Re: Clarification of as:content?

Kevin,

The definition I proposed and James merged (many thanks!) is:

The content or textual representation of the Object. Markup, including
references to visual elements such as images, MAY be included. The content
MAY be expressed using multiple language-tagged values.

See:
http://jasnell.github.io/w3c-socialwg-activitystreams/activitystreams-vocabulary/#dfn-content-term

Previously it had been specific about "natural langauge" and "HTML markup",
both of which would have ruled out SVG. This seems like a perfect example
of what would now be in scope that was otherwise unclear.

Rob

On Wed, Nov 4, 2015 at 1:10 PM, James M Snell <jasnell@us.ibm.com> wrote:

> Yes, it could, since SVG is a text based format.
>
> - James Snell
> jasnell@us.ibm.com
> Technology & Partnerships
> IBM Open Technologies Architect - Node.js, Mobile Backend & Cloud
> (559) 707-6331 (mobile)
>
>
>
> ----- Original message -----
> From: Kevin Marks <kevinmarks@gmail.com>
> To: James M Snell <jasnell@gmail.com>
> Cc: Robert Sanderson <azaroth42@gmail.com>, "public-socialweb@w3.org" <
> public-socialweb@w3.org>
> Subject: Re: Clarification of as:content?
> Date: Wed, Nov 4, 2015 1:09 PM
>
> I assume if the image is SVG then content would contain it directly?
>
> On Wed, Nov 4, 2015 at 8:31 AM, James M Snell <jasnell@gmail.com> wrote:
>
> This is one bit that is admittedly lacking in the current vocabulary
> definition and I appreciate the question on it.
>
> The fact of the matter is that `content` varies a bit based on the
> type of object. For a `Note` or `Article`, the `content` property
> contains the content of the resource itself. It's the actual markup
> text for the note/article.
>
> For an `Image` object, `content` is likely far less useful. In the use
> case you describe, using the `summary` field would be more
> appropriate:
>
> {
>   "@context": "http://www.w3.org/ns/activitystreams",
>   "@type": "Image",
>   "displayName": "Small Red Square",
>   "summary": "<p>This image is a small red square, for unknown use.</p>",
>   "height": 100,
>   "width": 100
> }
>
> Using `content` in this case wouldn't be appropriate because unless
> you've base64 or hex encoded it, there's no means of actually
> including the actual binary content of the image resource in the JSON
> serialization.
>
> In short: `content` == the actual content of the resource, while
> `summary` == the description or short summary of the content. I can
> play around with some alternative ways of describing that in the
> vocabulary document and I'm always open to suggestions on how to
> improve the text :-)
>
> - James
>
> On Tue, Nov 3, 2015 at 10:18 PM, Robert Sanderson <azaroth42@gmail.com>
> wrote:
> >
> > Apologies for the likely newbie question, which is likely the first of
> many.
> > Please bear with me, and hopefully they can be treated as a input from
> > someone without all of the background knowledge you all have ... like
> most
> > readers will be :)
> >
> >
> > The `content` of an Object is "A natural language description of the
> object
> > content.".  But not the content of the resource itself?  Traditionally
> one
> > would call the property "description" rather than "content"? Which is
> not to
> > start a naming discussion, just to make sure that `description` is the
> > operative word, not `content` :)
> >
> > So this:
> >
> > {
> >   "@context": "http://www.w3.org/ns/activitystreams",
> >   "@type": "Image",
> >   "displayName": "Small Red Square",
> >   "content": "<p>This image is a small red square, for unknown use.</p>",
> >   "height": 100,
> >   "width": 100
> > }
> >
> > Could be a 100x100 image, and `content` would describe the image, and
> > displayName give a label for it. (As per the example in the attachment
> > definition)
> >
> > So ... if you have a Note or Article without a URI, is there a way to
> > provide the actual representation of the resource, rather than a
> description
> > of that (err) content?  Example 135 / Fig 37 in -core, however, gives the
> > impression that the representation is in the content field?
> >
> > Could someone please show how to model this situation in AS:
> >
> > The Object is a Note with the uri http://example.org/note1.md.  It's in
> > Markdown and the note's representation is "You __won't believe__ what
> > happens at the end of [this video](youtube)!".  The description of that
> > content is the HTML: "<span>Clickbait in Markdown</span>".
> >
> > Thanks!
> >
> > Rob
> >
> > --
> > Rob Sanderson
> > Information Standards Advocate
> > Digital Library Systems and Services
> > Stanford, CA 94305
>
>
>
>


-- 
Rob Sanderson
Information Standards Advocate
Digital Library Systems and Services
Stanford, CA 94305