Re: digital-copyright Digest 2 Jul 2002 15:00:00 -0000 Issue 24

["John Erickson" <john_erickson@xxxxxxxxxxxxxxx>]

Patrick Durusau <pdurusau@xxxxxxxxx> wrote:

>>Date: Mon, 01 Jul 2002 07:51:55 -0400
>>From: Patrick Durusau <pdurusau@xxxxxxxxx>
>> :
>> :
>> The interests of the academic community in issues such as "fair use"
>> and allowing free (or at least non-commercial) use of texts and
>> research will not be well served by a standard that protects the
>> commercial rights in the "Lion King" and similar artifacts. Our
>> requirements are different and any standard for DRM should not
>> attempt a one size fits all solution. I am sure that the TC would
>> welcome academic input that would lead to a more nuanced standard
>> that meets a wide range of needs, one of the hallmarks of a
>> successful standard...

JSE: Please note that there are several of us on the TC who appreciate this and
are working diligently to ensure that a whole variety of non-content-industry,
non-technology-industry interests are heard and their requirements considered.
Indeed, how to accommodate concepts such as fair use are precisely the types of
things that a rights expression standard should take on --- expressing the other
stuff is easy.

The stated timetime is very much part of the concern, and some of us have been
working to ensure that a more reasonable period for collection and analysis of
requirements is established.

Those of us on the Requirements SC welcome and encourage the contributions and
concerns of the research and library community, the accessiblity community, and
any organizations that are working in the interest of end-users/consumers.

>>Note that a DRM standard will eventually find its way into
>>hardware/software and it will be too late to complain at that point
>>that it does not meet the needs of the academic community.

JSE: This is why standardization in the rights expression space is a critical
first step. The last thing we need is a world of policy-enforcement
architectures in which the policies are implicit (or "hardwired"). A preferred
architecture is one in which policies can be flexibly expressed by anyone with
an interest in controlling the behavior of information objects (and services).
The underlying platform should provides for the deterministic, distributed
enforcement of *applicable* policies, be they defined by the originator or the
end-user...

Domain--specific policy languages like XrML or ODRL start to do this, along one
particular vector. More general access control languages like XACML model other
concepts quite well. Ultimately, we will need a semantic combination of these,
as well as the underlying trust infrastructure that ensures that the policies
get applied in a deterministic manner.

Have a GREAT one!

| 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@xxxxxxxxxx         AIM/YIM/MSN: olyerickson