TEMPO.CO, Jakarta - Ainun Najib's eyes were glued to the iPhone he held, occasionally moving his thumb on the screen, when Tempo met him at the lobby of a well-known hospital in Singapore, on Saturday last week. Ainun had been reluctant to give an interview, given the notoriety of the kawalpemilu.org, the website he created, had gotten in his home country, these past few weeks.

He had lain low the past two days, on information that a certain person claiming to work for the campaign of one of the presidential candidates was hunting for him. "I must lie low before July 22," Ainun told Tempo. That was the date the General Elections Commission (KPU) was scheduled to announce the official final tally of the presidential election. He was willing to be interviewed after he checked out the background of the Tempo reporter Mahardika Satria Hadi, whom he met in Singapore. "My apologies for having to cross-check you first," said Ainun. But when he did agree to speak about it, he spoke intensely and passionately about the election vote-count website he created overseas, together with two of his colleagues.

Ainun, who currently works for a multinational information technology company, was raised among a santri (religious) family, in a village in the district of Gresik, East Java. His father, Abdul Rozaq, was a pesantren (Islamic boarding school) graduate who became the village religious leader and a local employee of the Gresik Ministry of Religious Affairs. His mother Rustinah was an elementary school teacher. "My parents and extended family are true blue NU (Nahdlatul Ulam-Indonesia's biggest Islamic mass organization)," explained Ainun.

Although he has lived in neighboring Singapore since his early university days, Ainun has not abandoned his home country. He felt compelled to do something when he saw the confused political development, following the July 9 presidential elections. He used his knowledge as a computer science graduate of Nanyang Technological University to create kawalpemilu.org, a website featuring the ballot recapitulation based on the KPU's formula C1 tabulation data. Excerpts of the interview:

Who first had the idea of setting up kawalpemilu.org?

The initiative came from me. After the election I initially intended to do nothing, just focus on the Ramadan fasting and wait for the KPU count. One day, I was struck down by a heavy dose of flu, forcing me to stay home. I turned to Facebook, and what I saw got me really worried, that both sides claimed to have won the election. I was very worried that the country could be divided because both candidates claimed to be right. One based it on the credibility of the quick count pollsters, which in the past always produced results that came closest to the official KPU count, because of the solid and militant network in the regions.

Why did you choose that name kawalpemilu?

Initially, we were going with kawalpemilu.com but changed it because it was not commercial. Our philosophy was to guard this election, so that it would always remain transparent to the public, prevent it from falling into the hands of people who wanted to rock the KPU, force public opinion, or whatever.

What inspired you to do it?

A friend posted on Facebook the status of the KPU data. I chatted with him and we talked about how we could do some good with the KPU data that could be downloaded. The three of us started talking about and conceived the idea of crowdsourcing.

The three of you? Who are the other two?

The three of us were the core team. One works in Silicon Valley in the United States and the other in Australia. They are all alumni of the Indonesian Computer Olympics, older than myself.

What was their role?

They created the website. I focused on producing the internal website with a secret address to enter the data. The database was over there. The other person created a site for the public-www.kawalpemilu.org-which was accessible to the public. These two people worked with another two of their colleagues.

Where was the website managed?

We never had a particular base. We never had meetings in the real sense of the word. The core team was in Singapore, and that was me.

There was really no headquarters?

(Laughing). Look, this is the era of 2.0. What is important today is virtual space.

After the team was formed, what did you do?

My colleague started creating the site on Friday (July 11). Meanwhile, I made the secret Facebook Group so no else could access it easily. Not everyone could join. I invited four other friends I trusted to start practicing programming the vote count. By that time, some 100 C1 forms (registering vote count in every polling booth) for Gorontalo province had been sent to the KPU. Gorontalo was the first province to have completed its C1 requirement. This was to be our pilot province. We practiced manually at first, opening the C1 forms one by one, typing the numbers in a spreadsheet. In three days, we were only able to complete all of Gorontalo's data, but only 50 percent of Bengkulu's.

Why did it proceed so slowly?

Because we did it manually, even though Gorontalo only had 2,000 polling booths. It wasn't until Monday afternoon that we found a system to enter the data and started on our project. After we felt comfortable with it, we invited others to make an assessment. By Sunday evening, the system was up and going. I appointed some friends to be the admin group and to seek help from other volunteers.

So in total, how many worked for admin?

I appointed 20 to work in admin, and they each sought the help of their friends.

So, how many volunteers were involved?

The total was 700, almost half of them in Singapore and the rest spread around the world. I don't know how it came to exactly 700 people.

How tight was the selection of volunteers to work at kawalpemilu?

Selection lay in the hands of the admin people. Personally, I chose to just look at his academic background. It was plain guesswork, but really, if they graduated from a bonafide university, they would surely have integrity. I also checked profiles, whether they were fanatics or not.

Who decided you should be the spokesman?

It was because I knew best the situation at home. My friends suggested I give the press a statement, to prevent errors. Then, it was best that we remain under the radar, not emerging too much. Once I did speak on their behalf, I just continued being the spokesman.

What is special about the kawalpemilu website?

It listed the national aggregate data, it can be seen from the perspective of the province, regency/municipality, village and polling booth. Every polling booth had its C1 form. People can check them out at a very fast speed. Anyone can make claims, but their claims had to be checked by a solid verifier, so it does not become unsettling rumors because it was stopped by an accurate verifier. There would be no more reason to mobilize masses to the streets in protest.

In your view, what distinguished your count with the one at the KPU website.

I think it's very good that the KPU was open to the public. We must congratulate them. Our website would not have been possible if the KPU had not opened up its data. The problem is that the KPU recapitulation could not be seen immediately. In fact when the DA1 form (count at the sub-district level), the DB1 (regencies/municipalities) appeared, as well as the DC1 (provinces), people would not have been able to go back to C1. For example, if there had been a mistake or a typo in the DA1 data, people would not have been able to track down the error, except if they were doing manually, one by one. Well, that's where kawalpemilu came in. Our data can be traced. We actually helped the KPU in making the counting more transparent. (*)

Read the full interview in this week's edition of Tempo English Magazine