Saturday, August 28, 2010

Agile 2010 - Part 2

Selecting which session to go after lunch was much more challenging... Many known industry experts such as Martin Fowler, Mike Cohn, Johanna Rothman, Alistair Cockburn, Robot Martin and many others had parallel sessions from 1.30 PM to 5 PM.
You can view the schedule here

Ok .. I selected to go to Agile Estimating & Planning: From Basics to Brain Stumpers by Mike Cohn because agile estimation and planning is never a “Done” topic and still challenging.
Now let me dig my memory a bit…
I can remember that I enjoyed the session a lot.. Audience involvement was so much that I thought he will run out of time to finish his presentation ( I expected it to be.. Because all of us who are in to agile estimation have our own share of confusion :) )
Many practical problems were discussed in the session.. Some stuff I remember even after 2 weeks are as follows;

- Talking about working with larger backlogs and estimating them upfront…
One of the ideas I picked at the session was that if the product is large, don’t go in to the entire user story capturing at once. First identify themes, epics and come up with a budget for the project. Not estimation. Because unless you are the “god of estimation” you will never come up with a right figure for any software project at this time.
Working with fixed price contracts was also discussed up to an extent, but not in detailed level.. still signing the contract with a budget of Min to Max would be a very good idea if you can get your customer understanding the same reality about software development as you are.
Once I noticed Mike was surprised that one attendee was mentioning about his huge backlog where they have realized user stories to be implemented for over a quarter of the year

- Estimation is not a commitment
I agree with his example to prove that estimation is not a commitment. You need to estimate. and you need to commit.. but its 2 different things and you don’t commit to the estimate. Still if you explain this to your manager or customer.. I want to know how long you have to find the next job……:-).. Anyway this is a lengthy explanation to do on how it works.. you can find some good explanation here.

- Estimating stories without hours
He went in to detail with the audience about estimating user stories only with story points without getting in to hours. Then he explained how to arrive at calendar estimation based on the average team velocity calculation.
There he got the question which anyone could foresee coming… How do you do this with new teams or new team members in a team.,. His answer was that in a situation like this you need few practice sprints to arrive in such facts.
I see a problem in this when we don’t have the same team continuing projects for a longer time.

In the evening we had the Ice breaking party again a network session provided by the conference. There I got involved over a discussion with some of the CSTS from Argentina, USA, Canada and Belgium. They had some discussions about how they think about agile alliance and their practical issues.



PROJECTIZED. Copyright 2008 All Rights Reserved Revolution Two Church theme by Brian Gardner Converted into Blogger Template by Bloganol dot com