On 10/11/13 10:43, Niklas Laxström wrote: > What is OWA? John refers to the "Open World Assumption". This is an informal concept used in knowledge representation to describe the assumption that statements that are not made are "unknown" rather than "false". This is contrasted with the "Closed World Assumption" that is common in databases, where we assume that unspecified information is false.
For example, if the ACME company has a database that contains a table for employees, and this table does not contain "Bob", then one would assume that Bob is not an employee at ACME. In contrast, if an OWL ontology contains information about people (e.g., using FOAF), and the ontology does not contain Bob, one would not assume that Bob is not a person (maybe Bob just has not created a FOAF file). If we want to say that Bob is not a person in OWL, then we can do this directly by using negation; in databases, this is usually not possible and we simply assume that omitted information is negated. OWA is closely related to what we call "monotonicity": the more information we enter into a system, the more informative it becomes. Databases are not monotonic in this sense: if I enter that Bob is an employee, then I add some information (that "Bob is an employee") but I also remove some information (that "Bob is not an employee"). OWL is monotonic, but other knowledge representation languages are not. What John refers to is that SMW assumes a default type for properties (Page). Therefore, the input behaviour is not monotonic: if you specify another type later, then this will make the formerly true type Page false. However, this is not a "violation" of the OWA. Wikitext is just a surface syntax and not the internal knowledge model of SMW. Wikitext can never be monotonic: for example, if a page contains "[[someprop::somevalue]]" and you add an input character "X" to obtain "[X[someprop::somevaue]]" then SMW will no longer contain the fact. Parser functions, templates, comments, etc. will all lead to similar effects. Clearly, it makes no sense for a surface syntax to be monotonic in this sense: a meaningful notion of monotonicity cannot refer to the character level. However, the only other structure that wikitext has is the DOM tree of templates and parser functions. Basic SMW fact syntax and MW category syntax does not feature there at all (it is just character data). So it does not work for defining a meaningful notion of monotonicity either. The simple conclusion is that wikitext is not a formal specification language, and that applying knowledge representation concepts like OWA and monotonicity to it is not meaningful. If you want information specified in a formal knowledge representation format, then you can use the OWL/RDF exports of SMW. These exports have a monotonic semantics that is based on the OWA. Markus ------------------------------------------------------------------------------ November Webinars for C, C++, Fortran Developers Accelerate application performance with scalable programming models. Explore techniques for threading, error checking, porting, and tuning. Get the most from the latest Intel processors and coprocessors. See abstracts and register http://pubads.g.doubleclick.net/gampad/clk?id=60136231&iu=/4140/ostg.clktrk _______________________________________________ Semediawiki-devel mailing list Semediawiki-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/semediawiki-devel
