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

Re: [scrumdevelopment] where specialization breaks down

Expand Messages
  • RonJeffries
    Hello, Lukasz, ... Please tell us a bit more about the relationship you see between the 18th century thoughts of Adam Smith about manufacturing, and 21st
    Message 1 of 5 , Mar 1, 2012
    • 0 Attachment
      Hello, Lukasz,

      On Feb 16, 2012, at 8:41 AM, Lukasz Szyrmer wrote:

      I’m just playing devil’s advocate, in order to try to isolate what is really new in scrum or even agile. Any thoughts or resources you might have on this would be interesting, particularly given the my growing interest in software craftsmanship.

      Please tell us a bit more about the relationship you see between the 18th century thoughts of Adam Smith about manufacturing, and 21st century team-based creative work.

      Ron Jeffries
      I'm not bad, I'm just drawn that way.  -- Jessica Rabbit

    • lszyrmer
      Ron, Before 18th century manufacturing, the main way people made a living was 18th century team-based (actually family-based) creative work. Manufacturing,
      Message 2 of 5 , Mar 1, 2012
      • 0 Attachment
        Ron,

        Before 18th century manufacturing, the main way people made a living was 18th century team-based (actually family-based) creative work. Manufacturing, when it was introduced, was the super-hyper-nano-bio-info-tech of that age. In Smith's thought experiment, specialization was directly responsible for a 400x gain in productivty, which is a little more than the upper end of productivity gains experienced with agile techniques (15-20x gains in distributed scrum-xp teams that Jeff Sutherland has case studies of). That's why I find this "manufacturing vs complex work" argument to be only part of the whole picture, and there's still something missing for me on a theoretical level. I don't really know what it is…

        Once possible difference is the existence of bureaucracy. Because bureaucacy creates information silos because of excessive specialization, bureacracy itself creates bottlenecks. By moving back to a collaborative and creative team model, you clear bottlenecks, and as a result, experence big jumps in productivity. So bureacracy in the Niskanen sense could just be "too much of a good thing", and scaling back gives you a productivity boost. I argued this in greater detail recently in a blog post: http://softwaretrading.co.uk/2012/02/18/heresy-alert-division-of-labor-doesnt-increase-productivity-in-software-engineering/

        But since that post, I've had another think through the logic. This argument itself has some problems with it also. What about agile within (presumably high performing) startups?

        And I still don't really feel like I've addressed this theoretical disconnect with this idea.

        Luke


        --- In scrumdevelopment@yahoogroups.com, RonJeffries <ronjeffries@...> wrote:
        >
        > Hello, Lukasz,
        >
        > On Feb 16, 2012, at 8:41 AM, Lukasz Szyrmer wrote:
        >
        > > I'm just playing devil's advocate, in order to try to isolate what is really new in scrum or even agile. Any thoughts or resources you might have on this would be interesting, particularly given the my growing interest in software craftsmanship.
        >
        >
        > Please tell us a bit more about the relationship you see between the 18th century thoughts of Adam Smith about manufacturing, and 21st century team-based creative work.
        >
        > Ron Jeffries
        > www.XProgramming.com
        > I'm not bad, I'm just drawn that way. -- Jessica Rabbit
        >
      • Michael James
        ... My guess: Standardization, mechanization, and automation turned out to be very powerful things. They worked so well last century that modern craftsmen can
        Message 3 of 5 , Mar 1, 2012
        • 0 Attachment

          On Feb 16, 2012, at 5:41 AM, Lukasz Szyrmer wrote:

          didn’t we have “craftsmanship” before the industrial revolution? Why didn’t that give us business growth on the scale that the above did? The middle ages had craftsmen organized in guilds for hundreds of years, but with very little economic (business) growth.

          My guess: Standardization, mechanization, and automation turned out to be very powerful things.  They worked so well last century that modern craftsmen can now delegate that kind of work to machines (or factory workers in Shenzhen who may be replaced by machines in a couple decades).

          --mj
          (Michael)
        • David Koontz
          Is this type of standardization & automation not what each of our developers do every day when they submit a bunch of code to the compiler? We have automated
          Message 4 of 5 , Mar 1, 2012
          • 0 Attachment
            Is this type of standardization & automation not what each of our developers do every day when they submit a bunch of code to the compiler?  We have automated and measured every single mili-second of processor time - that process step is very efficient.   They manufacture a product several time an hour, test that product, find it lacking, then destroy that product, and finally rebuild it better in a few minutes.

            What we have done in software is abstract way from the manufacturing - that process happens several times in an hour.  What we do is DESIGN.  What did Mr Smith say about the creative process of building a factory to meet a consumer need.  We can do that factory building phase of product development in about 3 months (order of magnitude).  

            Are we mixing up the scale of the problem domain?  Do you wish to use the same tools to solve all problems at all scales - I don't.

            D a v i d   K o o n t z


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