I think I understood the assignment...

Homework

While at the Learning From Incidents conference, Dr. David Woods gave us all an assignment and I have to say this is the first time I’ve "turned in" homework in a long while. There's many more gems of insights for me to mine from the recordings and my notes but I wanted to get my first impressions out in the ether.

What patterns do you see now that you did not see clearly before?

Individuals grasp the value of learning from incidents long before an organization or leadership does.

What elaborations of your patterns became visible?

What Laura Nolan calls “islands of knowledge” where experts respond to problems all the time creating a lack of opportunities for others to learn. 100000% agree. Especially when it comes to operations engineering.

I find this is common in the observability space both with siloed expertise in understanding the lifecycle of a metric/span/log as well as how to navigate enterprise observability tooling. It feels like many organizations do not get a decent ROI on their observability spend. Not because the data itself is garbage but due to inadequate training and the befuddling UX of most platforms. (OK and also because without a holistic telemetry strategy or even cohesive tagging/metadata story the data can be more like fool's gold).

How did you add to your repertoire?

Discussing with attendees and tech friends who couldn't attend

Taking notes for later reference

Catching up on talks I missed LFI Lineup on YouTube

Blogging about it

Reading white papers mentioned (this is huge for me as someone who's avoided academia since I left uni)

#AllOfTheThings

What facilitates learning?

Intentional institutional knowledge gathering and sharing (e.g. incident library to peruse, not laying off seasoned engineers en masse)

AND

Cross-organization/discipline knowledge sharing events like LFIConf

What hampers learning as updating/revision?

Not being able to openly ask questions and get a transparent answers for “why?” (Why do we have this incident retro format? Why do we have incident severities?)
Personally when I have a "why" I can cope better.... even if its an unsatisfying answer like "Because JimJoeBob who was the architect a few years ago really loved this esoteric infrastructure management tool and we are not going to invest the time to switch it this year"

Not enough slack in the project planning and day to day routines

I am very keen on further discussing and distilling learnings from conf attendees and other LFI nerds so please drop a comment below, reach out in the LFI slack, or find me on Mastodon to continue the discussion (@paigerduty@hachyderm.io)

CAT TAX