Tufte finds Conway’s law frequently rearing it’s ugly head. If you’ve worked on engineering projects, or taken over for someone else’s engineering project, you’ll recognize it immediately.
Any piece of software reflects the organizational structure that produced it.
That can be reworded to encompass any design project, software engineering or otherwise. It also is evident when an engineering team is slowly recycled with new members. The designs of the original team will be deemed crufty and most engineers joining the team will want to clean-slate the entire thing. This isn’t practical, so bolt-on and secondary systems tend to emerge. Extending existing systems is an impossible ideal.
Managers of engineering teams over this lifecycle must be understanding to guide this to fruitful ends.
Emphasis mine below.
Conway’s Law can be construed as humorous, to the extent that rigid organizations that are not willing to re-organize to generate an optimal design, can end up producing a sub-standard design that merely reflects the pre-existing organization.
But the essence of Conway’s Law also applies to flexible organizations that are willing to re-organize to produce an optimal design.
For example, consider a flexible company that is charged with designing a car, and the company does not yet have a car-design organization. The new car-design organization will probably consist of groups that correspond to the major components of a car: engine, body, transmission, interior, electrical, etc.
The essence of Conway’s Law applies even in this re-organization situation: the components and interfaces of the resultant car mirror the engineering groups and their interfaces.
And, significantly, any shortcoming with the interpersonal relationships between the engineering groups, may manifest itself in a shortcoming of the resultant design.
Wikipedia offers a corollary
In every organization there is one person who knows exactly what is going on at all times. This person must be fired.
Sounds like management!
Richard: If we had someone that knew exactly what was going on in all parts of UMBC, it’d be noteworthy, possibly worth a newspaper article in the student paper.
