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

Twenty Ways to Split Stories; Mini-Review; Word Stack Puzzle

Expand Messages
  • Bill Wake
    [Crossposted to XPlorations] Happy 2006 to you all! Twenty Ways to Split Stories http://xp123.com/xplor/xp0512/index.shtml When a story is too big, it s good
    Message 1 of 3 , Jan 2, 2006
    • 0 Attachment
      [Crossposted to XPlorations]

      Happy 2006 to you all!

      Twenty Ways to Split Stories
      http://xp123.com/xplor/xp0512/index.shtml
      When a story is too big, it's good to find ways to split it to extract
      the most value, or to reduce it to a size that fits in an iteration.
      This note outlines a number of ways to split a story.

      Mini-Review
      http://xp123.com/books/index.htm
      * Game Design - Theory and Practice

      For Fun
      http://xp123.com/g4p/0512/index.shtml
      * A "word stack" puzzle on games

      --
      Bill Wake William.Wake@... www.xp123.com
    • William Pietri
      ... That s a great list! I only could think of three to add: quality, reliability, and scale. Consider quality. Suppose you re making a digital camera. There
      Message 2 of 3 , Jan 3, 2006
      • 0 Attachment
        Bill Wake wrote:

        >Twenty Ways to Split Stories
        >http://xp123.com/xplor/xp0512/index.shtml
        >When a story is too big, it's good to find ways to split it to extract
        >the most value, or to reduce it to a size that fits in an iteration.
        >This note outlines a number of ways to split a story.
        >
        >

        That's a great list! I only could think of three to add: quality,
        reliability, and scale.

        Consider quality. Suppose you're making a digital camera. There are a
        zillion features to add, but there's one pretty basic one that seems
        hard to split: taking a photo. If you're going from scratch, that's a
        lot of work around image sensors, lenses, RAM, color fidelity, and who
        knows what else. It's way too much to do in a week.

        But suppose you start with a 1-pixel black and white camera? It seems
        useless: you point it at a lamp and get "1"; you cover the front with
        your hand and get "0". But from there you have a clear path of
        improvement along many axes, and each step can be a card. E.g., 9
        pixels, 256 pixels, 10,000 pixels. 3-bit color, 12-bit color, 24-bit
        color. 10% color accuracy, 50% color accuracy, 90% color accuracy.

        A similar qualitative-to-numeric trick can work with both reliability
        and scale. I often make teams break down uptime requirements so that the
        XP Customer can buy reliability as needed, with a series of cards
        ranging from one nine on up. And putting scale explicitly in the cards
        can radically improve agility: building a system that works for millions
        when your business plan calls for several months of closed alpha testing
        is hugely wasteful, slowing the team down for no initial benefit.


        I find these three splitting approaches especially interesting in that
        they often invert the relationship between developers and the XP
        Customer. Normally, the Customer has a bit of a kid-in-a-candy-store
        problem, while the developers act as a voice of moderation. But
        developers generally lean toward building high-quality, very reliable
        systems that scale hugely, while the Customer is perfectly happy to
        declare a weekly maintenance window if that gets him 25 points of shiny
        features.


        William
      • William Wake
        ... Nice! -- Bill Wake William.Wake@acm.org www.xp123.com
        Message 3 of 3 , Jan 3, 2006
        • 0 Attachment
          On 1/3/06, William Pietri <william@...> wrote:
          > That's a great list! I only could think of three to add: quality,
          > reliability, and scale.

          Nice!

          --
          Bill Wake William.Wake@... www.xp123.com
        Your message has been successfully submitted and would be delivered to recipients shortly.