Thomas Fischer (University of Göttingen, Germany)
In 2012 we started to use the BExIS system with the Collaborative Research Centre 990 on Tropical Lowland Rainforest Transformation Systems. CRC 990 is a broad project with 25 sub projects clustered as “Environmental processes”, “Biota and ecosystem services”, and “Human Dimensions”. Subject areas like ecology, forestry, botany, sociology, zoology, agricultural science, geography… had to fit into the system. Neither for sociological nor meteorological data the concept of “Taxa” is meaningful.
We found the structure of BExIS too rigid. Worst was the user interface to enter the metadata for the different projects with the interwovenness of personal data, project data, experimant description and data format. In particular the choice between reusing a metadata set and using the data set to upload the variable descriptions was hard to communicate (and to accept!).
To ameliorate the situation we decided to introduce at least two simplifications: First, to hide all the gory details of the data description in case of unstructured data and second, to provide the standard HTML interface elements for entering data and providing input like radio buttons and check boxes, since the “raw” BExIS system only gave textfields, textareas and dropdown lists, even for choices like “yes” or “no”.