copyright infringement case. But the source told CNET News.com that a heavy protective order is in place that will keep individuals' personal information cloaked."
Perhaps you should use the internet more. You might learn something.
Youtube keeps a record of views for a particular video, is that not good enough for them to know how often content is watched?
I can't believe how badly the court system screwed up here. It is clear what Viacom's interests are, they don't need IP addresses for their "study". This information can be clearly used to obtain identifiable information, that is why they want it.
You mention this article regarding a "protective order", but the judge ordered the IP addresses released too, which is it? With an IP address, they can just go to the ISP and demand that information, with no proof of anything, as they always have.
It's time to start class-action suits against VIACOM.
@newfenoix: I agree. My user name on Youtube is my real name (like here), so unless Google anonymizes it, then I think I fall in to the EFF's class of non-anonymous users:
If any single one of the YouTube users in the Logging database picked a Login ID that does identify that user (i.e. if my YouTube login was kurtopsahl), then the Logging database' information about viewing habits is protected by the VPPA, even if others pick anonymous pseudonyms. [www.eff.org]
If anyone could guess from my first post here, Viacom will soon know how much time I spend on Youtube searching for "sexy lesbian girls kissing" or "stripper babes pole dancing hot".
@Michael Belisle: When I say "I agree" I mean "Wait until Google announces their intentions with respect to this ruling, and then see why my lawyers say."
notably..or perhaps not...they don't mention the actual videos in there. I think gogole shouuld include them for free. Just fing brain dump all of youtube on viacom and watch them struggle to breath. Datamining may by a neat new tool but I highly doubt viacom is equipped to handle what would be petabytes of information.
You'd think from sheer volume of the information it would be useful.
Useless. Useless! Totaly changed my comment's message right there.
They're trying to prove that THEIR copyrighted videos are more popular than user-created videos, or that ALL copyrighted videos are? I read the Wired article twice, but the wording still confuses me.
Having said that, I think this sucks.