Filtering and Transparency
As Faris and Villeneuve document in chapter 1 of this book, states adopt a range of practices
in terms of how explicitly they discuss their filtering regime with the public and the amount that
citizens can learn about it. No state that we have studied in the past five years makes its block
list generally available, though partial information has found its way to the surface in a few
instances. In India, through freedom of information filings, some citizens have obtained information as to the list of those sites filtered. In Bahrain, citizens have compiled a partial block list
and posted it to the Internet. In Thailand, prior to the 2006 coup, a list of many thousand Web
sites had been posted online, plausibly leaked by the state, but not mapping closely to the
facts we have observed. A combination of citizen efforts and the circulation of Pakistan Telecommunication Authority blocking orders on the Internet have resulted in a partial list of
blocked sites in Pakistan coming to our attention.
22
Saudi Arabia is the most transparent state in terms of Internet filtering. The Saudi state sets
forth the rationale and practices related to filtering on an easily accessible Web site in both
Arabic and English. (In our first round of testing, in 2002, Saudi Arabia enabled us to run
tests directly against its system, but would not show us the list that it was using to determine
which sites it was blocking at any given moment; since publication of our first report on
this topic, the Saudis have disallowed us such easy and direct access to their system.) In
Saudi Arabia, citizens may suggest sites for blocking or for unblocking, in either Arabic or
English, via a public Web site. Access to most of these sites prompts a blockpage to appear, indicating to those seeking access to a Web site that they have reached a disallowed
site. Most states have enacted laws that support the filtering regime and provide citizens
with some context for why and how it is occurring, though rarely with any degree of precision.
However, among the states we studied, some of the central Asian states that practice justin-time filtering on sensitive topics—as well as China, whose officials sometimes deny the
presence of Internet filtering—obscure the nature and extent of their filtering regimes to
the greatest extent.
23
Some states, such as Saudi Arabia and UAE, make an effort to suggest that their citizens
are largely in support of the filtering regime, particularly when it comes to blocking access to
pornographic material. For instance, the agency responsible for both Internet access and filtering in Saudi Arabia conducted a user study in 1999 and reported that 45 percent of
respondents thought ‘‘too much’’ was blocked, 41 percent thought the amount blocked was
‘‘reasonable,’’ and 14 percent found it ‘‘not enough.’’
24 We have not delved into the veracity of
these findings.
Citizens may, in some instances, participate in the decision making as to whether a site
may be filtered or not. Three of the states in which we tested (Saudi Arabia, UAE, and Yemen)
respond to a request for a blocked site with a page that includes a mechanism for suggesting
that the particular URL may be blocked in error. However, to make such a suggestion requires
the user to have knowledge of the content of the Web page not able to be visited—and the
40
Jonathan Zittrain and John Palfrey
As Faris and Villeneuve document in chapter 1 of this book, states adopt a range of practices
in terms of how explicitly they discuss their filtering regime with the public and the amount that
citizens can learn about it. No state that we have studied in the past five years makes its block
list generally available, though partial information has found its way to the surface in a few
instances. In India, through freedom of information filings, some citizens have obtained information as to the list of those sites filtered. In Bahrain, citizens have compiled a partial block list
and posted it to the Internet. In Thailand, prior to the 2006 coup, a list of many thousand Web
sites had been posted online, plausibly leaked by the state, but not mapping closely to the
facts we have observed. A combination of citizen efforts and the circulation of Pakistan Telecommunication Authority blocking orders on the Internet have resulted in a partial list of
blocked sites in Pakistan coming to our attention.
22
Saudi Arabia is the most transparent state in terms of Internet filtering. The Saudi state sets
forth the rationale and practices related to filtering on an easily accessible Web site in both
Arabic and English. (In our first round of testing, in 2002, Saudi Arabia enabled us to run
tests directly against its system, but would not show us the list that it was using to determine
which sites it was blocking at any given moment; since publication of our first report on
this topic, the Saudis have disallowed us such easy and direct access to their system.) In
Saudi Arabia, citizens may suggest sites for blocking or for unblocking, in either Arabic or
English, via a public Web site. Access to most of these sites prompts a blockpage to appear, indicating to those seeking access to a Web site that they have reached a disallowed
site. Most states have enacted laws that support the filtering regime and provide citizens
with some context for why and how it is occurring, though rarely with any degree of precision.
However, among the states we studied, some of the central Asian states that practice justin-time filtering on sensitive topics—as well as China, whose officials sometimes deny the
presence of Internet filtering—obscure the nature and extent of their filtering regimes to
the greatest extent.
23
Some states, such as Saudi Arabia and UAE, make an effort to suggest that their citizens
are largely in support of the filtering regime, particularly when it comes to blocking access to
pornographic material. For instance, the agency responsible for both Internet access and filtering in Saudi Arabia conducted a user study in 1999 and reported that 45 percent of
respondents thought ‘‘too much’’ was blocked, 41 percent thought the amount blocked was
‘‘reasonable,’’ and 14 percent found it ‘‘not enough.’’
24 We have not delved into the veracity of
these findings.
Citizens may, in some instances, participate in the decision making as to whether a site
may be filtered or not. Three of the states in which we tested (Saudi Arabia, UAE, and Yemen)
respond to a request for a blocked site with a page that includes a mechanism for suggesting
that the particular URL may be blocked in error. However, to make such a suggestion requires
the user to have knowledge of the content of the Web page not able to be visited—and the
40
Jonathan Zittrain and John Palfrey