Submit Comment

show all (0)
There are no comments. Click the text to your left to make a new comment.
1 0

Advocacy for OGD release often focuses on ‘machine-readable data’, sometimes implicitly, if not explicitly, to the detriment of focusing on human-readable data. This can be based on assumptions that interfaces will inevitably be built onto data, providing non-technical users with access to the facts and information they desire.

2 0

In both E6 (Refugee Charity Research) and E7 (School Admission Appeals), although the data-users were technically skilled[1], their data use relied upon directly exploring spreadsheets. Either relevant facts were only available with enough granularity from these datasets, or they were searching for any relevant data they could find, as in E7, where the interviewee was pursuing a school admissions appeal:

3 0

“…when youre in that sort of situation you want to know about what happens to other people and how likely it is [that you will win]… you want information, youre desperate for any.”

4 0

Engaging with single datasets to extract facts can be a quick task, taking minutes and hours, rather than the hours and days required for interface development. Facts identified may empower individuals or organization in their interactions with the state, whether conflictual (in school appeals) or collaborative (in putting together funding proposals for charity-provided services). However, both the actors behind E6 and E7 noted it may have been useful to develop interfaces onto the datasets they explored, making it easier for their own organizations, or others in similar situations, to find relevant facts in future. As the interviewee for E6 explained:

5 0

“If there was some other larger purpose for it, what I might have done is take that Excel spreadsheet and create something a bit more …user-friendly from it …but, to be honest, … an Excel spreadsheet is interesting if you know what you are looking for.”

6 0

Interestingly, school admission appeals data has had simpler web-based interfaces to it provided by The Guardian (E8). Users of the Guardian Datablog can access admission appeal facts straight from a web-page without downloading data. However, there are no web-links between and this data re-use, in either direction. Whilst is seeking to become the point of access for government data, it is not, as yet, prioritizing the ease of end-user access to specific facts within datasets. In the case of COINS the task of providing simple access to data (e.g. CSV format) has been clearly passed to third-parties building on the original, non-standard, data.

7 0


8 0

[1] It is notable that even non-developer uses of data in the survey were carried out by people with wider interests in OGD and technology backgrounds. This is likely to be in-part down to sampling bias in the survey, but may also be an artifact of the developer-focused launch of and the way that certain high-profile dataset releases (such as the Ordnance Survey geodata) have caught the interest of particular groups. This contrasts with the US experience where was launched alongside the launch of citizen-focused, and which arguably accounts for PEW study figures indicating “Some 40% of  [US] adult internet users have gone online for raw-data about government spending and activities” (Smith 2010).