ISP’s vs. the State

The recent CRTC/Warman case has generated much discussion. One of the issues brought forward is that this case sought to have the CRTC authorize ISP’s to voluntarily block sites, something that distinguishes this from government-mandated censorship. While true, it rasises several issues.

Firstly, it is not neccessarily a positive thing. ISP’s have engaged in misguided vigilantism in the past, blocking content the ISP feels it is entitled to block (and doing so in a manner that blocked nearly 800 non-related sites in order to block just one site). ISP’s are not equipped or qualified to make judgements on content and will always default to the lowest common denominator which has serious repercussion on freedom of speech and expression. Putting “the issue in the hands of ISPs” is not necessarily a better option than state censorship.

Secondly, as the case of China illustrates, the government will often authorize certain topics to be censored leaving the decision on what is specifically blocked to private firms. As the N.Y. Times reported:

American Internet firms typically arrive in China expecting the government to hand them an official blacklist of sites and words they must censor. They quickly discover that no master list exists. Instead, the government simply insists the firms interpret the vague regulations themselves.

Domestic Chinese blog providers in China have told me how they create lists of what terms to filter themselves. Governments delegate responsibilty for censorsing to ISP’s and ICP’s (content providers). Often, ISP’s will become very conservative, blocking more than they are required to block. In Iran, the ISP ParsOnline is known for block more content than required by the government.

Thirdly, the filtering systems in use to block child pornography also suffer from serious problems. In addition to the fact that filtering does not contribute to the removal of child pornography from the Internet and does not contribute to the arrest and prosecution of criminals who traffic in this material it is also easily circumvented. At best, it prevents casual or inadvertent access to designated websites. (Websites are not neccessarily the most common way this material is transmitted, p2p systems for example are unaffected by this type of filtering). Also, there is often no review process (once on the list always on the list) or appeals process (if content is removed or a mistake made). And although the technology is getting better there is often significant over blocking — blocking content that was never intended to be blocked in the first place.

For example, a Pennsylvania state law that required ISP to block access to web sites suspected of hosting child pornography violated the First Amendment, according to the Federal District Court in Philadelphia. The ruling details the process, both legal and technical, that lead to the blocking of 1.5 million legitimate websites while trying to block access to approximately 400 websites suspected of containing child abuse images.

The protection of children, the elimination of child pornography and combatting hate speech are extremely important concerns. Actions taken to fight the production and dissemination of child pornography and hate speech should be designed to be effective while not impinging on the free speech rights of citizens or placing the responsibility to determine offending content on Internet Service Providers (ISP). To combat child pornography on the Internet, government, law enforcement the ISP/web hosting industry and civil society organizations must come together to work collectively on this serious problem.

One comment.

  1. […] The Citizen Lab has a post about the issue of ISP blocking of web sites and its relationship to state-mandated blocking, of the sort that we (together) cover via the OpenNet Initiative. […]

Post a comment.