OASIS Mailing List ArchivesView the OASIS mailing list archive below
or browse/search using MarkMail.


Help: OASIS Mailing Lists Help | MarkMail Help

xacml message

[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]

Subject: Minutes of OASIS XACML TC meeting: 17-Jul-08

Minutes of OASIS XACML TC meeting: 17-Jul-08

Time: 10:00 am EDT
Tel: 512-225-3050 Access Code: 65998

Proposed Agenda:

10:00 - 10:05 Roll Call & Minutes Approval
 Vote on Minutes from 3 July TC Meeting (Corrected)

Roll call:

    Rich Levinson
    Seth Proctor
    Anil Saldhana
    Erik Rissanen
    Tony Nadalin
    Bill Parducci
	6 of 8 voting is quorum

    Anil Tappetla
    Duane DeCouteau (gains voting status)

    Jamie Clark

Minutes approval:

    Corrected minutes of prev mtg (7/3) Approved.

10:05 - 11:00 Issues
 Attribute Designator Parameters

    Anil T. plethora of attrs - fcns used by CH vs PDP.
    Erik: doesn't think this level of detail in PDP, rather
	in the CH part of env
    Anil: concern about large number of config type attrs
    Seth: agrees context handler PDP - retrieve attrs different 
	ways; arg for config details as part of policy, if can't
	config ch w attr; interested in real world use case
    Anil: can provide examples; possible
    Seth: agrees w Erik that this is ch config info, not to be
	in policy; why config something for individual attr?
    Anil: ch config is separate issue
    Seth: ch has to get parameter, where does it get helper info?
    Anil: it is more than just ch issue
    Seth: never come across use case why config for ch must be
	part of policy
    Anil: example sent out
    Defer discussion to next meeting get more input

 #88 General Xpath functions

    Erik: proposed by Craig - new fcns - gave up on export/import,
	now proposing (should be 3.0) adding; couple open issues
	do we need uri variance? should substring be able to 
	extract to end of string? case conversion of international
	chars not easy
    Bill: URI handling different from strings, regex matches etc.
	- also value in neg index, chomping from back of string,
	back of string is often valuable for access ctl;
	no idea how to approach case sensitivity;
    Erik: case conv complex;
    Bill: maybe case conv too much;
    Bill: edit spec w these additions,
    Rich: what is xpath story on 3.0
    Erik: this is just new fcns
    Bill: more broad read
    Bill: Erik will repost proposal; impls have to figure out
	the case issue
    Bill: maybe include 3 digit lang code? 
    Erik: maybe locate parameter added to string
    Seth: Java used to be a little confused, but now locale-based
	and can optionally specify locale.
    Rich: maybe this is metadata similar to Anil's prev issue
	on "helper parameters"
    Seth: maybe default for whole policy; but this metadata
	is how to read strings as opposed to prev issue that
	was more explicit
    Bill: is there problem raising locale to policy level attr?
    Erik: diff attrs from diff locales
    Bill: Erik will propose

 Duration Data Types

    Bill: this issue derived from something Seth originally
    Seth: maybe
    Bill: could handle "today - 21 years > subj.birthday" type
    Erik: concerns that xacml incl data types that are restrictions
	on data types, but does not incl more general type; inclined
	to keep as is.
    Erik: why not ymd instead of just ym
    Bill: Seth's original note in 2003 - points out may or may not
	solve other needs but didn't have any then; so probably
	ok w what we have

 XACML Typos/ ipAddress, dnsName functions

    Erik: list of fcn identifiers in conformance section missing
	items that are in text that defines fcns themselves-
	probably typo; just add the "missing" identifier
    Erik: other part - do we need set fcns for ip addr and other?
    Bill: is regex match sufficient
    Erik: 2 use cases: policy want to spec ip addr ok w regex
  	but do we want intersection of tags of ip addrs?
	Probably not? scope?
    Erik: just update missing identifiers but don't deal w 
	missing qualifiers

 <Target> Optimization

    Erik: target has multiple levels; if one is indeterminate then
	whole is indeterminate; so if one element is indeterminate
	then can stop there rather than processing the rest;
    Erik: spec has break out possibility for match, but not for
	no match; 
    Bill: revisit when Hal gets back; went around w Polar on this,
	but maybe some issue remains - defer.

Bill: next meeting in 2 weeks: 31-Jul-08

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:

[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]