W3C home > Mailing lists > Public > public-dxwg-wg@w3.org > July 2017

Re: Agenda July 10, 2017 / Summary of Dataset related issues

From: Jaroslav Pullmann <jaroslav.pullmann@fit.fraunhofer.de>
Date: Mon, 10 Jul 2017 15:43:20 +0200
To: public-dxwg-wg@w3.org
Message-ID: <6a2b3521-968f-d7c2-af56-5e0e629b4215@fit.fraunhofer.de>

  Dear all,

   in order to support today's discussion I created an overview
  of related standards and their concept of "Dataset" [1].
  It is not meant to be exhaustive, just providing a starting point..

    Best regards
   Jaroslav

  [1] https://drive.google.com/file/d/0BwQz5mWMUijvLU05cGx1ZHFocDQ



On 07.07.2017 14:38, Makx Dekkers wrote:
> Jaroslav,
> 
> It is really necessary to assume a particular purpose for a Dataset? It seems to me that requiring a Dataset "to serve a particular, even hypothetical information need and any kind of digital processing and not simply the task of distribution (software)" creates a restriction that might render existing implementation non-conformant. For example, I know if work where Software Distributions are modelled as dcat:Distributions under a Software Solution modelled as a dcat:Dataset.
> 
> As to the CKAN definition of Dataset, I think that every CKAN Dataset is a dcat:Dataset but the inverse would not necessarily be true.
> 
> Makx.
> 
> 
> 
> -----Original Message-----
> From: Jaroslav Pullmann [mailto:jaroslav.pullmann@fit.fraunhofer.de]
> Sent: 07 July 2017 11:19
> To: public-dxwg-wg@w3.org
> Subject: Re: Agenda July 10, 2017 / Summary of Dataset related issues
> 
> 
>     Dear all,
> 
>       my initial assumption DCAT is about registries of (structured) bulk data was in deed challenged
>     by the mentioned use cases, since they consider any type of digital artifact including service endpoints.
>     As already discussed in context of ID6, Web Services are a sample of dynamic Distribution which exposes
>     a particular representation of the rather abstract Dataset (~ Information resource), so I'd exclude
>     them from the Dataset layer.
>     
>     I'd intuitively expect the content of a Dataset to serve a particular, even hypothetical information
>     need and any kind of digital processing and not simply the task of distribution (software).
> 
>     Here the working definition of a "dataset" in CKAN seems relevant [1]:
>    
>      "For CKAN purposes, data is published in units called datasets. A dataset is a parcel of data -
>      for example, it could be the crime statistics for a region, the spending figures for a government
>      department, or temperature readings from various weather stations."
> 
>     In next telcon I'll try to summarize the issues, but we may ask Andrea to comment on his detailed UC ID20.
> 
>     Best regards
>    Jaroslav
>      
> 
>    [1] http://docs.ckan.org/en/latest/user-guide.html
> 
>     	
> 
> On 07.07.2017 08:46, Makx Dekkers wrote:
>>
>> Karen,
>>
>> Not sure how to introduce the related use cases.
>>
>> As far as I see, the issue seems to be that people may apply, may want to apply, and are applying, DCAT to various types of data collections.
>>
>> Questions to consider could be:
>>
>> 1. is there a need to be able to identify the type of data(set) that is being described?
>> 2. if so, how can the type of data be identified -- i.e. in which property and using which (kinds of) controlled vocabularies?
>>
>> Maybe Jaroslav has other ideas?
>>
>> Makx.
>>
>>
>>
>> -----Original Message-----
>> From: Karen Coyle [mailto:kcoyle@kcoyle.net]
>> Sent: 06 July 2017 18:24
>> To: public-dxwg-wg@w3.org
>> Subject: Re: Agenda July 10, 2017
>>
>> Thanks, Makx. It would be helpful if you or Jaroslav (or both!) could introduce this set of use cases that you've suggested, with what you see as a best approach.
>>
>> kc
>>
>> On 7/6/17 9:05 AM, Makx Dekkers wrote:
>>> Karen, all,
>>>
>>> If I may, I would suggest not to start the discussion trying to define what a Dataset is. It seems to me that it would be more practical to look at what Simon mentions at the end of his problem statement "a way for a DCAT description to indicate the 'type' of dataset involved".
>>>
>>> Makx.
>>>
>>>
>>> -----Original Message-----
>>> From: Karen Coyle [mailto:kcoyle@kcoyle.net]
>>> Sent: 06 July 2017 17:28
>>> To: public-dxwg-wg@w3.org
>>> Subject: Agenda July 10, 2017
>>>
>>> https://www.w3.org/2017/dxwg/wiki/Meetings:Telecon2017.07.10
>>>
>>> This agenda follows Jaroslav's and Makx's suggestions by initiating the discussion of "what is a dataset?"
>>>
>>> kc
>>> --
>>> Karen Coyle
>>> kcoyle@kcoyle.net http://kcoyle.net
>>> m: 1-510-435-8234 (Signal)
>>> skype: kcoylenet/+1-510-984-3600
>>>
>>>
>>>
>>
>> --
>> Karen Coyle
>> kcoyle@kcoyle.net http://kcoyle.net
>> m: 1-510-435-8234 (Signal)
>> skype: kcoylenet/+1-510-984-3600
>>
>>
> 
> --
> Jaroslav Pullmann
> Fraunhofer Institute for Applied Information Technology FIT User-Centered Ubiquitous Computing Schloss Birlinghoven | D-53757 Sankt Augustin | Germany
> Phone: +49-2241-143620 | Fax: +49-2241-142146
> 
> 

-- 
Jaroslav Pullmann
Fraunhofer Institute for Applied Information Technology FIT
User-Centered Ubiquitous Computing
Schloss Birlinghoven | D-53757 Sankt Augustin | Germany
Phone: +49-2241-143620 | Fax: +49-2241-142146
Received on Monday, 10 July 2017 13:44:07 UTC

This archive was generated by hypermail 2.3.1 : Monday, 29 April 2019 13:44:55 UTC