Re: RDF Display Vocabulary Second Draft

From: Stefano Mazzocchi <stefanom_at_mit.edu>
Date: Thu, 23 Sep 2004 09:41:06 -0400

Chris Bizer wrote:

> Hi all,
>
> I have drafted a lens and a style vocabulary, reflecting some of the results
> of our discussion.
>
> The lens vocabulary allows you to group and order properties. Lenses can be
> related to each other for giving the browser hints which lenses to display.
> I use a RDFpath like path language for expressing conditions on properties.
> My RDFpath syntax is based on Demian's proposal, but still needs some work.
>
> The style vocabulary also uses RDFpath expressions to specify which styles
> should be applied. Styles are applied cascadingly. You can add additional
> content to styles and do substitutions e.g. with an image.
>
> Maybe the best is to first have a look at the FOAF example. The vocabularies
> are still far from perfect, but give an impression how two sperate languages
> might look.
>
> Maybe somebody likes to have a look, comment on the design or add features
> that I have missed.

I'm going thru it, here is the lists of things I noticed:

  1) style:group rdf:type rdf:Property ;

probably you mean ":group" here.

  2) Stlye -> Style ;-)

  3) why is ":label"'s label "property"?

  4) :DoNotShowIfValueMissing makes my 'conditional alarm' ring. This is
a "procedure" masked as a declerative value and it feels wrong. To be
honest, the whole section feels kinda weird to me, maybe it would be
good to discuss it a little more.

  5) another thing that strikes me as odd, is the fact that there seems
to be no distinction (at least in ontological grouping) between
"inline"s and "block"s. Shouldn't DisplayStyleBulletList be just a CSS
hooks and let CSS deal with the type of 'display' of this data? I
understand the need for, say, DisplayStyleImage, because CSS has no
notion of understanding how to "operationally interpret" the value of a
property or which one to consider between the various ones, but the
decision to style something as a table or as an ordered/unordered list
could be entirely delegated to CSS. Also because, to be honest, how
would you interpret DisplayStyleBulletList in, say, a 3D hyperbolic
graph visualizer? what about a speech synthetizer?

  6) :SortDESC rdfs:label "Sort ASC" ;

probably you mean "Sort DESC".

About sorting: we must at least differentiate between alphabetic sorting
and numerical sorting or the literals. But, then again, I'm not sure
having sorting at this level feels right at all.

  7) what is the cascading behavior of the included CSS definitions? do
they take precedence over the system one or are overloaded?

  8) what is the meaning of parts?

enough for now ;-)

-- 
Stefano Mazzocchi
Research Scientist                 Digital Libraries Research Group
Massachusetts Institute of Technology            location: E25-131C
77 Massachusetts Ave                   telephone: +1 (617) 253-1096
Cambridge, MA  02139-4307              email: stefanom at mit . edu
-------------------------------------------------------------------
Received on Thu Sep 23 2004 - 13:40:55 EDT

This archive was generated by hypermail 2.3.0 : Thu Aug 09 2012 - 16:39:17 EDT