The Sociotechnical Path to High-Performing Teams • Charity Majors • GOTO 2023

Рет қаралды 15,332

Күн бұрын

This presentation was recorded at GOTO Chicago 2023. #GOTOcon #GOTOchgo
gotochgo.com
Charity Majors - CTO at honeycomb.io & Co- Author of Observability Engineering ‪@CharityMajors‬
RESOURCES
/ mipsytipsy
github.com/charity
/ charity-majors
ABSTRACT
There is a yawning gap opening up between the best and the rest - the top few percent of elite engineering teams are making incredible gains year over year in velocity, reliability and productivity, and more teams are achieving elite status every year. Yet meanwhile the bottom 75% of teams are actually losing ground each year. Contrary to common belief, this has almost nothing to do with engineering ability, and everything to do with the sociotechnical feedback loops that are the beating heart of every engineering org.
Great teams make great engineers. So let’s talk about how great teams are forged. [...]
TIMECODES
00:00 Intro
00:50 Teams
11:00 How to build high-performing teams?
14:40 Insidious loop
15:30 Virtuous loop
21:51 ODD
25:19 A dirty little secret
29:16 Outro
Read the full abstract here:
gotochgo.com/2023/sessions/2517
RECOMMENDED BOOKS
Charity Majors, Liz Fong-Jones & George Miranda • Observability Engineering • amzn.to/38scbma
Charity Majors & Laine Campbell • Database Reliability Engineering • amzn.to/3ujybdS
Kelly Shortridge & Aaron Rinehart • Security Chaos Engineering • www.verica.io/sce-book
Nora Jones & Casey Rosenthal • Chaos Engineering • amzn.to/3hUmuAH
Russ Miles • Learning Chaos Engineering • amzn.to/3hCiUe8
Nicole Forsgren, Jez Humble & Gene Kim • Accelerate • amzn.to/442Rep0
/ gotocon
/ goto-
/ goto_con
/ gotoconferences
#Teams #HighPerformance #HighPerformingTeams #InsidiousLoop #VirtuousLoop #Observability #ODD #ObservabilityDrivenDevelopment #Accelerate #CharityMajors
Looking for a unique learning experience?
Attend the next GOTO conference near you! Get your ticket at gotopia.tech
Sign up for updates and specials at gotopia.tech/newsletter
SUBSCRIBE TO OUR CHANNEL - new videos posted almost daily.
kzfaq.info...

Пікірлер: 6

@NeilHighley 8 ай бұрын

Very good talk. Development is about understanding the system and how to monitor the system. Also, the point about the effect we have on the team versus the effect the team has on us is extremely useful.

@chris.dillon 8 ай бұрын

She's going to write another book and I will buy it. ❤

@barneylaurance1865 8 ай бұрын

This sounds like a very good path. I have a question about the key - given at 17:40 "if it's going to a place where real humans hit, that brings enough chaos to count". What counts as a real human hitting? Specifically does a person doing QA work on your software count as a real human, or do they only count if they're actually using the software as a tool to do something else. My situation is as a developer at a very small organisation where the main thing we do is run time limited annual online fundraising campaigns - and most of our traffic for the entire year comes in one week around the start of December. So it's hard to make sure real humans are hitting our code immediately after any work is done.

@prdoyle 5 ай бұрын

I think Honeycomb is somewhat unique in that they are themselves the observability tool. That means they are exactly the thing their own devs use to tell whether their changes are working. They don't need to do any artificial QA work; they just use the tool.

@barneylaurance1865 5 ай бұрын

@@prdoyle Right. They make a tool that they also use extensively themselves. Observability isn't the only market where that's possible - e.g. the makers of Slack probably have discussions about how to make it on Slack, the makers of Facebook probably use Workplace by Meta (aka Facebook at Work) a lot, CircleCI probably use their own product. But there are indeed many more products that can't be used much for real by the people making them.

@kalmarnagyandras 8 ай бұрын

1. AI doesn't solve any people problems, and technical problems are not the challenge. 2. AI/Robots can only take your work, if you're doing a robot's work. 3. Generative AI will also not take your work, because you can't encode intent and "sense" in a model. LLMs just spit out the most probable next word in a sequence.