I really enjoy the conversation with software architects on how to make technology choices. I polled the members of the New York CTO club on the topic and got amazing war stories – I’ll keep those for later. I gave a talk on this yesterday @ GeneralAssemb.ly. In this iteration I focused on what can go wrong in a traditional vendor evaluation approach taken by many corporations.
Have you ever been part of one of those? It goes something like this.
- Identify Candidates : bring in vendors that can solve your problem, on paper
- Develop Metrics : figure out how to compare the vendors, consider scalability, complexity and cost
- Gather Data : build a feature and metrics matrix
- Compare : SUM(A:Z)
- Pick a Winner : MAX(SUM(A:Z))
A bakeoff is generally worth it, but is both expensive and totally boring!
At the same time, corrosion is something that plagues this kind of process through corruption (lets discuss our product over a fancy dinner) or, sometimes, pure intimidation and escalation (lets discuss our product over a fancy dinner … with your boss).
This talk is about better, more agile options. IMO, it boils down to trusting people vs. process. Slides from my talk @GA here.