Rationale #

The continuing success of SRE implementations in diverse organizations must not be underestimated. However, these implementations do not necessarily enable the SRE community to study and define the SRE practice from a theoretical perspective.

These implementations are largely based on hard-earned and organization-specific experiences, therefore they do not guarantee convergence of ideas and essentially do not necessarily solidify into fundamental models that can be consistently utilized to express any aspect of SRE in an organization-agnostic manner.

For example, without a theory of service reliability, the meaning of SRE is difficult to untangle and will likely have varying degrees of subjective interpretations where its potential consequence is that SRE will likely end up as an organizational culture rather than a robust discipline that can be studied theoretically, taught in universities, and most importantly advance as a computer science field.

While insufficient theoretical foundation of a study doesn't always mean a problem from an organization's perspective, it does lead to implementation rut that is why there should be a good balance between theory and implementation. The former is the reason of SRE Science's existence.