Orphan Data and Datasets

Coordinator(s): DonaldHobern

Participants:

---

Description

This group will investigate the need for a repository to host orphan data or datasets containing GUIDs.

---

Discussion

In many ways this issue is part of a much larger question of long-term data curation (see also LSIDLongTermArchival). We should however consider the issue now because it may have implications for the best practices that we choose to adopt for LSID issuers. The use of an LSID rather than some other identifier should lead issuers to consider the long-term existence of the associated data object and of the LSID as a path to retrieve it.

Clearly no issuer can make an absolute guarantee that they will be able to serve their data far into the future. The issue here is therefore whether our community should establish any infrastructure which could be used to archive data sets to ensure that their LSIDs can be resolved even if the original issuer ceases to exist. If this is considered worthwhile, we need also to determine the implications for how issuers should manage their data (to ensure that such an archive, and ultimate transfer of ownership is feasible, see also especially LSIDResolverNamespaces).

---

Conclusions and Recommendations

---

Categories

CategoryWorkingGroup CategoryInfrastructureWG