presented by Eoin Woods
The paper is triggered by a very interesting question: “How do software architects prioritise their time and attention?”. So, what does an architect work on? In literature gives little answer. Philippe Kruchten (2008) gave an estimate: 50% architecting, and the remainder is communication. Fairbanks (2010) gives a risk-driven approach: “architecture as a risk-reduction cycle”. Poort and Van Vliet (2012) found architecture as a flow of decisions based on chaning risks and costs. Some use the SAFE framework, but only do cherrypicking, and then state they use that framework. So, the question remains: what do people really do?
so, they did a qualitative approach: semi-structured interviews with grounded theory to develop a model. They asked 7 architects in information systems: “how can architects concentrate their attention so that they are most effective?”. Somehow, three strong themes did emerge:
- Understand stakeholders needs and priorities;
- Prioritise time according to risks;
- Delegate as much as possible.
This results in a rather simple model, as the authors state themselves. But it makes sense!
One of the nice things is delegation: appoint or create groups to take architectural decisions. To validate the model, they start a survey, and everybody is encouraged to participate: http://bit.ly/arch-priorities.