W3C home > Mailing lists > Public > public-ws-policy@w3.org > August 2006

RE: policy framework questions

From: Yalcinalp, Umit <umit.yalcinalp@sap.com>
Date: Tue, 22 Aug 2006 15:52:15 -0700
Message-ID: <2BA6015847F82645A9BB31C7F9D6416501F8F3AD@uspale20.pal.sap.corp>
To: "Anthony Nadalin" <drsecure@us.ibm.com>, "Frederick Hirsch" <frederick.hirsch@nokia.com>
Cc: "Hirsch Frederick" <frederick.hirsch@nokia.com>, <public-ws-policy@w3.org>, <public-ws-policy-request@w3.org>
I do not get why you think 1 a is false. 
It seemed to me the question is whether the normalization strips all the
alternatives or not even if they are nested assertions with alternatives
in the non-normalized form. That is how I am reading Frederic's question
for 1 a. After the normalization step, there are no alternatives left in
the nested assertions even if the nesting is preserved. I believe this
was also part of the debate originally in one of Prasad's questions. 
What am I missing? 


	From: public-ws-policy-request@w3.org
[mailto:public-ws-policy-request@w3.org] On Behalf Of Anthony Nadalin
	Sent: Monday, Aug 21, 2006 7:13 PM
	To: Frederick Hirsch
	Cc: Hirsch Frederick; public-ws-policy@w3.org;
	Subject: Re: policy framework questions

	1) False, as assertions can be optional and thus no need to make
assertions mandatory (assertions may be domain dependant and have
different processing then the framework) not sure what you are gaining
here, if you state that a wsp:Policy can't be empty this also solve the
issue with less processing. What does it mean to have an empty assertion
	1a) False
	2) False, see #1
	3) False, as assertions are extensible
	4) True
	Anthony Nadalin | Work 512.838.0085 | Cell 512.289.4122
	 Frederick Hirsch <frederick.hirsch@nokia.com>

				Frederick Hirsch
				Sent by: public-ws-policy-request@w3.org

				08/21/2006 04:05 PM





Hirsch Frederick <frederick.hirsch@nokia.com>	


policy framework questions	

	Is there any disagreement with all of these statements being
	1) If a policy assertion is defined by the domain author to
	one or more nested assertions then it MUST contain a wsp:Policy,

	wsp:ExactlyOne or wsp:All element as a direct child (True)
	1a) In normal form any assertion that contains a nested
	will only have one nested assertion alternative (True)
	2) If a policy assertion is defined by the domain author to
	one or more nested assertions any instance of the assertion need
	contain any of these actual assertions. Their use may be defined
	optional by the domain assertion author. However, one of the  
	wsp:Policy, ExactlyOne or All child elements must always be
	but may have empty content (True).
	3) An assertion that is not defined by the domain author to
	nested assertions may never have a wsp:Policy,ExactlyOne or All

	child. (True)
	4) It is meaningless to compare <a /> and <a><Policy /> since
	case should never occur, since one of the two forms would be in
	when 'a' is a specific assertion (True)
	This would lead us to close issue 3548 with no action needed.
	regards, Frederick
	Frederick Hirsch
	[1] <http://www.w3.org/Bugs/Public/show_bug.cgi?id=3548>

(image/gif attachment: graycol.gif)

(image/gif attachment: ecblank.gif)

Received on Tuesday, 22 August 2006 22:49:39 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 20:38:27 UTC