On data curators
First published: 31st December 2013
Someone recently referred to the role of Spend Network (and other open data businesses) as data curators.
It is quite an elegant term for what we do.
We are a business, we exist to make money. We take a raw material (openly published government data) and manipulate it so that it can be used for insights and intelligence. We sell that insight to government and suppliers so that they can make better decisions about buying and selling.
In order for us to sell data it has to reach a professional standard that means it can be analysed and evaluated. Poor data causes problems, so we try to find the errors and weed them out. So, when Tower Hamlets spends three times more than average one month, we look to see why. When the same Council publishes transactions with Badenoch & Clarke for 23 months out of 24, it also pricks our curiosity. In both cases we’ve emailed the Council and asked them to check their records.
By using the data ‘in anger’ we spot anomalies, we find errors and we challenge publishers to put mistakes right. This work is necessary for our clients, but it also helps to make publishers more aware of their responsibilities when publishing data. Hopefully, knowing that they’re being watched will improve the quality of the data being published for everyone.
Originally published at spendnetwork.github.io.
