I think that RDFa provides the lowest barrier to entry. Using dcterms for
publisher, creator, title, etc. is a good place to start, and if your
collection (archival, library, museum) links to terms defined in LOD
vocabulary systems (LCSH, Getty, LCNAF, whatever), output these URIs in the
HTML interface and tag them in RDFa in such a way that they are
semantically meaningful, e.g., <a href="http://vocab.getty.edu/aat/300028569"
rel="dcterms:format">manuscripts (document genre)</a>

It would be great if content management systems supported RDFa right out of
the box, and perhaps they are all moving in this direction. But you don't
need a content management system to do this. If you generate static HTML
files for your finding aids from EAD files using XSLT, you can tweak your
XSLT output to handle RDFa.

Ethan


On Thu, Mar 6, 2014 at 12:56 PM, Eric Lease Morgan <[log in to unmask]> wrote:

> Let me ask a more direct question. If participating in linked data is a
> “good thing”, then how do you — anybody here — suggest archivists (or
> librarians or museum curators) do that starting today? —Eric Morgan
>