CTO of Telerik. Board member of the Scrum Alliance.
Stand on the shoulders of giants (a lot of his presentation was based on the work of: User stories applied / Mike Cohn
These are my notes from Stephen’s presentation
http://www.pluralsight.com/author/stephen-forte – Courses on Pluralsight (including Agile Estimation)
“The cone of uncertainty… is that early project estimates will always be wildly inaccurate”
Based on 50 years of data on software development.
How to do better?
- Have ‘stories’.. As a… I want to… In order to… to keep focussed
- Estimate toughness of each story
- Have a list ‘backlog’ of stuff to do and what I’m doing in next few weeks
- Keep on estimating after each sprint, and communicate back to client so they have no surprises
This means that:
- Know our release dates
- Will know what features will make it in as development proceeds
- Iteration 1 always goes wildly wrong (figuring stuff out)
- 5 iterations to get trust with client