W3C home > Mailing lists > Public > public-openannotation@w3.org > November 2012

Re: Body Description vs Title (RE: Textual Note example)

From: Robert Sanderson <azaroth42@gmail.com>
Date: Wed, 7 Nov 2012 08:52:14 -0700
Message-ID: <CABevsUHGgAVj9cS28mf9_gP3uC+J1cWHKby-1qCBRFb5XiKuBA@mail.gmail.com>
To: Leyla Jael García Castro <leylajael@gmail.com>
Cc: vladimir.alexiev@ontotext.com, public-openannotation <public-openannotation@w3.org>
Hi Vladimir, Leyla, all,

The Open Annotation model is based on web architecture, and hence the body
must be a resource not just a string.  ContentAsText is used to embed the
resource within the annotation document, but it's still a resource in its
own right.  If there wasn't a ContentAsText, then systems would try to
dereference the resource's URI to obtain the representation.

For multiple bodies, each resource that is the object of oa:hasBody is
individually related to each target. That doesn't seem to fit the use case
that Vladimir was describing where there's simply a resource that has a
title.

There are two possible models, in my opinion:

1.  The body has a title

_:x a oa:Annotation ;
  oa:hasBody _:b ;
  oa:hasTarget <someuri> .

_:b a cnt:ContentAsText ;
  cnt:chars "The description lives here" ;
  dc:format "text/plain" ;
  dc:title "Title of this resource" .

2.  The body includes the title
(same annotation)

_:b a cnt:ContentAsText ;
  cnt:chars "<dc:record><dc:title>Title of this
resource</dc:title><dc:description>The description lives
here</dc:description></dc:record>" ;
  dc:format "application/xml" .

And of the two, the first seems much more friendly to systems that don't
need to understand the title.

Rob




On Wed, Nov 7, 2012 at 4:34 AM, Leyla Jael García Castro <
leylajael@gmail.com> wrote:

> Hi Vladimir, Rob,
>
> Is it not possible to use multiple bodies here?
>
> Leyla
>
>
> On Wed, Nov 7, 2012 at 10:38 AM, Vladimir Alexiev <
> vladimir.alexiev@ontotext.com> wrote:
>
>> I could, but then why not use dc:description for the other field?
>> Why bother with cnt:ContentAsText at all?
>>
>> > From: Robert Sanderson [mailto:azaroth42@gmail.com]
>> > Sent: Tuesday, October 16, 2012 5:44 PM
>> > Can you simply attach a dc:title property to either the Annotation or
>> > the Body resource?
>> >
>> > eg:
>> >
>> > _:anno a oa:Annotation ;
>> >   oa:hasBody _:body1 ;
>> >   oa:hasTarget <some-uri> ;
>> >   dc:title "Title of the Annotation" .
>> >
>> > or:
>> >
>> > _:body1 a cnt:ContentAsText ;
>> >   cnt:chars "Here is the content of the body" ;
>> >   dc:title "Title for the Body" .
>>
>> > On Tue, Oct 16, 2012 at 12:23 AM, Vladimir Alexiev
>> > <vladimir.alexiev@ontotext.com> wrote:
>> > > ResearchSpace project has a need for Bookmarks with both Description
>> *and* Title
>> > > The current spec http://openannotation.org/spec/core/#InlineBodyusing ContentAsText allows only one string.
>>
>>
>>
>
Received on Wednesday, 7 November 2012 15:52:46 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Wednesday, 7 November 2012 15:52:47 GMT