W3C home > Mailing lists > Public > public-rdf-in-xhtml-tf@w3.org > January 2008

Re: Rethinking @src in the context of chaining rules

From: Mark Birbeck <mark.birbeck@formsPlayer.com>
Date: Tue, 8 Jan 2008 17:46:34 +0000
Message-ID: <a707f8300801080946h2905a56egd6e66cb23939b130@mail.gmail.com>
To: "Manu Sporny" <msporny@digitalbazaar.com>
Cc: RDFa <public-rdf-in-xhtml-tf@w3.org>

Hi Manu,

On 08/01/2008, Manu Sporny <msporny@digitalbazaar.com> wrote:
>
> Ben Adida wrote:
> > I think we're going to have to bring this to a vote.
>
> Before we do that...

Oops. I was obviously writing at the same time you were. :)

> ...I'd like to make sure we all know what each other is
> talking about, as I see several comments that still show some amount of
> misunderstanding. I'm starting to get confused as well. I believe Mark
> is correct... we're talking about two separate issues, here.

Thanks for this, since you are right that there is still confusion.


> Issues:
>
> #1: @src sets the subject.
> #2: @resource and @href do not set a new subject when chaining, only
>     @about and @src set a new subject when chaining.

That's correct.


> Here is a list of examples and the triples that they generate. I believe
> this is the viewpoint that Ben, Ivan and I have... does anything look
> out of place, Mark?:
>
> -----------------------------------------------------------------------
> <div about="#me" rel="foaf:knows">
>    <div about="#mark" />
> </div>
> --------
> <#me> foaf:knows <#mark> .
> -----------------------------------------------------------------------

That is now agreed.


> -----------------------------------------------------------------------
> <div about="#me" rel="foaf:knows">
>    <div resource="#ben" />
>    <div resource="#mark" />
> </div>
> --------
> <#me> foaf:knows <#ben> .
> <#me> foaf:knows <#mark> .
> -----------------------------------------------------------------------

In my model these triples would be generated, but both Ivan and Ben
are against this. If you believe that these triples should be
generated then we obviously don't have an Ivan/Ben/Manu agreement, so
it's lucky you sent this post. :)


> -----------------------------------------------------------------------
> <span about="#me" rel="foaf:img">
>    <img src="me.jpg" />
> </span>
> --------
> <#me> foaf:img <me.jpg> .
> -----------------------------------------------------------------------

I think we all agree on this triple, regardless of how @rel/@src is
resolved in the other thread.


> -----------------------------------------------------------------------
> <span about="#me" rel="foaf:img">
>    <img src="me.jpg" instanceof="foaf:Image"/>
> </span>
> --------
> <#me> foaf:img <me.jpg> .
> <me.jpg> rdf:type foaf:Image .
> -----------------------------------------------------------------------

As with everyone else, I would like to see this work. But I don't want
to support it at the expense of having @rel work with @src on <img>.
In the current proposal from Ben it's an either/or situation, so if
the vote goes against my approach to @resource setting a subject, then
we'll have to vote on this one, too.

Obviously if people find they're comfortable with my approach to
@resource setting the subject (when there is no @rel) then the same
rules applied to @src mean that we can actually support both
approaches.


> -----------------------------------------------------------------------
> <img src="me.jpg" instanceof="foaf:Image" rev="foaf:img"
>      resource="#me" />
> --------
> <#me> foaf:img <me.jpg> .
> <me.jpg> rdf:type foaf:Image .
> -----------------------------------------------------------------------

If @src is equivalent to @about then yes, this works. It's a clever
trick, and one I considered, but I'm not really keen on it, since to
me the first relationship we'd want to talk about would be that
between the 'current item' and the image:

  <img rel="foaf:img" src="me.jpg" />


> -----------------------------------------------------------------------
> <div about="#me" rel="foaf:knows">
>    <div about="#ben" rel="foaf:knows">
>       <div about="#mark" />
>    </div>
> </div>
> --------
> <#me> foaf:knows <#ben> .
> <#ben> foaf:knows <#mark> .
> -----------------------------------------------------------------------

I think we're all agreed on that one, now.


> -----------------------------------------------------------------------
> <span about="http://example.org/ben" rel="foaf:knows">Ben knows
>    <span href="http://example.org/mark" rel="foaf:knows">Mark, who knows
>       <span href="http://example.org/ivan">Ivan</span>
>    </span>
> </span>
> --------
> <http://example.org/ben> foaf:knows <http://example.org/mark> .
> <http://example.org/ben> foaf:knows <http://example.org/ivan> .
> [I believe that this is the point of contention - Mark thinks this
>  should be chained together so that Mark knows Ivan]
> -----------------------------------------------------------------------

No...definitely not! I clarified this misunderstanding before. In my
model @rel turns @href, @src and @resource into objects in relation to
the @rel. So to flip your mark-up slightly to make that clearer, we
have:

  <span about="http://example.org/ben" rel="foaf:knows">Ben knows
    <span rel="foaf:knows" href="http://example.org/mark">Mark, who knows
      <span href="http://example.org/ivan">Ivan</span>
    </span>
  </span>

I think everyone agrees that this would give:

  <http://example.org/ben> foaf:knows _:a .
  _:a foaf:knows <http://example.org/mark> .

Thanks again for taking the trouble to work through this.

Regards,

Mark

-- 
  Mark Birbeck, formsPlayer

  mark.birbeck@formsPlayer.com | +44 (0) 20 7689 9232
  http://www.formsPlayer.com | http://internet-apps.blogspot.com

  standards. innovation.
Received on Tuesday, 8 January 2008 17:46:48 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 8 January 2008 17:46:49 GMT