Re: Some thoughts on the Q&G vocab

Hi Phil,

I'm not sure if I understand correctly your questions, but I'm gonna try to
give my opinion :)

I think we should work to have data quality (and data usage) information in
a machine-readable format. Then, according to the data quality (data usage)
dimension, the data quality information itself may be a description
(subjectivity) or a value (objectivity).

Concerning CKAN, I think we can consider both vocabularies (data usage and
data quality) and it could be great to have data provenance as well. I
think it is important to identify what information concerning usage and
quality (and provenance) may be automatically gathered from CKAN and which
information needs to be provided by the user. Maybe we can get some
information from the system or from DCAT descriptions. Does it make it
sense for you?

For now, I'm gonna try to make a draft of how to describe data usage
considering the suggestions gave by the group in our last meetings.

Cheers,
Bernadette


2014-05-02 12:44 GMT-03:00 Phil Archer <phila@w3.org>:

> Dear all,
>
> As mentioned on today's call, I've been looking at the data quality and
> granularity vocabulary. Taking the discussion at the f2f meeting [1],
> Makx's work under the Eu ISA Programme [2] and the ODI Certificates [3] as
> my starting points, I worked through the issues and made notes in the wiki.
> Based on that I then created the diagram. All of which is available at [4].
>
> Eric - you kindly offered to help with the UML modelling, thank you. I've
> used Enterprise Architect for this - is that what you use by any chance?
>
> I think there are several high level talking points:
>
> 1. What are we trying to achieve - machine readability? Links to human
> readable documentation? Objectivity? Subjectivity?
>
> 2. How are we going to test this? Bernadette is building a CKAN extension
> for the data usage vocab - Bernadette - can it take on this vocab as well?
> (I hope so). The plan so far is for the two vocabs to be Notes, not
> Recommendations. That means we don't have to prove implementation.
> However... without implementation nothing is a standard and if we can take
> the vocabs through to Recommendation (i.e. prove multiple implementations)
> then they'll have a lot more weight.
>
> Any and all comments welcome. If focussing on a particular issue, please
> start a new thread.
>
> Cheers
>
> Phil.
>
>
> [1] http://www.w3.org/2013/meeting/dwbp/2014-04-01#Data_quality_task_force
>
> [2] http://www.slideshare.net/OpenDataSupport/open-data-quality-29248578(Slide 8)
> [3] https://certificates.theodi.org/overview
> [4] https://www.w3.org/2013/dwbp/wiki/Quality_and_Granularity_
> Description_Vocabulary
>
> --
>
>
> Phil Archer
> W3C Data Activity Lead
> http://www.w3.org/2013/data/
>
> http://philarcher.org
> +44 (0)7887 767755
> @philarcher1
>



-- 
Bernadette Farias Lóscio
Centro de Informática
Universidade Federal de Pernambuco - UFPE, Brazil
----------------------------------------------------------------------------