Rules that Give You Too Much Freedom

by Silvie Spreeuwenberg

Freedom should be removed by rules.  In fact, a rule that does not remove freedom in some way is not a rule according to the SBVR standard.  That being said, you may be confused by the title of this column.  Well, given this restriction on what a rule is and what you expect from a rule statement, there are still many ways to express such a rule.

RuleSpeak and SBVR's Structured English impose some restrictions on our use of language (see the box below), but due to the flexible nature of natural language you still have a lot of freedom ... consequently, we rarely see the same rule written in exactly the same way by different people using natural language.

There is a tendency to ask for a more restricted rule language to improve:

• consistency in writing similar rules by different people,
• consistent interpretation (non-ambiguity),
• support for validation & verification,
• translation to software systems.

Expressing rules as decision tables or decision trees greatly reduces your freedom of form but also your expressiveness.  So, when I need the expressiveness of natural language, I introduce very specific rule patterns to restrict my freedom and gain rigor.

Examples:

• ... must be calculated as ....
• A ... for ... must be calculated as ....
• A <data element> for a <timeframe> and a <location> must be calculated as <formula>.

Each pattern is more specific, provides more guidance, and (consequently) less freedom.  The last pattern above is even specific to a particular domain.

Sentence patterns seem to be the fashion today.  They are not only popular for rules but also used in the descriptions of user stories in agile projects:  As a <role> I want <something> so that <benefit>.

Rules must decrease freedom and so must rule sentence patterns.  I am curious about the sentence patterns you use or recognize in your organization.  Please send me an e-mail with your examples at silvie@librt.com .

Drs. Silvie Spreeuwenberg has a background in artificial intelligence and is the co-founder and director of LibRT.  LibRT helps customers assess and improve the quality of their business rules.  Silvie's experience with business rules modeling has resulted in the development of tools and techniques to increase the quality of business rules.  She writes, "We believe that one should focus on quality management of business rules to make full profit of the business rules approach."  LibRT is located in the Netherlands; for more information visit www.librt.com.