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

Re: [extremeperl] estimation

Expand Messages
  • Ged Haywood
    Hi Perrin, ... For the last two thousand years or so, people who have tried to change the way other people think have usually wound up nailed to something. ...
    Message 1 of 3 , Jan 28, 2002
    • 0 Attachment
      Hi Perrin,

      On Mon, 28 Jan 2002, Perrin Harkins wrote:

      > I am trying to convert the management at my organization to a more
      > XP way of thinking regarding estimates.

      For the last two thousand years or so, people who have tried to change
      the way other people think have usually wound up nailed to something.

      > I just had a meeting about a longer-term project (about 3 months)
      > where I told them that I can provide a working release every two
      > weeks, and they can choose the most important features they want in
      > the next release each time. They were very reluctant.

      They probably don't believe you.

      > They would much rather have me tell them a date when all of their
      > features will be done, and I think they would rather have this
      > regardless of whether or not the date is correct.

      They KNOW the date isn't correct, it's called 'experience'.

      > They just can't deal with the idea of working on the most important
      > features until the systems has enough to go into production, or even
      > with working toward a date and dropping features as necessary to hit
      > it. I wonder if this is the biggest obstacle to getting XP
      > practices adopted.

      If you put it like that, it will be an obstacle. You are in effect
      saying "we know we are going to fail, we just don't know by how much
      until after it's happened". Terrible sales pitch.

      > I also wonder if this is all the fault of consulting companies, who
      > always seem to give exact dates, even though they never hit them.

      Well, 'never' is arguable. I remember once estimating 30 days for a
      bit of assembly language coding and delivering it on day 29. (OK, it
      was in 1982, but I still have the PERT chart. On the wall.:)

      This has got nothing to do with the technical merits of the case, nor
      with consulting companies charging big bucks to make wild guesses.
      The point is they are saying "we will succeed". People like that a
      lot better, it gives them someone to blame when it all hits the fan.

      Don't forget that the people you are dealing with have their careers
      to think about. Most people are not high fliers, and it's really only
      high fliers who cope well with risk. Even if the chances of failure
      are small (and if that were the case you wouldn't get a kick out of
      success, would you?:) for most people it's much more important to have
      someone/something to blame in the case of failure than it is to have a
      high chance of success. That way it won't look bad on their CV.

      So be strong, make carefully reasoned estimates with confidence, and
      don't be scared of taking the risks - and the blame. After all, the
      worst that can happen is that you'll find yourself looking for work.

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