In what is bound to be a controversial element, the committee recommended that Internet service providers (ISPs) must be held liable for copyrighted material that goes through their systems. To be exempt from that liability, the ISPs must show they are acting as true "intermediaries," without actual or constructive knowledge of the content.Boing Boing's incomparable sysadmin, Ken Snider, a Canadian geek, wrote:
ISPs should be required to comply with a "notice and takedown" system against subscribers who violate copyright laws.
It is *extremely* important to me that our government not bow to CIRA. I have high hopes that the current minority gov't means they won't deal with this anytime soon, but I *want* to get the message out to every damned MP I can get my hands on. The problem is, I don't have any *specific* information on these provisions. I was hoping you would, or at least, could point me in the right direction (or even champion the cause with me! Woo!).So, Ken, here're some answers for you.
It's *critically* important to me that Canada doesn't follow the US in this process. I'm prepared to do whatever it takes to make the reasons *why* this is a shitty idea known, I just need some help making my points clear and concise, as well as containing the appropriate amount of "politik" that they'll make a difference.
Copyright is a system for regulating technology -- it regulates technologies used to make and distribute copies. We have lots of technology regulation in the world: there are rules that govern the operation of automobiles and rules that govern the marketing of electrical appliances. This isn't per se wrong.
But when the 20 horsepower locomotive was invented, the blacksmiths weren't able to successfully lobby to have 80 horseshoes welded to each engine, despite the rule that said that every "horse" used for transport needed four shoes. When you invent a railroad, you need railroad-rules for it, not horse-and-buggy rules. The facts that the railroad doesn't need shoes, or oats, or curry-combs don't reflect bugs in railroading: they are the feautres of railroading.
The Internet has one overarching feature that makes it superior to the technologies that preceded it: it can copy arbitrary blobs of data from one place to another at virtually no cost, in virutally no time, with virtually no control. This is not a bug. This is what the Internet is supposed to do.
It was really foresighted in 1996 for WIPO to sit down to update copyright law to suit the Internet. They recognised that the Internet was a fundamentally different thing from the technologies that came before it, and of course, a new technology needs new rules and regulation.
But WIPO got it horribly wrong. The approach that WIPO took to regulating the net was to create a set of rules that tried to make the Internet act more like radio, or TV, or photocopiers -- like all the things that it had already made rules for. The WIPO approach treated the ease of copying on the net as a bug, and set out to fix it.
Notice-and-takedown is an area where WIPO got it drastically, terribly wrong.
If you own a restaurant, you're not responsible for policing your customers to ensure that none of them are carrying stolen bank-loot. If someone burst in and pointed at the guy at the back table and said, "He's wearing my hat!" no one would blame you if you didn't wrestle the hat away from him and give it back to the accuser.
But under notice-and-takedown, this is what we ask of our ISPs. If you allow users to host stuff, you're responsible for what they host. If they put an infringing file on your server, you're required to know what they've put online, and you'll share in their punishment if you fail to block them from posting infringing material.
Now what is and isn't a copyright infringement isn't anything like a clearcut issue. ISPs aren't equipped to evaluate what's infringing and what isn't -- hell, even Supreme Court judges have a hard time figuring it out. Operating a server doesn't qualify you to understand and evaluate copyright law.
So there's a get-out-of-jail in notice-and-takedown. If you respond to accusations of infringement by taking your customers' materials offline quickly, you won't share in their liability. Now, given the kinds of penalties available to rights-holders for online infringment (in the US, it's $150,000 per infringement!), it's not surprising that most ISPs avail themself of this "safe harbour," removing stuff whenever a complaint comes in.
But a complaint isn't proof -- someone who rings up your ISP and says, "That file infringes on my rights" is like the guy who busts into a restaurant and shouts, "That guy is wearing my hat!" There's no way for an ISP to evaluate whether he's genuinely aggreived, whether he's nursing a grudge, whether he's just a nut. In the US, nuts, grudge-nursers and flakes all use notice-and-takedown to censor the Internet and get material removed.
Usually rights-holders will counter that this can be addressed through something called "counter-notification," where an ISP that's removed something is given the right to contact its customer and say, "This guy says you infringed his copyright. If you disagree, let us know and we'll put your file back online and you two can sort it out in court."
But in practice, counter-notification is a rare beast. Most ISPs just do the math and decide that sending a single counter-notification letter will cost them more in lawyer-hours than the customer in question will ever make for them. They just invoke the termination clause in nearly every ISP contract and shut down the account.
This is why notice-and-takedown is a near-perfect tool for censorship. Don't like what your critics have to say? Just sent a takedown notice and poof, it's gone! The Scientologists love this tactic -- they even get Google to remove links to sites that are critical of their "church" by asserting copyright infringement. Have a look at the truly chilling annals of ChillingEffects, which gathers up takedown notices and other nastygrams. The takedown notice is the favourite tool of the crank, the censor, and the bully.
Even when applied to genuine copyrighted works, takedown is dangerous to the point of unusability: the Business Software Alliance, MPAA, and RIAA send out automatically generated takedowns by the thousands, using software that does half-assed pattern-matching on files available on the net and then sending off letters to universities, ISPs and other entites demanding the takedown of book reports about Harry Potter, Linux distributions with the same names as movies, and academic work by professors with the same name as musicians.
What's more, notice-and-takedown is almost always accompanied by systems for peircing Internet users' anonymity: if you want to find out your stalking victim's new address and number, you need only find the message-board where she's posting about her troubles and write to the ISP as an infringed-upon rightsholder, demanding her info.
If Canada wants to "solve" the problems of the Internet, it should be looking to find "Internet-native" solutions. Canada's Internet laws should treat copying as a feature, not a bug. It should empirically evaluate which sectors are negatively impacted by file-sharing (mounting evidence suggests that almost none of the entertainment industry's woes can be blamed on the net) and then solve those industries' problems with blanket licenses and other tools that don't seek to regulate copying, something that's impossible to do without breaking the Internet.
Solutions that approach the Internet as a problem are no solutions at all.
Write to your MP, write to the Ministers -- This Slashdot poster collects the contact info for numerous government officials who are involved with this. Keep Canada's Internet free. Link (Thanks, Ken and Ian!)
Update: This post drew a lot of attention from Canadian activists and would-be activists. If you want to save Canada from the WIPO Internet treaties, a good place to start is the always-excellent Digital-Copyright.ca. On top of that, Will Pate has set up a petition to Parliament on this.