What does 'archiving' mean?

Archiving can have several meanings:

• Backing up data
• Storing data somewhere safe
• Preserving data for a specified period
• Keeping data in perpetuity
• Migrating data to newer formats
• Providing an emulator so that data in older formats, or that needs obsolete software to be understood, can still be used

Archiving research data means storing your data in a system or service where it will be:

• Safely stored
• Backed up regularly
• Discoverable
• Preserved for the future
• Made available to others for re-use, if possible

A data repository is probably the ideal place to archive your data.

Types of Data Repositories

Domain - examples include GenBank, Dryad, TreeBase

• Your data will stored with similar data
• Researchers will find your data easily
• Repository staff understand your kind of data
• The repository may offer computational tools to crunch or visualise your data
• But repositories may close down if funding ends

Institutional - examples include CSIRO's Data Access Portal, and university-based data repositories

• Repositories are backed by institutions
• Different types of datasets can be stored together, regardless of discipline
• Institutions generally guarantee support
• But not all research institutions have central data repositories

Find a repository at www.datacite.org/repolist

Archiving facilitates data re-use

Advantages of respository archiving

• The deposit process creates useful metadata to describe your data
• Repositories assign unique identifiers to data, e.g. DOIs or Handles
• They aim to preserve data long term
• They simplify data sharing
• They facilitate data citation
• Some repositories provide tools so researchers can interact with your data

To preserve data successfully

• Provide thorough documentation about your data
• Preserve a copy of any software required to create, manipulate or render your data
• Migrate your data to up-to-date formats when necessary