it's the title of a very nice editorial (pdf) by mike batty. i'm from the bit (and not from the beat) generation, and yet i like to cut & paste bits from the editorial (i would recommend to fully read it though):
I first wrote about `smart cities' almost as soon as I began writing these editorials in the early 1980s. ...Â In the 1980s the focus on instrumenting the city using network technologies was enshrined in the idea of the wired city. ...Â Many of these conceptions were based on visions of what wired cities might become rather than on the reality of what was actually possible then. ...Â What has changed these initial conceptions of the wired city is the development of ubiquitous devices of comparatively low cost that can be deployed to sense what is happening over very small time scales - seconds and faster - as well as over very fine levels of spatial resolution.
The idea of integrating much of this diverse data together to add value to our conceptions of how it might be linked to other more traditional data as well as focusing it on specific ways to make cities more efficient and more equitable, has come to define the `smart cities movement'.
Most urban theory and indeed planning and design fifty years or more ago was predicated on radical and massive change to city form and structure through instruments such as new towns, large-scale highway building, redevelopment, and public housing schemes. Planning was little concerned with smaller-scale development except its design, for nowhere was the function of the city understood in terms of how small spaces and local movements sustained the city. In short, the routine and short term were subsumed in the much longer term. New data and big data are changing all of this...
This is an issue that has barely been broached to date - how short-term big data informs longer-term data is part and parcel of our concern for how we might integrate traditional datasets from household interviews and so on with crowd-sourced data where there is less control, and remotely or directly sensed data.