W3C home > Mailing lists > Public > www-rdf-interest@w3.org > March 2000

Re: the chicken and the egg problem

From: Tom Van Eetvelde <tom.van_eetvelde@alcatel.be>
Date: Thu, 30 Mar 2000 10:40:06 +0200
Message-ID: <38E312E6.79A80F57@alcatel.be>
To: Pierre-Antoine CHAMPIN <champin@bat710.univ-lyon1.fr>
CC: ML RDF-interest <www-rdf-interest@w3c.org>
Hello Pierre-Antoine, Bill

Thanks for the explanation. I can conclude now that Appendix A is pure illustrative. Still, from a theoretical point of view, it doesn't seem right to me. Appendix A wants to make the implicit RDF semantics explicit by using the implicit semantics. This is a traditional example of a logical inconsistency.

I once read an article that a language should never describe its own language elements. This creates inconsistencies like the one below:

The following sentence is false.
The previous sentence is true.

It is impossible to decide if the first sentence is true or not (try out yourself). In order to overcome this, a meta-language should be used. For the same reason, self referential data structures are dangerous (cfr. Bill dehOra).

Anyway, I know that it is a hair-splitting issue. I guess nobody is really going to make a fuss about it as it is the hard coded semantics that is important, not the schema in Appendix A.

Regards,

Tom.

Pierre-Antoine CHAMPIN wrote:

> Tom Van Eetvelde wrote:
> > This doesn't seem right to me. Definitions have to be based on previously determined items, not on
> > what yet has to follow.
>
> that's right, but RDF statements about rdfs:range and rdfs:domain are no "definitions" of these concepts. They are just description of some properties of these concepts (namely, those properties that are expressible in RDF).
>
> In other words, no RDF engine could become an RDF validator only by beeing fed with the serialization of Appendix A, because this is not a "definition". To be a validator, it must "know" the implicit semantics of rdfs:domain and rdfs:range (i.e. the semantics must be hard-coded in the validator) since it can not be expressed in RDF. On the other hand, any property of rdfs:domain and rdfs:range expressible in RDF can be "learned" (read from RDF serialization) dynamically.
>
>   Pierre-Antoine
>
> --- Quid quid Latine dictum sit, altum viditur
>     Whatever is said in Latin sounds important.
Received on Thursday, 30 March 2000 03:40:42 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 7 December 2009 10:51:42 GMT