Comments on: Different World Views (TBox) for the same Structs (ABox) http://fgiasson.com/blog/index.php/2009/01/22/different-world-views-tbox-for-the-same-structs-abox/ Tue, 15 May 2012 14:41:46 +0000 hourly 1 http://wordpress.org/?v=3.3.1 By: Fred http://fgiasson.com/blog/index.php/2009/01/22/different-world-views-tbox-for-the-same-structs-abox/comment-page-1/#comment-242970 Fred Thu, 22 Jan 2009 21:58:16 +0000 http://fgiasson.com/blog/?p=921#comment-242970 Hi Valentine! Well, first the TBox defines the classes and properties (in RDF parlance) and their descriptions and relationships. So, the TBox is defined by these classes and properties descriptions. But As Mike put it: there is a problem with the semweb's "ontology" term since many people will describe everything in an "ontology": classes, properties, and instances of classes. So, I do think the classes and properties belongs to the TBox and the instances of classes belongs to the ABox. One of the point of this blog post is that we have to separate concerns between ABox and TBox. We do care about the ontologies at the level of the TBox. However we don't have to care about how the other system will think (tbox) about the data that is being exchanged (abox). So, sure that we care about the ontologies (at least I do) so that we can make some sense of data :) Does this help? Thanks! Take care, Fred Hi Valentine!

Well, first the TBox defines the classes and properties (in RDF parlance) and their descriptions and relationships. So, the TBox is defined by these classes and properties descriptions. But As Mike put it: there is a problem with the semweb’s “ontology” term since many people will describe everything in an “ontology”: classes, properties, and instances of classes.

So, I do think the classes and properties belongs to the TBox and the instances of classes belongs to the ABox.

One of the point of this blog post is that we have to separate concerns between ABox and TBox. We do care about the ontologies at the level of the TBox. However we don’t have to care about how the other system will think (tbox) about the data that is being exchanged (abox). So, sure that we care about the ontologies (at least I do) so that we can make some sense of data :)

Does this help?

Thanks!

Take care,

Fred

]]> By: Valentin http://fgiasson.com/blog/index.php/2009/01/22/different-world-views-tbox-for-the-same-structs-abox/comment-page-1/#comment-242957 Valentin Thu, 22 Jan 2009 20:54:51 +0000 http://fgiasson.com/blog/?p=921#comment-242957 Really enjoyed reading this post - but still think you made a crucial mistake: you equalize the TBox with 'how [...] properties will be processed, understood and managed'. But if you look at today's implementations of Semantic Technologies - how much of their behavior is really determined by the TBox? In the end - how much can you learn about Google's Social Graph implementation from the FOAF ontology? Even more - if you argue that the common case will be that applications exchanging data do not commit to the same ontology and will interpret relations and properties differently - why then bother with ontologies at all? Really enjoyed reading this post – but still think you made a crucial mistake: you equalize the TBox with ‘how [...] properties will be processed, understood and managed’. But if you look at today’s implementations of Semantic Technologies – how much of their behavior is really determined by the TBox? In the end – how much can you learn about Google’s Social Graph implementation from the FOAF ontology?

Even more – if you argue that the common case will be that applications exchanging data do not commit to the same ontology and will interpret relations and properties differently – why then bother with ontologies at all?

]]>