Wednesday, May 30, 2007

Wiegers on the Requirements/Design Distinction

Karl Wiegers and I have disagreed in the past on requirements terminology.

Now Wiegers has a good piece on the distinction between requirements and design. Or rather he attempts to sidestep the interminable semantic debate and focus on some important consequences of failing to understand the "why" behind product specifications.

In my view, the summary passage in the piece is:
When it comes to requirements specification and design, the essential issue is not one of what versus how. It’s a question of distinguishing the real customer need from just one possible description of how a clever developer might satisfy that need. Incorporating a solution idea into a requirement imposes a design constraint. The requested solution describes one way to satisfy some requirement but perhaps not the only way, the best way, or even a good way. Focusing on solutions masks the underlying requirement. This can make it difficult for a developer to understand what the customer is really trying to do, making it hard for him to devise the most appropriate approach to meet that expectation.
Actually, I think the "real customer need" is the what, and a "possible description of how a clever developer might satisfy that need" is the how, so the what versus how distinction makes perfect sense. Setting aside this quibble over terminology, however, Wiegers hits the nail on the head about focusing on underlying needs rather than dictating solutions to developers.

Wiegers elaborates:
The requirements analyst needs to detect when a requirement imposes unnecessary constraints on designers. This should lead to a discussion with the customer representatives about the underlying need that led to the customer proposing that specific solution. It’s important to respect the customer’s input. Don’t summarily dismiss the customer’s solution idea; important information is hiding in there somewhere. Use that input as a starting point to drill down to a deeper understanding of what the customer is really trying to accomplish. It’s possible that the customer’s solution idea will be appropriate, but don’t let the customer—or any other stakeholder—paint the development team into a constraint corner prematurely.
He then gives real-life examples (drawn from actual requirements documents). Here is one:
"A master power button shall be installed on the front panel." Further discussion might surface an explanation of why this precise design approach is necessary. Perhaps it’s required for compatibility with an existing product, or maybe it will conform to a pertinent standard or safety requirement. Or it could be an unstated ease-of-use requirement. If so, it would be good to know about any related usability requirements that could influence this, and possibly other, functionality or design issues.
In other words, customers don't care about "power buttons" per se. They care about compatibility, compliance, safety, and ease of use. If a requirements analyst fails to document these nonfunctional requirements in measurable terms, the product may end up with a "power button" but still end up being incompatible, not compliant, unsafe, and difficult to use.

2 comments :

David said...

Back in the late '80s, I went to a conference where a paper on formal requirements was presented. In that paper, requirements are decisions.

This doesn't change the WHAT of requirements.

Then, comes the notion that products enable and constrain simultaneously. Products used in the implementation phase are constraints that are ultimately mapped to requirements.

Design is the mapping of a media, software, with the carried content, the user's application domain (finance, markeing, not Oracle, or Java, or...).

rcauvin said...

Hmm, I little confused by the notion that "requirements are decisions", but that they don't "change the what of requirements".

After understanding problems in the marketplace, we do make decisions about which ones to solve. But once we make decisions beyond that, we are delving into HOW to solve them.