Proposed Student Projects
Dryad occasionally has the opportunity to work with student interns through projects like the Google Summer of Code and the DataONE internship program. This page collects ideas for projects that are suitable for student work. All of these projects provide valuable progress for Dryad. They are relatively self-contained projects, requiring a minimal amount of background knowledge before the student is able to make a meaningful contribution.
- 1 Processing data packages with SWORD and ORE
- 2 Metadata editing with HAMR
- 3 Improving the Dryad API
- 4 Generating Reports
- 5 Publish Dryad Metadata as LOD
- 6 Dryad Metadata DCAP (Dublin Core Application Profile)
- 7 HIVE and Dbpedia comparison for indexing Dryad holdings
- 8 Name authority control / Orcid work
- 9 Lingering Issues
Processing data packages with SWORD and ORE
Extend the DSpace SWORD interface to support BagIt data packages OAI-ORE descriptions. Upon receiving a package with an OAI-ORE description, it should be able to generate the equivalent data package and data files in the submission system. Extend the existing BagIt exporter to create equivalent packages for export purposes.
Proposed as a project for the 2012 Phyloinformatics Summer of Code.
- DSpace's Replication Task Suite has code for working with BagIt objects (beyond the BagIt transformer that is specific to Dryad).
- People who may be interested in collaborating:
- Stuart Lewis (SWORD)
- Hardy Pottinger (U Missouri-Rolla) was planning to build similar technology
- Marco Fabiani (Queen Mary University of London) is looking for similar technology
Metadata editing with HAMR
Build out the user interface for HAMR. Integrate HAMR with Dryad.
- Proposed as a project to the 2012 DSpace Summer of Code, but the organization was not accepted.
- Proposed (in slightly more general form) as a project for the Rose-Hulman Senior Project program.
Build a tool that allows curators to compare DSpace metadata with metadata from authoritative sources. The tool will allow curators to see DSpace metadata alongside metadata from a system such as CrossRef or PubMed. Individual metadata fields will be color-coded according to the degree of consistency. Curators will be able to click a button for each metadata field they wish to import from the authoritative source.
An initial design for such a tool was created during a hackathon at the 2011 Code4Lib conference. This design, called HAMR, provides a good starting point for the project, but there is considerable room for a student to work with the DSpace community to determine the final design of the tool.
Improving the Dryad API
Implement the proposal for a new Dryad API.
Develop more reports that are needed from the list of Curator Reports.
Improve the process for generating statistics associated with Dryad. There are two types of statistics, those associated with periodic reports to stakeholders (e.g., board meetings, annual reports to funders), and those associated with the Global Statistics Display.
Publish Dryad Metadata as LOD
This goal includes 1.) registering Dryad-specific + relevant properties at an appropriate Dryad name space (e.g., datadryad.org), so that Dryad metadata published as linked data can resolve. 2.) generating current Dryad metadata (where appropriate) as linked data, following on the DataONE LOD4DataONE work completed by Aida Gandara https://notebooks.dataone.org/lod4dataone/author/aida-gandara/ [summer 2011].
Dryad Metadata DCAP (Dublin Core Application Profile)
Follow on the initial DCAP work generated for Dryad AP 2.0 to the current 3.0 version and publish the work as a DCAP compliant with the DCMI Singapore Framework. This could be integrated into the LOD project noted above
HIVE and Dbpedia comparison for indexing Dryad holdings
Compare HIVE vocabularies and Dbpedia's underlying terminology for indexing Dryad content. A mapping experiment with Dryad's current search logs might also be considered.
The Dryad issue-tracking system contains many issues that have not been resolved. Many of these issues could be addressed by a student. A student could approach these issues in two ways:
- Identify individual issues that are tractable and solve them. These issues include problems like minor usability tweaks and documentation needs.
- Identify classes of issues that occur frequently. Develop tools/processes to either prevent these issues or to solve these issues as they occur.