Loading ...
Sorry, an error occurred while loading the content.

Re: [scrumdevelopment] Re: Sutherland-Scrum and the Toyota process

Expand Messages
  • David Starr
    Jeff, Do you see a correlation to the suitability of limiting WIP during the Sprint to the complexity of the work undertaken? For example, if the Development
    Message 1 of 3 , Oct 31, 2011

      Do you see a correlation to the suitability of limiting WIP during the Sprint to the complexity of the work undertaken? For example, if the Development Team is working closer to the chaos boundary, is there less need to limit WIP and focus instead of inspection/adaptation?

      And visa versa?

      David Starr
      Scrum.orgImproving the Profession of Software Development
      Blog: elegantcode.com | @elegantcoder

      On Sun, Oct 30, 2011 at 4:18 AM, jsutherland <jeff.sutherland@...> wrote:

      The reason Mary and Tom think I do Toyota Scrum is that I came to Scrum through Takeuchi and Nonaka. They were looking at the best teams at Toyota, Honda, and other so-called lean companies and saw the implementation of cross-functional teams working intensively together on continuous improvement. This is exactly what Taiichi Ohno implemented at Toyota. Lean practices are side effects of implementing Ohno/Nonaka/Takeuchi Scrum Project Management.

      Therefore you should see these on Scrum teams. Removing impediments is the same think as removing waste. If Scrum teams are removing waste their Scrum board should be a Kanban board. They should limit work in progess. They should stop the line, etc. Failing to do this is some kind of thing that is not Takeuchi and Nonaka Scrum. See: http://scrum.jeffsutherland.com/2011/10/takeuchi-and-nonaka-roots-of-scrum.html

      Ken and I are writing a book together right now and these issues will be described there. It is interesting that Ken talks about Professors Ogunake and Ray who educated him on empirical process theory. They showed him how traditional project management is totally unsuitable for software development. If it worked they said they would see 99.9% success rate whereas the failure rate is over 80% for larger projects. They told Ken to make sure he implemented empirical process management based on inspecting and adapting. So Ken doesn't talk much about lean and focuses on getting software done in 30 days or less and inspecting and adapting.

      However, this is just another view of what Taiichi Ohno calls continuous improvement.

      So I look at a Scrum board like Taiichi Ohno would look at it. There are a lot of horrifying Scrum boards out there, mainly because people are not limiting work in progress. People don't even know their velocity. They cannot demonstrate doubling, tripling, and then quadrupling velocity while simultaneously getting even better improvements in quality. Taiichi Ohno would never run a Scrum like that and neither would I.

      Another interesting fact is that Toyota does traditional project management for software development which is a disaster. It costs them $2B recently because of a software bug in the Prius anti-lock braking system. They are committed to introduce lean into software but they innovate too slowly. I talked to a former head of Toyota software development who says they will never get there in a reasonable time frame. Now that there are over 12 million lines of code in a Lexus they must turn into a software company. If they don't they will lose their number one position in the auto industry.

      Jeff Sutherland

      --- In scrumdevelopment@yahoogroups.com, "ashish_wrt" <traderashish@...> wrote:
      > "Then Mary singled out Jeff Sutherland as an exception, claiming he
      > doesn't do Scrum the way everyone else does, as he enforces all
      > necessary software development practices, has a lead engineer in the
      > team, does architecture up front, and has a Scrum Master who codes 90%
      > of the time. In essence he runs "the Toyota process". I am not
      > stating facts here, just repeating the gist of the conversation. I
      > concluded that in Mary's opinion only Jeff Sutherland (and those
      > trainers who work directly "for" him) understand what Scrum is,
      > or ought to be: i.e. Sutherland-Scrum rather than Schwaber-Scrum. "
      > from Tobias's 2008 blog
      > I stumbled on this in very old post by Tobias and found it very
      > interesting. I recently saw a post from Sutherland on this yahoo group,
      > and to think of it, it was very close to what Mary said above.
      > The process described above is the Scrum I learned and did at Yahoo and
      > curious to know if you agree with this approach or not and why?
      > What do you guys think?
      > (I do not know if Sutherland and Schwaber differ on Scrum or how they
      > differ. And I really do not care. That is not the point of the post
      > really. )
      > Ashish Gupta
      > http://www.bringingagility.com

    Your message has been successfully submitted and would be delivered to recipients shortly.