Dr. Peter Doorn

Age: 55
Interview: 10.12.2012

Profile

Peter Doorn  is director of the Data Archiving and Networked Services (DANS), an institute of the Royal Netherlands Academy of Arts and Sciences (KNAW) and NWO, the Netherlands Organization for Scientific Research. However, Peter was not interviewed in that function but as researcher, having as main field of study social and economic history. He studied settlement history in Aetolia (Central Greece) and worked on the digitization of Dutch census data.

Peter studied Human Geography at Utrecht University and received his PhD there. He taught Computing for Historians at Leiden University from 1985 to 1997, and parallel to that he was director of the Netherlands Historical Data Archive. From 1997 to 2005 he was head of department at the Netherlands Institute for Scientific Information Services (NIWI). He acquired and directed a considerable number of externally funded digitization projects and other projects in the field of humanities computing such as Historical Geographical Information System for the Netherlands (HGIN),  Life Courses in Context and Digital Research Infrastructure for the Arts and Humanities (DARIAH).

View on enhanced publications

Both the subject of research and its context (for example the community in which the project is carried out) may influence the preference for enhanced publications and for certain features. Some communities are more conservative than other. When it comes to the crunch, from a research perspective it is most important that the article is actually published in any form and that it counts.

Digital publications can easily contain an overwhelming mass of information. In particular in enhanced publications the user interface should be sparing and clean, without distracting elements. The reader must be encouraged to concentrate on the discourse itself, which is the basis of knowledge acquisition.

Explaining concepts is important, among others for identification of locations and for the classification of occupations. In earlier days these data sets have been stored in databases  and may be reckoned among research data as a whole. In the research fields mentioned universal ontologies are not feasible, due to essential differences according to time and place, and one must suffice with a limited agreement on the terminology used.

The complexity of perspectives such as in census data makes this material very suitable for visualizations (e.g. ‘Beroepen van Nederlanders in 1899’). 3D-models and 3D-graphs have a clear added-value and are ‘nice to have’ features of enhanced publications. Exploration by the user may be interesting in case of uncertainties: for example, certain occupations may be classified in more than one category, which may influence the aggregated data. Interactivity will allow the user to compare alternative choices.  

Top four

1. Footnotes and references directly linked to underlying information and data, in such a way that a simple mouse click takes the user to the publication concerned (or vice-versa, from the publication to the underlying data).

2. A facility for annotation and discussion is a general function, which can be used independently from the research subject. A user should be able to make his own annotations (preferably on a sentence level) as a gloss in the margin, but he may also want to share longer fragments  with others. The next step, eliciting discussion, requires a community of a certain size.

3. Visualizing linked information. A reader can easily get lost in the abundance of information. An overview (“a ground plan”) of what can be found is very helpful  before diving into detailed text. This overview does not so much concern the internal structure of the article (outlining), but rather the related content, as displayed by a tool such as the InContext visualizer.

4. Illustrations and animations are preferably embedded in the text itself. When the user clicks on a figure, the full image is displayed. Switching between a graph and a table is also a useful feature. A well-balanced presentation, for example through step-wise linking, prevents overload and makes it possible for the reader to view information when he is ready for it.

Audio fragments

· View on enhanced publications (1:27)

· Top four  (7:3)