• Create a plan. Ask questions such as
    • What data will be stored? For how long?
    • How much data? What is its growth rate?
    • How will the data be collected, backed up, archived, migrated, etc.?
    • Who is responsible for collecting, preserving, archiving, migrating, etc.?
    • Who controls it (PI, student, lab, funder)? Copyright?
    • Are there tools or software needed to create/process/visualize the data?
    • Any special privacy or security requirements (e.g., personal data, high-security data)?
    • What are funding agency requirements?
    • Where will the data be disseminated?
  • Determine how often data will be backed up and where it will be stored.
  • Plan a directory structure and use a standard file naming conventions and use plain ASCII text. For example, “oscilloscopeDI770_20120401_jjd” which is the equipment the measurements were obtained from, the date, and initials of who collected the data. Name the directories and files whatever makes sense for the research.
  • Document all data sets (e.g., describe variables, how the data was collected, on what equipment). Add metadata to data, that is data about the data.
  • Use nonproprietary file formats (e.g., .txt, .csv) to store data.
  • Keep an original copy of all data collected and perhaps even make it read-only. Make a copy of the original file to analyze.

Ask Michael Farmer (Geology and Physics), Debra Orr (Health Sciences), Andrew Stuart (Biology and Chemistry), or Megan Tomeo (Engineering) if you need further assistance.