Thread by @rdonoghue

Somebody messed up Jira today, but I'm not mad at them. The mistakes they made can all be traced to things that Jira makes hard and confusing, and the bad policy decisions around this particular project.

These problems are leading directly to specific actions we're going to take to make future problems less likely and more recoverable. I have a list, and will take care of most of them tomorrow.

I *absolutely* had a moment of strong emotional response to the problems. They were avoidable. They were rooted in mistakes that *I* would not have made. The traditional thing to do would have been to call the guy onto the carpet. That would have been *dumb*.

Yes, he messed up, but the underlying system was set up in such a way that it was not a question of if these things would be problems, only when. None of the decisions he made were foolish or ill-intended. They made sense in a bad context.

So we fix the *context*

I mention all this because it's easy to speak in the abstract about blameless post mortems and system resiliency, but when you have just spent more time than you like fixing a seemingly avoidable problem, emotions point more towards "revenge".

That is *not easy* to set aside.

And it's important to remember that. The person who had to shovel the crap has some genuine emotional reasons to be uninterested in your blameless post mortem, and their needs are critical to address as well.

That is to say, do not let a blameless post mortem be a "Let's pretend there wasn't a problem" most mortem, or a "Ignore that Indu was up all night patching" post mortem.

Blameless still leaves room for recognition.

This is mostly a reminder to myself. I am a genuine believer in systemic RCA and blameless post mortems, but in my attention to underlying systems it is easy for me to skim past the very real impact of the problem and the equally real work that went into fixing it.

Because here's the other thing:

If it IS a system problem
And I AM knowledgeable enough that I wouldn't have screwed up

Then there's a REALLY good chance that some of the systemic problems have my fingerprints on them.

(Which they totally do. I can point to several contributing factors that were on my "I'll get around to it" list)

Getting to the emotional point to be able to acknowledge that is hard enough from a clean start, much less when wrapped round the axel by an issue.

It is easier to blame someone than to look at a system, and it's easier to blame a system than to acknowledge that you have enabled or perpetuated that system.

You do it, because it's the work. But respect that it IS work.