RE: reference: a challenge for skos

<http://esw.w3.org/topic/SkosDev/SkosFaq> updated :)



> -----Original Message-----
> From: Dan Brickley [mailto:danbri@w3.org]
> Sent: 11 August 2004 16:38
> To: Miles, AJ (Alistair) 
> Cc: 'public-esw-thes@w3.org'
> Subject: Re: reference: a challenge for skos
> 
> 
> * Miles, AJ (Alistair)  <A.J.Miles@rl.ac.uk> [2004-08-11 16:31+0100]
> > 
> > > Thanks for this. I'm inclined to agree with his 
> conclusion, which is
> > > that we're not trying to solve the same problem. I wonder 
> if we should
> > > say something like this in the Web somewhere? eg.: SKOS 
> represents 
> > > thesaurus-like data structures in an explicit and 
> extensible format. 
> > > While these structures might be useful resources for researchers  
> > > engaged in automatic classification, parsing/interpreting 
> > > unstructured text, Natural Language Processing, etc., SKOS is not 
> > > expected to solve the difficult problems associated with 
> > > mapping from a 
> > > stream of characters to a structure which normalises them into
> > > references to uniquely identified 'concepts'. Machine 
> > > interpretation of 
> > > human-generated text is related to the general problems 
> of artificial
> > > intelligence (eg. common sense reasoning, background 
> knowledge, etc),
> > > ie. a known 'very hard problem'. SKOS attempts to address 
> an easier
> > > problem space: data sharing amongst thesaurus-based 
> applications. It
> > > does not make any grand claims regarding the utility of 
> home-grown or 
> > > specialist-maintained thesauri in everyday and scientific 
> life, beyond
> > > noting that they are widely used and that the lack of a modern,
> > > Web-friendly data model and syntax has hampered the exchange 
> > > and mapping
> > > of thesaurus datasets, and their use in Web applications.
> > > 
> > > Bit wordy, maybe?
> > 
> > Sounds pretty darn good to me.  
> > 
> > I did set up a wiki page for a SKOS FAQ
> > <http://esw.w3.org/topic/SkosDev/SkosFaq> a while back - I 
> keep meaning to
> > put in some work on that - this would be a good thing to 
> put there, although
> > we'd have to work out exactly what question it's answering :)
> 
> Q: I'm trying to build thinking machines that can read and understand
> Web pages, and hence absorb all of human knowledge and take over the
> world. SKOS is all I need, right?
> 
> (see also, http://www.w3.org/2003/11/15-tag-summary.html#closing
> 
> 	The World Wide Web is
> 	the Sum of Human Knowledge
> 	Click Here for Free Porn 
> 
> ).
>

Received on Wednesday, 11 August 2004 17:37:14 UTC