Overview

Crowd sourcing can be an efficient way to increase quality and availability of machine readable data, particular in cultural heritage institutions. On a policy level, identifying community crowd sourcing projects outside government institutions can also be an indicator of valuable datasets that should be prioritised for open release.

Why

Many institutions lack resources necessary to manually go through large collections of unstructured data that has been created over the years. By engaging and collaborating with external communities on this data it is possible to create more detailed machine readable data supporting a wider range of re-use cases.

Intended outcome

Relationship to PSI Directive

Possible Approach

Planning Phase

Implementation Phase

How to Test

Different tests can be undertaken:

Often quite short. In the case of Share-PSI BPs, it's likely that all tests will need to be carried out by people rather than machines but if something is machine testable, that's often more precise.

Evidence

Examples of crowd sourcing to replicate a government dataset that is not freely available:

  • Open Addresses (UK)
  • postnummeruppror.nu (SE)
  • Open Streetmap (Intl)
  • Examples of succesful use of crowd sourcing to create or improve PSI:

    Tags

    crowd sourcing, collaboration