Re: meta tag or EARL in page

On Wed, 30 Oct 2002, Charles McCathieNevile wrote:

> For people who have control over a document, and are interested in publishing
> the EARL metadata, the HTML link element is great.

Yep.

> But a couple of other possiblities arise.
> 
> People may not want to publish externally their internal assessment data. In
> this case they are unlikely to point out where it is in a document they serve
> publicly.

Well, that's only half the story.  They could publish internal data with
appropriate access restrictions, regardless of whether a URL to that
data finds its way onto a public page.

> Also, third parties may be interested in rating the accessibility (or other
> conformance characteristics) of documents or things with URIs.
> 
> In each of these cases something like annotea would be useful - using the
> concept of a ratings bureau that was introduced for PICS as a way of looking
> for RDF information abd a document. The annotea infrastructure allows for an
> annotea server to require authentication before giving out results, or for
> making results public, which would support the two use cases listed above.

That's another approach, but I don't think it really adds much value.
It adds an accessibility bar on the client side (which systems such as
HyperNews don't), and it is only effective within a closed community
who can all use the same Annotea database.

(talking of which, I wonder whether Annotea might benefit from pinching
anything from Hyper-G to work over multiple collaborating servers?)

> In general it isn't a problem for someone who produces metadata to find it -
> even if they don't have any control over the original document their data is
> about.

Indeed.  I think the main use for this within an organisation is in the
context of a QA process.  If you require every page to have a
<link rel="accessibility" type="application/rdf+xml" href="..."> -
and perhaps integrate this with a CMS, you can enforce Good Practice.

-- 
Nick Kew

Received on Thursday, 31 October 2002 19:42:32 UTC