Blindness is in some ways the most interesting of the three considerations in Ontological Design.
When solving any problem it is nearly always a requirement to define some conceptual abstractions that help to analyse the problem domain. But if these abstractions are not good then they will create unreasonable contraints when attempting to solve the problem, or simply hide very real issues.
Furthermore, to provide a solution in terms of a computer system, such concepts are then frequently further transformed to representations closer to expected computer constructs. Often this transformation is considered the main purpose of computer system design.
Continually ask questions. Test the chosen abstractions for suitability to represent the system from the users perspective. And most importantly consider how the chosen abstractions may increase the likelihood of Breakdown.
Refine.... test .. refine ... test.
The is-ness of a system is very closely related to blindness. Both in the ability of a set of concepts to be truly representative of a system, and also in the understanding of such concepts and their communication to others.