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] | [Elist Home]

Subject: [xacml] call for vote on re-definition of "successfully using"


It appears that we will not have at least three organizational
members willing to attest that they are "successfully using" the
XACML 1.0 Specification by 12/13/02.  This will prevent our
advancing the specification toward standardization for at least
another month.

I propose that we vote to redefine "successfully using" at our
meeting on 12/12/02.  Currently, by XACML TC definition,
"successfully using" means "having an implementation that passes
the XACML Conformance Test Suite".  My proposal is to drop our
TC-specific definition of "successfully using".  "Successfully
using" will mean simply that, in the attesting organizational
member's opinion, the organization is successfully using the
XACML 1.0 Specification.

I believe such a redefinition is reasonable.  My reasons follow.

A. No other OASIS TC has stated that "successfully using"
   requires an implementation.  The XACML requirement is
   self-imposed, not required by OASIS rules.

B. At least one member is unwilling to admit publicly to having
   an implementation for fear of exposing themselves to selective
   legal action by one or more purported IP holders.

C. At least one other member is unwilling to state, as required
   by the current OASIS IPR Policy (only where implementations
   are required)
   that they "have taken adequate steps to comply with any such
   rights, or claimed rights".

D. At least one other member has not implemented every feature of
   the language, although the implementation is well along.

You may not agree with the reasons stated in B and C, but the
fact is, there are at least two members who will not be attesting
this month for those reasons.

I think requiring implementations that pass conformance tests for
a specification to become a standard is a VERY good idea; so good
that I wrote the conformance tests myself.  However, my reasons
for requiring implementations have been satisfied without making
an implementation necessary for an attestation of "successfully

1) There are at least two full implementations of XACML, one by
   an organizational member (Sun), and one by an individual
   member.  Both pass the full Conformance Test Suite.

2) There several other implementations that are well advanced,
   but not yet complete enough to pass the full Conformance Test

2) There are many organizations that have carefully scrutinized
   the specification, as evidenced by the comments we have
   received during our public review period.  None of the
   commenters has stated that the specification is not
   implementable, especially not after we resolved the issues in
   the comments.

Anne Anderson
Anne H. Anderson             Email: Anne.Anderson@Sun.COM
Sun Microsystems Laboratories
1 Network Drive,UBUR02-311     Tel: 781/442-0928
Burlington, MA 01803-0902 USA  Fax: 781/442-1692

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

Powered by eList eXpress LLC