For example, consider the following as a plain English Acceptance Test:
Somebody goes to the page, enters a discount voucher and gets the right amount off?
The format for writing Acceptance Tests is normally like this:
A [named user role] can [select/operate] [feature/function] so that [output] is [visible/complete/etc.]
Now, as writing the test requires real data, and real results it could transpire to this:
A web user, John A Smith (who is flying with his wife Jane, 1 child called Jack, 1 infant called Janet) chooses economy class, one way to Paris Charles De Gaul, can enter a £25 fixed price discount voucher so that the full fare of £350 is reduced to £250?
The latter is a test which has a yes or no answer and is not ambiguous.
There can be a shorthand employed within the test case if everyone who reads the tests understands what that shorthand is, like so:
A web user, John A Smith, can enter a £25 fixed price discount voucher so that the full fare of £350 is reduced to £250?
This implies that John A Smith is a Persona understood by all to be:
John A Smith
Passenger Choice:
2 Adults: John A Smith and Jane Smith,
1 Child: Jack Smith,
1 Infant: Janet Smith
Flight: Economy, 1 way, CDG
In fact, when writing a scenario that doesn’t really care about certain variables, the Persona can be re-used as an archetypal good guy, with the atomic family, thus reducing waste in data set-up costs.
They also become data sets that everyone can talk about and form a common language for people to talk round.
