Predictions about the future of technology are notoriously tricky – today’s cutting-edge technology can quickly become tomorrow’s museum piece. They can also be misleading and a source of disappointment:
But when Tim Berners-Lee tells you about the ‘next big things’, you tend to sit up and listen a little more. At yesterday’s Open Data Institute’s Annual Summit, this was exactly what happened. So, what did Tim and others have to say about the future of open data?
- Internet of Things (IoT): This is an idea that’s been around for a while now. The ‘things’ part of the IoT refers to a huge range of devices with built-in sensors capable of communicating wirelessly via the internet. They can be sensors in train carriages, smart thermostats that can be controlled remotely, TVs, environmental sensors, medical implants or tracking devices fitted to wildlife. Each device generates its own data automatically, which it can send to another device or machine that can then be used as a basis for decision making. In traffic management, for example, road sensors could be used to identify traffic flows and that data could be used to make a decision on road closures or diversions. A more familiar example is an activity tracker, which can monitor your activity levels – calories burned, distances walked, heartbeat etc – and transmit them wirelessly to your phone or PC. From that, it’s a short leap to recommendations on improving your fitness or building more exercise into your daily routine. The advantage of this sort of automated data capture is that machines can quickly and routinely do this for us, unlike humans that tend to be slow and make mistakes. Of course, the IoT poses some huge moral questions around privacy, control and security. What becomes of the private sphere if all our movements and activities are tracked? How do we track who has access to our data? What happens if our activity tracker picks up a pattern indicating we’re likely to have a heart attack in the next six months? Does it tell us, our insurer, GP or nobody? But equally can we ever have the utility and ease-of-use we have come to expect from technology without the downsides?
- Machine learning: Rather than following precisely defined instructions inputted by humans, machine learning involves machines using algorithms to learn from data. The more data a machine has, the more it is capable of refining what it does. One of the more familiar examples is that of spam filtering. It takes humans a long time to read emails, so if we all had to individually separate out spam from non-spam it would be a laborious process. Having a machine to do the task for us is much easier. When you tell your email provider that a certain email is spam, you provide it with important information. The programme can then analyse the content, key words, layout and other information, which it can subsequently use to refine itself. As millions of users do the same, so the programme builds a vast knowledge of what is spam and what isn’t. Next time a similar message comes through, then, it should be easier to spot and filter it as spam.
- Fear: To be honest, this isn’t so much around the corner as smack bang in the middle of the block. And it’s been here for a while too. But fear is an important concept in the world of open data. Indeed, with new stories seemingly emerging each week about our data security, the privacy of our online conversations and an entertainment industry that has long feasted on our insecurities around humanity’s relationship with technology, fear is an inevitable part of any discussion on technology. It’s perhaps the single greatest barrier to the release of more open data. Companies fear that it may give their rivals extra intelligence or info. Governments fear the media picking through their data and finding embarrassing stories or huge errors in the dataset. And individuals fear releasing their data because of fears over their security and safety. All of these fears need to be addressed and tackled if more data is to be released openly; they can’t simply be swept under the carpet. But with appropriate safeguards, checks and balances the risks can be mitigated.
These are just three of the highlights from the conference: it would have been just as easy to write about any other aspect of open data from smart cities, to open data for art and open data and young people. The reality is that, perhaps without us knowing, open data has fast become a part of our everyday existence. Whether you use a smartphone to find out the next bus time or your PC to check the weather forecast or check in with Foursquare you’re using open data.