After blocking Telegram since July 14, the Indonesian government has lifted its ban on the encrypted messaging service that has become popular as a platform for radicals and terrorists to communicate and propagate their ideas.
The Ministry of Communication and Information Technology (MCIT) last Thursday unblocked Telegram's 11 domain names after the latter took down 166 channels containing information related to terrorism and radicalism, which the ministry and other agencies had reported to the company.
The move followed a meeting between minister Rudiantara and Telegram CEO Pavel Durov in Jakarta on Aug 1, during which Mr Durov apologised for not responding in a more timely manner to the ministry's repeated emails requesting action.
Mr Durov said Telegram was committed to blocking public channels that could be used for the propagation of terrorism globally, and particularly in Indonesia.
"We have discussed ways to make it more efficient," he said during a joint news conference with Mr Rudiantara.
Commuters use their smartphones while waiting for a train at the Tanah Abang station in Jakarta. Photo: Reuters
The meeting with Mr Durov was part of a series of meetings that the minister has had this month with other social media, content and video sharing and messaging services as part of a crackdown on "negative content" online.
The ministry had identified at least 55 Telegram channels that had terrorism-related content and according to police data, the majority of those convicted of terrorism-related activities between 2015 and July 2017 used the application to communicate.
Khairul Fahmi, a researcher from the Jakarta-based Institute for Security and Strategic Studies (ISESS), said the new breed of messaging services and devices had been a boon for terrorists seeking to communicate with cells in Southeast Asia, notably Indonesia, the southern Philippines, southern Thailand and Malaysia.
"Fundraising, recruitment and directions for action have become swifter but more obscure," Mr Khairul told Asia Focus.
Ade Bhakti, executive director of the Jakarta-based Centre for Radicalism and Deradicalisation Studies, said the use of online platforms by terrorists was hardly a new thing. Imam Samudra, who was executed in 2008 for carrying out the 2002 Bali bombings, used to communicate with his international network using an internet relay chat available on Windows, he said.
In addition to using messaging services, terrorists also hack commercial and business sites to steal money to finance their activities, he said.
"Bahrumsyah and Abu Jandal even shared their clips which contained speeches calling on their followers to join terrorist groups on YouTube," Mr Ade said, referring to two Indonesian nationals who went to Syria to join the Islamic State (IS). Abu Jandal is believed to have died in combat.
"Participants in a 2010 paramilitary training exercise in Aceh also posted clips of their training on YouTube."
Google, which owns YouTube, says there should not be any content that promotes hate speech on the world's biggest video-sharing site.
"But there's so much content uploaded everyday to YouTube in many different languages so we really rely on our communities and users to flag it," said Ann Lavin, Google's policy and government affairs director for Greater China and Southeast Asia.
She made the comment at a news conference on Aug 4 with Mr Rudiantara, who met Twitter representatives separately on the same day.
To identify disturbing content on YouTube, Google will introduce the "trusted flagger" programme in Indonesia in the next two or three months, said Ms Lavin. A similar programme has been introduced in France, the United States, Germany and Britain but Indonesia will be the first country in Southeast Asia to have it.
Content identified by flaggers, who believe it does not meet YouTube's community guidelines, would have priority to be reviewed by Google before a decision is made to take it down or not.
Ms Lavin said the process of reviewing and removing flagged content would be transparent and documented in Google's semi-annual transparency reports that list the number of requests for certain categories and content that is removed.
"The amount and the nature of requests will be absolutely transparent to the public," she said. Google, she added, was also cooperating with three civil society organisations as trusted flaggers to moderate clips and content on YouTube: The Wahid Institute which is led by former president Abdurrahman Wahid's daughter, Yenny Wahid; the safe internet advocacy group ICT Watch, and the Anti-Defamation Society of Indonesia (Mafindo).
Mr Rudiantara also met Jeff Wu, Facebook's Asia Pacific representative, on Aug 2. According to the ministry's director-general for information applications, Semuel A Pangerapan, Facebook agreed to install "geoblocking", which will allow users to control and monitor negative content in accordance with local culture and contexts.
"The geoblocking will have a special algorithm for Indonesian users and any flagged video cannot be accessed in Indonesia," Mr Semuel said.
He said that in the past two years the ministry had received 402 reports of negative content of varying nature, from terrorism to child pornography and hate speech. But Facebook responded to only half of those complaints because of differences in perception as to what constitutes negative content, he said.
The ministry urged Facebook to be more proactive about dealing with content deemed harmful to Indonesia when it receives information from a team comprising people who understand Indonesian values.
Frenavit Putra, the coordinator for the ICT Watch trusted flaggers, said their task would be to help supplement the work of the Google's algorithm in content moderation.
"Everyone can be a flagger, but the content they flag would have to wait in a queue before being reviewed, whereas the content that we flag would get priority to be moderated. Google will decide to keep it or to take it down," he told Asia Focus.
"When we flag content, we have to provide a detailed analysis and reviews, down to which minutes and which seconds the disturbing content occurs at. We have to give justifications for our analysis and what we flag isn't just the content on clips but also the comments," he said.
He said the group was not just providing input on content but also on context, and would be prepared to explain why a certain shot or comment on a clip uploaded to YouTube may have different meanings to different audiences.
For example, views on what constitutes pornography differ from one country to another. In some countries, some forms of nudity alone would not be considered offensive but in others they would be. "Google's algorithm cannot analyse that far," he said.
ICT Watch is in charge of monitoring content related to hate speech, child protection, age restrictions on certain content or pornography. The Wahid Institute will deal with content related to terrorism or radicalism, and Mafindo deals with content related to hoaxes or fake news.
"We are including civil society organisations to ensure that this is not a censorship regime but we really have to protect Indonesia from negative content, particularly radicalism, terrorism and drugs. They would be the priority for us," Mr Rudiantara said.
