Re: [tag] Action Item AI-Jan-7-09-0001

[Stephen Green <stephen.green@documentengineeringservices.com>]

Thanks Jacques.

Anyone else have any rewording suggestions? This might be a last
call before next draft. I'll put together a draft with '0.9.9.9-in-progress-1'
so we can change it without going beyond 0.9.9.9, to allow us to vote
0.9.9.9 as a committee draft, committee spec and public review draft
in case we want the option of making the post public review draft
committee spec (hopefully there'll be one!) version 1.0 and we don't
want to number anything 0.9.9.9.1 (a bit long for a version name!).

If we want to start voting after next meeting we'll want 0.9.9.9-in-progress
out before the meeting, won't we? Then the vote might make it become
0.9.9.9 after the meeting (if we keep to plan) as committee draft, etc, etc.

2009/1/15 Durand, Jacques R. <JDurand@us.fujitsu.com>:
> Stephen:
>
> I think you are right. I just read more carefully the current (new) text
> in .998 , and I think that overall it is fixing the problem.
>
> Editorial suggestion: the two first sentences seem convoluted to me (the
> point they are making is not quite clear, at least to me).
>
> Suggest to reword:
>
> "The specification will often have a clause called a conformance clause1
> [CONF1][CONF2] which defines a conformance profile [VAR] for the
> normative statements from which the specification is comprised .Testing
> such conformance involves matching test results to specification
> normative statements for which there is directly or indirectly a
> conformance requirement."
>
> As:
>
> "The specification will often have one or more conformance clauses
> [CONF1][CONF2][Appendix A, Glossary] which define (one or more)
> conformance profiles or levels [VAR]. A set of test assertions may be
> associated with a conformance clause in order to define more precisely
> what conformance will entail."
>
> Other rewording might be OK too, just that these two first sentences
> (especially the second one  on testing) could use some rewording I
> think.
>
> Jacques
>
>
> -----Original Message-----
> From: stephengreenubl@gmail.com [mailto:stephengreenubl@gmail.com] On
> Behalf Of Stephen Green
> Sent: Thursday, January 15, 2009 12:19 PM
> To: Durand, Jacques R.
> Cc: TAG TC
> Subject: Re: [tag] Action Item AI-Jan-7-09-0001
>
> Jacques,
>
> I'm not sure. The wiki text is out of date and this text is already
> different in draft 0.9.9.8 which doesn't concentrate on fig 1. Does the
> 0.9.9.8 text still need replacing like this?
>
> Secondly, we did decide earlier against this amount of text to contrast
> TA with Conformance Clause and Test Case - deciding to put more into the
> glossary. Plus we added the diagram to clarify this. Is the addition of
> the text to accompany the diagram enough for this paragraph? That's what
> I tried to do in 0.9.9.8 - does this need replacing?
>
> Best regards
>
> Steve
>
> 2009/1/15 Durand, Jacques R. <JDurand@us.fujitsu.com>:
>> in wiki, under section 1.1, suggested replacement for the intro
>> paragraph that is illustrated by Fig 1:
>>
>> http://wiki.oasis-open.org/tag/TestAssertionGuidelines
>>
>> Suggested new text:
>>
>> The notion of test assertion is to be distinguished from the following
>
>> often related concepts:
>>
>> Conformance Clause: The specification will often have one or more
>> conformance clauses [CONF1][CONF2][Appendix A, Glossary] which define
>> generally what is required from an implementation to claim conformance
>
>> to the specification or to one of its profiles. Test assertions are
>> generally defined separately from conformance clauses. They may
>> however be associated with a conformance clause in order to define
>> more precisely what conformance will entail. For example, a set of
>> test assertions may be assigned to a conformance clause, or may even
>> be directly referred to by the conformance clause.
>>
>> Test Case: Test cases [Appendix A, Glossary] are derived from test
>> assertions, meaning they implement the related test expression(s),
>> thus providing an executable way to verify that an implementation
>> fulfills the normative feature addressed by a test assertion. Several
>> test cases may be derived from the same test assertion, e.g. to verify
>
>> fulfillment in different contexts. Conversely, a test case may be
>> derived from several test assertions
>>
>>
>>
>> Jacques
>
>
>
> --
> Stephen D. Green
>
> Document Engineering Services Ltd
>
>
>
> http://www.biblegateway.com/passage/?search=matthew+22:37 .. and voice
>



-- 
Stephen D. Green

Document Engineering Services Ltd



http://www.biblegateway.com/passage/?search=matthew+22:37 .. and voice