- From: Frederick Hirsch <frederick.hirsch@nokia.com>
- Date: Fri, 20 Apr 2007 13:59:29 -0400
- To: ext Maryann Hondo <mhondo@us.ibm.com>
- Cc: Frederick Hirsch <frederick.hirsch@nokia.com>, public-ws-policy@w3.org
It seems to make sense to chose option a) so that this works for clients that are not aware of the EndOfLife assertion. The text in the proposal following the choice reads as if a) were chosen which is consistent with that choice. Thus propose > policy alternative offered should (!!The WG much choose which to > use). > a) not contain the EndOfLife policy assertion even with an > ignorable attribute. This is because an alternative with the > EndOfLife assertion with an ignorable attribute will only intersect > with a client operating in strict intersection mode, IF the client > also has an EndOfLife policy assertion. > b) contain the policy assertion type. If Company-X adds > theEndOfLife policy assertion even type to a subsequent > alternative, then requesters using strict mode will not understand > the assertion type and the alternative be replaced with "policy alternative offered should not contain the EndOfLife policy assertion even with an ignorable attribute. This is because an alternative with the EndOfLife assertion with an ignorable attribute will only intersect with a client operating in strict intersection mode, IF the client also has an EndOfLife policy assertion." regards, Frederick Frederick Hirsch Nokia On Apr 17, 2007, at 9:36 AM, ext Maryann Hondo wrote: > > All, > I worked with Dave O. on a revised proposal for this bug for which > I attach the following diff of section 3.8. > > Justification: The current text does not completely explain the > difference > between extending the policy framework itself, and using > extensibility points > to allow policy expression authors to version policy assertions. > > Target: Primer > > Proposal: Change text in section 3.8 as follows: > > > > > > > > <ws-policy-primer-section3.8text-diff-04016.doc>
Received on Friday, 20 April 2007 22:30:21 UTC