Re: First version of requirements on Linked Data and Morphology

John, the requirements for morphology and terminology are basically ok with me.
Those for lexical linked data need to reuse existing models: not much sense to remake wordnet or framenet (lexical) ontologies, while they already exist.
I would say that "the model should have a simple mapping to existing lexical vocabularies such as WordNet-OWL and FrameNet-OWL".

Moreover, I would like to add that "the model should allow mappings between existing lexical vocabularies". This additional requirement opens up for two possible approaches/requirements:

1) "the model should have primitives that make it a lingua franca between different lexical vocabularies"
2) "the model should provide a basic mapping vocabulary between classes/properties of lexical vocabularies, e.g. by reusing SKOS

Let's discuss this in a while
Aldo

On Jan 18, 2013, at 1:16:22 PM , John McCrae <jmccrae@cit-ec.uni-bielefeld.de> wrote:

> Hi all,
> 
> I have written up initial versions of the requirement for publishing language resources as linked data, and for modelling morphology. Does anyone have any comments?
> 
> http://www.w3.org/community/ontolex/wiki/Specification_of_Requirements/Linked_Data
> http://www.w3.org/community/ontolex/wiki/Specification_of_Requirements/Morphology
> 
> Regards,
> John McCrae