The International Open Data Conference 2015 (IODC 2015) kicked off this morning with an opening plenary that featured prominent speakers with different viewpoints (the optimists, the critical), roles (evangelists, suppliers, intermediaries, users), and from different sectors (government, media, INGO, private sector, research institute). The opening plenary, and the sessions within the day, raised very important issues. Despite the fact that we have travelled long enough as an open data community, we still struggle with definitions, questions about what open data is, what can be made open, and how we define and operationalise data ownership.
From a developing country perspective, the opening plenary struck me significantly in terms of how we frame the open data discourse. To an extent, the discussion highlighted the differences that do not only exist in the open data sphere but also elsewhere – differences between countries, differences within countries, differences across disciplines. There are countries we categorise as North or South, countries that are developed and developing, countries with stable and functioning democracies and countries we inappropriately call fragile states. There are people with fast, stable, and affordable internet connections and people whose internet connection is so slow that downloading a dataset takes longer than cooking rice. There are people with advanced technical open data skills and people without any. There are disciplines that argue regarding open data’s economic value and others that question the very value and motivation of open data itself.
It’s quite disturbing that often we forget that these differences exist; that when we speak of the trajectory of open data, from Sebastopol then to Ottawa now, and when we describe the future of open data in the world, we assume that different countries, cities, organisations, or individuals have a shared view, or a shared experience. For example, it is understandable that when DJ Patil talked about “Open Data – The Revolution Ahead”, it focused largely on a developed country perspective like the US. However, it made me feel like the revolution he is talking about is so far from reach by the region and country I come from, and even close to impossible at the current state of things where we are experiencing extreme poverty of quality data, both at the national and subnational levels.
This theme seems to flow all throughout the day. For example, when Maciek Hawrylak remarked regarding Open Data Lab Jakarta’s presentation on open data and offline dissemination in a panel on “Data + Public Money” that it is important to remember that there are people in developing Asia without internet access, it sounded like an afterthought, and not really something woven into the very fibre of the discussion.
Also, the discourse on open data during the conference often becomes techno-centric and seems to reside within the domain of those who are able to understand what CSVs are, what metadata means, or how maps are generated. The panel on “Companies: Putting Open Data to Work” focused on how open data made possible the mushrooming of companies and start-ups. But one participant remarked that he missed examples from the global South, and a panelist replied that technical capacity is deficient in these contexts. This does not only highlight differences and inequalities in capacities, but also the deficiency in our understanding of open data as a value chain or infrastructure, as what Nigel Shadbolt argued in the end-of-the-day plenary. After all, open data is beyond portals, hackathons, and applications; one thing that the Kenya Open Data Initiative tellingly reminds us.
Will open data recreate the same inequality that some development interventions have had in the past? Will the powerful and the wealthy appropriate for themselves open data’s benefits? Or will open data be used, as what Sam Pitroda argued at the session “Open Data Around the World”, to level-off the inequality that exists among citizens in countries like India?
We cannot answer these questions if we frame our analysis devoid of context and propose that open data is neutral. We cannot also answer these questions if in conferences such as IODC 2015, we focus on open data initiatives that worked and forget to reason out why certain things worked the way they do. We also cannot answer these questions if we define open data impact as use cases and not changes in lives of people.
There is a serious need, I would argue, to rethink the way we frame the argument for or against open data and this should start from the recognition that, as we launch and spread open data internationally, there are specific cultural values and norms, laws and systems, actors and practices that define the field where open data operates in countries. This recognition should include the acceptance that directions may not be necessarily similar, open data initiatives will largely be differentiated, and that outcomes, if not impacts, of open data will not be the same for different types of intermediaries and end-users.
We need to build the evidence that open data will lead to concrete results in people’s lives. While this is difficult, as recognised by the panel on “Emerging Impact of Open Data”, we should not backtrack and propose to use stories instead to provide positive evidence. While it is correct to recognise that impact at this stage may not be something that can be measured yet, as the phenomenon is entirely new for most countries especially in the developing world, we should start to think of methodologies that can capture how open data has changed people and the communities where they live.
More than anything, the first day of IODC 2015 highlights the need for action – not only for governments to set standards of disclosure or policies of use, but also of other things. We need actions to create an enabling environment for open data including technology, political leadership, economic resources, socio-cultural institutions, among others. We need to build capacities of governments, intermediaries, and users, to supply, use, and understand data. We need research that would critically analyse open data initiatives and ecosystems as well as research that would help us measure open data use and impacts.
The list is long. The second day of the conference will hopefully provide solutions. With more than a thousand people helping together frame an agenda for action at IODC 2015 Day 2, there is a reason to hope.
Excellent review! I support many of the points raised here.
Interestingly though, one of the really interesting sessions for the conference was a one-to-one with public servants from Canada and the Caribbean. What came forth was that while one country may be further ahead in terms of policy, some of the same issues remain in practice.
Some of these really are as you note – context based. But here the context is not based on country to country but actually sector to sector. The focus here is on the public sector as a specific actor with a particular modus operandi, history, culture, etc. Much of the change that the Open data movement is trying to accomplish relates to the public sector and governments. This raises some important questions about the nature and operations of governments and the public administration machinery that have implications for the success of open data as a movement within government and government as an instigator for opening data in the wider society. So, what are the specific features of the public sector (regardless of location) that can impact positively or negatively on open government or the open data movement more generally? What are the antecedents to the Open data movement within public administration and what does the success or failure of these efforts say about how governments (developing and developed) should or can be engaged in the open movement? The latter is both as a sector in need of transformation as well as an instigator of change in the wider economy. As the public sector has gone through a number of reform trends over the years, what do these suggest about how this particular reform agenda should be governed? How can open data be embedded within the public sector? Admittedly, delving within this category (i.e. sector level) will show up nuances though these may not all be based on levels of development but also regulations, extent of regionalism, administrative culture, etc.
The kind of sharing experienced in this session may not have been facilitated in the larger conference given that while much of the experiences of the developed world were being presented, these were sometimes in an unquestioning way, while those in the developing were being interrogated sometimes through the same value-laden language that Tim Davies alluded to in an earlier blog.
Ultimately though, one of the benefits of the open data movement is that developing countries have an opportunity to acutally shape the emergence of practices, models and theories, which has not always been the case in other waves of reform. In some cases, it may take a good 20 years before reforms initiated in the developed actually reach the developing world but as seen at the 3rd IODC, developing countries are integrally involved in the movement (even if through a few actors). As such, there is the potential to benefit in real time and shape the discourse and actions and not simply be policy-takers.
Although day #2 did provide us with some success stories, it didn’t necessarily gave us any solutions. I do think we shouldn’t seek solutions in conferences like this, instead we took the lesson learned from the success stories and try to apply and adjust it in our own context to create our own solution. We then bring our solution to next year event as our success story, and so the cycle continues.