- From: Ed Summers <ehs@pobox.com>
- Date: Tue, 15 Jun 2010 21:32:05 -0400
- To: Karen Coyle <kcoyle@kcoyle.net>
- Cc: public-lld <public-lld@w3.org>
On Tue, Jun 15, 2010 at 5:11 PM, Karen Coyle <kcoyle@kcoyle.net> wrote: > Can you say a little more about what you mean by "vocabulary usage"? In the context of library-linked-data by 'vocabulary usage' I guess I just meant using an RDF vocabulary to describe library resources. By 'guiding principles' I was thinking of stuff like: - how does a vocabularies authority factor into use? - when do you reuse a vocabulary? - when do you extend a vocabulary? - when do you create a new vocabulary? - is it ok to select terms from multiple vocabularies? - what's the best way to document the use of multiple vocabularies? - how should you connect your vocabulary to other vocabularies? - when should you use RDFS? - when should you use OWL? - how do you put labels on resources? - what are the pros & cons of literals/resources as objects in a triple (strings vs things)? - when/how should you connect your resources to other linked data resources? - at what granularity should we model things: e.g. records for People vs People? - how important is an understanding of httpRange-14? - should we consider formats like atom and json with typed links between web resources to be linked data? Maybe this is asking too much--but I think that unless we have some consensus about these sorts of things, choosing what vocabularies to document as best practice will be difficult. //Ed
Received on Wednesday, 16 June 2010 01:32:37 UTC