On Friday I attended a fascinating series of keynote talks, followed by a panel discussion, about the opportunities for entrepreneurs to use open data. The talk was part of a series of events organised by the group Silicon Valley comes to the UK (#svc2uk). The panel was moderated by local angel investor Sherry Coutu and included:
- Andrew McLaughlin, Former deputy CTO for US Gov’t and head of ‘Code for America’
- Hans-Peter Brondmo: Head of Social at Nokia
- DJ Patil: Data Scientist in Residence at Greylock
- Megan Smith: Google
- Reid Hoffman: Greylock
First up was Megan Smith from Google, who talked about the opportunities for Google to open up some of their (huge) data sets via APIs, adding “providing privacy is maintained”. One example Megan gave was how the Google Books API had been used by Harvard researchers to analyse the trends in language in literature from 1750 to the current day. The word ‘women’ was apparently not used in literature extensively until the 1970′s, when its frequency of use suddenly rocketed. Megan also highlighted the potential for crowdsourcing data through social applications and the benefits of using Google data to inform the bio and life sciences. The example of trends in flu outbreaks indicated through searches for flu related topics in Google sprung to my mind.
Out of all the speakers, Andrew McLaughlin really stood out for me. Andrew was an advisor to the US government when he worked at Google. He then progressed to deputy CTO at the White House and is now the Head of Code for America. Andrew reminded us that we live in phenomenal times where there have been profound changes in the economics of information. He referred to the impact of internet based information on the relations between states and citizens. This is all too true when you consider how the ready access of information via mobile devices has fuelled significant changes in world politics in recent times, particularly in African and Middle Eastern countries.
Andrew cited Moore’s law and the impact that computing power has had on what we can do with vast data sets, both in terms of storage and processing speed. Coupled with the explosion in cloud computing capability, Andrew advocated the power of data to improve public services and drive positive changes for civic life. He also highlighted some of the challenges, for example data sharing and privacy.
Most interestingly Andrew recalled his experiences when he first started working in government. He expected to be able to drive performance management improvements using available data sets, but quickly found that no agencies showed an interest in making changes to enable data sharing and mash ups. However Andrew found there was appetite to open up data at a City level.
When the panel was asked the question, “which government data sets are most useful?” Andrew was quick to state that all public data should be freely available, so that people can find out themselves which are the most useful data sets, rather than having the decision made by those who manage the data.
At the start of the panel I sent a question by Twitter which I see as one of the biggest issues for anyone working with open data:
My question wasn’t answered directly by the panel, but DJ Patil did cover the topic to some extent. He talked about the challenges of turning data into a meaningful product. In other words, how do you make data useful and usable by people in their every day lives. DJ advised that application developers need to make big data “small and actionable”. He also advised enabling data to facilitate conversations with users, in order to improve data products.
The questions from the audience to the panel indicated a general concern about privacy issues. Although I can see that there are huge opportunities for young entrepreneurs, I also hope that they use public data wisely.
As a User Experience practitioner I believe there are considerable obstacles to overcome to ensure that real world problems and user needs can be identified and solved using data driven applications. There is already an overwhelming amount of data available these days via APIs. But there is far, far more data that still to be unlocked and provided via API. This is most definitely going to be a growing area for UX researchers and designers to be involved in. And events like hack days and design jams can facilitate learning about how to realise the opportunities in data.