RE: [office] Not specific to this attribute, example only

["Dennis E. Hamilton" <dennis.hamilton@acm.org>]

+1

I would say this qualifies as a defect and should be treated as such.  

It might be that these are (producer) recommendations to consumers that
blah-blah-blah, and in that sense, if they say should (or recommended or
recommendation to avoid normative language), we might also need to say
whether the consumer behavior is implementation-defined or
implementation-dependent.  I suspect the latter, since we provide a blanker
permission that Conformant Document Consumer support of features is elective
(though presumably, if supported, conformance provisions kick in). 

 - Dennis

-----Original Message-----
From: robert_weir@us.ibm.com [mailto:robert_weir@us.ibm.com] 
Sent: Tuesday, August 03, 2010 07:50
To: Patrick Durusau
Cc: ODF office
Subject: Re: [office] Not specific to this attribute, example only

Patrick Durusau <patrick@durusau.net> wrote on 08/03/2010 09:51:52 AM:

[ ... ]
> My question is that in a future revision, should attribute settings 
> be advisory? 
> 

What is the alternative?

"disabled: texture filtering shall not be enabled."

But this is not really testable unless the semantics of "texture 
filtering" is fully defined.

or we could say:

"disabled: texture filtering is disabled." 

But that is just tautological.

In other words being more precise about the force of the clause (mandatory 
versus optional behavior) is a secondary issue compared to developing a 
testable description of the semantics.

-Rob

---------------------------------------------------------------------
To unsubscribe from this mail list, you must leave the OASIS TC that
generates this mail.  Follow this link to all your TCs in OASIS at:
https://www.oasis-open.org/apps/org/workgroup/portal/my_workgroups.php