Re: ISSUE-4 (Alien Ranges): Should we define ranges for other people's vocabularies?

Hi Irene,

On 6 Jan 2012, at 20:43, Irene Polikoff wrote:
> I do not believe that adding definitions to resources that have originated in someone else's ontology is, even slightly, rude. The ability for anyone to say anything about anything is fundamental to the Semantic Web. 
> 
> Different people and systems have different requirements and points of view. Not wrong or right, just different. In making a new statement, you are not forcing this change on everyone in the world. Only people who agree with your viewpoint will use the DCAT ontology including the statement about foaf:primaryTopic.

This viewpoint might be reasonable advice for general ontology modelling, but I don't think it is appropriate in W3C standardisation work.

Using dcat together with other data where foaf:primaryTopic is used with non-dataset values should not make the data inconsistent. The modelling of dcat (or any other W3C-recommended ontology) should be compatible with the common usage of any existing ontology referred to in the standard.

> Since Semantic Web standards are about formal, machine processable definitions, putting a usage note is not the right solution.

I disagree. It may not be as formal as you'd like, but it is certainly not wrong.

> A person can read it, but a program based on DCAT ontology would not know that the value of :primaryTopic must be dcat:Dataset.

What is the use case that requires a program to know this?

Best,
Richard




> 
> Regards,
> 
> Irene Polikoff 
> CEO, www.topquadrant.com
> 
> 
> -----Original Message-----
> From: Government Linked Data Working Group Issue Tracker [mailto:sysbot+tracker@w3.org] 
> Sent: Friday, January 06, 2012 11:01 AM
> To: public-gld-wg@w3.org
> Subject: ISSUE-4 (Alien Ranges): Should we define ranges for other people's vocabularies?
> 
> 
> ISSUE-4 (Alien Ranges): Should we define ranges for other people's vocabularies?
> 
> http://www.w3.org/2011/gld/track/issues/4
> 
> Raised by: Phil Archer
> On product: 
> 
> DCAT defines a range for foaf:primaryTopic of dcat:Dataset. FOAF of course defines no such range restriction.
> 
> I asked about this on the Sem Web IG [1]. Jeremy Carrol and David Booth both agreed (mark the date - JJC and David Booth AGREED!) that this is no big deal. If people want to take on board our assertion that foaf:primaryTopic has a range of dcat:Dataset then they can but they don't have to.
> 
> It just feels wrong to me, if not downright rude to Dan and Libby? My preference would be just to leave the range as is (owl:Thing).
> 
> My preference would be simply to provide a usage note to say that when used with DCAT the expectation is that foaf:primaryTopic will point to a dcat:Dataset but not to actually define a range within the schema.
> 
> 
> [1] http://lists.w3.org/Archives/Public/semantic-web/2011Nov/0150.html
> 
> 
> 
>

Received on Tuesday, 17 January 2012 12:17:51 UTC