I got caught up in this link tonight from the Agile Management Blog. The post is about what the "right conditions" for agile adoption are and as a corollary why it seems to be so hard to make that adoption complete.

I found it really interesting that the author used the book Collapse by Jared Diamond as an explanation. I've been through many an agile adoption fight, most I've lost or at best come out bloody and beaten. Personally I think the difficultly lies in the very nature of software development. The incessant need to "schedule" and "manage" always seems to get in the way.

In the past I've had discussions with stake holders that when something like this:

"I need to know when you can release version 2.5 so I can schedule the resources appropriately."

My answer: "Well, the release is currently estimated at 6-10 months."

"I need a specific date."

My answer: "What features can I pull from the release if the date approaches and we find that the estimates weren't accurate?"

"If you say something will be in the release it needs to be there."

My answer: "I can't promise to release every feature on a specific date when we don't know exactly what the features are or how much time and effort each will take."

"So you need more time to make the estimates accurate?"

My answer: "No. The estimates are unlikely to be significantly more accurate..."

How do you explain that the very nature of the absolute schedule and variable work will never meet at a specific point and time? The issue isn't so much one of people unwilling to change [as the Norse apparently were], but rather people unable to communicate why the change will benefit everyone.

From the perspective of a software developer, I need to know what the project priorities are, when the hard stop date is, and who is working on the project. To me, the 'who' and the 'stop date' are the two important pieces of the puzzle. From the business standpoint the priorities are the most important. The piece in the middle, the fact that without the right people, the priorities will never happen, always seems to be overlooked.