Re: measuring the database

"R.Menhennett" <r.menhennett_at_lmco.com> wrote in message news:cn2mcn$p4n4_at_cui1.lmms.lmco.com...
> Folks,
>
> Where I work the software developers (usually using C++) often use SLOC
> to measure their output for the purpose of trying to predict the amount
of
> development effort required for future projects. This, however, has not
> been
> applied to our database objects, like stored procedures or tables, that
are
> part of
> the development effort.
>
> Does anyone have any experience with measuring database objects for
> predicting
> future effort? If so, what are you measuring and how successful is this.
>
> Thanks in advance.
>
> --
> Roger
>
>

There are two ways to look at this:

1. The amount of effort to program/build something is directly related to the completeness and accuracy of the specifications derived from the users. Being that these requirements are never complete nor accurate, it is impossible to determine the amount of effort needed. Past performance is _not_ an accurate enough indicator. As has been demonstrated for decades, it is not possible to accurately predict how much code can be written in a given amount of time. Pardon me- you _can_ predict how _much_ code can be written, but not if it will be bug free or perform the necessary functions. Anyone who tries to do so should be spending the time coding. Or drinking coffee- at least you get a buzz for your effort.
2. To predict future effort, you need to get out your crystal ball.

If you don't believe me, ask yourself this question: Why are I.T. projects always late and over budget?

P.S. Don't bother trying to convince your programmers/management that they are wasting their time trying to predict how long it will take to do anything. Received on Fri Nov 12 2004 - 12:11:24 CST