RDF browser/editors

From MWCSWiki

Jump to: navigation, search

Page Is Unavailable Due To Site Maintenance, Please Visit Reserve Copy Page

RDF browsers are tools that allow the user to browse linked data on the Semantic Web, based on the functionality of the World Wide Web. An RDF browser displays the RDF triples gathered from a resource, then allows the user to follow any links to other RDF resources, found by following explicit links (such as rdfs:seeAlso) or by dereferencing URIs. RDF browsers can be pointed at any URI on the web, but do not store the RDF data (beyond maintaining a local cache).

RDF browsers typically render RDF either inline, in which all triples involving a resource are listed on the page (and other resources involved in those triples can either be navigated to directly or expanded in the current page to show a nested view), or as a graph.

OKM also needs to represent RDF data to the user in an understandable way. It uses an inline view, like most RDF browsers.

Unlike RDF browsers, OKM stores all its RDF data on the server.

OKM also keeps track of which users assert/deny each statement. Similarly, some RDF browsers keep track of the source of a statement (provenance).

Some RDF browsers are domain specific, allowing the display to reflect a particular ontology.


Browsers

  • FOAF Explorer [1]
    • Domain-specific
    • Inline view
  • Marbles [2]
    • Represents provenance with color-coded circles; statements asserted by multiple sources have multiple circles.
  • Explorator [8]
    • Displays triples inline, grouping statements with the same subject
    • Faceted browsing -- filter based on certain properties
    • Build complex queries by using the result set of one query as part of another query, and using set operations
    • OKM distinctives:
      • Not an editor
  • ODE / OpenLink RDF Browser
    • Cited by multiple papers as an RDF browser, but couldn't find a paper focusing on it
    • Inline view
  • objectviewer [3]
    • Seems inactive
    • Couldn't find a paper
    • Cited by Explorator paper [8]


Editors

  • Protege [4]
    • Primarily an ontology editor, allowing users to author information about classes, properties, etc.
    • Does allow the creation of individuals in an ontology.
    • Available as a web application (Web Protege) or a standalone application.
    • Multiple users can access a Collaborative Protege server and modify the same ontology. The system tracks changes and comments. Users can create and vote on proposals.
    • http://protegewiki.stanford.edu/index.php/WebProtege
    • OKM Distinctions:
      • Primarily ontology, not instance based
      • Designed to get consensus, not diversity
      • Metadata first (at least in spirit)
    • Other Protege notes

<font color=grey>

  • RDF Editor (http://www.cs.rpi.edu/~puninj/rdfeditor/)
    • Allows the user to modify an RDF document and view the corresponding triples (see screenshot).
    • Couldn't find any papers about this one.
    • (looks like not worth pursuing)
  • RDFAuthor [2]
    • Graph view
    • Cited by Tabulator paper [6] and others
    • Couldn't find a paper focusing on RDFAuthor
    • Site: http://www.xml.com/pub/r/1271
    • (looks like not worth pursuing)

</font>

  • SMORE [1]
    • Allows users to annotate text with semantic data
    • OKM Distinctions:
      • free-text cake
    • <font color=red>I think this is actually a semantic wiki, even though it doesn't classify itself as such. -SD</font>
  • LOOMP [5]
    • Allows users to annotate content with semantic data, stressing an intuitive interface
    • Looks Rhizome-ish (like its ZML language to annotate wiki content with semantic stuff)
    • OKM Distinctions:
      • free-text cake
    • <font color=red>I think this is actually a semantic wiki, even though it doesn't classify itself as such. -SD</font>
  • Conzilla [4]
    • Began as a "concept browser", was then modified to use an RDF backend
    • Allows users to create graphical context-maps and annotate them with metadata
    • From the paper:

"There are several reasons why we have chosen to express context-maps in RDF. First, this enables good integration with the information triples using internal referencing techniques such as URIs and the reification mechanism. Second, it allows inference engines to easily make use of the combination of information and presentation triples. Third, it allows context-maps to be extended and reused in other contexts. Fourth, it allows flexible authoring and annotation of the context- maps themselves, effectively allowing statements like, "I agree with what was said about that information"."

    • OKM distinctives:
      • Based on graphical, visual layout and presentation of the graph
      • Context-map-centric
      • Limited provenance
  • Tabulator [6, 7]
    • Inline view
    • Javascript application, available as Firefox extension
    • View/edit multiple graphs at once
    • Paper [7] talks about provenance:

"In the Semantic Web, primarily, users read aggregated information in the graph, ignoring the fact that the data about them may have been assimilated from many sources, possibly with inference. The original tabulator experience [5] demonstrated that readers must also be able to determine the source documents, and so understand the provenance of the data (we use the term document, though the source may be the sort of thing more often referred to as a store, and may be accessed using SPARQL rather than a simple HTTP dereferencing; the same social aspects of the information apply in either case). The reader can then ask questions such as: Who wrote this? Who is maintaining it? Can I trust it? May I re-use it? and related social questions. These attributes follow from the source of the data. Just as, to trust a document on the web, one peeks at the domain name of the web site, so to trust a statement in the graph, one peeks at the URI of (and metadata about) the document."

    • Paper [6] talks about displaying provenance:

"The user explores an abstract web of data, which is the conjunction of all the graphs of documents that have been read. However, at any time he can check the provenance or source of any piece of information. Clicking on any data cell highlights the source in a pane of sources and their status. Double-clicking on the source allows the metadata about the source itself to be explored."


  • IsaViz [3] [5]
    • Java application
    • Graph view
    • single-graph editor, model-centric (according to Tabulator [7] and Conzilla [4] papers)
  • RDF Instance Creator (RIC) [6]
    • Java application
    • Forms-based interface
    • From UMD, not really maintained
    • Cited in SMORE paper

1. A. Kalyanpur et al., SMORE-semantic markup, ontology, and RDF editor (Defense Technical Information Center, 2006).

2. D. Steer. RDFAuthor. http://rdfweb.org/people/damian/RDFAuthor/ (2003).

3. E. Pietriga. IsaViz. http://www.w3.org/2001/11/IsaViz/.

4. M. Palmer and A. Naeve, “Conzilla-a Conceptual Interface to the Semantic Web,” Lecture notes in computer science 3596 (2005): 136.

5. M. Luczak-Rosch and R. Heese, “Linked Data Authoring for Non-Experts.”

6. T. Berners-Lee et al., “Tabulator: Exploring and Analyzing linked data on the Semantic Web,” in Proceedings of the 3rd International Semantic Web User Interaction Workshop, vol. 2006, 2006.

7. T. Berners-Lee et al., “Tabulator Redux: Writing Into the Semantic Web” (2007).

8. S. F. C. de Araújo and D. Schwabe, “Explorator: a tool for exploring RDF data through direct manipulation..”

Retrieved from "http://rosemary.umw.edu/mediawiki/index.php/RDF_browser/editors"
Personal tools