I started my career in software immediately after leaving university in 1990. I started out as a software developer and worked on a wide variety of different projects and platforms including: Automated Mapping/Facilities Management (AM/FM) systems for Hongkong Electric; billing systems for Telecom NZ; and a collections management system for Te Papa Tongarewa (the National Museum of New Zealand in Wellington).
After a few years I noticed that a large percentage of my projects (about 20%) were never released into production; either the project was cancelled for political reasons; or the code that I wrote was poor quality; or the customer had simply changed his mind. I thought about this problem for a while and I decided that I need more control over what I was doing. I can to the conclusion that the only way to obtain more control was to move into project management.
Throughout the 90′s (especially the early 90′s) waterfall was the predominate methodology. Rational Unified Process (RUP), only became popular in the late 90′s even though Ivan Jaccobsen’s book on Objectory (the pre-cursor to RUP) was published in ’92.
And so, I became a waterfall Project Manager.
One of the most successful waterfall projects that I was involved with was Land Online a project for Land Information New Zealand (LINZ). LINZ is a department of the New Zealand government that maintains a central register of land title transactions and cadastral surveys. This was medium sized project (of about 100 people) and the initial phase spanned about 3 year. The LandOnline project was finally rolled into production in 2000.
LandOnline was a success in may ways, and I personally learnt many valuable lessons from it. I learnt that customer involvement is crucial and I was luck to work with some to the best land information people in New Zealand including Neil Early, Ralph Winmill, Dave Morris, Heather Dougherty, Phil Davison, Don Grant and Tadeusz Dawidowski.
I also realized that it is absolutely possible to successfully deliver a waterfall project, but that a lot of compromises both personal and professional need to be made in order to do so.
The final waterfall project that I was involved in was an Operational Support System (OSS) from Rhythms.Net in Englewood, Colorado. This project had every single problem that we now know are common to waterfall projects and the end result was predictably disastrous. If you haven’t heard of Rhythms here‘s why.
After my Rhythms experience I decided that there had to be a better way in which to deliver software. I started looking around and was lucky enough to discover the Agile software development community. I was working with ThoughtWorks when I first worked with Ken Schwaber in 2001, and this article was the result. It’s also available in a print friendly format here.
For several years I travelled the US working on Extreme Programming (XP) projects for a number of clients including Capital One, MetLife, NationWide, and TransCanada to name just a few clients. It was a time of great change and learning. I had a difficult relationship with ThoughtWorks and I met some great people there, so I won’t say anything more on the subject.
The last few years have been very productive and busy for me. I started this blog (2005), became a Certified Scrum Trainer (CST, 2006) , became a Certified Scrum Coach (CSC, 2007) and started Seattle Scrum (2007). I’m looking forward to 2008; it’s going to be another busy year. I’ve got several articles in progress that I think are going to be interesting. At least, I hope you enjoy them as much as I do!