Navigating Stormy Waters — g0v.tw’s Lessons for Global Open Government Advocates
Welcome back to the second part of our interview with Chia-Liang Kao of g0v.tw. In the last edition of Open & Shut we spoke about the influence of the Open Knowledge Foundation rankings upon Taiwanese open data policies, and the general state of the open data landscape in Taiwan. This second part of the interview explores the work, achievements, and organisational structure of g0v.tw, and how it has been instrumental in helping them gain leverage over government policies.
The g0v manifesto Source: http://g0v.tw/en-US/about.html
Harsha: g0v zero practices an interesting model of a leaderless organization. What’s that about?
Chia-Liang Kao: We model this after the open source communities general idea. We try to encourage collaboration between different kinds of professionals and have things as decentralised as possible. People can host their own g0v zero hackathons or projects as long as they fit within the framework of g0v’s principles, the things we’re trying to achieve, and how we’re doing it. The objective to design as decentralised a community as possible was always part of our manifesto.
Harsha: Are there political aspects to the work g0v zero does? Or is g0v zero politically neutral?
Chia-Liang Kao: From the manifesto, we state that we are nonpartisan. The project organisers have different views of how things should be. What we’re trying to do is produce data that is seen as neutral, and we tell people to convince others with that data. But we also want to show how that data was produced. If people have a different view from you, at least you can have a debate based on the same data. They can fork your project and come up with a different interpretation. Allowing people to have different views to coexist, but sharing the underlying data or code is important to facilitate these conversations. As you can see, people are comfortable with open source and our style of organising, and the community atmosphere is more liberal, I think.
Harsha: What project have you worked on that you’ve enjoyed the most?
Chia-Liang Kao: I did a few in the beginning of my stint here. My project is trying to make the community work, scale, engage in international collaborations, and connect people doing similar things across the world. The first project we did were the visualisations for the Taiwanese general budget. That was a quick hack, even before g0v zero. The idea we had for g0v zero was substituting the “o” with “0″ in gov. The catchy thing is that if you change one character from a really bad government website URL, you end up at our better designed website. If the government wants to take it away from us, they sure can because it is open source. That was the original idea. And right after we built the visualisation tool, we had thought that the government would adopt the work we’re doing — that didn’t happen for three years, when the Taipei city government took our code to officially tell people what the city budget was like. This was a huge success.
A screengrab of the budget.g0v.tw website
This was a great implementation of the fork and merge model. In software terms, you are free to fork from my code and make changes and merge it back — the Github way. When we think of g0v projects, our hope is that we fork the government and one day merge it back. Of course, with help from someone in the community, something similar happened in a project we were working on. There are complicated policies and its hard to debate between the different versions of the legislations. Someone in the community built a tool to visualise how you’re affected by the different versions of the legislations based on your typical work day. The Ministry of Labour took this code and used it to communicate about their official versions of the tool.
Harsha: What was g0v zero trying to achieve during the Sunflower Movement?
Chia-Liang Kao: The Sunflower Movement was when a lot of contributions flooded into g0v zero. The community had quite a few projects. Are you familiar with the background of the movement? The trade deals that happened between China and Taiwan were passed in the legislature without due process. That was the common ground people were against. The interesting thing is that people were not against the trade deal — everyone was against the lack of due process in passing it through. People were not happy with how the evaluation was done, how the decision was made and how it passed through Congress.
Live streaming the Sun Flower Movement // Source : http://flipthemedia.com/2014/07/social-media-taiwan/
So, there was an occupation that lasted three weeks. A lot of people went to the scene, we provided live screens, infrastructure and internet support to allow for content to be streamed. Online, the community developed a search tool to let you see if you’re affected by the trade deal, seven months before the protest. It was an issue that the professional community saw happening and no one was talking about it. They were trying to build a tool to get people to understand the real deal. It is a really complicated deal in reality. If you read the custom tariff, there were billions of quotes for the services and there were different versions of the list, and you could barely match them. You had the UN’s list for industry, the Taiwan census’ list and the trade deal’s own list of committed industries. They spent eight months trying to integrate this into a Google-like search engine — a box where you can put your company’s name and it tells you whether you’re affected by the trade deal or not. That’s a neutral information-centric tool. This was before the protests. The protests really accelerated things.
Harsha: What have you learned from working at g0v zero? Do you have any advice for open data practitioners across the world?
Chia-Liang Kao: It is tough to give a general lesson that will be useful in all countries. In authoritarian governments, you probably don’t want to collaborate with the government to do things. But I think you can find allies inside the government, to push things towards a more democratic and open process. I guess you need to find allies in different parts of the whole open government movement. If you’re in a less crazy country, the agencies will have their own goals and if you can somehow convince them that their goal is aligned with yours and how open data can help them achieve theirs, then maybe things could happen. With trust being built, it’s possible you could affect their future goals.
I guess if we’re talking about open government data, government collaboration is definitely important. But for more closed societies, I think crowdsourced data collection, depending on the circumstances, could possibly be dangerous. The big lesson is that for NGOs that are focusing on particular issues, they should collaborate more with the tech community. Most of the time they’re not into these issues because they’re unaware of them. If you convince them that it’s important, the open-source community can build something; they can build something that could magnify your reach. It could also be a new way of engaging the general public with these issues.
Harsha: In your opinion, is there a tool or a process that g0v zero has done that you think is innovative?
Chia-Liang Kao: One massive crowdsourced project we built was the campaign finance digitisation project. The detailed campaign finance project with corporate donations is only available on paper, but there was a movement with a dozen people moving into a building and printing out everything they could. They went to a scanner and uploaded everything to Google Drive and the engineers turned 3,000 pages of tabular data into images and converted it into a captcha service so people could type it in.
Scanned copies of the Campaign Finance Documents // Source : https://hackmd.io/s/rJefb-hi
Harsha: I find it interesting that in a county like Taiwan that ranks so highly in the open data rankings, you still had to deal with paper.
Chia-Liang Kao: Like I said earlier, freedom of information (FoI) and expression are two different things. We have good freedom of expression here. While we have a freedom of information act that was passed over twelve years ago, before the internet took over. So now the focus is on providing information in more easily accessible formats, which has been tricky. FoI is also dependent on agencies. They might try to be really slow in providing the data if they don’t want to, say, by asking you to clarify things, allowing them to delay the answer. Or sometimes it could be a technical delay. This is still a challenge — convincing people to be transparent. The whole movement around open data is to make this information open even before such a request is passed. FoI only guarantees that when you ask, the government shall respond. For things to be more proactively open… well that’s another story. How the government agencies releases the data is yet another story.
Harsha: What is g0v zero working on right now and what’s coming up in the near future?
Chia-Liang Kao: Right now, my personal interest is in community participation and technology. How do you scale and turn deliberation into a larger scale decision-making process? It’s not possible right now. We’ve been doing a lot of experiments for over five years. We believe in the idea of liquid democracy. I think it is an interesting problem to solve in this space.
We haven’t answered the question: if technology can bring a half a million people to the street, then how can technology make them decide on something? That’s something I would like to see in future projects; doing that. We’re trying to do some grant making this year. We fund civic tech projects that are open source — about $15,000 per project. We’re experimenting with the community, and the people in these projects are very committed. We want to help them make more polished projects, and that’s something we’re trying this year modelled on the Knight Foundation’s Fund.
We also have the Civic Tech Fest in Taiwan from the 10th to the 16th of September. If you’re interested in coming over, you definitely should!
Harsha: Thank you so much for taking out time to talk to us!
For more from Chia-Liang Kao, follow him on Twitter @clkao.