OpenData has been a hot topic for the past few years, it means many different things to different people, the commercial sector see’s, (should see?), the potential to leverage OpenData for the creation of new products and services which they can sell to make money, e.g AirBnb and Uber both rely on access to other companies spatial data, (maps). The public service sector should be able to use the data to create better public services, and government decision makers should be able to use the data to make informed decisions.
Note the use of words like “use” and “leverage”, both imply the ability to access the data, and more specifically to be able to access the data in a Machine Readable format. As you can imagine the data we are talking about comes from many different sources, and will actually exist in repositories/databases all over the internet, maintained and contributed to by many different actors. The different user groups who need to use the data will most likely do so with their own specialized tools, be it ArcGIS, QGIS, STATA, R etc, so it is critical that their tools can talk directly to the data they need, thus the need for the data to be in a Machine Readable format.
One of the issues, especially in Tanzania, (and probably for many other regions of the world), is who/what is going to seed that initial collection of data AND who/what is going to make sure that the people who could potentially use the data are aware of it and can actually access and use it. Increasingly donor funded projects are facilitating the collection of huge datasets, the Zanzibar Mapping Initiative, is an excellent example of that, where the whole island was mapped using small scale drones. This for the most part by Zanzibari’s for Zanzibar, and the data has been made available as OpenData via the ZMI Geonode, and OpenArialMap. This in itself is a great win, especially as the approach taken was to train Zanzibari students and surveyors so if nothing else it has increased capacity. We here at Uhurulabs rely on our commercial contracts as they allow us to subsidize our public/innovation sector work and for our commercial work we rely on professional human resource, see our pilots, two of which are from Zanzibar and products of the ZMI project.
So in the Data collection has been seeded and done, and it is available as OpenData, how has it been used so far? The data has been used for things like building footprint digitization, we hope to provide a link to that soon. It has also been used by the Zanzibar Commision of Lands for land use and city planning, we also hope to provide links to information about that. Another very exciting use case has been in the Open AI Tanzania Challenge. which invited data scientists to develop feature detection algorithms that can automatically identify buildings and building types using high-resolution aerial imagery. It is our hope and expectation that the classifiers that were developed will be released as OpenSource and as such we have asked on in the Challenge Forum.
Another very exciting initiative is Ramani Huria…
“Ramani Huria is a community-based mapping project that began in Dar es Salaam, Tanzania, training university students and local community members to create highly accurate maps of the most flood-prone areas of the city. As the maps have taken shape – their benefits have multiplied and their potential magnified, now serving as foundational tools for development within all socio-economic spheres beyond flood resilience. The project is supported by funding from the UK Department for International Development through the Tanzania Urban Resilience Programme“
We were lucky enough to be involved in this project, most of our work involved collected data using drones. It always seemed that to some degree the drones stole away some of the attention from the _real_ work that was going on, on the ground. Everyone was very impressed with the quality, speed and accuracy that drones were able to capture high quality aerial images, but the real rich, actionable data came from the boots on the ground. The data includes things such as if a particular house has been flooded, and if so to what height, if a structure is public or private etc. What is really special about this data is its OpenData, the collection of which has been funded by public money, and done for the most part by specially trained university students and now available for free for public servants to make informed decisions, and business to create new services from.
In the maps below you will see the standard OpenStreetMap basemap, all the data from Ramani Huria was contributed to OpenStreetMap. However you will notice more detailed layers on top that give you more information and allow a deeper zoom level. You can then use the feature info tool (i), to get detailed information on the various assets, e.g clicking on a building will give you information such as if it is residential or not.
It is not however just donor lead projects that are releasing OpenData. Tanzania joined the Open Government Partnership, (OGP) in 2011.
Tanzania declared its intention to join OGP during the launching meeting in September 2011, one of six in Africa that qualified to be involved in the OGP.
Make the data available – in bulk and in a useful format. You may also wish to consider alternative ways of making it available such as via an API.
Make it discoverable – post on the web and perhaps organize a central catalog to list your open datasets.
Note the words in RED, again these words get thrown around quite a lot but what do they mean in this context? 99% of the people who use an API will never know they are using it, it simply means that the data is stored in such a way that the software YOU use to get your job done can access it, without you needing to know the technical details of HOW its being accessed. An example of this is when you go to google maps and search for a location, and then tell the app to give you directions to there, you don’t need to know the technical details of HOW that is happening, or how it know’s that there is high traffic on a certain route. You simply want to enter a location and click a button, and it is the fact that many systems make their data available via an API, that allows it to be possible.
What does discoverable,mean in this context? Simply that a user or system is able to find the data when it needs it. This can be as simple as a search box on a website, or more complex e.g system that pushes you information based on your location and previously identified preferences.
Which now actually brings us the main point behind this post, yes it is true that Tanzania has made great strides in its ability to collect, valuable geospatial and other data. It has also made good progress in its ability to then make that collected data Open. Where I think it is now struggling is in the ability to store, manage and reliably make that data available in easy, useful, discoverable systems using things such as API’s and well crafted interfaces that can be accessed by all stakeholders. This however is a struggle that is not Tanzania’s alone, the whole world is seeing an explosion in the availability of data, it is coming from all kinds of sources, drones, new low orbiting satellites, weather stations, from government, private sector, individual citizens etc. At the same time there is an increasing awareness that good decision making comes from the utilization of good data. Last month we were invited to speak at the Smart Land Administration forum in Finland, where the topic of the day was Spatial Data Infrastructures, (SDI),
A spatial data infrastructure (SDI) is a data infrastructure implementing a framework of geographic data, metadata, users and tools that are interactively connected in order to use spatial data in an efficient and flexible way. Another definition is “the technology, policies, standards, human resources, and related activities necessary to acquire, process, distribute, use, maintain, and preserve spatial data”.[1]
A further definition is given in Kuhn (2005):[2] “An SDI is a coordinated series of agreements on technology standards, institutional arrangements, and policies that enable the discovery and use of geospatial information by users and for purposes other than those it was created for.”
Note that from the above definition it is clear that the weight is on the institutional policies, standards, and agreements which need to exist. Without good strong institutional policies, standards and agreements an SDI or anything like an SDI cannot exist and as such the public will NEVER realize the true potential of the Data revolution. Data will continue to be collected as the market understands it has value, but rather than a tool to liberate the general citizen, it will be a tool to control and oppress.
Keep going to the government websites I mentioned above, use the data, and when you have questions about it ask the relevant body which is very often NBS, the reality is for the most part people in Government are their and willing to help. It is only by the use of such data do people start to understand its value and are motivated to make sure it is kept up to date and available. If you find a website or service to be down then inform the owner.
At Uhurulabs we are committed to servicing and helping anyone who wants to use data and technology for the betterment of Tanzania and its people, as such we are committed to maintaining a number of services, the first of which is now live and is a Geonode that is available for anyone to host geospatial data, and for anyone who wants to access the data.
https://geonode.uhurulabs.org
Our resources are limited and as such we will try to scale the service according to demand, if you experience any problems please let us know by emailing to geonode@uhurulabs.org. For the most part what we will be trying to do is aggregate datasets that we find in other places, especially
data that is not available via API
data that we believe is at risk of being lost
data we have collected ourselves.
The second is one which came to mind while writing this article, and it will be a page that monitors the various websites and services that provide access to data on Tanzania.
