is (could be?) a WikiWikiWeb
including the ideas from SemanticWeb
from a WikiPage
to another can be an RdfProperty
between an RdfSubject?
and an RdfObject?
. A WikiPage
can be seen as an RdfResource
The goal is a collaborative approach on writing ontologies using ResourceDescriptionFramework
(or better WebOntologyLanguage
) as OntologyLanguage
As I know there aren't implementations of a SemanticWikiWikiWeb
, if you like to work on PlatypusWiki
, please, contact us. -- PaoloCastagna
Seems to be a SemanticWebLog?
. -- PaoloCastagna
My student project (Master degree) is related to the SemanticWeb
. Yet, I have a personal project that uses WikiWikiWeb
, and I love the idea behind Wiki. So I started thinking upon an extension of WikiWikiWeb
that would use RDF. Since I thought it was a very new idea, I searched the Web to know whether people were already working on it. I'm pleased to find it is not already thoroughly analyzed and implemented. Given the fact it makes the link between my student work and my personal project, I would appreciate collaboration in designing a SemanticWikiWikiWeb
. -- AntoineZimmermann
has some ideas: http://webseitz.fluxent.com/wiki/z2003-04-16-WikiwordAsTopic
I will work on the semantic wiki idea in my PhD thesis. All info will be put on http://www.freshbrain.com
is a semantic wiki that exposes the entire site -- content, structure, and metadata -- as editable RDF. It uses a Wiki text format varient that lets the user write arbitrary RDF and XML. And WikiNames
can name any arbitrary RDF resource, whose presentation depends on the context.
Just found this page - I agree that Wiki and SemanticWeb
could be a very useful combination, and wrote some notes on this on our local wiki - I called it MetaWiki
for the sake of a shorter name, but I see that term is already taken here for Wikis about Wikis. --DavidAllsopp
These are interesting developments -- RalphHodgson
I did some work about SemanticWiki
in my master thesis. I analysed a intra net wiki for data that could be expressed in OWL. Idea was to have the wiki (markup) as common data basis for the human readable HTML representation and the machine readable OWL.
- Most interesting data is already prepended by its meaning (name: Myself, email: email@example.com)
- The one (OWL) instance per page approach simply doesn't work
- Instances are normally not marked up by the user.
So my approach was:
- Have a OWL-Ontology/Schema that is used to parse wiki pages and to find data within them
- Using the local names of the OWL Properties as key words within the wiki pages (better use a special Description-Property to give them a name)
- Ignore keywords within text
- Selected the data using the structure of the Wiki page.
- Tables create one instance per row/column
- Use the Property-Type to further analyse the found data
- Use the Domain of the Properties to group them to instances.
- Use wiki links as Object-Properties and mangle them to the "right" instances on the remote page.
- Offer markup to give instances within the page names.
- Render as HTML anchor
- use links to these anchors as links directly to the instances
- Color the data found within the wiki page so the user can see how it is interpreted by the engine and correct errors easily
There is a very basic implementation using MoinMoin
and Jena. It converts the wiki markup to an XHTML like XML file and passes it to a Java program. Right now the results don't go back to the wiki.
See also ReflectiveWiki