W3C home > Mailing lists > Public > public-dwbp-wg@w3.org > March 2014

Re: Scope: data vs. metadata vs ontologies

From: Steven Adler <adler1@us.ibm.com>
Date: Mon, 10 Mar 2014 11:32:52 -0400
To: Giancarlo Guizzardi <gguizzardi@gmail.com>
Cc: Antoine Isaac <aisaac@few.vu.nl>, Public DWBP WG <public-dwbp-wg@w3.org>
Message-ID: <OF5F5122C6.B072B75B-ON85257C97.00554533-85257C97.0055688C@us.ibm.com>
Are you saying we need to document the lineage of lineage metadata to 
assure it isn't fake?  Fascinating idea but kind of like a hall of 
mirrors.  Is this even technically possible?

Best Regards,


Motto: "Do First, Think, Do it Again"

Giancarlo Guizzardi <gguizzardi@gmail.com>
Antoine Isaac <aisaac@few.vu.nl>
Public DWBP WG <public-dwbp-wg@w3.org>
03/07/2014 10:49 AM
Re: Scope: data vs. metadata vs ontologies


If there's a case proposed for it, the quality of the vocabularies 
(ontologies, thesauri, etc) may be on the menu, why not.

Good. My point is that it is hard to talk about quality of data without
talking about the quality of the things which must be used to define
what this data mean.

But please don't use the bare word 'meta-data' for it! This word has a 
different meaning in wider communities, where it roughly mean what you'd 
call basic data, describing something (that is, pretty much any RDF data) 
[1]. We shouldn't introduce more confusion. Talking about vocabularies 
will already bring us enough headaches!

Sure. I was speaking very loosely. I just wanted to stress the importance
of also considering information that refers to the semantics of data
(one of the key aspects of data quality). 

Your concern to avoid confusion is of course important.
I meant indeed primarily what you could call "structural metadata".
(although, technically speaking, ontologies are not really "data", 
structural or not).

I trust the intended meaning is conveyed now.


On 3/7/14 3:47 PM, Giancarlo Guizzardi wrote:> Dear all,

I would like to raise a point of discussion that
will certainly impact the kind of contributions we
can make to the use cases catalog but also to
the final deliverables.

Up to now, we have focused our discussion on data in the more traditional 
sense of the word.
This is of course understandable. However, as well all know, data per se 
is devoid of
meaning and one of the most fundamental aspects of data quality is data 
So, since the meta-data is also data (and I am using the word meta-data
in a general sense to include vocabularies, ontologies, etc...),
my question to the group is: Is the group also interested in the quality 
of meta-data?
If so,  this can most certainly have an impact on the concepts
to be contemplated in the data quality vocabulary.

best regards,
Received on Monday, 10 March 2014 15:33:27 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 20:24:12 UTC