Getting Started with SRE - Stephen Thorne, Google

[@mjrulesamrat]

"50% on Toil and other 50% working on things today that'll make tomorrow better". Couldn't agree more!

[@TheCALMInstitute]

"SRE is when software engineers define an operations function". This is exactly backwards, dismissive, and pedantic.

[@cultivei73]

why?

[@jerub9363]

Yep. It can come across that way. Ive written a long form post about this a while back medium.com/@jerub/site-reliability-engineering-is-operations-17a213ff88bc as well (which serves to expand on the concept, but not directly address your criticism) To acknowledge how you have reacted: we don't mean that software engineers are better ops people or can design a better ops function. Operations is an interesting and well understood job function, and it would be hubris to believe a group of software engineers can do it better. By letting software engineers design and implement an operations function, we intend to highly incentivise the folks who have to do the ops (who aren't keen or good at it) to replace the ops with software (which they are motivated and capable to do). I hope that addresses your concerns. If you want to read more, especially putting the quote in context, then I recommend the introduction to the SRE book, where Ben Treynor (who said the above line originally) writes at length about how and why he started the SRE practice at Google.