As a newcomer to the ANDS program I thought it would be useful if I listed some of the material I was recommended to read in my quest to gain an understanding of what might be common and recommended practices in the areas of both describing and archiving research data. In some cases they contained very detailed guidelines for researchers. Some of them are already listed on the ANDS web site in the resource guide for data management planning.
I didn't go about this in any systematic way and there will be other resources worth reading if anyone would like to suggest more.
One of my starting points was the final report of the two-year UK Datashare Project. This was a two year project (finished March 2009) funded by the JISC's Repositories and preservation programme. The projects overall aim was to contribute to new models, workflows and tools for research data sharing in the UK and there were three project partners (sound familiar?). I found the partner reports in the appendices illuminating and each had implemented a data repository in different ways. The report is at http://ie-repository.jisc.ac.uk/336/
and here's an example of a dataset in the Edinburgh DataShare http://datashare.edina.ac.uk/dspace/handle/10283/16
"Best Practices for Preparing Environmental Data Sets to Share and Archive" is published by the US Oak Ridge National Laboratory Distributed Active Archive Centre and contains excellent practical instructions for description of research data for future re-use in a cross-displinary environment. It is very detailed and thus refreshing as it really gets down to the nitty-gritty http://daac.ornl.gov/PI/bestprac.html
"Managing and Sharing Data - a best practice guide for researchers" published by the UK Data Archive also contains very good and quite detailed advice for researchers. Because it is targetted at all areas of research it is not as specific as the previous document which was able to go into more specifics because it was specific to Environmental Data Sets. http://www.data-archive.ac.uk/news/publ ... haring.pdf
"Guidelines for Depositors" published by the Archealogical Data Service in the UK also has detailed guidelines for a specific discipline subset, but these can be generalised to research data sets in many areas of the humanities and sciencehttp://ads.ahds.ac.uk/project/userinfo/deposit.cfm
"Data Documentation Initiative: Toward a Standard for the Social Sciences" is a journal article in the International Journal of Digital Curation July 2008. It is a good introduction to DDI which is the mostly widely-used standard for social science data description and thus of interest. The Australian Social Science Data Archive (ASSDA) currently uses DDI 2.0 but are in the process implementing new systems so they can move to DDI 3.0 which supports hierarchies and relationships via flexible grouping of studies and will be developing a mapping to RIF-CS.http://www.ijdc.net/index.php/ijdc/arti ... File/66/45
"We Need Publishing Standards for Datasets and Data Tables" OECD Report 2009. Strong business case but also lots of recommended metadata elements for datasets and dataset collectionshttp://dx.doi.org/10.1787/603233448430
Standard, Standards, Standards
After understanding the RIF-CS information model that underpins the ANDS registry, I found it necessary to also understand the Dublin Core Collections Application Profile. Although currently most of the Australian Institutional Repositories use simple or qualified Dublin Core for resource description, increasingly Institutional Repositories and Data Archives and Data Registries are using the Dublin Core Collections Applications Profile (DCAP). DCAP supports relationships and hierarchies and is designed for both collection level descriptions as well as resource descriptions. As these will be sources of collection records (though not Party, Service or Activity records) for the ANDS Registry it is useful to be familiar with DCAP. http://dublincore.org/groups/collection ... n-profile/