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

cooper on agile

Expand Messages
  • Robert Biddle
    So as usual for me after a big conference, I am still mulling over what happened at Agile 2008 in Toronto. This year for the first time, the UI+Agile issue was
    Message 1 of 69 , Aug 12, 2008
    • 0 Attachment
      So as usual for me after a big conference, I am still mulling over
      what happened at Agile 2008 in Toronto.

      This year for the first time, the UI+Agile issue was addressed well,
      and I am only sorry I did not get to see all of the relevant
      presentations.

      I was especially looking forward to Alan Cooper's keynote, and a bit
      disappointed it was at the end of the conference, because that gave
      limited time to discuss it with others.

      I don't know Alan personally, but I have read his books and admire his
      work. I think he overstates his case, but that's how things are for gurus.
      He was at the conference the whole week, but Agile is now a huge
      conference, and I didn't get to talk to him. Especially because of the
      early interview/debate with Kent Beck, I was interested to hear his
      views.

      Anyway, I'd like to discuss the ideas from the keynote. His slides
      and notes are here:
      http://www.cooper.com/journal/2008/08/alans_keynote_at_agile_2008.html

      Perhaps the most important thing in the talk is that he showed a
      positive attitude to agile methods, and suggested that an agile
      approach to software development was an ideal partner for UI design.
      I was pleased at this, and say much the same thing myself.
      tally, and in collaboration with the
      UI people and customer. And the again in a production step, which did
      not need to be iterative, incremental, or collaborative. I did not
      understand why.

      He said a number of things about software development and programming,
      and I think they were in harmony with the views held in the agile
      community. He went out of his way to make clear that, while his
      emphasis has been on UI design for some time, he is an expert
      programmer, and still have passion and insight for such work.

      At the more detail level of his talk, there were some things I
      strongly agreed with, and others that puzzle me.

      For example, one point he made was that it sometimes seems agile
      methods suggest that developers just have to ask for stories, with the
      unrealistic expectation that the articulated stories will reflect what
      is really needed. I agree that is a problem.


      One point that puzzled me was the suggestion, I think, that iterative
      and incremental programming was necessary to deal with the
      technical challenges in writing the code. He seemed to suggest that
      the "business" aspects could be worked out in advance. I realize he
      advocates research and design before programming, but this seemed a
      heavier line than I expected.

      (He generally merged the "business" and UI aspects, but I'll leave
      that issue for another time.)

      Then, as I understood it, he suggested that the software be written
      twice, once iteratively, incrementally, and in collaboration with the
      UI people and customer. And the again in a production step, which did
      not need to be iterative, incremental, or collaborative. I did not
      understand why.


      So: comments from anyone on the talk? And especially, any advice on
      the details that puzzled me? Alan?

      Cheers
      Robert Biddle
    • Steve Freeman
      ... His last section essentially described the phases of RUP. That s not necessarily bad in itself but it suggests either he didn t want to say it out loud
      Message 69 of 69 , Aug 23, 2008
      • 0 Attachment
        On 12 Aug 2008, at 17:33, Robert Biddle wrote:
        >
        > One point that puzzled me was the suggestion, I think, that iterative
        > and incremental programming was necessary to deal with the
        > technical challenges in writing the code. He seemed to suggest that
        > the "business" aspects could be worked out in advance. I realize he
        > advocates research and design before programming, but this seemed a
        > heavier line than I expected.
        >
        > (He generally merged the "business" and UI aspects, but I'll leave
        > that issue for another time.)
        >
        > Then, as I understood it, he suggested that the software be written
        > twice, once iteratively, incrementally, and in collaboration with the
        > UI people and customer. And the again in a production step, which did
        > not need to be iterative, incremental, or collaborative. I did not
        > understand why.


        His last section essentially described the phases of RUP. That's not
        necessarily bad in itself but it suggests either he didn't want to say
        it out loud because he thought it would upset us, or that he didn't
        know.

        His contention that, having done it once, us coders could sit in quiet
        rooms and type it in (because by then we'd know everything we needed
        to about the design) was not popular on my table.

        I prefer Don Norman's line. I saw him talk years ago just after he
        joined Apple. "Gee", he said (more or less), "until I joined a product
        company, I'd no idea writing software was so hard."

        S.


        Steve Freeman
        http://www.mockobjects.com

        Winner of the Agile Alliance Gordon Pask award 2006
      Your message has been successfully submitted and would be delivered to recipients shortly.