My experience is that many big, complex organizations need an approach based on Information Management Transformation – a radical change to how they manage their information that is referred to as Information Development in MIKE2.0. But transformation is hard and there is a need to show value quickly. I’ve seen agile development techniques work for Software Development; can they work for Information Development?
The idea behind Agile Information Development is to provide an approach to most quickly deliver Information Management engagements using the MIKE2.0 Methodology. Agile development processes can be difficult for information management engagements due to the complexity of historical issues. Agile Information Development makes use of the techniques in XBR, Continuous Implementation and Continuous Improvement and accelerates them further. These techniques are from strategy through to implementation.
To find out more, go to the Agile Information Development Solution Offering. It’s in the early stages, so please jump in and help out!
May 27th, 2008 at 4:14 pm
Continuous improvement could also be called do-overs. If initial stages are rushed in the interests of being fast and agile then mistakes in data models, ETL templates and assumptions about the data could result in a lot of rework in the next phase. A lot of information development documentation is aimed at making sure the development work is of a high quality the first time around. There are tools to make this documentation process faster - such as metadata tools and profiling tools - that may eventually take the sting out of the early phases of Information Development work.
May 27th, 2008 at 11:18 pm
Hi Vincent - you are right - if you don’t get the foundational stuff in place first you’ll have lots of rework. I think of continuous improvement as “good” re-work/re-factoring that you should expect and explicility plan for. Bad re-work is when you are fixing a bunch of issues that you should have avoided in the first plan. There is of course some grey areas in this definition but I think the key is that if you’re going to be re-factoring you should plan on it ahead of time and that a lot of ineffeciency will be introduced if you don’t get the foundation right.
One of the things I hope we can acheive with MIKE2.0 is to give users some foundations that they can start with for any project, including templates, etc.
New technologies are making this process easier as you point out - making documentation a by-product of the process is extremely valuable.
The idea of agile information development is also to try and introduce methods that we apply in other areas of software development that can work well, like XP and Scrum.