Use Case Dimensions

  1. What is the user's relationship to the addressed site?
    1. No previous relationship between website and user
    2. User has previously interacted with the website
      • May or may not have established account credentials (e.g. password)
      • May or may not have shared private information (e.g. bank routing number)
  2. How does the user arrive at the site?
    1. Manual entry of address
      1. Site address may be memorable brand (e.g. amazon.com)
      2. May have learned from advertisement
    2. Manual entry of part of address with browser heuristics completing the address
      1. Many browsers will automatically add .com to domain name (e.g. amazon => amazon.com)

    3. Bookmark or other relationship stored in browser or OS
    4. Link provided by external application
      1. From email client
      2. Instant messenger
      3. Other
    5. Web link
      1. From search engine
      2. From partner site (e.g. citi.com to accountonline.com)
      3. Advertisement
      4. Other
  3. What task is the user intending to perform? (Intended interaction)
    1. Retrieve information
      1. Even simple information retrieval may be security-critical (e.g. looking-up instructions and address mailing a banking deposit)
    2. Interact with others (e.g. interpersonal email, social networking, web-based IMs)
      1. Trust in data received affects how user chooses to act upon it.
      2. Secure entry of data into system important for user's reputation (e.g. adverse effects if user falsely associated with NAMBLA)
    3. Transferring funds (or other valuables) within own accounts or to others
      1. Trust motivation here is obvious

    4. Exchange goods or services
      1. User should understand implicit or explicit contract terms (by bidding I'm agreeing to buy this item)
      2. User will want to know whether he/she is really communicating with the other transacting party (where to ship goods)
  4. What can go wrong? -- See ThreatTrees