Data for Africa, by Africa: Rebooting open data and the High Level Conference on the Data Revolution

The original post by Savita Bailur, Open Data Research Lead of the Web Foundation, could be found here.

In this extended blog post, Savita Bailur, the World Wide Web Foundation’s Open Data Research Lead, reflects on last week’s ‘Data Revolution in Africa’, held in Addis Ababa. The World Wide Web Foundation was delighted to be in attendance to bring together a range of African civil society organisations, open data advocates and technology pioneers. The event resulted in the creation of a new ‘Africa Data Consensus’, which is intended to support evidence-based inclusive development policies across the continent, and states that official data should be ‘open’ by default.  We want to thank everyone who has taken part in the discussions that have made this step forward possible and we look forward to seeing how the African Data Consensus will lead to action being taken across the continent.

 

We’re all still buzzing from the energy and excitement from the events of the Data Revolution in Africa (27-29 March), particularly the Web Foundation’s Rebooting open data in Africa workshop in Addis Ababa on Friday (27 March). Following on from the sister Regional Agenda Setting workshop for open data in Asia in Jakarta last month, our rebooting open data workshop had a similar aim of understanding the state of open data, this time in Africa rather than Asia, and scoping future research and lab opportunities. As with the Jakarta workshop, this workshop was supported by the IDRC-funded Harnessing Open Data to Achieve Development Results in Africa and Asiaproject and organised by the Web Foundation. The workshop also had another aim – we formed part of 17 data communities who were invited to form an African Data Consensus and present it to the High Level Conference (HLC) on Data Revolution in Africa. The HLC was being organised by the United Nations Economic Commission for Africa (ECA), the African Development Bank (AfDB), the African Union Commission (AU), and the United Nations Development Programme (UNDP). The Web Foundation was an official partner.

The open data community’s workshop agenda was packed and the room full with almost 60 participants from civil society, academia, government and private sector across Africa, including representatives of African Development Bank, USAID, ActionAid, Africa Freedom of Information Centre, Tax Justice Network Africa, Code for South Africa, IDRC, Social Justice Coalition, Friends of Lake Turkana, iHub Kenya, the Pan African Parliament, Ethiopian Ministry of Communication and Information Technology and many more.

We kicked off using the hashtag #ODAfrica2015, asking “where do you see the value of open data?” This started a flurry of tweeting, projected onto a Tweetwall left up throughout the day: sectors were mentioned such as agriculture, financial transparency, and healthcare, as well as outcomes such as better decision making, more efficient expenditure, overall transparency and accountability.

These were picked up by powerful lightning talks on the importance of open data in tax justice (Savior Mwambwa, Tax Justice Network- Africa (TJN-A); in environmental and infrastructure transparency (Angelei Ikal, Friends of Lake Turkana) and parliamentary transparency (Jessica Musila, Mzalendo). Leo Mukutu and Adi Eyal, respectively from iHub Kenya and Code for South Africa dug deeper into these cases on how open data added value. Next was a session on research and capacity building, from on-the-ground experiences in Mali and Uganda, the IDRC perspective, and Web Foundation researchers on the Open Data Barometer in Africa as well as the Open Data in Developing Countries research, reiterating key issues of intermediaries and capacity building discussed in the earlier session. Former Permanent Secretary of Kenya’s Ministry of Information and Communication, Bitange Ndemo, also dropped in to share his experiences from Kenya’s open data portal and way forward.

The final substantive issues discussed were on the need for more policy and investment, respectively by Daudi Were (Ushahidi) and Ivo Njosa (African Development Bank). Eko Prasetyo, Policy Manager at the Open Data Lab Jakarta (co-funded by the Web Foundation and Ford Foundation) also gave a presentation on the practical methods of engagement with governments that the lab is employing in Indonesia.  All presentations from the day can be found here.

Towards the end of the day, we went relatively high tech again, asking participants to fill in 2 quick survey monkey questions along with overall evaluation of the workshop. The questions asked:

1. If you had $1 million what would you invest it in terms of building capacity of open data in Africa? and
2. If you had $1 million what would you invest it in terms of research in open data?

The overall answers are here in the raw form but the key themes which emerged from both these questions and overall discussions through the day were:

• Funding in capacity building of CSOs, researchers, training of government officials
• Prioritising research into impact, action research, hybrid methods of open data (going beyond portals) and other methods
• That citizens across Africa should be able to access, understand and engage with open data. Data collected by African governments, belong to the people: it is aright to data.
• An agreement that open data should be accessible and free and timely (“data delayed is data denied”)
• High-level political commitment to proactive disclosure of public sector data, particularly the data most critical to accountability
• Sustained investment in supporting and training a broad cross-section of civil society and entrepreneurs to understand and use data effectively
• Contextualizing open data tools and approaches to local needs, for example by making data visually accessible in countries with lower literacy levels.
• Support for sub-national data initiatives as a complement to national-level programmes
• Legal reform to ensure that guarantees of the right to information and the right to privacy underpin open data initiatives

In particular, we discussed the need for emphasis on:

• Geospatial data/georefencing data
• Use of satellite data (e.g. agriculture, climate change – increase capacity of countries to use this data and open this up – most of these data is available (sold) outside the continent – we need to access freely within Africa)
• Open budget data
• Municipalities to be given opportunity to analyse and collect data to improve local urban governance
• Not just capacity building but also more “power” to NGOs
• Starting with the champions within governments
• Respecting privacy and sensitive datasets
• Knowledge and awareness around licensing
• Working on and making “operational data” open e.g. how many nurses per hospital
• Working more with governments, private sector and research institutes
• Ensuring more communication and coordination between all data communities
• Working more with the “excluded” so far
• Working with governments to offer our expertise on open data – an opportunity such as this to feed into the HLC was critical
• Understanding and facilitating between different types of govt data (e.g. parastatals)
• Driving the demand for data which should create ownership, use, add value to data
• Involving the media who will communicate to citizenry
• Championing capacity building
• Working with community radio stations which will be good tools for grassroots awareness and advocacy as well as other actors/hybrid methods for disseminating
• Engaging with technocrats in governments
• It doesn’t stop at “Africa”: contributing to international data communities
• Championing governments on timeliness of data released: “Data delayed is data denied”

There was also an additional question in our survey on “what was new that you learned about open data today” that generated much discussion, including the point that we needed much more time and space than one day for the workshop, as well as learning from concrete examples (e.g. Indonesia, Uganda and Mali).  In the final session, we worked together to present the open data community’s vision to the HLC on the data revolution. This was presented under a tight deadline by the end of the day to the drafting committee, who worked this into a preliminary African Data Consensus.

This was opened up with the other data community contributions in a plenary discussion the next day (28 March) and we were happy that our colleague Eko Prasetyo was again invited from the Open Data Lab Jakarta to give a lightning talk on the Jakarta Lab model. Given the number of questions and audience members who came up to engage with Eko, we know that the Lab will receive a lot of interest and requests for more information.

And we were finally delighted when the press release of the African Data Consensus emphasised open data, beginning with: “African governments should acknowledge open data provided by recognised communities as sources that complement the work of national statistics offices for better decision making, service delivery and citizen engagement, concluded a group of data activists today in the Ethiopian capital”. A big win for the open data community is that the final adopted Data Consensus clearly states that official data should be open by default.

Of course, the discussions at our workshop were not entirely positive. Amongst challenges, we discussed:

• Governments’ fear of the unknown makes them perceive risks associated to the open release of data, especially in areas that can lead to increased transparency and accountability. Small scale, short term, narrow/sector focused initiatives, should lead to quick wins for the government to better understand the risks, challenges but also benefits of embarking on open data initiatives, and allow them to embrace open data across the whole of government showing strong leadership and commitment.
• Technology and data infrastructure is scarce across Africa, data is scattered, and there are very low skills to collect, analyse and disseminate data. Capacity building across the whole data value chain is needed for data to be made useful and to be ultimately transformed into information that can improve people’s lives. The role of intermediaries must be strengthened as well as hybrid methods of extending the range of the web (e.g. community radio; wall paintings etc).
• Releasing data without clear purpose leads to poor take up and disillusionment. Data release should be driven by demand and relevance. People’s everyday issues should be properly identified through participatory and inclusive processes and data release should be prioritized accordingly to address them.

There was also a concern that multilateral institutions are resourcing governments to spend significant funds on open data portals, without analysis on who is using them and for what. As one participant stated “open data is beyond portals and beyond just the technology”. At the HLC itself, it was clear that governments were reluctant to go beyond official statistics as sources of data, which was reflected in the eventual AU/UNECA Ministerial statement. And finally, Twitter made clear there was much work to do in traditionally “non-open” governments (such as Ethiopia itself), particularly when we were faced with Tanzania passing a Bill in the midst of the workshop to limit publication of data to only those from the government’s own Bureau of Statistics – a serious setback to the country’s Open Government Partnership commitments.

So, where do we go from here? There is much to digest and we will produce a more comprehensive workshop report in the next few days. We’ll also make sure we have all the presentations and photographs available in a public folder as soon as we have consent from all. From a research perspective, we will start identifying open data research partners and priority areas as soon as possible to work with on open data in Africa. We are architecting a lab model for Africa we intend to start implementing in the next few months. We will also continue to directly engage with members of the HLC panel to ensure that open data remains high on the agenda… After all, as Daudi Were stated during the workshop “any data collected by governments belongs to the citizens of that country … whether African governments or any government”.

Note: Thank you to all those who attended and participated, and special thanks to Nnenna Nwakanma for overall organisation, Irungu Houghton for facilitating and Emmy Chirchir and Denise Karunungan for fantastic support. Please continue to use  #ODAfrica2015 to keep the discussions going!