A Thai military-led coup on Sept19, 2006 overthrew the government of prime minister Thaksin Shinawatra. Websites critical of the military action were blocked and became inaccessible in the country.

During the run-up to the Bahrain general election in 2006, the government denied access to a number of key opposition websites.

In the Western world, where the belief is that Internet content is not filtered, the situation is far from being unregulated.

In Canada, one Internet service provider (ISP) blocked access to a website run by a union member during a labour dispute.

In the United States, the government is filtering Internet content in another way. Stealthily, it is surveilling Internet content to identify potential terrorist attacks.

That states filter content deemed sensitive to their countries has been observed and discussed since the Internet became popular in the 1990s.

But there had been no concrete proof that states were actively doing so. Until now.

For the first time, empirical evidence has been collected from a global research project to show that governments do deny citizens access to information deemed sensitive.

Conducted by the Open Net Initiative (ONI), the research of 40 countries sought to find out the scope and depth of Internet censorship.

ONI is an international collaboration of four leading academic institutions, namely the University of Toronto, Harvard, Cambridge and Oxford.

It seeks to identify and document Internet filtering and surveillance so as to discover potential problems and help in the formulation of better public policies.

Findings from the research were published in a book in February. Called Access Denied: The Practice And Policy Of Global Internet Filtering, it documents and analyses Internet filtering practices in more than three dozen countries.

The 450-page book is edited by Ronald Deibert, John Palfrey, Rafal Rohozinski and Jonathan Zittrain, who are also ONI's principal investigators.

They were assisted by in-country researchers who accessed local ISPs to check on the extent of Internet censorship.

Motives for Internet content filtering, they said, can be grouped into three areas: politics and power; social norms and morals; and security concerns.

The first one-third of the book discusses global Internet filtering, the politics and mechanisms of control, filtering techniques and ethics.

The rest of the book comprises regional overviews of areas like sub-Saharan Africa and the United States and Canada, and Australia and New Zealand.

These are followed by country-specific summaries of filtering policy in countries such as Singapore, China, Morocco, South Korea and Venezuela.

Singapore did not fare too badly. Researchers found no evidence of content being blocked on political and security grounds.

As expected, there was selective filtering for socially sensitive topics. Seven websites related to pornography were tested on SingTel and StarHub's Internet services and found to be blocked. But these websites were available on another ISP, SysTech.

Singapore is not alone in denying access to such content. The book points out that Australia, for example, has strict Internet content laws for a Western country.

While child pornography is a no-go, it has a system to rate violence and sexually violent scenes to limit viewership.

The Australian authorities can also issue take-down orders if it finds content sensitive to and unsuitable for its society.

In the case of Singapore, the book omits to explain that the seven websites found blocked from Singapore's Internet network probably belong to a list of 100 banned by the Singapore Government.

The ban, introduced in the late 1990s, is a symbolic gesture to reflect its stand on safeguarding the values of society. The list of 100 blacklisted websites is an evolving one.

When this policy was unveiled together with Singapore's lighttouch Internet regulatory regime in 1997/98, it was greeted internationally with indignation as it went against the common belief then that Internet content should not be censored.

Whatever the world online community and freedom-of-expression supporters say, the lighttouch policy has worked for the Republic.

No access is denied to any information, idea or news needed for business deals or important work.

So there is merit in Singapore's approach to Internet content regulation. Its rich experience has become part of its e-government effort and ought to be shared with other states.

Going forward, the light-touch policy can be tweaked to make it more relevant to current needs. For example, a person whose website has been banned and who wants to be heard in a formal appeal should be allowed such a recourse.

It may be time for the Government to consider setting up a formal appeal structure.

Ultimately, each country has the right to block sensitive online material to protect its society's values and mores.

Understandably, all states will want to regulate the online world to have law and order like in the physical world so that people and communities can live together safely and harmoniously.

The only caveat is that the Internet is the new tool of business and censorship must not be at the expense of business operations or the lives of ordinary citizens.

chngkeg@sph.com.sg

This article was first published in The Sunday Times on 8 June 2008.