Re: Tin Man Design Question 6 - to vote Jan. 9th?

[Carmelo Montanez <carmelo@nist.gov>]

Hi All:

    Although it may be trickier/harder to create reports and write a harness
with all
those sub elements in there, I guess it does makes sense for the reason
David
states below.  I will say lets go with it.

Carmelo
----- Original Message -----
From: <David_Marston@lotus.com>
To: <xslt-conformance@lists.oasis-open.org>
Sent: Thursday, January 04, 2001 3:16 PM
Subject: Tin Man Design Question 6 - to vote 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:
> 6. Is there any operational problem with having all
> discretionary items lumped under one <Discretionary> element?
>
> DISCUSSION:
> I'm sorry that this is such a dry question. Tony, Carmelo, or
> anyone else thinking about how the test harness will use the
> catalog data needs to consider whether it makes a difference.
> In Tin Man, it's proposed that there is zero or one element
> called Discretionary, which contains as many sub-elements as
> needed to represent all the discretionary items that affect
> the applicability of this test case. The alternative is to
> allow ~50 optional elements directly inside TestCase, each
> named for a discretionary item.
>
> This question is not about the naming scheme for the ~50
> items themselves. Keep in mind that the list is likely to
> expand when XSLT 1.1 and 2.0 come under our consideration,
> plus expansion based on supporting more languages in the
> test system.
>
> I advocated the encompassing Discretionary item because it
> helps in managing the data. It might also provide a filtering
> hook for the catalog as a whole.
>
> We could discuss this and vote only if there seem to be two
> schools of thought.
> .................David Marston
>
>