Taiwan’s Digital Minister without Portfolio is a new Cabinet position created to promote open government. Using internationally recognized methods, we are promoting open data to increase government transparency, encourage public participation, and create records that will hold government accountable in the policy-making process. We also want to encourage people who are not normally part of the policymaking process, or who perhaps have limited literary or legal training, to bring their strengths to that process.

Korea’s Government 3.0 initiative has earned high praise from the U.N., the Organization for Economic Cooperation and Development, and the World Economic Forum (WEF). What have you done in recent years to enable civil servants to release government data as they’re doing their work, rather than releasing the data separately from current processes or paperwork?

The U.N. has ranked Korea No. 1 for e-government, but the International Monetary Fund and the WEF have not ranked our government efficiency as high—around No. 23 or so. One of the goals of Gov 3.0, therefore, is public service reform: to promote transparent government and open data while getting the private sector involved in those efforts. First we changed the structural foundations and combined different government databases into big data, then we used data analysis to build a more scientific decision-making process. We had to change the way civil servants worked so they could release as much data for public use as possible.

As for enabling civil servants to release data at the same time that they’re working, we are not at that point yet. But at least we can release the original official documents, and a lot of the documents have already been released under “right-to-know” open government data initiatives. So as some of the information is being used, it can be released right away. We also have laws that require central government officials at or above the department director level to disclose their official documents.

Taiwan passed the Freedom of Government Information Law in 2005: On one hand, a member of the public may ask the government for information concerning himself or herself—what we call “my data” or “portable data”—but not information on other people. On the other hand, the government should take the initiative to disclose official documents and other government data.

Questions:

First, many of the documents are paper copies. Without very good scanning or artificial intelligence (AI) recognition software, it may be difficult to release documents in a way that the public can easily use, and the documents usually come in image file formats. How does Government 3.0 address this?

Second, the Freedom of Government Information Law stipulates that an official document that has not yet been approved—even a document drafted by a department director—may not be disclosed before it has been approved, unless disclosure is necessary and in the public’s interest. Does Korea have this kind of restriction?

Korea also does not disclose documents that have not yet been approved, and we don’t release the official document itself but a kind of summary report. The government system has an “official documents” section for agencies to refer to, and documents created on computers are stored in a cloud system. These are the foundations of open data.

All of our official documents are now maintained on electronic systems, so releasing them is not difficult. In Seoul, for instance, every stage of the architectural or construction license application process is open, from the point of submission through the approval process. The Electronic Government Act requires administrative agencies to handle all official documents electronically so they can be digitized right away.

Regarding public service reform and electronic systems, are these capabilities you look for when hiring civil servants or something they receive training for after being hired? And how do you promote this kind of thinking in the more senior civil servants? Third, what role does Government 3.0 play?

The Government 3.0 committee is not responsible for selecting or hiring civil servants. These abilities are relevant but not required for civil servants, and they can learn on the job after entering government work.

The PDIS has hired a consultant with work experience at the U.K. Policy Lab to study ideas from the U.K. Civil Service Reform program, inject “design thinking” into Taiwan’s government processes and data-driven policy, and bring all agencies together in the decision-making process. This “participation officer” is being promoted by the Taiwan government now.

Does Korea model its public service reforms on similar programs in other countries? What are some specific objectives over the next one to three years for reforming administrative efficiency?

Korea takes examples from many countries like the U.S. and the U.K., especially regarding open data. Unlike Taiwan, we do not engage outside experts, but we do maintain frequent contact and experiencial exchanges with other countries.

Open data can be implemented in a very short time, but data-driven policy will take longer and depends on the abilities of civil servants. So we’d rather continue promoting these ideas in existing projects than start new projects.

What is the relationship between the Government 3.0 Committee and government agencies? Can the committee assign tasks to agencies directly, or is there a contact window? Do you hold meetings to coordinate and solve problems? How frequently? And is there a reward/penalty system?

The Government 3.0 Committee was established in July 2013. Prior to 2013, work relating to Government 3.0 was carried out by our Ministry of the Interior. We discovered it wasn’t easy to bring about change because open government is not only about open data, but about how civil servants work, and every government agency had its own civil service culture.

At the time, we wanted to bring private-sector thinking into the civil servant work model, so we used this concept to set up the committee and invited experts from the private sector. Now we have six chairpersons to help promote various aspects of this task.

It was quite difficult to get government agencies to coordinate with each other, provide information, share information, and provide new services, so we used the private-sector approach and convened individual agencies for coordination meetings. If the agencies couldn’t coordinate the work, the committee would step in, and if the committee couldn’t do it, Blue House secretariat officials would join the coordination effort.

Committee meetings are held once every three months among the chairpersons and government officials at the deputy minister level. However, chairpersons from the private sector who want to discuss new concepts or ideas can call a meeting at any time. Meetings with the Blue House are basically held once every two months.

As for contact windows, the government agencies send their representatives directly. The committee will also set important goals for the agencies while the agencies set their own goals as well, and all of these efforts are regularly supervised.

How effective is the supervision? What if they don’t reach their goals?

There’s no penalty. We look at whether the goals are achieved. If there is a delay, we’ll press them to pick up the pace.

As for work assignments, either the chairpersons will assign the work from the top down, or the agencies will propose work objectives from the bottom up to the committee. The committee will also reward and recognize successful cases.

Taiwan has a public policy participation platform (join.gov.tw) consisting of three sections: The first is “Propose,” where any policy proposal that gathers 5,000 signatures must be discussed among government agencies and responded to.

The second is “Talk.” In principle, government agencies must announce all proposed legislation on this platform in advance to allow 60 days for public discussion. If the public finds problems or difficulties with a piece of legislation, it can be blocked or sent back for redrafting.

The third is “Supervise.” The Executive Yuan requires government agencies to disclose how much money they spend on projects. This section uses graphics to show the amount spent on each project on a monthly basis, the progress, and what public tenders have been issued.

This website doesn’t have a reward or penalty function, but lining up all the agencies for comparison does create a psychological effect, and they have to respond to online questions from the public.

Minister without Portfolio Audrey Tang previously served as an external expert to Taipei City in 2015 to make the budget system more visual. The bureaus and departments were very resistant at the outset, but after some education and training, they asked their staff to reply to hundreds of online questions one by one. This created quite a stir because city residents discovered they didn’t have to go through the media or public representatives to have direct exchanges with government officials, and it deepened the public’s trust in those officials.

This budget.taipei website is still in use. After successful testing in 2015, this website was used to upload all budgets to the city council in Excel format starting in 2016.

Does the website publicize information on the overall projects, or does it publicize the individual items as well?

The graphics show information for specific bureaus and departments, but people can click on any of the individual items and leave messages there.

In Korea, the central government, local governments, and education are the three main pillars of public finance. In 2015, figures for all three were made public. Last year, there was some thought to consolidating the funding utilization figures. Expenditures by local governments and the educational sector are released immediately, and in the future, information regarding all government expenditures will be made public immediately to further public understanding.

When we make spending information public, we also record the related tender bid information. Currently, the government wants to make all the information public, including the bid-winning vendor’s proposal, as well as the mid-term and final reports. Does Korea also do it that way?

In Korea, some of that information can be made public, and some disclosure information is more detailed, and the government is now allowing various government organs to move in that direction.

In Taiwan, the biggest attitude adjustment for government organizations is letting go of so-called “face” issues. Sometimes, when things don’t go well, the media will use information that was made public as grounds for criticism. But we feel that making such information public is the only way to invite public participation. Is there a similar issue in Korean culture?

Getting the approval of another government agency to disclose information isn’t that easy, and the person responsible for the case is also under considerable pressure. But for the benefit of the public and the nation, and to protect the public’s right to know, the government is now actively promoting disclosure.

When government data is, in principle, made open to the public, the volume of data is incredible. It’s not just numerical data; it also includes written information like official documents, reports, and supporting materials attached to those reports. But when most citizens utilize that information, they don’t know what government agency or entity it’s from. And it’s hard to correlate the data simply by doing a full document search. So many countries are trying to use Knowledge Graph or Web Ontology to correlate data from different sources. But to set that up, all government entities have to coordinate their efforts, and input data using the same categories. In Taiwan, that’s extremely difficult.

We’re now considering using machine learning and AI to do that, like in foreign countries, where AI is being used to find precedents for many legal issues. To resolve issues surrounding the use of natural language, the government has already set up a test lab to research this issue. Does Korea have any ideas about fusing AI technologies with search functions?

Using AI to retrieve data is already part of the Government 3.0 plan. Using AI to read documents is also being promoted, but it will take more time. While promoting these ideas over the past two or three years, we found the biggest question has been whether AI can read documents accurately. The next administration will probably continue to promote this idea. Most people don’t really understand how AI is being utilized in Government 3.0. In Taiwan, how is the government convincing the public that AI is feasible?

The government’s promotional strategy is just to use AI in tasks where the outcome is the same regardless of who performs the task. Examples include typing out what someone is saying, or retrieving research—no matter who is involved, the only difference is speed, and no personal value judgment is required. So most likely it will save time for civil servants, but it doesn’t replace their judgment.

The way this came about was “top-down.” For the government, it became a policy, which was subsequently promoted by individual agencies. It’s more of an information and communication technology issue, and essentially technical, so it’s being promoted as a technology. If the head of the nation decided to implement the technology, then there’s no problem, and it can be disseminated.

But the most important thing is: What does the public think about these changes? How do they feel? The biggest concern is how the public feels. How should they be informed about the change? Most citizens simply aren’t concerned about the technology the government uses. What they are really concerned about is what kinds of services are being provided to make their lives more convenient and comfortable.

Here is a good example. The Ministry of the Interior has a concept called the “civilian design corps.” It’s not government-led; it’s promoted by the public themselves. For every project, somewhere between eight and 15 people form a group, including government officials, citizens, experts and designers. So average citizens participate and take part in promoting initiatives, and the groups have been pretty successful.

In Taiwan, we have a kind of regional design called “co-design” or “co-planning.” This is an important concept for mayors in many cities, who want to find a way to get residents to participate in the design process for urban renewal or urban planning, letting people discuss things during public forums. And they’re now trying to allow people to join in discussions even before the public forums are held.

The biggest problem is that people aren’t used to waiting for a year or a year-and-a-half before discussions produce a plan that everyone can accept. So the development process is always criticized by people who are looking for efficiency. Is that the way it is in Korea?

Government 3.0 represents the third stage, or “service design.” In the first stage, the government simply made all the decisions without inviting citizen participation; this was how things were done during the developmental era. The second stage was about making government information public, and sometimes collecting public feedback. In the third stage, citizens can demand that the government do something. For issues like urban development or land development, getting the public to participate and voice their opinions is imperative—and this is what service design is all about.

In France, an independent authority called the National Commission for Public Debate (Commission nationale du débat public) pushed for a law in 2002 requiring that any development project receiving a certain level of funding must devote a specific percentage of its construction funding to service design. This involves bringing related parties, whether civic or other groups, and other stakeholders into the process, and it would take at least a year to implement any innovation plan. That meant statutory budgets would come from development projects, and legal initiatives or projects that don’t involve development would not receive funding.

In Taiwan, on the other hand, public funding is generally allocated to hire academics and experts to implement plans. In the course of a year, each city can only choose three or four projects, but there is no funding or budget to do anything beyond that. Does Korea have a method to allocate budget funds for service design like France?

In that situation, the participants will think of a way to convince the non-participants, some of whom will oppose it, and some will be indifferent altogether. There are many examples in foreign countries. In the U.S., during the Obama administration, he hoped for more citizen participation. The government first set up a framework, which was actually a way to get experts, citizens and interested parties to participate and execute the plan.

Korea’s digital technology is pretty good, so it wasn’t hard to promote the e-government idea. Using digital capabilities to improve service quality is a lower priority, but more important is promoting concepts like openness, sharing, communication and collaboration. The standards and norms for strategic decision-making have changed, so this is now being given greater consideration.

Organizations like the Internet Engineering Task Force and the Internet Corporation for Assigned Names and Numbers are frameworks with multiple stakeholders. They’re not affiliated with the U.N., and aren’t managed by any particular government. They operate based on communication with stakeholders and transparent procedures, so developers around the world trust them. Their procedures are continuously changing, and the change process is transparent and shared. This is their core value.

In the Taiwan government, from the very beginning, we have done our best to transform the processes and procedures for communicating with the public, and share that with the world as best we can. We thank Korea for sharing their experiences to date, and are very interested in continuing our cooperation.

Are the responsibilities under Minister Tang something new in Taiwan or had they been promoted before?

The “open government” concept in Taiwan has four pillars: Transparency, Participation, Accountability and Inclusion.

During the previous administration, the Cabinet avidly promoted transparency and participation. But that transparency and participation didn’t always relate to practical implementation and public benefits. In the past, the government depended on stakeholders with expertise in law or data, while excluding stakeholders with expertise rooted in individual experience or local wisdom. So for the current Cabinet, inclusion is most important. The new “digital minister” position is promoting the four pillars we just talked about, but the first two pillars—transparency and participation—have been advocated for a long time.