Where Agile Clashes with Modeling
Michael Mahlberg, TCG
Some folks say that Agile and modeling don’t go together. The reasoning goes something like “Big Design Up Front – BDUF – is dead. This is the age of agile. Therefor no more big models, therefor no more modeling.”
Obviously, somewhere along the road something went wrong.
Just as the agile manifesto doesn’t propose to drop planning (on the contrary – most approaches make planning a cornerstone) Agile in itself does not have to clash with modeling. Yet there is a lot of discussion wether “those huge modeling efforts” can be a good idea in the age of Agile. In my experience (and I started agile stuff in 2001 since before then what I did was just called FDD or XP) most of the time something “agile clashes with <X>” it really clashes with preconceptions about <X>. So in this talk we’ll walk through a number of those preconceptions with regard to modeling and how to address them.
This is not a talk on data modeling – this is a talk on agile approaches to dealing with the dichotomy presented by the necessity to quickly adapt to changing circumstances while simultaneously providing a stable view of the world for a multitude of reasons (from regulation through business interests).
Michael has been running his own company in Germany since the mid eighties and moved his focus from implementation to organization in the mid nineties. Working in a all kinds of industries from pure software development companies through to public administration and from companies with less than to dozen employees to the big multinationals, he has not yet seen a project where architecture and processes where not of crucial importance. As an ‘early adopter’ of agile – and later co-founder of the Limited WIP Society Cologne – he now spends most of his time supporting clients in their quest for more effective ways to work, mostly by applying lean and agile concepts.