Re: AW: AW: Considerations on RDF presentation

From: <>
Date: Fri, 6 May 2005 18:00:35 +0200 (CEST)

Actually, I'd be happy simply with a better label to the resource itself.
Sesame's explore mode has "visit this resource on the Web" as the
link text, slightly better (I think) than the URI. I like Semame's
approach: one button with one label that means the same thing in each
display. Though I could use a smaller label, such as "document web"
or "media web".

In a semantic display, this button takes you to the same point in the
"Document Web". Similarly, you have one button with one label to take
you from any Document Web display to the Semantic browser display of
the meta data for that document. Thus you switch between two browsing
worlds, both with portals at the document resources.

In Noadster papers we've stated that a browswer can check the
mimetype of a resource to know if it a media resourse, and thus give
such a button, or a conceptual resource, and offer no such button.

Originally, Noadster displayed both the media/document and metadata
together. Worked well for small and medium-sized images, which would
appear as part of the display. Large images and documents would not
work so well, so we stopped the approach.


> Emmanuel Pietriga wrote:
>>> That URI shouldn't be interpreted as a RDF
>>> Resource,
>>> meaning that the browser looks in the repository for information
>>> about the
>>> resource, but as an external UR*L* which is retrieved and displayed.
>> Isn't the browser smart enough to figure that out by itself?
> Jacco van Ossenbruggen wrote:
> Nope. There could be loads of interesting RDF metadata attached to
>, so the user might be interested in viewing
> as a concept in the RDF repository. But in other
> cases the user might want to browse to the page itself.
> We encountered this problem in Noadster too. The current "solution" is
> that for the current focal point, we add a link at the bottom of the
> page that links to the uri itself (and not to another Noadster page
> displaying information about the resource with that URI). We use the
> URI itself as the anchor of the link. We never liked this solution
> ourselves.
> Lloyd wanted to remove this feature completely, because in many cases
> the URI provides no clue about the destination of the link. In
> addition, most pages have a URI as a focal point that only exists as the
> URI of an RDF resource, and resolving it results in an non-existing page
> (often) or leads to the source RDF file defining the page (only slightly
> better). Only RDF-literate users would understand this behavior,
> another good reason Lloyd wanted to delete this feature.
> I only argued for including it because it is the only general way to
> "get out" of the system and go to the Web - an important aspect of a
> SemWeb application.
> So I would be in favor of adding a hint that allows an application to
> make a clever decision about offering only a link to repository
> information about the URI, a link to the URI itself or both. It would
> make me happy, because it allows me to offer sensible "exit" links when
> appropriate, and it would make Lloyd happy becuase it would remove all
> those nonsense links with ugly anchors.
> Jacco
Received on Fri May 06 2005 - 15:59:37 EDT

This archive was generated by hypermail 2.3.0 : Thu Aug 09 2012 - 16:39:18 EDT