From: G. Liebisch (liebisch@iabg.de)
Date: Thu Mar 14 2002 - 00:27:46 EST
Ivy Hooks wrote
"... document any assumptions with each requirement, as part of the
rationale for the requirement. For example, "I assumed that only one person
would use the system at a time", "This requirement assumes that the
spacecraft will land in the ocean upon return", and anything else like
this."
I did consider this kind of statements as requirements that were not
confirmed sufficiently. This idea seems to be confirmed by the fact that
"many assumptions aren't compatible with other constraints/requirements" as
Andrew Gabb wrote in an earlier message. "Constraint" is another concept in
this sentence, that has to be considered.
Assumptions may be assigned to the system itself or they may be assigned to the context (see mails of Andrew Gabb, Don Mills and of M.S.Feather). And It seems, that "context" is used differently in these texts: purpose/reason here and environment there. Another observation by Andrew Gabb is, that "a number of initiatives are being pursued at the same time".
I must confess, that I have problems with the terminology. Could you agree
that:
1 there is a system and it's environment; the border may move
2 there are required characteristics of the system
3 there are assumed characteristics of the environment
4 Characteristics may be more or less dependable
5 Characteristics may be ultimately fixed or there may be alternatives
6 Characteristics may be constraints or not
Characterising statement in the above way could have some benefits: If a requirement is not dependable, you have to do sth about this requirement (not about the whole document). If there are alternatives, you may decide whether you can live with them. If the system border moves, you change the attribute of a statement that tells whether it pertains to the system or to the environment.
Goetz Liebisch
This archive was generated by hypermail 2.1.6 : Fri Feb 21 2003 - 11:30:39 EST