Episode Notes

In this podcast episode, JUXT CTO Malcolm Sparks, JUXT Head of Delivery Joe Littlejohn, and XTDB Head of Product Jeremy Taylor spoke with guest Mark Burgess, an independent researcher and writer.

Formerly a professor at Oslo University College in Norway and the creator of the CFEngine software and company, Mark was invited to write the foreward (https://sre.google/sre-book/foreword/) to Google's 2016 book: "Site Reliability Engineering - How Google runs production systems".

They discuss Mark's journey to developing Promise Theory and explored techniques to 'scale simplicity' in the creation of large, reliable systems. One common (yet false) assumption is that all components of a system can be trusted to be 100% reliable. This misconception can lead to costly workarounds in production.

They touch on the 'congruence' debate, considering whether and to what extent we should be concerned with the inherent inefficiencies in 'the automated building of things from scratch.'

They also discuss the counter-intuitive observation that digital systems are far more complex and less resilient than analog systems, and how this may be due to the absence of an error-correcting mechanism in digital systems to maintain equilibrium.

Please let us know if you have any points to add or if you were inspired by any part of the discussion. Happy listening!