OASIS eXtensible Access Control Markup Language (XACML) TC

  • 1.  Proposed Agenda for 17 November TC Meeting

    Posted 11-16-2011 22:42
    Time: 13:00 EDT (GMT-0400) Tel: 513-241-0892 Access Code: 65998 Proposed Agenda for 17 November 2011 TC Meeting: I Roll Call & Minutes Approve Minutes: 3 November 2011 TC Meeting http://lists.oasis-open.org/archives/xacml/201111/msg00003.html II Administrivia RSA InterOp status - timing of prep XACML v3.0 Public Review feedback summary http://lists.oasis-open.org/archives/xacml/201111/msg00004.html III. Issues Intellectual Property Profile WD-05Comments & Discussion http://lists.oasis-open.org/archives/xacml/201111/msg00005.html XACML Implementation for UNIX File System (xacml-users) http://lists.oasis-open.org/archives/xacml-users/201111/msg00000.html


  • 2.  Draft Minutes for 17 November TC Meeting

    Posted 11-18-2011 14:57
    XACML TC November 17, 2011 Attendance Voting members The Boeing Company Crystal Hayes The Boeing Company Richard Hill Oracle Rich Levinson Oracle Hal Lockhart Axiomatics Erik Rissanen The Boeing Company John Tolbert Bell Helicopter Textron Inc. Paul Tyson Quorum was achieved > Approve Minutes: > 3 November 2011 TC Meeting > http://lists.oasis-open.org/archives/xacml/201111/msg00003.html Minutes approved unanimously > > II Administrivia > RSA InterOp status - timing of prep Hal stated that we are still in the process of organizing the Interop, but a meeting has not been scheduled yet. Some of the members reported receiving notice that the Interop mailing list had been activated. > > XACML v3.0 Public Review feedback summary > http://lists.oasis-open.org/archives/xacml/201111/msg00004.html All the comments during the recent Public review came from TC Admin. However an important comment was received just after the review concerning the degree to which the core spec should specify what is meant by URI equality. Erik pointed out that the comment covers only the case of URIs used as identifiers, defined by the TC or other profile authors. In these cases, the initial definition of the identifier should be such that no special processing, such as escape characters is required. (In other words, the identifiers should be defined and used only in canonical form.) In these cases, equality comparison can be done very simply by comparing code point by code point. Erik further pointed out that the more general case of comparing, for example two web URLs for equality is not only complex, but deployment-specific, because in different situations, different components of the URL may have to be compared or ignored. Erik recommended that we respond to the comment by adding wording to warn Profile authors that any URI identifiers that they define should be in canonical form. There was general discussion of the issue. It was agreed to take it to the list, with the objective of resolving the issue by the next call. > > III. Issues > Intellectual Property Profile > WD-05Comments & Discussion > http://lists.oasis-open.org/archives/xacml/201111/msg00005.html John announced that he had posted WD-05, but that apparently he had misunderstood the comments made previously, so in effect he was withdrawing it and had WD-06 nearly ready to go. There was general discussion of use of Boolean Attributes vs. enumerated values and the use of URIs (specifically URNs) rather than strings for enumerated attribute values. The result of the discussion was that John agreed to immediately post WD-06 so others could comment and mark it up. John promised to try to respond with a new draft in advance of the next call. > > XACML Implementation for UNIX File System (xacml-users) > > http://lists.oasis-open.org/archives/xacml-users/201111/msg00000.html There was a little discussion of this issue and disagreement as to whether it implied that there is something missing from XACML. Discussion will be continued on the list. Hal