Recovered from the Wayback Machine.
Danny Ayers points to Leo who says that A generic RDF browser is not possible. He further clarifies his statement that you would need a stylesheet to render the data correctly. And this means you would, according to him, need a separate stylesheet for every schema.
Danny disagrees partially, saying:
From a random snippet of RDF/XML you can still infer quite a few things – what are properties, what are classes. Barest minimum is that you know something is to be treated as a resource or as a literal. That’s infinitely more that you get with arbitrary XML alone (you may know the datatype of something thanks to a schema, but even then you won’t know what the something represents).
If you have the RDF Schema/OWL ontology for the term then you should be well away.
So why not have visual components that reflect the RDF class system
I would assume that both gentlemen are discounting BrownSauce because, technically, it shows information about objects but doesn’t necessarily render the objects visually based on some criteria or constraint.
However, it does provide a human interpretable view of the data contained in the RDF files, based on specific rules and criteria defines as part of the data model associated with RDF. But it doesn’t render anything.
Stylesheets aside, one has to ask, why would one want this in the first place? We can’t render the data in a relational database without a program providing an interface and being knowledgeable about the business model constraining the data. Yet relational data has proven itself to have a use. Or two.
Why would we want to have an RDF browser?
Leo provides a nice example of what he means by rendered RDF data, with the FOAFNaut, an online utility that allows one to query a database of FOAF data, and then depict relationships discovered in a visual manner.
Keen. Now, why would we want to have an RDF browser?
Seriously, there is nothing unique or special to RDF than there is to any other data model that has been defined in the past; whether a data is visually rendered is dependent on the data, not the general model used to define the data. Geographical data in a relational database can, and does get used for visual models; FOAF data transformed from RDF to relational could be just as compellingly displayed.
BrownSauce gives us a way to read RDF without having to understand RDF and that, to me is sufficient. Add in a little visual sugar, with bubbles and all, and that would be nice, though not essential.
But what RDF provides that a relational model does not is a way of interrelating bits of data from one model with bits of data in another model and have it make sense without having to break either model. Not just store the data together – have it work together. And does so incredibly easily by publishing simple little text files that even the most moronic bot could consume if slightly trained.
So I create my poetry schema/vocabulary and encourage people to use it to annotate their online publications that contain poems. Someone else defines a schema/vocabulary for weblog posts, capturing information such as author and CC license and category and so on, and encourages people to use it to annotate their posts. A third person comes along and defines a schema/vocabulary for web objects that captures information such as whether the object has been moved, where to, is it obsolete, or other specific types of information going way beyond HTTP status codes. This person, again, gets people to use these vocabulary.
One file is now described by three separate schemas and a bot comes along and swoops it all up and is able to combine it because of one important factor: they all identify a specific resource with a given URI.
I then come along and ask for a poem that uses a bird as metaphor for freedom and my collected data then returns a selection of posts that reference a poem, giving the name of author of the post as well as other information, such as the name of the poem and its author, the fact that the post has now been moved to that location, and there’s a photograph associated with the post.
Sigh. The beauty of it brings a tear to my eye.
But I digress, badly, from the initial conversation point: why would we want a generic RDF browser. Beats heck out of me. I’m still trying to figure out how one achieves critical mass for these three vocabularies, much less how to render them.
