W3C home > Mailing lists > Public > www-qa@w3.org > December 2001

[www-qa] Re: Conformance Requirements Guideline

From: <david_marston@us.ibm.com>
Date: Tue, 4 Dec 2001 14:23:02 -0500
To: www-qa@w3.org
Message-ID: <OFDC00D676.9634553A-ON85256B18.00654AC7@lotus.com>
Taking Section 3 and Section 7 together, a specification is required to
have certain conformance verbiage and "should" have it in the form of a
conformance clause. I would like to upgrade the verbiage about clear
identifiability to SHALL level. Section 3 would thus begin:
In order to conform to this Conformance Requirements document, a
specification shall contain a conformance clause. The location of the
conformance clause shall be clearly identifiable from the table of
contents and any relevant index. The conformance clause should exist as
a separate section within the specification, so that the reader can find
all conformance provisions from a single starting point. A specification
that conforms to this document shall:....
Having done that, Section 7 should not say that "the conformance clause
is a section of a specification that..." but rather that it's a part of
the spec or a collection of parts and that there should be one place
where all such parts are rolled up into a total set of conformance
provisions, possibly using links to dispersed provisions (e.g., for
profiles or levels).

I think we need to add a subsection 8.1.3 that mentions equivalence
standards. In the XML world, we have the InfoSet Recommendation as a
standard for equivalence of two XML documents. Any specification for a
processor that emits XML as an output can define conformance by saying
that the XML emitted must have an InfoSet equivalent to the InfoSet of
the XML in the test assertion. By so stating, the specification cites
as normative the XML Information Set Recommendation. Encouraging the
use and reuse of equivalence standards will encourage the development
of convenient tools for measuring equivalence.
.................David Marston
Received on Tuesday, 4 December 2001 14:25:24 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 20:40:29 UTC