From: Andrew Gabb (agabb@tpgi.com.au)
Date: Fri Oct 31 2003 - 13:29:08 EST
Brett Boschma wrote:
> I do not know who it was who said that a well-written
> requirement would always be around the same size in
> function points (perhaps 5 points), but I tried to see
> things from that person’s point of view. I tried to
> think of a reason why somebody might say that.
Perhaps the statement was taken out of context, or the person involved has little breadth of experience, because it makes little sense when you look at it. It also is potentially dangerous if someone actually believes it (and too many of those modern-managers / numbers-are-everything types will want to).
The first issue is the *level* of requirements are you talking about. I've seen projects where there are 300 top level user/operational requirements, but over 20,000 more are derived as the development proceeds. So which level(s) do you choose? I've also seen single requirements that result in 1000s of FPs, eg incorporation of a wordprocessor in a system.
However, I guess it might be possible (in a very narrow application field) to train REs/analysts to write requirements which are (a) more aligned with FPs, and/or (b) result in a consistent FP/req ratio, with a hell of a lot of analysis, trials, training and practice, but I'm not at all sure that the payback would be useful. I think it's more likely that it would skew the expression of requirements in a way that could hazard the project.
Having analysed a number of quite different projects regarding the number of requirements used, I found that it would be very difficult to predict (or even justify) the number of requirements in each case, simply because there were so many different factors involved. Just a few of these are customer domain and tech knowledge, developer domain and tech knowledge, type of application, new/upgrade, quality required (inc. safety, security), standards used, compatibility with other apps, etc. Size and complexity are the main drivers, of course, but these are by no means totally dominant.
Another illustration of this is the following question: What would be the effect of using 100 or 1000 requirements as a basis for the same project, in terms of performance, cost, schedule, risk, effort, final FPs? Assume that the requirements set is reasonably 'good' in each case - the only difference is detail.
Andrew
-- Andrew Gabb email: agabb@tpgi.com.au Adelaide, South Australia phone: +61 8 8342-1021, fax: +61 8 8269-3280 ----- ------------------------------------------------------------------------------- To send a message to this mailing list send it to re-online@it.uts.edu.au. To unsubscribe from this mailing list, email majordomo@it.uts.edu.au with the message `unsubscribe re-online' in the BODY of the mail.
This archive was generated by hypermail 2.1.6 : Mon Nov 03 2003 - 09:00:13 EST