Preserving Your Data

Solutions for Long-Term Preservation

Where you decide to put your data is fundamental to preservation. Generally data stored in personal and departmental servers has a short lifespan. These systems may not be designed for the long term, and in the event of a failure, your data may not be recoverable.

Institutions have a vested interest in the long term preservation of your research products. The University of Toronto Libraries has a mandate to help preserve the outputs of U of T researchers, including research data. We have a number of services that can help archive and preserve your data:

If these repositories do not meet your needs, please contact us to explore other solutions.  

Preservation Considerations

Preservation File Formats

The file formats you use have implications for the preservation of your data. This includes what type of files you are archiving and how they can be accessed by authorized users in the future. 

Some resources for identifying good long-term preservation formats include:

While it is hard to have a specific recommendation for every type of data you may have, good long-term file formats typically have the following characteristics:

  • Non-proprietary
  • Open, documented standard
  • Common usage by the research community
  • Standard representation (e.g. ASCII, Unicode)
  • Unencrypted
  • Uncompressed

Regardless of which standards or formats you use, you should document software necessary to access and use the data.

Metadata

Your metadata should be preserved along with your data. 

If your data is already well-organized for your own use, or has been shared with others in a data repository, much of this work has likely already been done. Required metadata can often be retrieved from existing repositories or other sources where the data has already been described. If you have been storing your data on departmental file servers, your local IT staff may be an important point of contact too.