Empiricism and values alone are not enough

 

There's an old marketing story about a cannery with pale salmon that was not doing well against superior, pink salmon. They created the slogan "guaranteed not to turn pink in the can." Great marketing - turning a weakness into a strength.

It occurred to me that this is what Scrum subtly does when it says it's based on empiricism - as if this is a strength. But every method uses empiricism. Some more than others. Ironically, Scrum uses it at the lower end of the spectrum. Let me explain.  

The scientific method is based on four pillars:
1-Empiricism
2-Theories that explain our experience
3-Validating or invalidating our theories in the face of new evidence or upon reflection of old evidence
4-Running experiments to gather new evidence to test our theories

A key aspect of the scientific method is that no theory is sacrosanct. Practices can be considered to be a theory that this is a good way to work. Since it’s been proven that no practices are universal, methods should avoid making any practice immutable. Instead, each theory about a practice being useful should include where it is useful. While objectives may be virtually universal (e.g., managing the amount of work being done) how you achieve them depends upon the situation you are in.

Approaches that require specific practices only work in certain situations. It is important to understand what these situations are when adopting such an approach. Unfortunately, because we live in a “caveat emptor” world (the buyer beware) don't expect proponents of an approach to provide this, although they may provide a statement that sounds like they do. For example, one hears Scrum proponents say “it works when teams are working in complex situations” but notice that does not really provide for which complex situations it works in.

Approaches can be made considerably broader if they
- are defined by a set of objectives
- provide a set of optional practices that can be used - this avoids starting from scratch
- provide a way to tell if a new practice is better than a current one - allowing for the creation of better practices

Frameworks that don't provide this will see practitioners drop a required practice when it’s not applicable but won’t know how to find a better one. This is a common occurrence in Scrum, which provides neither objectives for its practices nor a theory on which to see if a new one is better. The common phrase for when this occurs is “Scrum but.”

Empiricism is great. But it's not enough. This is why Disciplined Agile's Scrum provides what's stated as being required above, enabling a fit for purpose solution right at the start. If you like a preset, prescribed approach, you can also use DA Scrum to start with a default set of options that have you be consistent with Scrum practices. But then you’ll be able to change them as you learn and discover there is a better set – one meant for your team(s).