W3C home > Mailing lists > Public > public-rdf-comments@w3.org > May 2013

Re: reification (editorial)

From: Kingsley Idehen <kidehen@openlinksw.com>
Date: Wed, 29 May 2013 09:55:19 -0400
Message-ID: <51A608C7.3090005@openlinksw.com>
To: public-rdf-comments@w3.org
On 5/29/13 9:38 AM, David Wood wrote:
> On May 29, 2013, at 08:29, Sandro Hawke <sandro@w3.org> wrote:
>
>> On 05/28/2013 03:41 PM, Kingsley Idehen wrote:
>>> On 5/20/13 11:39 AM, Henry Story wrote:
>>>> In Turtle there is no way of specifying a graph (other than through reificiation which
>>>> is understood to be broken).
>>> Why is reification assumed to be broken? I think the perception is that its cumbersome due to data bloat. Like blank nodes, this is a feature of RDF that's often misunderstood and then in the process maligned.
>>>
>>> Reification is the powerful mechanism for granular descriptions of triples (statements) without exiting existing RDF semantics.
>>>
>> I wonder if there's some useful guidance we can give about when/how to use reification, or if it's best to just remain silent.   There will be a very strong temptation for some people to represent datasets in graphs using reification, but (as Henry pointed out) that's very hard to do because datasets are referentially opaque while reification is referentially transparent.  (Of course it's possible to make a referentially opaque reification vocabulary (as I did in [1]), but it's pretty awkward.)    Anyway, since we decided not to deprecate reification, I think (as an editorial matter) it would be excellent if we could give some helpful advice about it. If someone thinks they could draft that advice such that it would get WG consensus, that'd be great.
> +1 to including some description or guidance for users.  I think this would best fit in Concepts and thus become my problem.  It would be very helpful if someone would like to propose some language.  I'll throw something in if nobody else proposes.

Simple practical usecase (outside Uniprot and YAGO2): how would you 
model LinkedIn endorsements using RDF? Bearing in mind, each endorsement 
is a sort-of-verified-claim, so one needs to be able validate 
endorsements across dimensions such as endorser authority, endorser 
identity etc..


>
> Regards,
> Dave
> --
> http://about.me/david_wood
>
>
>
>>         -- Sandro
>>
>> [1] http://www.w3.org/TR/rif-in-rdf/
>>


-- 

Regards,

Kingsley Idehen	
Founder & CEO
OpenLink Software
Company Web: http://www.openlinksw.com
Personal Weblog: http://www.openlinksw.com/blog/~kidehen
Twitter/Identi.ca handle: @kidehen
Google+ Profile: https://plus.google.com/112399767740508618350/about
LinkedIn Profile: http://www.linkedin.com/in/kidehen







Received on Wednesday, 29 May 2013 13:55:46 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 20:29:56 UTC