Agile 2006 – Day 2

Today was a tough day with so many good sessions and OpenSpace events going on. After much debated, I started the day off in Todd Little’s Context Driven Agile Leadership. Just before his presentation started, I was talking with a genetleman from a consulting firm out of D.C. who said something interesting: “Anybody can rearrange furniture and have people stand in a circle” targetting who he called “light-weight consultants”.

A lot of Todd’s presentation was data points about his company’s implementation of Agile. It was fairly dry, though he mixed it up with some humorous clips of bull fights and the such. Some key points I pulled out were:

• Anything more than “barely sufficient” is a waste from a process perspective
• Don’t look to technology for silver bullets – you’ll more likely find them in people, values, communication and expectation management
• Mike Cohn’s presentations are really loud

Afterwards, I headed over to OpenSpace. I ended up chatting with the fine folks at ObjectMentor about a vehicle they just purchased and the CARFAX report they got (and doing a great job of completing misidentifying Dave Thomas as PragDave and looking like an utter fool).

One interesting thing did come up. We were discussing Ruby and he said it seemed like it had the ability to be the next big language – like Java or C# or C++ was. When I asked if they thought there were any other languages in the running, he said that as far as he could tell Ruby was the only one he could see. That, folks, is some cool stuff.

Over at OpenSpace we talked about Behaviour Driven Development with the founders Dan North and Chris Matts. It was very enlightning, and Dan talked about estimating scenarios instead of stories. In other words, take your stories, break them into scenarios of implementation, and estimate those. We happen to already do this – except that we just turn the scenarios into individual stories and toss the original card.

Over lunch the CARFAX team got together and chatted for a bit. After lunch, I headed to Mike Cohn’s Agile Estimating and Planning. Mike is an outstanding speaker, and the room gets very loud (which explains my third point from Todd’s talk). He covered the topics of Ideal Hours versus Story points, estimating, and planning poker. Mike developed a playing card deck which contains enough planning poker cards for 4 team members. It was incredibly enlightning, and while I pulled a lot of points out, here is only a sample:

• I’m the most ignorant today then I ever will be
• He’s never been asked to estimate a project in number of lines of code (neither have I)
• The 4 most important words are “Inspect and Adapt” + “Range” (which is what you should give someone when they ask for an estimate

I also enjoyed the exercise of taking the past iterations and calculating the means for the last 8 iterations, the mean of the worst 3 from the same period, and our last iteration, and showing where we can get to in the iteration for each number.

After Mike’s talk we had a social event before the Agile Alliance meeting. After the meeting a group of us got to head out on double-decker buses to a river boat cruise sponsored by NetObjectives and VersionOne, where I won a tres cool Agile 2006 Rugby shirt (even if it is a bit heavy duty), and got some great pictures which I’ll be posting as soon as I can borrow a USB cable.

Tomorrow is the Leadership Summit all day (which is a shame because there are some really good discussions) followed by the Google event and the IBM BoF event on the new Eclipse Process Framework. Until then, have a great night!