Semantic Werks

Thoughts on people, machines and systems.

The requirements problem defined

leave a comment »

Here’s a useful way of thinking about how requirements, implementation, and preferences interact:

the core problem, defined

Below are the ways these things are defined. Essentially, we are seeking a set of plans, P, that will satisfy the functional requirements, G, the qualities, Q, according to stakeholder attitudes, A, without violating the domain assumptions, K.

  • Definition 1. Believed content, i.e., ϕ in Bϕ, communicated by way of assertive, declarative, or representative declarative speech acts is a domain assumption, denoted generically k.
  • Definition 2. Desired content, i.e., ϕ  in Dϕ, communicated by way of a directive speech act is a quality constraint, denoted q, if and only if ϕ describes qualities and constrains quality values. Described qualities must have quality space with a well-defined and shared structure.
  • Definition 3. Desired content, i.e., ϕ  in Dϕ, communicated by way of a directive speech act is a goal, denoted g, if and only if ϕ neither describes qualities nor constrains quality values.
  • Definition 4. Desired content, i.e., ϕ  in Dϕ, communicated by way of a directive speech act is a softgoal, denoted ˆq, if and only if ϕ describes qualities or constrains quality values, whereby the described qualities must have a quality space with a subjective and/or ill-defined structure.
  • Definition 5. There is a justified approximation, denoted jApprox(ˆq; q) if and only if there is a justification for the claim “q approximates ˆq” and there is sufficient correlation between values in the quality space of q and the quality space of ˆq.
  • Definition 6. Intended content, i.e., ϕ in Iϕ, communicated by way of a commissive speech act is a plan, denoted p.
  • Definition 7. Attitudinal content communicated by way of an expressive speech act, i.e., ϕ in Aϕ, is an attitude, denoted a, if and only if it evaluates in terms of favor or disfavor one or more elements constituting K, P, G, Q, or ˆQ.

This is from Jureta, Mylopoulos and Faulkner, 2008, ‘Revisiting the Core Ontology and Problem in Requirements Engineering“.

Advertisement

Written by Neil

2008 September 22 at 16:17

Posted in Uncategorized

Tagged with

Leave a Reply

Fill in your details below or click an icon to log in:

Gravatar
WordPress.com Logo

You are commenting using your WordPress.com account. Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Connecting to %s

Follow

Get every new post delivered to your Inbox.

Join 198 other followers