The Fine Print on Open Data

Gartner analyst Andrea Di Maio filed an interesting blog post recently pointing out some pitfalls of the new open data revolution.

The most interesting part of Di Maio’s argument, at its most basic, is this: if government isn’t analyzing the data, then you’d better start worrying about who is.

Under the old model, the public usually only saw the cleaned up product of raw government-gathered data about weather patterns, demographic movements, soil contamination and other issues. That meant there was only one set of biases to worry about: the government’s.

With dozens or hundreds of public and private organizations parsing this data, we -- as consumers of those data products -- had better become much savvier or, occasionally, we’re bound to get snowed, say by an interest group’s analysis of global warming data.

A standard argument is that the vast majority of open data analysis is and will be done on figures the government wouldn’t have parsed itself -- or wouldn’t have parsed in that way or to that degree -- so this is all a net gain. That’s undoubtedly largely true. But, in a time of declining budgets, the government also is likely to crowd source some analysis it might have once done itself.

None of this is to say open data is a bad thing. It has already proved itself a boon to the transparency and commercial worlds. But, as Di Maio points out, it’s important to insert a note of caution into the revolution.