No Ba, No QA, No Managers: Anarchy Works!

We dont have these management roles and dont miss them

Q: Why did you try this

I was experimenting, if you can spend all you money on programmers, do you not get more value

We know the roles of managers and dont like them - pm's dont really do project manager - the have roles like ambassidors - managers feel that they should make decisions as they are "in charge"

Have roles that are ambassidors, arbritrators between the software development and the customers

BA's principles
Knowlage transfer does not create bisiness value - agile is about empowering people - rather have the programmers talk to the business directly


BA is a single point of contact - having one person holding that in the head
- this leads to single point of failure, with team approach you get knowledge sharing and redundancy


How is the schedule of work managed?
- resource rumbles - talk about projects at high level and make sure a project is check-boxed
- be quite strict with the stakeholders in regards priorities - if the stakeholders ask for 8 things then which one is the most important
- business people keep coming up with ideas for business, throw up ideas that could be a winner - company accept that some projects will fail
- leadership is becoming more prevalent within the culture - Fred is a facilitator that is demonstrating leadership, getting coaching training and encouraging leadership in others


Who picks the teams ?
- the developers themselves, they know who can do what and how well - a resource rumble
- adjudicator to talk about overall priorities


avoid one guy being everything - that's a resource failure

still need domain experts to give input at a high level - separating the what from the how

programmers a pooled and move around the project to keep them fresh and remove

no need for testing, peer review, etc all the time as you leverage the skills and responsibilities of the team

use techniques such as blue green - have a quick mechanism to release and fix issues that may arise

taking the story concept and break into even smaller pieces and delivering them quickly

story tirrany - when is this story going to ready... kills off innovation


minimal risk involved
failures happen often enough so the testing of role backs happen often

Recommended book: Kent Beck - Smalltalk best practices

Notes by John Stevenson - @Jr0cket