Q: My understanding is that the Dataverse would be hosted at UNC, under Johnathan Crabtree's guidance. Is that correct? Do you know how technical customizing and branding a Dataverse environment is? I guess ultimately my question is, given that I am the Research Data Librarian and can focus a fair amount of my time on a project like this, would it be necessary to have someone from IT locked in to participate in the pilot, in order to be considered? Or can a librarian, on their own, participate, referring to IT only when needed?
A: The lions share of the work would be that of a research data librarian describing and cataloging files. If you already have good metadata formats that may not be bad either. The DVN customization is all done in style sheets CSS and small amounts of java script. The difficulty depends on the degree you want the DVN site to identically match your home site and how complex the home website is to imitate. It would be ideal to have input from the home website IT group but not necessary. I suspect that this will only be a very small amount of the labor required. We can handle most of the technical side but as you mention it would be great to coordinate this to prevent websites from changing and messing up the customization code. I think you would be OK participating on your own if you have guidance from the website designers or had some css/html experiences.
Q: Is there any information about capex, opex, processes and best practices from institutions that are current DVN users?
A: Typically users provide 80% of the labor it takes to ingest the data so ongoing expenses for the users depend on the shape of the documentation they get as input. Some studies require little metadata work where others require heavy research to document the files.
Q: Are current users integrating their dataverse services with other publicly-facing services like Dspace or Omeka?
A: The DVNs built in OAI-PMH server allows for the harvest of meta-data across platforms. We are working with our library to sync our Odum collection in the public Endeca catalog here at UNC. I know that Harvard's Library is doing the same. The DVN can export MARC, DDI, and Dublin Core records. We have also tested an iRODS connector as well. Participants planning to sync with an existing system other than Endeca would allow us to try a new area of collaboration. Dspace or Fedora might be good ones.
Q: Are current users integrating their dataverse services with back-end services like REDcap (Vanderbilt) for providing anonymized copies of primary data sets?
A: We have never worked with systems like REDcap but that would be an interesting test case, however that may be a larger effort than planned for this pilot.
Q: Are Dataverse Network guides available on-line?
A: Written guides for users, installers and developers of the Dataverse Network software are available at: http://guides.thedata.org/. Video guides on use of the Dataverse Netork software are available from the Odum Institute at: http://www.odum.unc.edu/odum/contentSubpage.jsp?nodeid=667#dataverse.