Dear Colleagues of the DCMI Education Community,
As a member of the IMS Global Learning Technical Advisory Board I am in
particular concerned with the development of application profiles of
specifications. As such I am following the development of the DCMI Education
Application Profile with great interest. I am far from trying to give you
any advice how to solve the issues you are discussing today. The only thing
I can do is to inform you how these issues are now seen by a number of IMS
members who have reflected about them before. I emphasize that the following
are strictly my personal views, though I believe they are very much in line
with current IMS activities on application profiling.
I think a number of my colleagues within IMS is sharing Andy’s concerns that
complex metadata systems have failed to gain the importance we all expected.
This is a major reason why there is actually no observable activity on
metadata specifications within IMS. Even IMS’ own metadata specification is
no longer pursued – instead IMS is using the IEEE LOM metadata
specification, profiling it where necessary. The reason for this is the
insight that the community is served best if there are as few overlapping
specifications in use as possible.
A second IMS experience, which relates to your discussion of description of
item vs. description of compound entities, is, that specifications have much
broader applicability than just describing certain entities, when they are
considered in context, in particular in their interrelation with other
specifications – metadata within content packages containing interactive
items whose behavior must be described using information about the learner
and about other roles in the educational process etc…
Taking both positions together led IMS to a new way of creating
specifications: Creating new specifications is avoided whenever possible.
Instead, relevant use cases are analyzed to determine which existing
specifications (from inside or outside IMS) are relevant. Then these
specifications are studied and profiled to capture what is the foreseen need
of the use cases and these application profiles are combined into what is
called a domain profile in IMS jargon. I think the process of domain
profiling is by now quite well understood. Apparently Stuart’s view of the
DC AP as a module for re-use goes into a similar direction.
In IMS we found it quite useful to let the specification efforts be informed
by some fairly concrete application models. It seems to help confining the
problems when one has to deliver within a restricted time frame something
that is useful for concrete applications. In fact, this has been made part
of the IMS specification process to the extent that new specifications will
be released to the public only if there is a specified number of
implementations available for demonstrating that the new specifications meet
the intended application models.
I wish you all success in your ongoing discussion which I shall continue to
follow.
Ingo
Ingo Dahn
University Koblenz-Landau
Knowledge Media Institute Koblenz
[log in to unmask]
-----Ursprüngliche Nachricht-----
Von: DCMI Education Community [mailto:[log in to unmask]] Im
Auftrag von Andy Powell
Gesendet: Montag, 13. August 2007 16:28
An: [log in to unmask]
Betreff: Re: DC-Education Application Profile: update + feedback requested
Note to self: stop mentioning FRBR! ;-)
Jon,
so what does history teach us? Why are we where we are now? I would argue
that the "effort aimed at distilling semantics & simplifying them through
delivering sufficient consensus across a significant community of practice"
essentially failed. It failed because the approaches reached thru that
consensus cost more to implement than the benefits they realise in the
context of the original use-case (resource discovery on the Web).
When was the last time you found something because it had been described
using DC?
What history tells us is that DC is too complex for the 'simple' resource
discovery scenarios envisaged when the initiative started. Those scenarios
now tend to be catered for by full-text indexing and social tagging of one
form or another. At the same time DC is not complex enough for the
scenarios typically found in digital libraries, scholarly communication,
elearning, commerce and the like.
Yes, the DCMI Abstract Model tends to move us more towards the latter. Yes,
explicitly modelling the entities in the world that we want to describe is
more complex than not doing so.
Complex but necessary. All IMHO of course.
Re your specific question: Why would or should the AP be limited to the item
level? Apologies, I perhaps wasn't clear enough. I was arguing that I
suspect that most people interpret the current DC Ed AP as applying to
resources at the item level (because it doesn't say anything about other
kinds of entities - learning objects as 'works' for example). I'm
suggesting that we need an AP that explicitly acknowledges that there are
different kinds of entities being described, and that provides an explicit
model of what those entities are and what properties we are going to use to
describe them.
Andy
--
Head of Development, Eduserv Foundation
http://www.eduserv.org.uk/foundation/
http://efoundations.typepad.com/
[log in to unmask]
+44 (0)1225 474319
> -----Original Message-----
> From: Jon Mason [mailto:[log in to unmask]]
> Sent: 13 August 2007 14:40
> To: Andy Powell; [log in to unmask]
> Subject: Re: DC-Education Application Profile: update +
> feedback requested
>
> Andy,
>
> I think history is an excellent teacher. It is really
> interesting to consider how the DCMI began -- through an
> effort aimed at distilling semantics & simplifying them
> through delivering sufficient consensus across a significant
> community of practice. Where are we at now? There's been lots
> of tweaks & enhancements since but also an incremental move
> toward getting to an over-arching abstract conceptual model
> that not only accommodates the earlier simplicity but is
> somehow more elegant in an Internet-architecture sense. And
> in this move there's the added recent effort to try &
> harmonise with a number of other abstractions on offer, that
> have been developed via different standardisation processes.
> Not that such efforts are not worthwhile; but I'm concerned
> about how real the conversation can be & to how many people.
>
> One of the hazards I now see emerging in the evolution of DC
> application profiles & within the DCMI more broadly is the
> pursuit of an abstract model that may never be sufficiently
> abstract & is somehow at odds with its roots. The problem
> emerging is that in trying to harmonise all the various
> abstractions on offer the conversation is becoming
> increasingly one that can involve fewer & fewer stakeholders
> (ie, if they want to participate in an informed way). I
> wonder how many people in the DC-Education know enough about
> FRBR to chime in? Of course, there'll be a few -- but I doubt
> there will be a critical mass. In our community there will
> always be newcomers (as well as a few jaded ones) trying to
> make sense of all this. I think one of the deeper problems
> here with pursuit of the abstract is that "one person's
> precision is another's blur".
>
> I also feel that the term "learning object" has now been
> appropriated in so many conflicting ways that it has almost
> been rendered meaningless. But I can tell you one thing for
> sure, the DC-Ed AP started out (in 1998!) as a way to specify
> those properties of a resource that are most meaningful in
> educational settings.
>
> I don't really understand why your abstractions limit the
> DC-Ed AP to a single-entity application. Maybe I just don't
> understand your interpretation of entity-relationship models.
> But the power of the original DC approach in my view was (&
> is) its capacity to apply in a generic (non-specific) sense
> -- ie, the object is determined by context; it could be an
> information resource, a collection, a course, anything
> really. Why would or should the AP be limited to the item
> level? What's an item in net space these days? And why
> should there have to be a multi-entity approach? Maybe
> sometimes; but please explain because I'm really beginning to
> wonder whether this is just scope creep.
>
>
> Jon
>
> At 07:10 PM 13/08/2007, Andy Powell wrote:
>
>
> As far as I recall, the current DC-Ed AP is a
> single-entity application
> profile - i.e. it can only be used to describe a single
> resource, a
> 'learning object'.
>
> I assume that the 'learning object' in this case is a
> FRBR 'item'?
> Whether or not I'm right, it would at least be sensible
> to be clear what
> is being described in terms of the FRBR model?
>
> My gut feeling is that any useful AP in this area needs to be
> multi-entity - covering (at least) the 'learning object' and the
> 'agents' associated with that learning object? So I
> think some work
> needs to be done on the entity-relationship model that
> underpins the DC
> Ed AP. This needs to be done before any more work is
> done on properties
> and other metadata terms.
>
> Andy
> --
> Head of Development, Eduserv Foundation
> http://www.eduserv.org.uk/foundation/
> <http://www.eduserv.org.uk/foundation/>
> http://efoundations.typepad.com/
> <http://efoundations.typepad.com/>
> [log in to unmask]
> +44 (0)1225 474319
>
> > -----Original Message-----
> > From: DCMI Education Community
> > [ mailto:[log in to unmask]
> <mailto:[log in to unmask]> ] On Behalf Of Sarah Currier
> > Sent: 12 August 2007 20:19
> > To: [log in to unmask]
> > Subject: DC-Education Application Profile: update +
> feedback requested
> >
> > Dear all,
> >
> > The Dublin Core Conference 2007 in Singapore is fast
> > approaching. The next face-to-face meeting of the DC-Ed
> > Community will take place there.
> >
> > Diane Hillmann, Stuart Sutton and I have been working hard to
> > get draft materials related to the proposed DC-Ed Application
> > Profile ready for this meeting. We would like to disseminate
> > the results of this work prior to the meeting so that anyone
> > who can't attend can give feedback and input. We'd be very
> > grateful if you could spend some time having a look and
> > feeding back to us. Discussion on this list is particularly
> > welcome. If you only have time to look at the areas that
> > particularly interest you, that's fine! I have tried to
> > delineate the different areas we'd like feedback on below, to
> > make it easier for you.
> >
> > We now have a wiki page linking to a new draft Application
> > Profile document:
> >
> http://dublincore.org/educationwiki/Working_20Draft_20of_20DC
> <http://dublincore.org/educationwiki/Working_20Draft_20of_20DC> _
> > 2dEd_20Application_20Profile
> >
> > This draft gives an overview of the background and purpose
> > of, and requirements for, the AP. It lists the metadata
> > elements for inclusion in the AP, along with IEEE LOM
> > correlations, and notes aiming to make a start on guidance
> > for use of the elements involved.
> >
> > It also includes a list of additional IEEE LOM elements
> > related to educational use of resources, and outlines a
> > workplan and community acceptance plan for taking the
> AP forward.
> >
> > Any feedback gratefully received. Here are some particular
> > areas you may like to look at:
> >
> > 1. Expression of the AP using the DCAM Description Set
> > Profile. A draft of this document is available at:
> >
> http://dublincore.org/architecturewiki/DescriptionSetProfile
> <http://dublincore.org/architecturewiki/DescriptionSetProfile> .
> > Volunteers to help with this would be very welcome.
> >
> >
> > 2. Best practice guidance notes for the properties covered
> > within the AP. NB: There are comments for many of the
> > properties in v0.3 of the AP, which are noted in the new
> > working draft- do these meet your needs / seem reasonable /
> > useful? Any further suggestions?
> >
> > 3. Conforming to educational achievement standards. Stuart
> > Sutton has done some valuable work on the use of conformsTo
> > as a refinement of the Relation property. NB LOM users:
> > conformsTo is not in the core DCMI vocabulary used in the LOM
> > Relation fields- it is a more recent refinement of this
> > element. The intention of conformsTo in the DC-Ed AP is to
> > support relating of educational resources to curricula,
> > competency frameworks, learning objectives and the like. See
> > Stuart's notes at
> >
> http://dublincore.org/educationwiki/Correlation_20Resource
> <http://dublincore.org/educationwiki/Correlation_20Resource%A0>
> > for ideas about how to express the closeness of fit of a
> > resource to a particular educational achievement standard.
> >
> > 4. Use of further IEEE LOM elements in the DC-Ed AP. The
> > working draft has a table listing relevant LOM elements and
> > giving information about them. Do any of these look like
> > properties you would wish to describe for educational
> > resources? Have we missed anything important?
> >
> > 5. Vocabularies for Type and Instructional Method. After
> > taking on board discussion and feedback on the DC-Ed
> > Community email list, and at the DC-themed meeting of the
> > JISC-CETIS Metadata SIG earlier this year, we now have an
> > updated wiki page summarising the approach of DC-Ed to
> > recommending vocabularies for the AP, including a page on the
> > four draft criteria for recommending vocabularies:
> > http://dublincore.org/educationwiki/Vocabularies
> <http://dublincore.org/educationwiki/Vocabularies>
> > < http://dublincore.org/educationwiki/Vocabularies
> <http://dublincore.org/educationwiki/Vocabularies> >
> >
> > We also have a document listing all of the vocabularies we
> > have so far identified, along with notes on how they meet the
> > four draft criteria discussed earlier. This is
> available here:
> > http://docs.google.com/Doc?id=dhbqfq9m_0f6mdc2
> <http://docs.google.com/Doc?id=dhbqfq9m_0f6mdc2 > (NB: It is
> > also linked to from the DC-Ed wiki vocabularies page).
> >
> > If you have a vocabulary that isn't listed, or if you have
> > further information on any of the vocabularies listed, please
> > let me know.
> >
> > Equally, if you have any further ideas on recommending
> > vocabularies in the AP, please let us know.
> >
> > 6. Workplan and Community Acceptance Plan. These are noted
> > near the end of the AP document, and are admittedly a bit
> > sketchy. If anyone is of a particularly administrative frame
> > of mind and has feedback/suggestions, we will be interested
> > to hear them.
> >
> > Many thanks all. Please disseminate this email to
> > individuals or fora you know of that may be interested, and
> > do encourage people to join the DC-Education list as well.
> >
> > Best wishes,
> > Sarah
> >
> >
> > --
> > Sarah Currier
> > Co-Moderator, Dublin Core Education Community
> >
> > Product Manager, Intrallect Ltd.
> > http://www.intrallect.com <http://www.intrallect.com/>
> >
> > 2nd Floor, Regent House
> > Blackness Road
> > Linlithgow
> > EH49 7HU
> > United Kingdom
> >
> > Tel: +44 870 234 3933 Mob: +44 (0)7980855801
> > E-mail: [log in to unmask]
> > --
> >
> > --
> > Sarah Currier
> > Product Manager, Intrallect Ltd.
> > http://www.intrallect.com <http://www.intrallect.com/>
> >
> > 2nd Floor, Regent House
> > Blackness Road
> > Linlithgow
> > EH49 7HU
> > United Kingdom
> >
> > Tel: +44 870 234 3933 Mob: +44 (0)7980855801
> > E-mail: [log in to unmask]
> > --
> >
>
>
|