LinkedEarth Wiki

The LinkedEarth Wiki leverages semantic wiki technology to crowdsource the curation and standards development of paleoclimate data. Like traditional wikis, they enable the collaborative authoring of content. Secure access and time-stamped content also enable the tracking of changes and the accountability of users, as well as moderation capabilities by community members of recognized expertise. In contract to traditional wikis, semantic wikis allow contributors to assign meaning to their content, specifying relationships between the objects they describe. This enables artificial intelligence reasoners to parse, process, and aggregate data into more useful forms. The LinkedEarth wiki then automatically translates this information into Linked Open Data, a universal format to share data across the web.

Our Vision

LinkedEarthWiki_design_v3
Conceptual vision for the LinkedEarth Wiki. Click on the picture to enlarge.

The LinkedEarth Wiki is based on Semantic MediaWiki , itself built as an extension of MediaWiki (the wiki software used for Wikipedia). There are hundreds of installations used for very different purposes and communities most of whom gather people with no background in computer science. The center of the project is the RDF/Linked Open Data paradigm, allowing to archive data and metadata (see here for background on semantic wikis). The inner core is the Semantic MediaWiki framework, while the outer core is the proposed semantic data curation wiki, which mediates the relationship between the core and the periphery (users, existing databases, ...).

 
How to start on the LinkedEarth Wiki?

Follow our startup guide on the LinkedEarth wiki to be able to start uploading and editing your own dataset as well as contribute to the ongoing community discussion about the standardization of paleoclimate datasets.

LinkedEarth Wiki Features

Structured Properties: The user can specify site name, archive, domains, genus, interpretation, and measurement information. There is a link to download the dataset itself.

Crowd Curation: The interface is designed for on-the-fly creation of metadata properties while encouraging reuse and normalization of properties. When a user adds a new property, they can choose a new name for the concept they want to specify. The system uses a command line completion search (similar to Google's search completion) to show properties defined by other users that match what the user is typing. This capability gives users freedom to define their own metadata terms, while encouraging normalization and convergence.

Adoption: The metadata properties need not be predefined in advance, instead each user will be able to add new ones or may adopt existing properties.

Credit: The system automatically tracks the contributions of each user, and points to their pages. This gives users visibility, status, and reputation.

Query: Special wiki pages can be created whose content is dynamically generated through queries based on Metadata.

Open Publication: All the metadata can be exported as RDF triples and made publicly available on the web.

Background on Semantic Wikis

Wikis are one of the most popular frameworks for collaboration on the Web. Wikis are easy to use, track the provenance and the history of user changes, and scale well to thousands of users. Semantic wikis augment regular wikis with the ability to structure information though semantic annotations, and with reasoning capabilities the exploit that structure to organize the wiki's knowledge (Gil, 2013). In wikipedia, "infoboxes" provide a very basic structure which is a form-based organization of some parts of the page. However, the system cannot exploit this structure or reason about its content. The infobox is just a form for users to organize content, but the system cannot understand its structure. In contrast, in semantic wikis users can define classes, properties for those classes, and restrict the values that those properties can take. For example, the Wikipedia page for Los Angeles is linked to the page for California via Los Angeles is in [[California]], while in a semantic wiki, the hyperlink would be Los Angeles is in [[state California]], where state is a structured property. Each of these semantic properties is turned into a logical assertion in the form of a "triple", such as Los Angeles state California; all triples take the form object property value. As content is added using these structured properties, the semantic wiki can use logical reasoning to organize the information and make inferences. For example, if Los Angeles is in California and California is in the US, then it can infer a new triple that Los Angeles is in the US even though that was not explicitly stated as a triple. Users can then query the content to generate dynamic content for wiki pages. For example, a page could be created as a query for all cities in the US, and it would be generated from the current contents of the site.

Semantic wikis may be seen as a microcosm of the Semantic Web, since users exploit semantic web technologies while retaining a very accessible Web collaboration interface. The triples in the semantic wiki are turned into Web objects that are openly accessible and easily interlinked with other triples. This is done by using the W3C Semantic Web standards specifically the W3C Resource Description Framework (RDF) standard. The many billions of interlinked RDF triples that are now on the Web are known as the Web of Data or Linked Data.