Re: [rights] Simple question...

[John Erickson <john_erickson@hplb.hpl.hp.com>]

Carlisle asks:
> With the RL TC, from what I heard on yesterday's TC call,
> it seems that steps 1 and 2 have happened, and step 5 is
> happening.  My question is:  did steps 3 and 4 happen?  I
> have not been able to attend every call, and not all the
> meeting minutes are posted on the Web site, so I honestly
> don't know the answer.  Was there a TC meeting (concall or
> F2F) in which the requirements to be met in the 1.0 specification
> were voted on and approved?

JSE: No. If you think back to the beginning of the XACML process (which is the
only other OASIS activity that I have followed closely), you'll remember that
there were essentially two key founding "specs" --- the Univ. Milan work and the
IBM-Tokyo work (there might have been others as well, and certainly there was
the experiences of the members, but those are the two I knew about; I had been
following the two of them quite closely prior to XACML). The XAML group agreed
to essentially "drop the tranny" and create, through a concensus process, a work
product that would be far better than the submitted pieces.

To do this properly, the various constituencies had to agree that none of them
knew all that was needed. The XACML group employed a very deliberate use cases
and requirements-gathering process, adopted a requirements document, and with
this in hand started the REAL work. The group deliberated on the "policy model,"
which is the equivalent of the authorization algorithm of the XrML core spec,
and then moved on to developing a schema. As far as I understand, there were no
"implicit requirements" --- features haven't been inherited simply because they
were in one of the contributed pieces.

What we have seen in the RLTC case is something entirely different; an analog
would have been if the XACML group had launched the requirements-gathering
process as described, but also had said, "oh by the way, the Univ. Milan
submission will be version 1.0, and it won't reflect the open part of this
committee's work..."

There are innumerable, successful industry standards in place today that
required adopters to re-implement upon their introduction, typicially migrating
from an implementation based upon progenitor specifications to those standards.
There are often changes to features and even the loss of a few features, but
usually this is a SMALL price to pay for the greater capabilities that the
openly-authored, concensus-driven standards offer.

| John S. Erickson, Ph.D.
| Hewlett-Packard Laboratories
| PO Box 1158, Norwich, Vermont USA 05055
| 802-649-1683 (vox) 802-371-9796 (cell) 802-649-1695 (fax)
| john_erickson@hpl.hp.com         AIM/YIM/MSN: olyerickson