W3C home > Mailing lists > Public > www-qa-wg@w3.org > September 2003

Re: SpecGL Test Assertions

From: <david_marston@us.ibm.com>
Date: Fri, 5 Sep 2003 22:52:57 -0400
To: www-qa-wg@w3.org
Message-ID: <OF53D488BA.14ECA2B8-ON85256D99.000E3D1E-85256D99.000FDE31@lotus.com>

Mark Skall wrote:
>One open question still remains. Ckpoint 10.1 asks for test assertions
>to be provided, but says nothing else about them.  In my test assertion
>for SpecGL I asked for the test assertions to cover all the requirements
>in the specification.  Lynne seems to think that SpecGL may have
>purposely avoided doing that but couldn't remember why....

I don't remember why, but this is an example of one of the steps in my
famous table of QA levels. Just like the difference between having a
suite of tests for some of the assertions and the higher level of a test
suite that meets some standard of completeness against the assertions,
having assertions that meet a standard of completeness is a higher level
than just having some assertions.

Historical note: I composed the original table [1] on the way home from
the QA Workshop, as I thought over all the different expectation levels
that were brought up during the discussion. Notice for example that all
levels 5 through 10 talk about the set of tests not being complete, but
the completeness of the list of needed cases improves as you go upward
through that range. Since the list of needed cases derives from the set
of test assertions (there can be more cases than assertions), you can
picture how the set of assertions follows the same progression.
.................David Marston


[1] http://lists.w3.org/Archives/Public/www-qa/2001Apr/0004.html
Received on Friday, 5 September 2003 22:53:30 GMT

This archive was generated by hypermail 2.2.0 + w3c-0.30 : Thursday, 9 June 2005 12:13:14 GMT