- From: Alex Rousskov <rousskov@measurement-factory.com>
- Date: Sat, 7 Sep 2002 22:50:37 -0600 (MDT)
- To: Mark Skall <mark.skall@nist.gov>
- cc: www-qa@w3.org
On Sat, 7 Sep 2002, Alex Rousskov wrote:
> On Sat, 7 Sep 2002, Mark Skall wrote:
>
> > Again, nowhere does this definition say "There is a finite
> > cost-effective process to enumerate all valid inputs." like Alex
> > seems to think it says.
>
> You need to re-read the thread more carefully. The enumeration
> problem was not with the original definition. It is related to the
> discussion of the definition proposed by Kirill Gavrylyuk. The
> original definition does not have the enumeration problem. It has a
> problem of marking all behavioral specs non-testable (assuming a
> what I would consider "expected by an average person" interpretation
> of the word "check" in the context in question).
I guess I need to change/clarify that statement. The original
definition DOES have an "enumeration" problem (it's just that the
problem has been discussed at length in the context of Kirill's
definition and used as the most obvious/simple example in my message).
Alex.
--
| HTTP performance - Web Polygraph benchmark
www.measurement-factory.com | HTTP compliance+ - Co-Advisor test suite
| all of the above - PolyBox appliance
Received on Sunday, 8 September 2002 00:50:39 UTC