Establishing an identity
Creating an account gives a service a record to work with. Establishing what that record can be trusted to say requires separate checks, each with a particular purpose and limit.
Creating a record
Mia wants to join a photo-sharing service. She enters a name and email address, chooses a password and submits the form. The service can now begin building a record that represents her.
At this point, the name is whatever Mia typed. It could be a legal name, a nickname or an invention. The email might belong to her, someone else or no one. Creating a record gives the service something to manage; deciding what to trust in it takes further work.
Account record
- Record
- Something to manage
- Name
- Whatever Mia typed
- Might belong to her, someone else or no one
Deciding what to trust in it takes further work.
What verification proves
The service sends a code, or a link carrying a one-time value that proves the message was opened, to the submitted address. Mia opens the message and completes the check. There is now evidence that someone taking part in registration could access that mailbox at that time.
That is useful when the service needs to send account messages. It doesn't establish Mia's age or legal name, and it doesn't show that nobody else can read the same mailbox.
How much evidence is enough depends on the service. A photo discussion group may welcome a chosen display name. A service holding sensitive personal records may need a much stronger connection between the account and a particular person.
Account record
- Email access
- Checked at this time
- Legal name
- Not established
- Age
- Not established
- Only mailbox user
- Not established
Identity proofing
Suppose Mia is opening an account that will let her retrieve her medical records. A verified email alone isn't enough to decide which patient's records she should receive.
Identity proofing is the process of building confidence that someone is the person they claim to be. It can involve checking evidence, confirming information with trusted sources and establishing that the evidence belongs to the applicant. Collecting more information is useful only when it helps answer the necessary question.
Which name to show beside her photos?
- At stake
- A display name
- Verified email
- Not needed
Proofing checks needed
- Check the evidence
- Confirm with trusted sources
- Evidence belongs to the applicant
A self-chosen label may be enough.
Which patient's records should she receive?
- At stake
- Sensitive personal records
- Verified email
- Not enough on its own
Proofing checks needed
- Check the evidence
- Confirm with trusted sources
- Evidence belongs to the applicant
Evidence suitable for that sensitive access.
Sources of trust
Mia may be the right person to choose the name displayed beside her photos. Her employer is a better source for whether she still works in a particular department. Those two facts can belong to the same account while relying on different sources.
For each value, ask who supplied it, what was checked and why the service should rely on it for this purpose. Trust in a source is specific: verifying an email doesn't turn every field on its owner's profile into verified information.
Mia's accountThose two facts can belong to the same account while relying on different sources.
- Why rely on it
- She is the right person to choose it
- Still current?
- Whatever she last chose
- Why rely on it
- A better source for where she works
- Still current?
- Updated when her status changes
Check your understanding
Try these questions before moving on. If an answer isn't right, use the feedback and try again.
Try it in the Lab
- Welcome Mia and find out what her account provesHelp Mia register, compare her account with Ava's, and follow the difference between a name someone supplies and an email address they verify.