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

 


Help: OASIS Mailing Lists Help | MarkMail Help

rights-requirements message

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


Subject: Re: [rights-requirements] Parallel or Complimentary System


Thanks, Pete, for this posting last week. Please accept my apologies for not
responding sooner (spontaneous travel to UK, etc, etc)...

Pete wrote:
> per our conference call this morning, let me frame this notion.
> I am not going to get hung up terms but try to express the concept
> of inherent rights versus granted rights and how we might resolve
> a means to provide technology that can enable each of these. Call
> it fair use call it rights granted by the US constitution or by
> EU directive... The notion is:

JSE: The notion of "inherent rights" *may* include one's "role," as Pete
suggests, but it will also include (as I suggested in another email) the context
and attributes of use. Indeed, in certain scenarios, "role" might indeed *not*
be a parameter for consideration.

I can certainly see why one might consider role as a simplification --- a number
of years ago I proposed just such an approach --- but I'm suggesting here that
"role-based fair use" is an *over-simpliciation*.

> ASSUMPTIONS.
> 1. Unless there is a continuous connection between provider
> and consumer, even if it exists it will be impractical to modify
> an expression on how a digital item can be used each time its
> "consumer" changes roles and when the expression traverses
> geographies as well.

JSE: Again, a fundamental problem with this is that Pete is assuming that "fair
use" can be adequately cast as, or reduced to, a role-based access control
problem --- for example, given an affiliation with a particular group, a
particular use becomes a "fair use."

He does bring to mind one of the difficulties pointed out in the Dan Burk/Julie
Cohen paper [Dan L. Burk and Julie E. Cohen, "Fair Use Infrastructure for
Copyright Management Systems," http://www.cfp2002.org/fairuse/burkcohen.pdf] ---
that of *spontaneity*. The risk that transactional approximations of fair use
have is that they fail the spontaneity test if the client isn't networked when a
decision needs to be made (and the applicable policies are not resident on the
machine). Thus we have to consider the notion of somehow capturing policies ---
policies expressed using an REL --- that adequately evaluate. These would of
course be based upon conditions (attributes describing context of use, etc) that
must somehow be gathered.

An interesting aspect of Pete's assumption is that, if he didn't assume a
role-based simplification, he wouldn't necessarily have this problem. Again,
this is the question of what the attributes of the usage are, not simply (or at
all) who the user is, even their role...

> 2. Roles are important when dealing with inherent rights. As expressed
> today the role of teacher vs consumer, production executive vs consumer,
> etc have inherent rights and for expressions to embody all possibilities is
> also not practical. Each of us changes roles several times each day from
> worker, to digital item creator, to consumer, to parent to.... and the list
> goes on

JSE: Role-based authorization is certainly important --- indeed, critical --- in
doing distributed "DRM" correctly; it is arguably a superior approach for DRM
applications serving the enterprise and for inter-organizational requirements,
especially within the research and education communities.

But, again, it is not clear that "role" is an acceptable surrogate for a
particular use, aspects or parameters of which might have been unanticipated
(which must be one of the entry points for making a "fair use" determination).

> POSSIBILITIES
>
> 1. define a complimentary set of expressions to those currently under
> consideration (or maybe we have what we need and its only the context that
> changes) and the behavior of an associated "enforcement" function that is
> bound to a person. Allow it to define roles that that person plays, such as
> children, parents, teacher, office manager, ticket agent etc... and define
> inherent rights associated with each role. This will vary some not only by
> role but by native geography.

JSE: First, again, simply going to "role based" is not the answer to fair use or
to other copyright limitations.

Second, if the role-based capabilities Pete describes are *not* present in the
REL in the first place, I would argue that it is a not sufficiently flexible
"access control language..."

> 2. The enforcement function would then be responsible for resolving
> rights/permission (whatever) expressions delivered with the digital item
> verses inherent rights and thus what can or cannot be done with a digital
> item.

JSE: This is an important notion. This also ties into the David Parrott document
that was sent to the w3c-drm list
two weeks ago [David Parrott, "When is a Right not a Right: When it is a
Permission"], esp. the part in which Dave describes a possible implementation of
the policy-based enforcement mechanism I was talking about in my 2001 Dlib
article [John Erickson, "Fine-grained Policy Enforcement for Digital Information
Objects"].

| 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



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


Powered by eList eXpress LLC