W3C home > Mailing lists > Public > www-qa-wg@w3.org > September 2005

Re: Test Metadata Document

From: Tim Boland <frederick.boland@nist.gov>
Date: Mon, 12 Sep 2005 09:07:43 -0400
Message-Id: <5.1.1.5.2.20050912084441.020213b0@mailserver.nist.gov>
To: www-qa-wg@w3.org

At the beginning of the document ("Abstract"), there are the
terms "test case", "test suite", and "test(s)".    What is the
difference between a "test case" and a "test"?   In TestFAQ
the term used is "test" not "test case".   I think it's good
not to use additional terms for similar concepts unless warranted..

Is metadata only useful for conformance or also for interoperability
(that is, why not just refer to test suites in general - drop "conformance"
word in front of test suite)?   I don't see where the word adds any
value unless I'm missing something, and in TestFAQ the term used
is "test suite" without qualifier in front..   If the term "test suite" is
used without qualifier then generality is added, if that is appropriate..

  Do we need definitions of "test", "test case", "test suite" upon first usage
in "Abstract"?

What objective evidence is there that "defining and providing metadata
has proved helpful in a variety of instances.."?    In other words, are 
there some
examples we could reference of the usefulness of metadata?   Is the XML Query
example cited supporting TestFAQ or the use of  metadata (just says "these
principles"..)?

The text in "Abstract" and that at the beginning of "Introduction" is 
duplicative..

Thanks and best wishes..
Tim Boland NIST
Received on Monday, 12 September 2005 13:08:38 GMT

This archive was generated by hypermail 2.2.0 + w3c-0.30 : Monday, 12 September 2005 13:08:38 GMT