Come and get it! OpenOil publishes first open API for oil rights
OpenOil is pleased to announce what we believe is the first open API applied to oil extraction rights around the world: the 20,000 “blocks”, “fields” and “concession areas” (we’ll get into naming problems a little lower down) from some 69 countries involving some 2,142 unique corporate structures as operators and contractors.
We hope this Application Programming Interface is a significant step towards a formal, open data standard around extraction rights.
So what exactly is an API and why do I care?
I’m glad you asked!
APIs are popping up all over the Web. As our developer Dan says in his documentation, the OpenOil API will then allow smarter queries of the data with questions like “What concession blocks exist in Uganda”, or “Who holds exploration licenses in Brazil”.
All OpenOil data is open – you can already download the entire data set. But with large data sets – and we hope to grow this data set in depth and breadth fairly rapidly – you may not want to be overwhelmed with it all at once, and be left poking through vast spreadsheets. Using the API allows you to frame a question to get just the data you want.
The second thing an API offers then is automatic updating. Say you want to include, for your members, clients, citizens on your own website, a list of all offshore oilfields known to be active in West Africa. You can build a query which retrieves the current data, embed it in your web page, and it will retrieve all the (oil)fields in the database. And when we update, your page will update automatically.
These are fairly simple queries. But the API function achieves more and more value the more data you add to the system. Imagine, for example, you want to be automatically informed three months before any designated exploration period is coming to an end in Brazil, and the company either has to declare commerciality or relinquish the field? Or that you would like to know when any of BP 1,200 affiliate companies farm into or out of operations anywhere in the world?
But our purpose in publishing the API, supported by the Shuttleworth Foundation, is not just to make our own data easier to access. It’s to suggest a step towards an open data standard around extraction rights. And the point of that in turn is to enable automatic information exchange between any two parties.
This is hugely important. No one organisation can – or should – be responsible for gathering and updating data around extractive industries. But if we have agreed standards, agreed ways of referring to how extraction rights are granted, then all interested parties can exchange information with each other quickly and painlessly.
When Web developers can easily set up mechanisms by which the Nigerian government can exchange information with the UN Environmental Programme, ExxonMobil, and EITI, – without the need to refer to us or any other institution, that is when we will have an open data standard.
And it’s that frictionless exchange of data which will be really necessary to allow the growth of analysis and understanding at a higher level of extractive industries as entire systems at a global level. To fulfill the why of transparency initiatives, in other words, and create independent analysis in the public space which genuinely empowers everyone touched by these industries.
We have based this API, then, on the idea that we can jointly create a namespace which uniquely identifies every spot on the planet as it relates to extractive industries, and in a way which does not rely on OpenOil or any other party as an authoritative reference point or information provider.
This follows the simplest possible logic of units and measurements which are uncontested and runs as follows: all extractive rights are awarded by sovereign governments. Each government uses unique naming procedures within its own jurisdiction to demarcate areas, and all rights have a start and end date (if current, the end date is some point in the future). Put those elements together and you can uniquely identify any rights granted to the smallest level.
Once you have a way to unambiguously define these lower-level “atoms”, anyone can take them and layer them into composite and more interpretative structures – such as implementation of EITI’s 2013 data standard to make license registries available, for example, or NRGI’s initiative to define extractives projects. In fact, if a data framework is successful, it tends to “disappear into the furniture” of higher level applications which make use of it.
There are a few methodological wrinkles to our approach – which is why this API and blog are also an invitation to join a public debate about how best to build a formal data standard. The sovereign state principle is not absolute, for instance. Iraq’s central government in Baghdad challenges the right of the Kurdish regional government to demarcate areas and award contracts. And although we haven’t found a case in practice in the 20,000 records we currently have, it would be possible in theory for there to be two blocks with the same name in the same country.
Geospatial co-ordinates, of course, are in theory unambiguous. But in practice there is no way of predicting when such data will become very broadly available, so to make a naming convention depend on it would effectively be to delay its implementation indefinitely. And there are also complications around cases where the physical space allocated for rights changes, even while its name remains the same.
The oil industry itself has had a data sharing project for almost two decades called Energistics. But this has concentrated on standards for exchange of geological information on reservoirs, and drilling and production data. We were not able to find an open standard referring to the granting of rights.
Just as last year, the draft convention on naming contracts was proposed – and a decision taken to adopt and evolve the standard together by colleagues such as NRGI, we hope this API can also serve as the start of a discussion around how the governance community can agree and implement the open data standards which we all need to take our work to the next level.