Thoughts on personal data vs non-personal data

There are opportunities to collect data from sources that aren’t being considered, inside digital government conversations. The separation of personal data from non-personal data is up for debate, meanwhile, companies profit from strategic information made proprietary

I’ve been thinking a lot about how the government, or regulators, can start thinking about privacy and the data economy before even developing any system requirement. The steps involved in procurement, for example, should consider some framework to demand digital systems, since these systems will impact directly in the life of citizens. On the other hand, some systems collect and consume data that isn’t related to any citizen at all — but still, affects them, and these data could be generating new ideas, both for public policies and local business.

Lots of efforts within governments have been made to collect all the data they can, embedded in projects that see data as unquestionable value, but leaving behind 1) considerations about citizens privacy (and subsequent human rights); 2) costs to maintain data stores at a long run, or even 3) costs to keep personal data secure and free from predatory activity. I understand that these efforts should be partially left behind, and a new perspective should be adopted in the public administration, especially where there is no regulation in place. It is understandable that the first devices to be considered as data collection tools are people’s cellphones, cars or smart devices that could be inside residencies, since they are already there, yet cheaper technology for data collection might be included in the conversations about systems for digital Government.

Cameras and microphones can be replaced by a more affordable technology, considering storage and infrastructure, proportioning better and accurate data, preserving privacy and human rights. But what is a reasonable way to start thinking about data collection for digital government systems if we don’t have regulations in place yet? We can’t wait for the regulations, because software has been developed for the last ten years, at least. Where there is software, there is data collection.

I had to make use of a Venn diagram, of course, to draft the idea that some data might be cheaper to collect, and it should be open, by default, to create an ecosystem where the government also provides the data, as a platform. Some other types of data should not be used by the government at all, and still regulated with strong guarantees for the users. E-commerce and social media are examples of these data. I recognize it is a gray area. Smart houses, for example, could contain data about water consumption but should not be used as a source in every case. To the contrary, less granular data, collected at general distribution points, could serve for a Federal Agency willing to oversee water management between cities, for example.

The “special data” — personal data managed by public systems by default, like transportation, financial or health data, should have strong regulations in place, with interoperability rules that are clear and debated with the society as a whole. Beyond that, the data collection rules regarding the special category need the understanding of the user, and the clear consent inserted on the system framework. As an example, I’ll give the link to the framework described by John Wilbanks & Stephen H Friend on their article for Nature called “First, design for data sharing”.

Venn diagram to verify before demanding a system for digital government

I understand that there might be skepticism on the use of a similar framework to classify systems even before they are demanded, officially. However, every day I see cases where simple, cheap solutions should be in place, like in the management of dams, avoiding environmental crimes like the one that happened in Brazil this year, killing more than 200 people and one entire river. Another example is data about infrastructure, obtained in not orthodox manners, like laser or sonar, for example should be strongly considered as a priority for a governmental data strategy. The governments are building their digital infrastructure — the arrangement that will support the sovereignty and capacity of management of resources in a close future, that might even be in place already.