A new skepticism on open data? | Opensource.com

A new skepticism on open data?

Posted 06 Aug 2012 by 

Paul Brownell (Red Hat)
Rating: 
(5 votes)
Venn diagram of data types
Image by : 

opensource.com

submit to reddit

Resistance to open data is much older than the concept of open data itself. Those who control—and/or benefit from the control of—data have traditionally resisted its open dissemination.

This resistance is being steadily eroded by government policy (see open data policies in the US, UK, and a long list of other national, state and local governments), by growing social and political movements in Europe, by technological advances such as the move to “Big Data,” and by the continued work of the broader open source, open content, open access community.

However, an additional line of resistance (or, at least, skepticism) is appearing that breaks from traditional positions based on vested self-interest and on public policy arguments such as the efficacy of releasing government-funded R&D. A lengthy June blog posting by Tom Slee entitled “Seeing Like a Geek” calls out an “open government data doppelganger—the shadow of commercial interests that follow civic hackers wherever they go.”

Slee argues that the disruption caused by open government data can have the effect of “empowering the empowered,” particularly “in countries where the gap between rich and poor is large.”

He points to compelling cases of data projects that have greatly benefited self-interest groups. In one example, he criticizes Brandon, Manitoba for releasing property tax and assessment records for every property in town, allowing commercial groups to target wealthier individuals in the area. In another, he points to the Tamil Nadu (India) government program that digitized land records where property lines were contested and clear land titles didn't exist. The poorer residents contend that the wealthier and more tech-savvy residents used their influence to “fix” the records to their advantage. Through these & other examples Slee concludes that he is “not convinced that a coherent case can be made for ‘open data’ as a public good, independent of the social changes that must accompany it, until the movement confronts its doppelganger.” He ends by calling for more and stronger policy “experimentation” in standards, licensing, and other areas.

As governments and other institutions digitize more public records. there are serious policy questions to answer regarding privacy and data integrity, which are beyond the scope of this blog post. However, Slee misses the mark in laying the blame at the feet of the open data movement. Mainly, he confuses open data with publicly available data.

As has been previously posted, Open data does not mean that a government or other entity releases all of its data to the public. Rather, open data means that whatever data is released is done so in a specific way to allow the public to access it without having to pay fees or be unfairly restricted in its use.” The problems in the previous examples stem from what data (or portion of data) is made available, when open data is really about how data is made available. Simply put, a citizen or consumer should not have to buy a particular vendor's product in order to be able to open, use, or repurpose public data, having already paid for the data collection through their taxes. Open data does notand cannotguarantee that the data being released by the government is useful or of high quality. Policies do also need to be in place to address privacy concerns, but that is a hurdle regarding what data should be released before the government decides to release it.

Arguments such as Slee's are being put forward at a time when much of government is engaged in such—or similar—discussions. In Europe alone, the European Commission is in the midst of a major proposed data protection regulation and just released an ambitious new strategy to promote cloud computing.

This underscores the importance of engagement—as individuals and as a community—in the important decisions being taken by our policy leaders. The data definitions we use will be critical as these policies are crafted and as we finally begin to see the impact of open data on society.

submit to reddit

3 Comments

Greg P

A starting point might be for more companies to work on opening up their own data. I work in a hospital using proprietary software for its medical records, and we are totally dependent on the software to access and collate our own information. You can only do what they manage to build into the software. Creative, useful searches? Forget it.

Vote up!
0
Vote down!
0
Unidentified

Governments aren't the only ones. Corporations (especially substantial ones) should operate openly and transparently so that whistle-blowers and the people can hold them accountable and call out corruption.

Vote up!
0
Vote down!
0
Unidentified

i am not a native english speaker, but so far, my english was always good enough for opensource.com articles.
with this one, i had some problems. to be precise: where you explain the specific cases of open data that Slee critiques.. i did not understand them. i somehow understood that rich people had a better way to exploit poor ones, but did not get how.
... it feels a bit like you have to spend too much time talking to lawyers. ;-)

Vote up!
0
Vote down!
0

Paul Brownell is EMEA (Europe, Middle East, Africa) Public Policy Director for Red Hat. While currently located in Europe, he previously led Dell’s Federal Government Affairs team in Washington, DC. He also represented the National Venture Capital Association (NVCA) and the American Electronics Association (AeA) before the US government. Paul splits his time between Geneva, Switzerland (where he lives), Brussels, and other EMEA capitals.

What is open government (promo)

Reader favorite