The end of piecemeal research

According to the Netherlands Organisation for Scientific Research (NWO), both publications and the data underlying them should be accessible to everyone. Where will you store your data? How will you make them searchable? How will you make them suitable for re-use? Scientists will have to answer these three questions in order to qualify for the € 1.5 million NWO grants known as Vici grants.

Atmosphere researcher Prof. ir. Herman Russchenberg (CEG) approves of the NWO’s philosophy. ‘But it takes time to make data accessible. Data cannot be used by others unless they are equipped with good meta-data. At Cabauw in Lopik, we are conducting research on the atmosphere. We have to make sure to provide good documentation of the context in which we perform measurements (e.g. wind direction and the settings on the instruments). This is quite complicated’.

He continues, ‘A 20-year measurement programme allows us to research good agreements with each other’. ‘When individual researchers develop new measurement techniques, however, it can quickly take on a piecemeal character. Then you have to decide for yourself how to ensure that the data can be used by others. And we should also give much more consideration to file types. No one works with floppy disks anymore’.

Prof. Alan Hanjalic (EWI) predicts problems. He is developing algorithms for making video files searchable. ‘We need huge mountains of video files in order to test algorithms. We are dependent upon Google, Yahoo
and other companies in this regard. They make data available to us if we can help them. We can subsequently make the algorithms that we develop open to the public, but not the data underlying them, as these data belong to the company’.

Dr Johan Molenbroek, an ergonomist in Industrial Design Engineering, is a staunch advocate of open data. Last year, he was awarded the data prize from the  Research Data Netherlands foundation for developing Dined, a database for designers concerning variations in human dimensions within the Dutch population.

Privacy is a thorny issue. ‘How can we ensure that people who have been photographed are not recognisable? We can’t simply post these three-dimensional images on the internet. I’ve made the people unrecognisable by using Photoshop to remove certain layers from the photos without affecting their usability’.

Stay informed about the research

Receive the Delft Outlook newsletter 4 times a year