Re: Tin Man Design Question 3 - to assign Jan. 9th?

[Carmelo Montanez <carmelo@nist.gov>]

Hi All:

    I think the test categories listed in Tin Man and in David's message
should fit
 quite well the rest of tin man model.  It also make sense to have
categories
 so when any reports are produced there is a sense or organization and
structure to the tests that is obvious and visible to the users.  Remember
also that
some people may be interested only in a subset of the tests and having them
catalogued will make it very easy for anyone to target a set of tests.

Will examined the other issues submitted by David later today and in the
morning.
Sorry for the lack of comments, these are issues relevant to the areas I am
focusing.

Carmelo

----- Original Message -----
From: <David_Marston@lotus.com>
To: <xslt-conformance@lists.oasis-open.org>
Sent: Thursday, January 04, 2001 3:14 PM
Subject: Tin Man Design Question 3 - to assign Jan. 9th?


> Here is one of the issues identified in my message titled
> "Test Case Markup & Cataloging, Tin Man edition" that I sent
> to this list on 12/1/2000:
> 3. Whether to have categories, and if so, what the categories
>    are (Carmelo's proposal or other)
>
> DISCUSSION:
> In the Tin Man proposal, there *is* a category attribute with
> a short list of categories....
> <!ATTLIST TestCase
>   Category ( XSLT-Structure | XSLT-Data-Model | XSLT-Template
>   | XSLT-Result-Tree | XSLT-Data-Manipulation
>   | XSLT-Extendability | XSLT-Output | XPath-Location-Path
>   | XPath-Expression | XPath-Core-Function
>   | XPath-Data-Model | Mixed ) #REQUIRED >
>
> IMHO, the only way we really care about categories is if we use
> them to divide up the submitted tests for our own analysis. For
> example, if each of us will take a bunch of test cases (pooled
> from all submitters) and review them alone, then report back to
> the committee, then it would make sense to have exactly nine
> categories and give one to each person. This brings up the big
> question of how we intend to review the tests. If someone has
> some other justification for having categories, speak up now or
> they might not be required. (I also haven't been able to think
> of a reason to have categories but make them optional.)
>
> I suspect that we can discuss and vote on whether to have a set
> of OASIS-defined categories, then assign follow-up work if we
> will have them. The follow-up could start with the above set.
> .................David Marston
>
>