Against the V Model

The V model aims at offering a global reference framework for tracking dependencies.

It is conceived from a "development from scratch" point of view.

• it will not cope naturally with maintenance, when the inputs will be bug reports or enhancement requests, part of them beyond our control and thus requiring conversion, merging, pruning and purification to be compared to the initial requirements
• it doesn't suit 3rd party products being integrated or updated during the process; it will lead to freezing, and thus missed opportunities.

It is a waterfall model, highly depending on early decisions, made with limited knowledge and poor verification ability.

It is based on verification (comparing results to intentions) instead of on validation (checking that the results are satisfying) even when the latter could be achievable.

It leads to duplication of information (between the specifications and the products) with only informal (and time-consuming) consistency checking.