Welcome to SPN

Register and Join the most happening forum of Sikh community & intellectuals from around the world.

Sign Up Now!

Time Estimate

Discussion in 'Information Technology' started by Daniel, Jul 28, 2006.

Tags:
  1. Daniel

    Daniel
    Expand Collapse
    Guest

    Good morning,

    I finally managed to get some specifications from my boss as to the type of
    database that he would like me to develop. The issue is that he asked me to
    provide him with an time estimate to accomplish the work. I have tried to do
    this in the past and have had some horrible surprises.

    - Are there any general guideline that could help me make better estimate as
    to the development time?
    - Any good websites that cover this subject?

    Thank you,

    Daniel
     
  2. Larry Daugherty

    Larry Daugherty
    Expand Collapse
    Guest

    How long is a piece of string? How high is up?

    There are so many variables that go into the creation of an
    application ... It is true of all human endeavors. You can't give an
    accurate estimate until a product specification is complete. Then you
    would base your estimate on your expectation as to your ability to
    complete the analysis, the design, the coding and the testing. That
    has to be based on your recollection of how long it has taken you to
    do those things for a similarly sized task before now. Another thing
    is the level of sophistication that will go into bullet proofing your
    application and making it ever so user friendly.

    There are a couple of truths that go along with any project of this
    kind.

    1. It will always take longer to complete a project than you think
    it could or should. At the end of a long overdue project you'll be
    able to tell where the time and effort went but you couldn't see it
    coming at the beginning. Make what seems to be a reasonable estimate
    and then double or triple it.

    2. Once you deliver your application, having accomplished
    everything in the specification, the requests for changes will come
    pouring in. So, to the extent possible, involve your user community.
    Hold regular meetings with them and let them know you value their
    contributions. It can be valuable to get prototypes of the
    application out at regular intervals. Getting them out a few days
    before you meet with your users is a good idea. When you do that
    you'll have to continually remind your users that this is the
    direction the project is going and you need them to take it for a spin
    to see if it looks like you're going to meet their needs. You will
    have difficulties with your boss who will look at the first prototype
    and figure that the project is all done.

    3. Make sure you have a fairly complete product specification and
    functional specification before you start designing with Access. Keep
    at it until you can't think of anything else the application has to
    do. Publish it to your user community and your boss. Once you get
    into the actual designing and coding in Access you'll find other
    things but if you start without having those two documents in hand
    your project will take even longer because you won't know where you're
    going and you won't know when you're done. Most people who aren't in
    a formal development system fail to recognize just how important the
    specifications and design are to the success of a project. Until
    recently, IBM was the biggest manufacturer of software in the world.
    For software development projects large and small the rule of thumb
    was that 2/3 of every resource expended in the development was
    expended in the specification and design. The coding and testing only
    consumed 1/3 of the resources!

    I believe that Stan Leszynski referred in one of his books to some
    rules of thumb that he used for each element of an Access project
    (circa Access 2.0 or 97). He had a team of developers who were well
    practiced with Access and who sere doing the same things repeatedly.

    One other thing to bear in mind as you work on estimates: You have a
    lot of other things going on. You tend to estimate work on how much
    devoted effort it takes to do a given thing. You tell your boss "It
    looks like 4 weeks of effort". He immediately blocks out one calendar
    month on the calendar. Exactly half the time you'll probably need to
    be able to devote 4 weeks of effort.

    Once you have some experience developing applications in that
    environment you'll be better able to estimate. Until then it's kind
    of a crap shoot.

    HTH
    --
    -Larry-
    --

    "Daniel" <Daniel@discussions.microsoft.com> wrote in message
    news:BBC3BA28-92CB-4F5C-8C4A-3A96B0171098@microsoft.com...
    > Good morning,
    >
    > I finally managed to get some specifications from my boss as to the

    type of
    > database that he would like me to develop. The issue is that he

    asked me to
    > provide him with an time estimate to accomplish the work. I have

    tried to do
    > this in the past and have had some horrible surprises.
    >
    > - Are there any general guideline that could help me make better

    estimate as
    > to the development time?
    > - Any good websites that cover this subject?
    >
    > Thank you,
    >
    > Daniel
     

Share This Page