Folks:
We may have something soon that will help...
The original design goal of the NSDL Metadata Registry was to provide a
place to register, as well as manage Application Profiles (DSPs). The
ultimate goal of the project is to be able to provide a DSP management
interface that will allow the expression of the DSP as both an XML
schema for data validation and an OWL Ontology to express semantics.
We recognized that in order for this to work we also needed to register
the building blocks, so we started with registering/managing
SKOS-described value vocabularies and have been finishing up
managing/registering Metadata Schema (Element Sets). We've been working
with ALA Publishing to provide a limited version of the DSP management
tool for use in their RDA Online product and expect to be able to have
at least an initial version of this added to the Metadata Registry by
sometime in August.
If you're interested in participating with us in this project please
contact either of us.
Jon Phipps: <[log in to unmask]>
Diane Hillmann <[log in to unmask]>
> On 27/6/09 18:31, Karen Coyle wrote:
>> Mikael considered it a draft. Perhaps it just needs to be tested (as I
>> said, I ran into problems with it, which haven't yet been worked
>> through). Also, if there are other serializations, they should also be
>> linked to the specification.
>>
>> And, as I have found, a DSP in XML isn't directly usable for folks who
>> are wanting to implement... It may be that we have to create another
>> layer that translates easily from DSP to application. This has NOT been
>> the practice in the DCMI community from what I can tell. Instead, there
>> has been tendency to steer clear of implementation details. But the
>> world needs to implement. If there is a desire to keep underlying
>> standards and implementation separate, perhaps we can share
>> implementations on an area of the DC site. That doesn't, of course,
>> address issues that may arise between the standards and implementation,
>> but I for one am willing to move forward and ad lib a bit in order to
>> move things forward. We can address conflicts (or alleged conflicts)
>> over time, without having them hold up some interesting new
>> developments.
> Dan Brickley wrote:
> From my perspective, DCMI is fundamentally a community with common
> goals and interests, rather than a generator of dry, abstract
> standards. It happens that most of the work historically done here has
> been on making standards, but making tools is equally important. If
> DCMI can host that kind of collaboration (especially tools that track,
> implement and improve the standards), so much the better! The recent
> focus at DC on describing application profiles, rather than on
> defining new metadata terms, shows there is still plenty of
> flexibility and responsiveness in the organization. If people have the
> time and energy and inclination to collaboration on tools here, that'd
> be great...
>
> cheers,
>
> Dan
>
|