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

122590Re: [XP] Let's make things worse...

Expand Messages
  • glbrown@inebraska.com
    Sep 8, 2006
      Quoting Willem Bogaerts <w-p@...>:

      > Imagine a death march project. You are way behind schedule. You have to
      > deliver _something_ now. Anything. Or your client will rightfully walk
      > away without paying. Oh, and you know that your client will not accept
      > what you deliver. You deliver mainly to buy time. He will very clearly
      > tell you what is wrong. And then you'll have to deliver something again,
      > but something better. He may call it off if this one feature does not
      > work right....
      >
      > Sounds awful. But somehow also very familiar. This is the point where
      > waterfall projects get shorter releases. Where the customer's voice gets
      > more important than the design document written by the development
      > company. Where communication is painfully clear and quite frequent.
      > Where it becomes clear that it is the customer who decides if the work
      > is accepted, not some document.
      >
      > So let's see a software project as in your worse nightmares. Assume
      > every project to become a never-ending story. And unpredictable. And,
      > instead of becoming a farmer or something, you keep on doing those
      > horrible IT projects.
      > Now if you _know_ that it is a never-ending story, you would probably
      > adopt some very agile tactics: you'd split the project (and the payment)
      > up in chunks. You'd make acceptation a condition for further
      > development. And if the customer always has something to complain, make
      > him watch you, so he can't complain afterwards that it is "obviously" wrong.
      >
      > So I can see an agile method as a homeopathic treatment of a software
      > project. It is worsening the symptoms to stimulate the self-healing
      > processes.
      >
      > This line of thought could explain aversion against XP with managers.
      > Those managers would try anything in the world to suppress the symptoms
      > of a never-ending story, as soon as possible, to cure it.
      >
      > So instead of best practices turned to the extreme, I sometimes see XP
      > as a preparation for the worst.
      >
      > ... 'cause then it can only get better.

      I think that XP promotes shared risk between Customer and Developer.

      Fixed scope, fixed price puts the majority of the risk on the Developer.

      Time and materials puts the majority of the risk on the Customer.

      Sharing the risk promotes a healthy and productive partnership between
      the Customer and Developer. Anything less is almost certain to result
      in both parties failing to meet their expectations.

      GB.
    • Show all 14 messages in this topic