ONI Country Profi le
305
.blogspot.com ) without blocking the host domain, http://blogspot.com , or any of the
other blogs hosted on it.
While BSNL blocked more sites than the other ISPs, the ONI found that in general
the level of fi ltering among ISPs was consistent with only slight variations. Further, the
contents of fi ltered Web sites are similar across ISPs. For example, each ISP blocks a
variety of extremist sites, such as Web sites of Hindu extremist groups ( http://
hinduunity.com and http://hinduunity.org ) or Web sites with critical or extremist
political (particularly “ American right-wing ” ) commentary ( http://mypetjawa.mu.nu ,
http://mackers-world.com/ ). The OpenNet Initiative has found that these sites are consistently targeted for fi ltering by India. Web sites with information on human rights in
India, Internet tools such as proxies, and content related to free expression are also
targets of fi ltering. Data also showed that ISPs consistently fi lter pornography, but
compared to other types of content, the number of blocked pornographic sites is small.
Technical analysis revealed evidence of collateral fi ltering on two ISPs: Bharti Airtel
and MTNL. Collateral fi ltering is a result of IP-based blocking and refers to Web sites
that are unintentionally fi ltered as a result of sharing the same IP address as a Web
site that has been intentionally blocked. For example, 2006 – 2007 testing found that
a site about American-Israeli rabbi Meir Kahane ( http://kahane.org) was blocked
because it shares the same IP address as the Hindu Unity Web site ( http://hinduunity
.com , http://hinduunity.org ); 2009 – 2010 testing confi rmed that the block was still in
place. Similarly, during 2008 – 2009 testing a Web site for travel agents ( http://www
.positivespace.com ) and a system administrator resource Web site ( http://gwsystems
.co.il ) were found blocked as a result of sharing that same IP address with the Hindu
Unity Web site.
Conclusion
Indian ISPs continue to selectively fi lter Web sites identifi ed by authorities. However,
government attempts at fi ltering have not been entirely effective because blocked
content has quickly migrated to other Web sites and users have found ways to circumvent fi ltering. The government has also been criticized for a poor understanding of
the technical feasibility of censorship and for haphazardly choosing which Web sites
to block. There are still parts of the IT (Amendment) Act, including absolving intermediaries from being responsible for third-party-created content, that have not been
tested since its enactment. This lack of action could signal stronger government monitoring in the future.
Notes
1. Internet and Mobile Association of India (IAMAI), “ I-Cube 2009 – 2010: Internet in India, ” April
2010, http://www.iamai.in/Upload/Research/icube_new_curve_lowres_39.pdf .
305
.blogspot.com ) without blocking the host domain, http://blogspot.com , or any of the
other blogs hosted on it.
While BSNL blocked more sites than the other ISPs, the ONI found that in general
the level of fi ltering among ISPs was consistent with only slight variations. Further, the
contents of fi ltered Web sites are similar across ISPs. For example, each ISP blocks a
variety of extremist sites, such as Web sites of Hindu extremist groups ( http://
hinduunity.com and http://hinduunity.org ) or Web sites with critical or extremist
political (particularly “ American right-wing ” ) commentary ( http://mypetjawa.mu.nu ,
http://mackers-world.com/ ). The OpenNet Initiative has found that these sites are consistently targeted for fi ltering by India. Web sites with information on human rights in
India, Internet tools such as proxies, and content related to free expression are also
targets of fi ltering. Data also showed that ISPs consistently fi lter pornography, but
compared to other types of content, the number of blocked pornographic sites is small.
Technical analysis revealed evidence of collateral fi ltering on two ISPs: Bharti Airtel
and MTNL. Collateral fi ltering is a result of IP-based blocking and refers to Web sites
that are unintentionally fi ltered as a result of sharing the same IP address as a Web
site that has been intentionally blocked. For example, 2006 – 2007 testing found that
a site about American-Israeli rabbi Meir Kahane ( http://kahane.org) was blocked
because it shares the same IP address as the Hindu Unity Web site ( http://hinduunity
.com , http://hinduunity.org ); 2009 – 2010 testing confi rmed that the block was still in
place. Similarly, during 2008 – 2009 testing a Web site for travel agents ( http://www
.positivespace.com ) and a system administrator resource Web site ( http://gwsystems
.co.il ) were found blocked as a result of sharing that same IP address with the Hindu
Unity Web site.
Conclusion
Indian ISPs continue to selectively fi lter Web sites identifi ed by authorities. However,
government attempts at fi ltering have not been entirely effective because blocked
content has quickly migrated to other Web sites and users have found ways to circumvent fi ltering. The government has also been criticized for a poor understanding of
the technical feasibility of censorship and for haphazardly choosing which Web sites
to block. There are still parts of the IT (Amendment) Act, including absolving intermediaries from being responsible for third-party-created content, that have not been
tested since its enactment. This lack of action could signal stronger government monitoring in the future.
Notes
1. Internet and Mobile Association of India (IAMAI), “ I-Cube 2009 – 2010: Internet in India, ” April
2010, http://www.iamai.in/Upload/Research/icube_new_curve_lowres_39.pdf .