This is the fourth conversation of the 100+ Conversations to Inspire Our New Direction (#OKFN100) project.
Starting January 2023, we are meeting 100+ people to discuss the future of open knowledge, shaped by a diverse set of visions from artists, activists, academics, archivists, thinkers, policymakers, data scientists, educators and community leaders from all over the world.
The Open Knowledge Foundation’s team wants to identify and debate issues sensitive to our movement and use this effort to constantly shape our actions and business strategies to deliver in the best possible way what the community expects from us and from our Network, a pioneering organization that has been defining the standards of the open movement for two decades.
Another objective is to include the perspectives of people from diverse backgrounds, especially those from marginalised communities, from dissident identities, and whose geographic location is outside the world’s major financial powers.
How openness can speed up and strengthen the fights against the complex challenges of our times? This is the key question behind conversations like the one you can read below.
Today’s conversation is with David Eaves, Associate Professor of Digital Government at the Institute for Innovation and Public Purpose at UCL and SVP of Investment Strategy at Co-Develop. David was one of the key instigators of Open Data Day more than a decade ago. This is the reason why the conversation is published the week after Open Data Day(s) 2023.
David is a public policy entrepreneur, open government activist and negotiation expert, involved with open data communities since their beginnings. In 2009, as an adviser to the Office of the Mayor of Vancouver, David helped draft the Open Motion, after which he went on to advise numerous governments at the local, provincial/state, national and international levels. He is also a Lecturer of Public Policy at the Harvard Kennedy School of Government.
We spoke to David at the end of February when the Open Knowledge Foundation team had our hands full with Open Data Day logistics and was in the middle of the process of selecting 20 organisations that received funding. In 2023, a total of 175 self-organised community events took place on 5 continents. In some ways, David is responsible for the initial spark.
With each conversation, more people are joining the #OKFN project. This time, we had the pleasure of having the participation of:
- Renata Ávila, CEO at OKFN, based in France
- Nikesh Balami, International Open Data Lead at OKFN, based in Nepal
- Oleg Lavrovsky, Board Member at Opendata.ch in Switzerland, a member of the Open Knowledge Network
- Lucas Pretti, Communications Lead at OKFN, based in Spain
We hope you enjoy the read.
Renata Ávila: Let’s start with Open Data Day (ODD). We would love to hear the story from the source. Having arrived recently at OKFN, we started reviewing internal documents and archives, and there are names of key people who were moving processes in the past. Your name is there, and Wikipedia also features you as the initiator of ODD. How did it start and what was the idea behind it?
David Eaves: It was very early days in open data, around 2010. We had just done the open data portal for the city of Vancouver, and we knew Ottawa was doing something similar. Washington DC had done a hackhaton to bolster their open data portal and I thought, “Well, everybody’s gonna need to do their own hackdays and activities to build energy in their cities”. The challenge won’t likely be has hard for a city like DC or Vancouver, but it may be challenging for smaller cities or those that have yet to build a community around this. So if we could create something global that people could connect with and leverage, it would give more energy to people who were trying to do local things.
That was the original idea. I was in touch with Edward Ocampo-Gooding Mary Beth Baker who had been supporting all things open data in Ottawa. “You have a crew organising events in Ottawa, and we have a community doing some things in Vancouver. If we both did something on the same day, it would be bigger”. I also knew someone, Daniela Silva who was working on this issue with her colleague Pedro Markun in a Brazilian hack collective. Then I thought: “If she does something on the same day, then it’s international, as we would have three cities in two countries”. Everyone agreed and that was the base ingredient to make me feel like we could move forward. So I wrote a blog post, then people started to email asking for instructions and ideas, so I spun up a wiki. That was the way we started trying to coordinate things.
To me, the key to the success of the Open Data Day was that it had to be very light organizationally. In that first blog post, I tried to set some highly inclusive cultural norms, like “everybody should be allowed to come”. Also, there was an idea of setting a goal to mobilise people to develop stuff and make use of open data. It became an arbitrary deadline for cities to get data out by. And it also became a community activation moment, where government officials could show up and have a conversation with developers and activists.
But I can claim no credit for any of the local successes that have happened. We just tried to create a framework that would empower local people to be successful. Any local success that’s happened in any country is really all a result of those local activists and volunteers who did all the work. There was a lot of thought about establishing a framework to allow others to build on top of the platform but the real work of making it happen was all local.
Nikesh Balami: It’s a very inspiring story for someone like me, from the Eastern side of this world following the open data movement. The Open Data Day project has been inspiring us for many years. What difference do you see today in the open data field compared to how it started? Where is the open data conversation heading today?
David Eaves: A few things come to my mind. The addition of new open data portals is interesting but way less interesting than the establishment of open data standards and schemas. For example, the transparency standard for extractive industries (EITI) was a really interesting step in that direction. Same for things like Where does my money go? and other open source repositories. I’m very excited about those types of tools (and sad that some of been deprecated).
As a general rule, getting another American or Canadian city to do an open data portal is positive but has diminishing returns. A lot of policy victories have been won, and probably the marginal benefit of getting more players on is low today, although I think that open data portals might actually help solve something in specific jurisdictions with real problems. So I think the battle is much more now on demonstrating value, and that’s harder (but not impossible) to do.
One alternative is trying to share specific code bases to pull people around standards. This is particularly relevant to me, as I’m now on the investment committee and a member of Co-Develop, which is really thinking about digital public infrastructure and particularly through a digital public good lens. How can we create reusable code bases that are addressing common or shared problems… and that could potentially drive open data standards. That would be of real value.
Oleg Lavrovsky: It’s great that you mentioned platforms and impact, because Open Data Day has also become a platform somehow. It started with just a bunch of Wiki pages with the names of cities on them to turn into this global initiative. I would say ODD is almost like an international ritual, it’s very important that the community has a day of celebration and express it in the form of work, hackathons, and people contributing data or code to the movement.
And I want to ask you a question within the framework of global fairness and equity topic, which I think the Co-Develop project also builds upon. Do you agree with the funding model that was made available in Open Data Day over the past years? It’s the same micro grant, a flat rate, for the entire globe no matter if you’re in Switzerland or in Lagos or what kind of project you’re trying to do. Do you think this model is aligned with parameters such as Sustainable Development Goals (SDGs)? Or what else could we do in the future to generate more impact?
David Eaves: I think it is good that these grants are Global South oriented. We should ask ourselves what that funding accomplishes and what are the strategic goals of that funding. What is the purpose of that funding – to just perpetuate the event, or funding to advance some specific objectives?
I think the current hypothesis is that micro grants are a lever to allow more things to happen in the community. But maybe there are better ways. One option would be choosing an app, or a given database, and setting a goal around Open Data Day to get everyone involved to understand the tool, get it scaled, and then build a community around it. I also love the notion of linking things to the SDGs, as you mentioned. You could try to be a little more prescriptive and choose one SDG goal to nudge people around. Or even just picking a theme.
Again, I think standards matter a lot and lead to the question I asked before. How is open data creating value? That’s the key thing. It was a promise for which we have some real successes but need to deliver still more value. The funding has to help drive that value.
Renata Ávila: I fully agree on open data creating value, and the next part of that question should be: to whom? In Latin America, for example, we have seen that many of the targeted data sets opened by cities and governments were very interesting for the international private sector, and not for locals. Locally, many countries are now facing skills and computing capabilities gaps. I know that you are involved in many initiatives involving skills gaps, so I’m curious to learn your perspective on the following scenario.
The cost to play seriously with data has been massively reduced and it’s easier to create public interest value. But, in my experience, the people working today in the public sector seem to be less tech aware than the generations of 10 years ago. In the past, it was common to find very young enthusiastic people inside the public administration, and now it seems that everybody only cares about the misinformation thing and has neglected to be an active “partner in crime”. Have we missed the link for public interest value?
David Eaves: The open data as a hot topic inside government is way lower now. And there’s a lot less tech utopianism than there was 10 years ago.
I remember now that one of the most popular blog posts I ever wrote was an argument around “open data as libraries”. One point I tried to make was that libraries tried to be inclusive with a diverse range of books in them (even books that upset political elites!). The idea is to get people to find things that they want to read in order to learn how to read. And that was the second point, that, libraries had a schooling function: to teach people to read. We built libraries before people knew how to read! Not after! And the third is that libraries are still needed after the vast majority of people learned how to read. For me, the open data portals maintained by public administration should be thought of as a new type of library providing data stats to people, so they can go play with and learn.
On the topic of government officials, I remember that Open Data Day at some point became a place where totally different people met. There were journalists, civic hackers, government people, transparency nuts, and all these different people who actually don’t even agree. We were also joined occasionally by the the FOIA/ATIP generation of the 1960s and 1970s. But they were quite different as they were influenced by the Nixon scandal and fundamentally didn’t trust the government, so they wanted transparency to hold the government’s feet to the fire. We were interested in that too! But also in helping government get better. But the larger point is we had a big tent with many actors motivated for different reasons. Some of the public value we create should help governments, and some should hold it accountable. It’s a difficult mix to maintain, but both are important. And neither always appreciated by government officials!
Lucas Pretti: “From Open Data to AI” was the theme chosen for distributing this year’s Open Data Day’s micro grants. What are the challenges on the artificial intelligence front? In which ways open data can contribute to or guarantee an ethical AI that impacts societies for the better?
David Eaves: AI is further away from everyday users. The number of people who can understand, work and play with AI is even small than the number of people who can play with a spreadsheet and data more generally. So one risk I fear here is that it would shrink the number of people who might be able to engage. Conversely, it could be an opportunity to educate and engage people around AI… and that would be a positive.
I do think Open Data is a powerful tool to enable equity in AI. If large data sets are needed to train AI, and those data sets are all privately held, it will increase inequality. So having large open data sets could be of critical importance.
Renata Ávila: Building on top of your point about value, I wanted to discuss the setbacks on the open data movement in the last 10 years, which is leading to renewed challenges with secrecy and opacity. One first thing that happened was a huge confusion between personal data and data. People embrace the general idea that “it’s bad to collect data”. There has been a bit of data demonization. A second thing was the obvious tensions around data and privacy. Now we have clear rules and frames, but I’m not sure if the so-called “gold standard” GDPR was truly informed by the open data community. It doesn’t seem to me it was.
And a third thing is the new opacities embedded in the data sets. We come from having a good faith belief when an institution or a company published a dataset. But now it requires more steps now to believe in that data set: verify how the data was produced, by whom, and who paid for it. It is producing new tensions in the environmental community, for example, not to mention the war in Ukraine, when both governments took the information down.
It seems to me that we are entering an era of new challenges for transparency that cannot be solved with access to information laws, and more strict, refined requirements for public data sets to be validated or challenged by the civil society communities. What do you think about that?
David Eaves: I’m less worried about the war piece. When governments go to war, it’s a different political environment and questions of transparency – particularly if it could help an enemy combatant – become more difficult to assess. Instead I would focus on laying the foundation for what will happen in Ukraine after the war and to think about if and how data could create value for them.
On the legislation piece, it would be interesting to ask it directly to the access information community. There was open data legislation passed in the United States. It would be great to being evaluating it impact.
Again, I think the standards are much more interesting and compelling, especially where they create value. There has been a lot of thought put into the strings that come with public money today. If you get public money, that has to come with public benefit and public value. There should be open data elements to any policy paper that whatever government is trying to get approved. That would help monitor and create more transparency and accountability, as well as competition.
For me, a great example of the success of this logic has been scientific publishing. Many governments now stipulate that funded scientific research must be published in open journals. Prior, these public dollars were actually ending up creating private and proprietary content, restricting a public good that the public dollars were supposed to be creating. That’s not right.
There are some incredible things happening tough. I don’t know if you are familiar with the eGov Foundation in India. They’re basically creating an API schema for every government activity. The goal in India is to make it available to anyone who produces something with the public interest: a schema for reporting a problem, another for a parking ticket, one for everything. Their intention is that every vendor’s system will be interoperable because the inputs and outputs of those systems are all schematized. They’re in 1,500 cities already…
Renata Ávila: When you combine open with scalability, then the impact is widespread, of course.
Lucas Pretti: That’s also a great case for Frictionless, one of OKFN’s main projects today trying to make data more interoperable. That’s exactly the point.
David Eaves: That’s great to hear. I think we should be interoperability and openness are both critical – different, but critical.
Nikesh Balami: I would like to follow up on something you said earlier, when you mentioned the different types of open data initiatives, and how they evolved over time. Is there any initiative that you have been closely following recently, and that you think people in the open data community should know about? In other words, what are your favourite open data projects?
David Eaves: You’re probably familiar with the Open Up South Africa team. They have a whole range of projects. I like to show the Municipal Money project to my students as an interesting example. Again, how much is it used? I don’t know. I can imagine it’s used a lot by government officials, and journalists. But is it used by ordinary citizens? Probably not that much. Should it be? I’m not certain – I think the project is hugely valuable even without adoption by a lot of everyday citizens.
Getting back to the Municipal Money project, I think it’s great and opens the way for many questions. How can that website be scaled? How to make this port available in Brazil, for example? How do we figure out a sustainability model for this site? Maybe the way is to find a funder with a truly global view, start with a small grant and pick some countries we care more about, or where this is more urgent at a local level.
Oleg Lavrovsky: I have a final thought to share, not actually a question. While we were talking, I set up an Open Collective project to try to encourage David to get back to writing. Please have a look. Hopefully, this can get you back to expand this exact kind of wise and forward-thinking discussion that we’re having today. Let’s not stop here.
David Eaves: I would love to actually just troll through my blog and find a whole bunch of old pieces and assemble them together. It would probably inspire a whole bunch of new thoughts about how they need to be updated and rethought. That would be a great project. Thank you for that!