On Montag, 9. Juni 2008, S Page wrote: > Jeff Thompson wrote: > > At the beginning of SMW development, the idea was that, to make everyone > > happy, anyone could create any property that they want and give it > > whatever meaning they want in their wiki. That's great for making > > everyone feel included (even the wiki users who just want to jump in > > without thinking about what they are doing). > > > > But as soon as SMW has to do inference, even the basic inference for > > transitive and inverse properties, then this requires not only a > > syntax, but also a semantics. Some people don't like (or don't > > understand) the semantics developed for OWL, but at least it is > > well-defined. > > > > Does SMW use the semantics defined for RDF/OWL by default? > > SMW is "just" PHP code that parses wiki annotations and makes DB > queries. It implements some useful semantics, e.g.: > * querying for category membership also finds members of subcategories, > * querying for a property finds also finds pages with its subproperties > * querying for a page also finds #REDIRECT pages. > but not (currently) others (transitive properties, identical properties, > etc.). So there's no "default" semantics, just code, though I'm sure > the main developers were guided by ideas in RDF/OWL.
Remarks: * Any OWL implementation is "just" code. Its behaviour can still conform to the OWL semantics, and SMW does that for the limited OWL sub-language it covers. So OWL is really the default semantics (of course, to be honest, the semantics of class and property hierarchies are the same in most knowledge representation formalisms anyway!). * Identical (equivalent) properties are supported in current SMW by means of mutual subproperty statements (if P is subproperty of Q and Q is subproperty of P, then P and Q are equivalent). The upcoming SMW will also support #REDIRECTS for properties to assert identity. > > Special:ExportRDF does use RDF, RDFS, OWL, and XSD semantics to > represent the annotations in SMW; export some pages and see! As you > note, many SMW sites don't care about this. As we've discussed on the > mailing lists, other sites do care about the semantics that RDF Export > "promises", especially when it comes to MediaWiki categories. Yes, but the export in a way provides an official definition of how SMW constructs can be interpreted and reused in other systems. It is crafted in such a way that it leads to the same results as the one obtained by SMW internally, so that the wiki semantics becomes indeed "transportable" via the export. > > > If a wiki > > user wants a different semantics (for what defines equivalent topics, > > etc.) > > (Equivalent topics isn't a good example because its sort-of covered by > making one page a #REDIRECT to another, or to link to an external > concept see <http://semantic-mediawiki.org/wiki/Property:Equivalent_URI>.) > > > does SMW support a language for them to define it (pretty difficult)? > > FYI, to me SMW 1.1's "language of semantics" is, roughly: > * MediaWiki's articles, categories, and #REDIRECT > * SMW's property annotations > * and SMW's 8 special properties > <http://semantic-mediawiki.org/wiki/Help:Properties_and_types#Special_prope >rties> > > > That's insufficient to define semantics for things like transitive, > inverse, broader, etc. You're free to add additional properties and > categories to your wiki representing any semantics, but they won't mean > anything to SMW, and it's up to you to make your queries respecting > them. You can see lots of experiments along these lines on the sandbox > site, e.g. check out > http://sandbox.semantic-mediawiki.org/wiki/Category:Transitive_relation Exactly, that was what I referred to in my earlier mail. > > > One thing you can do is use > http://semantic-mediawiki.org/wiki/Help:Import_vocabulary to associate > terms in your wiki with an external ontology. Then RDF export will use > that ontology in place of <property:Your_name>. The classic example of > this is importing FOAF so Property:Name can export as > <rdf:RDF ... xmlns:foaf="http://xmlns.com/foaf/0.1/"; > > <foaf:name> > which means something to Sir Tim's "Semantic web"; > instead of > <!ENTITY property 'http://semanticweb.org/id/Property-3A'> > <property:Name> > which doesn't mean anything outside your wiki. Maybe put less strongly: it always means something *semantically* (in the sense of "semantics = formal semantics" used for "semantic technologies") but it has no meaning *pragmatically* to most other tools. Foaf is known to various tools, while your personal wiki vocabulary is not (yet? ;). This is really the only difference. > > At one point I wanted to take all the properties that people invented in > experiments with ontologies on the old ontoworld.org site and say > [[Imported from::owl:inverseOf]], [[Imported from::skos:narrower]], etc. > But as Markus wisely pointed out, doing this would suggest that SMW > supports these features. On our own wikis at least, we try to keep the OWL/RDF export in synch with the built-in semantics of SMW. One could of course also import OWL vocabulary that would then appear in the exported data, but the export would then no longer reflect what SMW considers it to be, but rather have a different semantics. > > > You see where I'm going with this question. Doesn't SMW need to get > > past the "anything goes" and officially commit to some well-developed > > semantics such as OWL? > > What's wrong with what I've described above? What would you like to see > instead? I imagine that as SMW does support additional features of > ontologies, each will effectively have certain semantics in its PHP > implementation and promise certain semantics in RDF export, just as say > <http://semantic-mediawiki.org/wiki/Property:Subproperty_of> does now. > > I hope this simplistic overview helps. BIG DISCLAIMER: I haven't > inspected the new features in SMW 1.2, Currently no new semantic features are really scheduled for 1.2. The new code prepares new reasoning mechanisms, but these are not really there yet. > nor have I used SMW with > SMW_RAPStore (RDF API for PHP) instead of the default SMW_SQLStore. RAPStore sits on top of SMWSQLStore, and it uses the OWL/RDF export of SMW internally. Hence it is also compatible with the SMW built-in semantics, even though it has some limitations (RAP is an RDF system that does not support all OWL/RDFS features, so some schema information might be ignored by RAP). -- Markus > > As ever, I'd love to hear of people's experiences using SMW with > external ontologies. > > Some of these topics are touched on by > http://semanticweb.org/wiki/Help:Suggestions , which people can edit. > > Regards, > -- > =S Page > > ------------------------------------------------------------------------- > Check out the new SourceForge.net Marketplace. > It's the best place to buy or sell services for > just about anything Open Source. > http://sourceforge.net/services/buy/index.php > _______________________________________________ > Semediawiki-devel mailing list > Semediawiki-devel@lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/semediawiki-devel -- Markus Krötzsch Semantic MediaWiki http://semantic-mediawiki.org http://korrekt.org [EMAIL PROTECTED]
signature.asc
Description: This is a digitally signed message part.
------------------------------------------------------------------------- Check out the new SourceForge.net Marketplace. It's the best place to buy or sell services for just about anything Open Source. http://sourceforge.net/services/buy/index.php
_______________________________________________ Semediawiki-devel mailing list Semediawiki-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/semediawiki-devel
