Business rules analysis invites you to state facts about a business in concise language sentences. Each of those statements expresses a single point of truth.
This has a technical reason. It’s because of the way business rules are rooted in propositional logic, as explained in great detail in OMG’s SBVR specification.
But this format also has a practical spin-off. Each rule applies in and of itself. It does not need a surrounding narrative:
- At make-time, decision makers in the business can take an individual rule and say if it’s how they want it to work, or not.
- At use-time, designers and operators can know from the rule alone whether their conduct complies with it or not.
All are in need of a good glossary, a terminological dictionary that specifies the meaning of the business terms used in the rule. Otherwise, to make sense of a rule they don’t, strictly, need any neighbouring rules. Much less do they need overarching narratives that somehow try to provide extra information.
This has a profound practical consequence. You can put rules in a document, of course – they’re text. But rules don’t NEED the document format. They don’t take part of their meaning from being in a particular place in a Table-of-Contents structure, or from being placed in the context of a section that has a particular theme and maybe an introduction that sets the scene. They don’t need summarizing in conclusions. Rules can be placed instead in a central repository: a database that stakeholders can query over the Internet.
The fact that such a repository is possible and documents are not necessary, is a great opportunity. The truth about documents is that they were never an ideal format for business rules. In a document, you must give a rule ONE place in a narrative. This goes against an essential property of business rules, which is that they’re relevant in multiple situations. Ronald Ross gives an example similar to the following in his 2011 book Building Business Solutions:
“To a client who spends over $ 50,000 in a booking year must be appointed a personal account manager in that year and the next year.”
This rule, by definition, applies to situations where:
- A client starts spending over $ 50,000.
- A new booking year starts.
- An account manager leaves the company.
(Ross calls these situations “flash points” for the rule. They are events that should cause someone in the organization to look at the rule.)
Document writing makes you lazy. Likely, you’ll only mention the rule in the section talking about what to do when customers become preferred customers.
If you do remember that the rule is relevant in the other two places as well, when using documents you are reduced to copying the rule at least three times, because there will be different document sections for each “flash point”. But copying rules is very unattractive. When the rules change, it’s all too easy to overlook some of the places where they appear. This is how inconsistency is introduced in organisations.
Documents have some unbeatable advantages caused by the advance of electronic word processors. Information in them is easy to package and label, present visually on the page, send through the mail, and (extremely important even in 2013) print on paper. Repository tools must do their best to keep up. In particular, they must offer good reporting options.
But digital documents are also very easy to copy (clone). This is NOT one of their strengths. It’s how inconsistency comes about, and incidentally, it’s how organizations produce enormous unwieldy mountains of badly organized documents that could be replaced by lean-and-mean, single-point-of-definition central repositories of business rules.
Leave a reply