[annotator-dev] AnnotateIt.org Webpage Metadata Storage

[Randall Leeds]

On Wed, Feb 20, 2013 at 2:21 PM, Abel Corver <contact at abelcorver.com> wrote:
> Hi everyone,
>
> I have recently started using AnnotateIt.org - both the plugin and the API -
> and am amazed how well it works. I have two questions, but if you want
> background on my project, please see the P.S.
>
> (1) Is there currently an extension to "link" two comments - which might be
> on different webpages - together? Is anyone writing this extension, or would
> anyone be interested in collaborating to create one?

Not that I'm aware of. I am interested in collaborating on this,
though. I think it's a particularly challenging to think of good ways
to snip content across page transitions and then stitch them together
in some composition interface, but it's also one of the most obviously
powerful uses of annotation.

> (2) <AnnotateIt.org/api> already stores as metadata for every comment the
> website URL, but stores no information about the URL itself. For my project
> - if a webpage represents an article, documents, video, etc - I want to
> allow users to enter the author, title, published date, document type, etc.

The reference backend (https://github.com/okfn/annotator-store)
currently uses ElasticSearch for storage, which has no problem
ingesting arbitrary objects. A plugin to do what you're looking for
merely needs to hook into the "beforeAnnotationCreated" event and add
the appropriate fields to the annotation object. Something as simple
as "$(body).annotator().annotator('subscribe',
'beforeAnnotationCreated', function (annotation) { /* Add properties
to the annotation object here */ });" should do it! If you run into
trouble, follow up here on in #annotator on freenode IRC.