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

Re: Designing for incremental delivery vs. big-bang delivery

Expand Messages
  • Jeff Patton
    ... that ... caught ... design, ... I d agree those are also critical strategies to apply - no need to just have one. My biases are a function of my past
    Message 1 of 102 , Aug 3 7:19 AM
    • 0 Attachment
      --- In agile-usability@yahoogroups.com, William Pietri <william@s...>
      > On Mon, 2004-08-02 at 08:13, Jeff Patton wrote:
      > > While Alistair was telling the story I was thinking to myself
      > > any interaction designer that was worth his salt would have
      > > this problem very early by doing just a little bit of contextual
      > > analysis.
      > Interesting! My reaction is that in addition to better up-front
      > this also would have been caught by better collaboration or frequent
      > releases. It only became a big problem because they made big bets on
      > their initial guesses about what the user would need.

      I'd agree those are also critical strategies to apply - no need to
      just have one.

      My biases are a function of my past experiences - as are everyone's I
      believe. Since discovering XP and agile approaches in general, I've
      delivered releases directly to customers in 6-8 week increments for
      the last several years. My customers generally installed this
      release in lab environments and had their expert users pound the heck
      out of it. They were unwilling to deploy each release into a
      production environment because of the costs of installation, rollout,
      and training. On more than one occasion I've been caught by
      surprises when the application was actually deployed in the field.
      The app seemed fine to the expert users, but once it was used in the
      real work context, we found big things that needed to change. In
      those cases I ended up kicking myself because I received some warning
      about these issues from the U-CD models we'd constructed and the role
      analysis we'd done. We'd placed so much stock in the expert user
      feedback, that we ignored the messages we were getting from our
      contextual analysis.

      So, I'd assert that we maintained great collaboration with genuine
      customers. We did do frequent releases and mid release
      demonstrations. We discounted our interaction design work in favor
      of input from those customers. Things didn't go as planned. In the
      future I'll work closer with our customers to help them understand
      the interaction design work so they can make more informed choices.

      Now - after all this, as a side effect of the high collaboration, the
      customers - the organization we were doing work for - felt
      responsible for the problems, and didn't blame us for the results.
      They were happy to pay for us change the product. So, by most folks
      assessments, things were successful. The actual end-users who had to
      struggle with the product for a couple months were pretty ticked off


    • Phlip
      ... Thanks; I made a note of it. But InternetExploiter and Windows XP don t support MNG, so I can t use MNG without introducing new systems. ===== Phlip
      Message 102 of 102 , Aug 7 7:55 AM
      • 0 Attachment
        Anthony Williams wrote:

        > I'm sure there's a variant on PNG that can animate
        > ..... checks ImageMagick
        > docs ..... MNG.
        > IIRC, Mozilla has optional MNG support.

        Thanks; I made a note of it. But InternetExploiter and
        Windows XP don't support MNG, so I can't use MNG
        without introducing new systems.


        Do you Yahoo!?
        New and Improved Yahoo! Mail - 100MB free storage!
      Your message has been successfully submitted and would be delivered to recipients shortly.