W3C home > Mailing lists > Public > public-gld-comments@w3.org > April 2012

Re: Vocabulary alignment

From: Dave Reynolds <dave.e.reynolds@gmail.com>
Date: Tue, 10 Apr 2012 11:51:45 +0100
Message-ID: <4F8410C1.1050101@gmail.com>
To: Simon Reinhardt <simon.reinhardt@koeln.de>
CC: public-gld-comments@w3.org
Hi Simon,

Thanks for your comments. Here I'm only going to address your questions 
on org [4].

On 05/04/12 23:49, Simon Reinhardt wrote:

> The "Data Catalog Vocabulary" [2] uses a term from FOAF, namely
> foaf:Organization. "Terms for describing people" [3] then also seems to
> be largely based on FOAF, and based on "An organization ontology" [4]
> which, however, defines its own class org:Organization. Was there a
> particular reason for that?

Yes.

Originally we believed there were differences in intention between 
org:Organization and foaf:Organization. The foaf definition talks 
particularly about "social institutions". However, following discussions 
with the FOAF maintainers, particular Dan Brickley, we concluded the 
intentions were identical and so added the equivalence relation.

At that stage it would have been possible to deprecate org:Organization 
in favour of foaf:Organization. However, since by then there were public 
sector bodies starting to use org:Organization we felt the change would 
be disruptive.

There are also some advantages to having all the key properties for the 
ontology be in the same namespace, though that is certainly not a 
universal rule.

> And also for defining org:hasMember as an
> equivalent property to foaf:member instead of re-using that?

Originally there was only org:memberOf. In increasing the alignment to 
FOAF Dan suggested that adding an explicit inverse (org:hasMember) would 
be symmetrical (we have inverses for most properties) and would allow us 
to map that to foaf:member.

This and the earlier history are noted in the change history [7].

> (Btw: The section about org:hasMember mentions an inverse org:memberOf
> which is not defined separately.

An omission in the documentation, it is there in the ontology.
That has been fixed in the source text.

> But I'm actually in favour of not
> defining inverses.

I can sympathize with that but one has to adopt one style or the other. 
See [8].

> And the domain and range of org:hasMember seem switched.)

They were correct in the ontology but swapped in the documentation.
Now fixed in the source text.


Please could you let us know if this response is acceptable.

Thanks,
Dave

> [1] https://plus.google.com/102497386507936526460/posts/Xswyq5GxdvL
> [2] http://www.w3.org/TR/vocab-dcat/
> [3] http://www.w3.org/TR/vocab-people/
> [4] http://www.w3.org/TR/vocab-org/
> [5] http://www.w3.org/TR/void/
> [6] http://www.w3.org/TR/vocab-data-cube/

[7] http://www.w3.org/TR/vocab-org/#change-history
[8] http://www.w3.org/TR/vocab-org/#notes-on-style
Received on Tuesday, 10 April 2012 10:52:22 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 10 April 2012 10:52:23 GMT