I do agree with Valentine regarding the terminology problem.
I think it is a metadata community (as a whole) problem. Many times we give different names to the same things or slightly the same things.
This does not help implementors.
The metadata community (as a whole) should make an effort in order to unity/standardize names.
Still on Karen's email. I think we really need a way to express an Application Profille using RDF. That was one of the issues/complains of the developers I have interviewed during my PhD work, DSP was not interesting since in the end there was no use for it (more than the description), no way to automatically validate data modeled with that AP.
regards, Mariana
-----------------------------------
Mariana Curado Malta | maltas.org
[log in to unmask]
PhD, Associate Professor | http://www.iscap.ipp.pt
http://algoritmi.uminho.pt/research-teams/semag/
Trade Union delegate | snesup.pt
________________________________________
De: DCMI Architecture Forum <[log in to unmask]> em nome de Valentine Charles <[log in to unmask]>
Enviado: 23 de julho de 2015 08:14
Para: [log in to unmask]
Assunto: Re: [RDF AP] Darwin Core, SHACL, and APs
Hello Karen,
I have looked at the Darwin Core documentation which is indeed very good and I looked at the SHACL document which I have to admit had lot of trouble to understand.
My problem is that when I looked at the XML or RDF version of the Darwin Core I do understand the application profile and what I have to do. I think it is mostly due to the fact that it refers to a framework we are familiar with (such as DCAM). I was completely lost when I looked at SHACL.
My feeling is that if a DC application profile is a particular implementation of SHACL it should provide the framework we are familiar with on top of SHACL. The framework might not be DCAM but I think it will be important to re-use some of the terminology. I think my main problem with SHACL is that I didn't understand the terminology and had therefore trouble to associate it with my EDM use case.
My confusion might come the fact that I'm not so familiar with languages like SHACL but I thought I would share my thoughts with you.
Best wishes,
Valentine
________________________________________
From: DCMI Architecture Forum [[log in to unmask]] on behalf of Karen Coyle [[log in to unmask]]
Sent: 22 July 2015 20:24
To: [log in to unmask]
Subject: [RDF AP] Darwin Core, SHACL, and APs
I recently discovered Darwin Core[1], a sort of an extension of Dublin
Core for biological data. It is well-documented, and has both RDF and
XML versions. What struck me is that the XML version is based on DCAM
and is essentially an application profile.
We still have the question before us of whether we can create an AP in
RDF. The issue (and it is one that is tormenting the W3C SHACL group as
well) is how to define a bounded data set (aka: a record) using RDF.
Assuming that the W3C group will resolve that, it may mean that we can
use SHACL as the implementation language for an application profile, and
that a DC application profile could be a particular implemention of SHACL.
Thoughts? Comments?
kc
[1] http://rs.tdwg.org/dwc/index.htm
--
Karen Coyle
[log in to unmask] http://kcoyle.net
m: 1-510-435-8234
skype: kcoylenet/+1-510-984-3600
|