Re: Annotations and the Graph

Hi, Randall–

I do agree with you that RDF (and probably other formats, with various 
tweaks) provide a way to describe an annotation using generic 
capabilities, as you suggest.

There's nothing new about annotations, about RDF, or about describing 
links on the Web.

Yet, in the 25 years since the Web was created, despite there being 
numerous annotation projects or annotation capabilities in various 
formats, there has not emerged a popular, widely-agreed way to 
characterize and exchange annotations. Annotations have not yet reached 
critical mass as a method of exchanging targeted comments in all this 
time, despite there being significant benefits from them over similar 
structures (like traditional comments, or referential posts).

The working hypothesis (no pun intended) of this group is that we can 
help annotations reach this critical mass by breaking the problem space 
down into several functional components that work independently or 
together to provide a framework for annotations. These components 
include: an interchange model with various serializations; protocols for 
publishing, reading, searching, exchanging, and federating annotations; 
one or more robust anchoring mechanisms; and client-side (DOM) APIs and 
events.

The interchange format (e.g. the Data Model) is an important part of 
that, because it distinguishes annotations as a distinct Kind of Thing, 
a specific data abstraction with specific functionality (including 
targets and bodies) that lends itself to particular modes of 
presentation (including an expectation of contextual display with the 
target content). We could define it simply as a generic RDF graph, but 
that doesn't serve the goal of helping people think of annotations as a 
specific approach to a common problem (which is the decontextualization 
and centralization of user-generated content, among other things).

Further, if all we did was to define an RDF graph, that is unlikely to 
capture the interest or needs of the JavaScript-centric community, which 
would be a pretty large omission.

As an intellectual exercise, and an attempt to define the very minimum 
we need to define an annotation, this thread is interesting, but I 
personally think we need more than that if we're going to clearly 
communicate our vision for how annotations can improve the Web.

Regards–
–Doug


On 10/27/15 7:56 PM, Randall Leeds wrote:
> On Tue, Oct 27, 2015 at 4:41 PM Robert Sanderson wrote:
>
>     And finally ... we already have rdf:Statement (and the
>     recommendations against its use) and named graphs. If an Annotation
>     were restricted to just a single statement, I'm not sure that we
>     would need a new specification :)
>
>
> Thanks for making my point for me.
>
> SpecificResource and the selector vocabulary is great and I don't see
> anything that exists quite like that.
>
> But we have mechanisms to distribute statements with attribution. I
> don't understand what the model adds to that other than, it seems, a way
> to obfuscate the semantics and avoid actually creating triples that
> relate the body and target.
>
> In a sense, it seems to me like the purpose of much of the model is to
> escape from having to model that which we want to model.

Received on Tuesday, 3 November 2015 15:58:49 UTC