Highlights from USI 2010 and Lean SSC London
June and July have been busy months. I'd like to share a few highlights from other speakers in USI 2010 in Paris and the Lean Software and Systems gathering in London on which I had the honor of presenting.
Long term sustainable releases with 99% backward compatibility [USI 2010]
In "The challenges of long-term software quality in open source" Jürgen Höller described how they worked in the Spring team to achieve 99% backward compatibility by avoiding revolution and using evolution, even when radical new features are fit in. During the last seven years the Spring team have absorbed 4 major JDK's and 4 generations of J2EE. I was sure this was possible and Jürgens team shows it is. A challenge to all of us the next time we want to restart from scratch :)
Learning to Learn - becoming a Lean startup [Lean SSC]
In this presentation Damon Morgan shows how they as a company now have reached a level where they continuously do set based engineering of business ideas. He showed using their Quote web page how experimenting with not so obvious changes lead to a jump in business leads. I noted another experience which I have seen - when you get flow going, estimation is redundant.
Using Kanban to get knowledge and continuously improve [Lean SSC]
Benjamin Mitchell blew me away with his presentation. I had some seriously great laughs :) But there are some serious learnings as well. Benjamin has done some great efforts in experimenting with statistical process control in software. For example, he could demonstrate that a bulk part of the product portfolio wasn't generating value to cover the complexity it brought by. But what does help if there isn't a thinking process in the organization capable of absorbing these learnings? I will highlights his takeaways, which we all can improve on:
Long term sustainable releases with 99% backward compatibility [USI 2010]
In "The challenges of long-term software quality in open source" Jürgen Höller described how they worked in the Spring team to achieve 99% backward compatibility by avoiding revolution and using evolution, even when radical new features are fit in. During the last seven years the Spring team have absorbed 4 major JDK's and 4 generations of J2EE. I was sure this was possible and Jürgens team shows it is. A challenge to all of us the next time we want to restart from scratch :)
Learning to Learn - becoming a Lean startup [Lean SSC]
In this presentation Damon Morgan shows how they as a company now have reached a level where they continuously do set based engineering of business ideas. He showed using their Quote web page how experimenting with not so obvious changes lead to a jump in business leads. I noted another experience which I have seen - when you get flow going, estimation is redundant.
Using Kanban to get knowledge and continuously improve [Lean SSC]
Benjamin Mitchell blew me away with his presentation. I had some seriously great laughs :) But there are some serious learnings as well. Benjamin has done some great efforts in experimenting with statistical process control in software. For example, he could demonstrate that a bulk part of the product portfolio wasn't generating value to cover the complexity it brought by. But what does help if there isn't a thinking process in the organization capable of absorbing these learnings? I will highlights his takeaways, which we all can improve on:
- THINK for yourself in your context
- Get KNOWLEDGE by studying your process as a system, end to end from the customer's point of view
- RUN EXPERIMENTS to learn while you work
Loading ...
Recent Comments