Take a seat, relax and enjoy this demonstration¹ in which Egbert Gramsbergen shows how 3TU.databrowser works in about 50 minutes time (Dutch). The movie repeats what digital objects are and may serve as a sound fundament for understanding the remainder of this paragraph and performing the assignments. 

With 3TU.Databrowser you search in the metadata of datasets and digital objects, no less and no more. A good quality of metadata makes it easier to find what you are looking for. 

All metadata are located above the gray line. If more metadata are available they can also be stored as data. The other way around: data may also contain metadata. These metadata may be manually extracted and put above the gray line. De datasets themselves are below the gray line. 

The metadata with an indentation are from different objects than the object you are currently looking at. For example: Below you see map (location).

Within 3TU.Databrowser an explanation is available about each field. Just click on the little question mark next to the item you want to get more information about. 

Besides this short explanation, we will practice with 3TU.Databrowser hands on during the second face to face meeting in this course. For one, we will upload a dataset. 

Search 

Via the search interface you can search within the metadata describing the datasets. You can limit your search to object type (like location, measuring instrument etc.), creator, year and data format. 

When you click location Luxemburg you are searching for this location within all available datasets within 3TU.Databrowser. A location object is quite different from a dataset object. Several location objects exist: points (cheep locations in terms of computational work) and areas (expensive locations in terms of computational work). 
When clicking Home Data, World, map of measurement locations you get 33 points². Areas like Ubbergen for example aren't included here. These areas can only be displayed when evoking the concerning map via the corresponding object. The 33 points are the locations where measuring instruments are located. These points are thought to be most important. If you do not want to frustrate the speed at which data are displayed, such choices have to be made. Only by looking carefully at your search results and taking different angles, you can gain understanding how choices made are influencing search results. 

Self uploads of datasets

From June 2011 onwards researchers may upload their datasets themselves. Here you can have a look at the datasets which were uploaded by researchers. (collection:user_uploads). Files for self upload should be single datasets less than 1GB. If researchers want to upload more files, they should compress them into a zip-file. When uploading you can choose the MIME-type (See data formats). 
After upload the data package in unzipped and a bagit is made.  In this way the files get an inventory describing what's in a dataset. Afterward the total is compressed to a zip-file again. This all happens fully automatic.

Before upload you have to sign in with OpenID. You will get a personal page showing you how often your dataset was viewed by others. If you only want to look at datasets or download them you do not have to be signed in. 

OPeNDAP

Datasets with the formats NetCDF and HDF5 are not located at the server of 3TU.Datacentrum (Fedora) itself. They are located at another server called "OPeNDAP". A large part of the data resting on the OPeNDAP server has been made visible via 3TU.Databrowser. (see Data processing)

<sup>1. TUDelft Library. (2011). 3TU.Datacentrum [video]. Retrieved 9-12-2011 from
 http://collegerama.tudelft.nl/mediasite/SilverlightPlayer/Default.aspx?peid=42ab8ce8ca0146d29884e533abab3a2c1d
2. Search performed 17-12-2011</sup>