Open Data is a philosophy and practice that makes data easily available in order to enable re-use of the data in new and unforeseen ways. Open Data relies on (1) a liberal licensing model that encourages re-use, (2) data discoverability and (3) data accessibility.
- Liberal licensing – enables third parties to re-use data with minimal or no legal or policy constraints. This may range from an open license but copyright is maintained (e.g. http://geogratis.gc.ca license encourages use, but copyright is retained by the Government of Canada) to copyleft approach of the Creative Commons initiative.
- Data discoverability – Given that data files can be numerous and may not be easily opened and viewed, it is important that data files are catalogued. Hence Open Data relies extensively on some form of metadata to catalogue the data.
- Data accessibility – The value proposition of open data is the more data is used the more valuable it is. Data accessibility options range from simply putting unenhanced “raw” data on the web to offering it in a wide variety of formats for diverse audiences requirements. Data needs to be in some format that enables reuse by programmers who develop new applications. Typically this is a structured XML format or equivalent. While web pages that summarize data in a tabular format allow for accessibility of data for the human eye, they are less useful for machine to machine application development and are not considered to not meet the criteria of “Open Data”.
In terms of Web 2.0, Open “Data is the Intel inside” that drives mashups and applications. Source: Tim O’Reilly, 2007.
The Four Panton Principles for Open Data in Science are worth consideration.
Filed under: Uncategorized |