Ontiki explores the notion of ontological wikis: user-editable sets of web pages which have strong ontological support. The current (WIP!) incarnation of Ontiki is based on a graph database, some publicly-available data sets, and a GUI-flavored web app. The result is (largely) a graph-aware "wrapper" for Wikipedia (WP).


Wikipedia is an awesome resource. I peruse it daily, edit and link to it frequently, and donate often to the Wikimedia Foundation. That said, I don't think the WP experience is all that it could or should be, particularly when multiple pages are involved.

Although some WP pages have problems (e.g., edit wars, lack of citations or introductory material), most of the ones I use are clean, correct, etc. Not perfect, of course, but very respectable attempts to codify the current consensus on individual topics.

However, there is very little obvious structure (let alone contextual information) above the level of pages. So, although WP's pages and links form a (huge, densely linked) graph, users have little way to take explicit advantage of this fact. Nor does WP provide any support for the creation or sharing of reading lists, ontological views, etc. Ontiki is, in part, an attempt to remedy these deficiencies.

However, Ontiki's charter extend far beyond the domain of WP. Many collections of text (e.g., scientific papers) might benefit from a collaborative tool of this sort. A large, graph-based ontology should serve well as a skeleton for most of these collections.

Existing Work

HTML-based links only connect pages in one direction. However, because WP is backed by a database, it can provide supplementary linkage information. So, for example, it offers links to related pages in Infoboxes, on Tools > What links here pages, etc. DBpedia (and DBpedia Live) extract information from WP, making the result available as a graph-structured knowledge base.

Based on DBpedia (along with GeoNames and WordNet), the YAGO (Yet Another Great Ontology) project maintains an enormous, mechanically-generated ontology (encoded, in Turtle, as a set of RDF triples). YAGO-SUMO extends this work, based on information from SUMO (Suggested Upper Merged Ontology), a highly axiomatized formal upper ontology.

In addition, various knowledge engineering projects use WP as a testbed for data mining technology. In summary, a vast amount of WR-relevant information is available. YAGO itself contains about 300M facts, so something like a billion facts (and growing :-) seems plausible. See Data Sets and Resources for more information.


Clearly, a large amount of contextual information is available for WP topics. I would like to make this information useful to (and extensible by) its users. Although there are clearly some crocodiles on the path, it isn't difficult to sketch out a plausible approach:

  • Load database from bulk data sources.
  • Enter / edit lists of inquiry-related topics.
  • Suggest some "plausibly relevant" topics.
  • Accept user feedback, suggestions, etc.
  • Generate / edit "ontological views".
  • Enable multi-user collaboration.

Getting Started

This web contains a fair amount of material, most of which will not be of interest to casual readers. The best starting points are probably:

Detail Pages

Here, for convenience, is a full list of the detail pages:

And here is a list of related project webs on this site:


I have been exploring the idea of ontological wikis for quite a while. Several years ago, I created this (Ontiki) web as a way to explore ideas about combining ontologies, wikis, etc. The Ontiki web now contains information on the current (2014-2015) version of Ontiki. For information on the previous version, see Ontiki_2009.

This wiki page is maintained by Rich Morin, an independent consultant specializing in software design, development, and documentation. Please feel free to email comments, inquiries, suggestions, etc!

Topic revision: r97 - 24 Sep 2015, RichMorin
This site is powered by FoswikiCopyright © by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
Ideas, requests, problems regarding CFCL Wiki? Send feedback