Re: LC-59

On Thu, 8 May 2003 david_marston@us.ibm.com wrote:

> >LC-59.6
> >"sufficient" is not testable (CP 7.2)
> >LC-59.7
> >'all' is not testable (CP 7.3)
>
> I think those words are testable. In the latter case, lawyers can give
> you a definition of "all issues are resolved" that should be suitable.

"All" is not testable, in general. "All identified" or "all listed"
is. "All identified issues are resolved" makes sense. "All issues are
resolved" makes sense only if the set of "issues" is well-defined,
known a priory, and finite. For example, "isolate all criminals" or
"fix all bugs" is not testable unless you define a "criminal" as a
"convicted person" (and not just "somebody who broke the law") and a
"bug" as a "registered problem report" (not just "existing
malfunction").

> The former case, in context, is a little more dangerous. The Ckp says:
> >Identify sufficient staff resources to meet the needs of any
> >transferred test materials.
> Which can only be fulfilled by a guarantee by the WG that they will
> perform as much work as is needed. WGs aren't really able to commit
> that way, either from their own members or the proverbial "staff
> resources", since members can drop off the WG. I think the real goal
> is to make a commitment that appears to be sufficient at the time it
> is made, and thus "appears to be" is the non-testable part, not
> "sufficient".

True. Just removing "sufficient" would probably have the same effect
and may sound better. (Obviously, we should expect the WG to identify
sufficient/appropriate resources when they are identifying resources,
but we cannot test that qualifier without knowing the future). We had
similar problems with other checkpoints (e.g., "define specification
scope"  is more-or-less testable while "specification scope MUST be
easy to find" is not).

Alex.

-- 
                            | HTTP performance - Web Polygraph benchmark
www.measurement-factory.com | HTTP compliance+ - Co-Advisor test suite
                            | all of the above - PolyBox appliance

Received on Friday, 9 May 2003 00:40:37 UTC