W3C home > Mailing lists > Public > public-swd-wg@w3.org > May 2008

Re: SKOS comment: naming of hasTopConcept

From: Antoine Isaac <aisaac@few.vu.nl>
Date: Mon, 05 May 2008 21:01:27 -0400
Message-ID: <481FADE7.7010004@few.vu.nl>
To: Kjetil Kjernsmo <Kjetil.Kjernsmo@computas.com>
CC: public-swd-wg@w3.org

Hi Kjetil,

Nice to see you back with your always interesting use cases :-)

I have actually just finished to re-read the SKOS reference, it says 
that you can use hasTopConcept to target concepts that have parents. The 
idea I think is that these concepts can start as entry points for 
browsing sequences, no matters where they are in the hierarchy.

But that's true, it will be difficult to avoid saying that the typical 
(and advised) use would be to use hasTopConcept to explicitly target the 
top concepts...


> Hi SWD members,
> We're developing an application where the Knowledge Organisation component is 
> based on SKOS. Other Semantic Web technologies are used throughout as well.
> We anticipate that associative relations will be used extensively, possibly 
> also sub-propertying skos:related. We think of the system as a graph rather 
> than a tree. It is thus not natural for us to name any top concepts, but we 
> have a requirement that we identify certain concepts as concepts of 
> particular importance, starting points, or something like that.
> skos:hasTopConcept seems almost right for our use, but it has the shortcoming 
> that it seems to tie strongly in with a hierarchical structure. This seems 
> odd to us, as SKOS does support associative relations, and they seems left 
> out.
> We would therefore like to know if the SWD has looked into a having a similar 
> construct as skos:hasTopConcept for non-hierarchical structures, and if not, 
> if the SWD could look into it. Perhaps calling it "skos:hasMainTopic" and 
> dropping the mention of a "hierarchical" from the spec would be OK?
> Also, for our application, it would simplify some queries if there were an 
> inverse of this property.
> Kind regards 
> Kjetil Kjernsmo
Received on Tuesday, 6 May 2008 01:33:00 UTC

This archive was generated by hypermail 2.4.0 : Friday, 17 January 2020 17:31:51 UTC