Post-conditions?

[stephen.green@systml.co.uk]

Reality/sanity check:

Are post-conditions actually different enough in a TA from a 'Test Effect'
to warrant their inclusion in the TA Model?

The logic that the post-condition has to be true or the test is inconclusive
seems now, in hindsight, a bit weak: if the post-condition              
                                has to be true then
why isn't it a part of the test and therefore part of the Test Effect? If not
part of the test, included in Test Effect then how does one know whether the
test is inconclusive?

It fits nicely with UML contracts but they have a different purpose - they
use the post-condition as near enough (it seems to me) the same as what
becomes a Test Effect.

Personally I'd be happy with their removal since we have Test Effect

-- 
Stephen Green

Partner
SystML, http://www.systml.co.uk
Tel: +44 (0) 117 9541606

http://www.biblegateway.com/passage/?search=matthew+22:37 .. and voice