Monday, April 04, 2011

#NArchief evaluates #GLAM experiment

The Dutch #Wikimedia chapter and the Dutch National Archive experimented with a selection of 719 pictures of Dutch politicians. Given that the NA collection contains some 526.557 images, the graph to the right illustrates nicely what room for future cooperation there is. As it is an archive, there is more, much more to the National Archive then just images.

This evaluation is in Dutch. It is written by Kennisland, a Dutch open source and open content powerhouse. They have build themselves an enviable reputation. The Dutch chapter has collaborated regularly with them.

The following list is cherry picked and full of pertinent and/or actionable conclusions.
  • material of particular relevance to a culture will get most attention by that culture
  • the files made available to Wikipedia reached more viewers than all the files at the archive
  • Europeana gives access to the whole collection and provides only twice as many referrals as the nl.wikipedia
  • Wikipedia traffic does not generate traffic to the archive - lack of attribution for pictures is one reason
  • there is a lack of reliable statistics provided by Wikimedia
  • this lack of reliability makes collaboration less interesting for a GLAM
It is quite ironic that we are asked to improve the attribution in our articles. What will likely do the trick is something similar to the attribution we build in for "use on the web"

Wim Kok van de PvdA in de Tweede Kamer - NL-HaNA Anefo 934-4587 WM645
on the picture
In the discussions about GLAM partnerships the role of the chapters are given a pre-eminent role. This is a good thing because the best results will be achieved when the relations are long term and personal. It is for all these reasons best when the chapters get their responsibility as they are the "owners"  of the GLAM relations.

The statistics should be presented with the same level of quality and sophistication as the Wikimedia statistics and the numbers should be as timely, accurate and complete.

No comments: