The open data movement is a worldwide growing trend and has become a global phenomenon setting a new agenda on access and public services delivery. Its impact on the way we build community life is undeniable, too. A hot topic crossing the borders of the first advocates and early adopters. It has become one of the issues of the day.
Of course, those working more directly on projects related to open data, both from public management (fighting, much of the time, against visible and invisible walls slowly falling down) and from private and civic sectors, creating solutions and tools for collective use of public data for different purposes, are well aware that this widespread of open data initiatives throughout the world is not a good indicator to measure the success. It simply reflects a trend. But celebrations are always tempting and might let us forget the final objective of open data.
I could not attend Future Everything last week and I fully regret it, as it gathered some panelists and speakers that are among my most favourite names regarding the intersection of technologies and urban living, open data and related topics. One of them, Usman Haque, shared some ideas worth remembering and I hope we can soon watch the video. In the meantime, the notes from his talk, In Praise of Messy Cities / Grub Street & the super wicked, seem to be a great link to keep in mind. Telegraphic but food for thought, here are some of them:
- assumes if we had enough data we could make perfect decisions
- attempt to understand, explain, control »> “It’s not me, it’s the data, which is impartial”
- Borges metaphors: map, library, encyclopedia »> implies infinite data, merely need an index and our problems will be solved
- often only the least controversial data is ‘opened’ »> transportation data, not wikileaks (japan has efficient transport and has done for decades, not an open data/technology issue)
- is ‘open data’ just letting off steam?
- society of the data spectacle, discourages participation, no accounting for the curator of the data, their goals, or the feedback mechanisms for propagating the principles
- city is a mess. not infrastructure. data infrastructure. praying to the algorithm god isn’t going to solve this.
- embrace super wicked urban problems, don’t reverse-engineer problems based on existing solutions
- the spectacularisation of data, revelling in complexity only so that ‘experts’ can rescue us from the cacophony: scientists, urban planners, yes, even artists
- the concerning thing about this neo-postivism is when it’s applied to the design and manipulation of our cities because these processes have their own ‘god fantasies’:
- efficiency (those big biz initiatives that use “Smart” throughout their PR material)
- all the things that go counter to the sustainability of what makes a city a city
- social goals that rarely have anything to do with technology and sound suspiciously like the sorts of things urban planners were saying in the 50s and 60s when they gave us highways and highrises/tower blocks
Among other ideas, Haque depicts a fundamental aspect when translating the predominant discourse on smart cities to the reality of urban living, where complexity and unpredictability are basic features while some visions on big data pretend to have all the answers and even constitute a new science of cities. On the other hand, he is remarking a counterintuitive idea: while availability of data seems to be growingly accessible and create new freedoms there is the risk of turning this into a big barrier for the non-expert.
Some months ago, David Eaves also raised in his article Lies, damned lies and open data some related issues that deserve to be mentioned, as he points out that we cannot remain paralyzed celebrating the success of a movement so strong and influential when, in fact, open data does not eliminate the need for political debate. This idea fits with the risk of neo-positivism of data (also mentioned by Haque) or technological determinism: we made to make public information available, accessible and transparent, so the data are already there and are clear, objective and unbiased, and yet, they are only the material-and enough progress is access to it- to critically intervene on reality. Again, a new warning against the risk of depoliticization and the asepsis of raw data. Access to evidence of government activities and to use and manipulate the data deriving from them to create tools, applications and services is not the end, but the beginning:
Quite the opposite. Open data will not depoliticize debate. It will force citizens, and governments, to realize how politicized data is, and always has been.
How do public institutions get public data? And what kind of use they make of them? What kind of underlying bias are they using to choose some topics and not others when processing the generated information? These and other questions have always been part of political confrontation and will remain.
In this post, Slee addresses the role of big companies and digital entrepreneurs in selecting what and how is released and reused, how this ecosystem works as a market, the always relevant demographic and cultural bias ("empowering the empowered"), etc. This position can be seen as critical or cynical from the point of view of open data enthusiast, but should not be underestimated because it is the key to the whole thing. In fact, David Eaves responded to Slee in a worth a read article and both articles reflect the fundamental debate that needs to be maintained to give a political horizon to what it cannot be merely a technical issue but a political one.
These inputs put a little wary of the risk of triumphalism and techno-determinism. Open data, along with other movements, have challenged traditional public management logic, creating new ways for collective and concerted action, but there is a long way forward and a less reductionist vision is needed to avoid the trap of thinking a perfect ideologically data-driven neutral future is here and algorithms will be working on our behalf. On the contrary, complexity, confrontation and power-based conflicts are, as always, relevant.
Puedes leer una versión previa más corta en castellano en Ya tenemos open data, las preguntas son las mismas.