Re: Action-47 Proposed Data Usage Vocabulary Definition

Thanks for writing this Eric, I like the text.

I think we've discovered in the course of our work here that the 
division between data description and data usage is never clear and the 
more we think about these issues, the more I see the usage and Quality & 
Granularity aspects as being all part of the same thing.

I think like Antoine, I am a little concerned about the ideas of 
limitations being so explicit. If limitations are, as some of your 
wording suggests, that "dataset X really only makes sense if you use it 
with dataset Y or in the context of Z" then, that sounds v useful. 
Licences and rights are another form of limitation, but instinctively I 
recoil from the idea of "you can only use this data to do A, B and C" - 
as the saying goes, the best uses of your data will be thought of by 
someone else. Describing what it was original compiled for, yes, that 
makes sense, but that shouldn't then lead to a description of what it 
can be used for - if indeed that's what you meant?

Phil

On 25/07/2014 13:16, Antoine Isaac wrote:
> Hi Eric,
>
> I think this bit of text is great. It may miss some cases, but of course
> in our current state it's difficult to do better. And your text is good
> acknowledging this limitation, and the need for later customization.
>
> The only part I'm not happy with was triggered by the last bullet point:
> "Limitations:  specific constraints or limitations on data including:
> lifespan, availability, and audiences.". this is not about an
> application or data usage. This seems to be more about properties of a
> dataset being used (or not).
>
> The charter [1] is pretty clear about Data Usage Description Vocabulary:
> "This will describe the use made of one or more data sets." It talks
> about "discoverability of the application", not discoverability of
> datasets proper.
>
> In fact the problem may appear earlier in the text. A literal
> interpretation of the charter would indeed rule out all the "can be"s
> ("can be used", "can be processed"). the charter's view on usage and
> application is really about "was" or "is".
>
> Coming back to the last bullet point: perhaps "availability', 'lifespan'
> are also relevant for applications. But it require a slightly different
> wording then.
>
> Best,
>
> Antoine
>
> [1] http://www.w3.org/2013/05/odbp-charter.html
>
> On 7/24/14 7:37 PM, Eric Stephan wrote:
>> Hi everyone,
>>
>> Its been a couple weeks since I sent out this request, and I haven't
>> received any responses.  Could you take a few moments prior to our
>> DWBP meeting tomorrow to review the data usage vocabulary definition
>> [1]?  If possible I'd like to close this action item.
>>
>> Does this capture  the scope of data usage?  Is there anything
>> incorrectly stated?  Is it missing anything?
>>
>> Many thanks,
>>
>> Eric Stephan
>>
>> References
>> [1] https://www.w3.org/2013/dwbp/wiki/Data_Usage_Vocabulary_Definition
>>
>>
>>
>> On Fri, Jul 11, 2014 at 8:10 AM, Eric Stephan <ericphb@gmail.com> wrote:
>>> As a part of this action item [1], feedback is required from the
>>> working group on the definition of the data usage vocabulary.  A
>>> proposed definition [2] off of our Data Usage Vocabulary Meeting notes
>>> is now on the wiki.
>>>
>>> This proposed definition is a culmination of ideas we've had in our
>>> Data Usage Vocabulary meetings and a recent DWBP meeting.
>>>
>>> Please let us know if you think the draft definition is complete
>>>
>>> Eric and Bernadette
>>>
>>> References
>>> [1] http://www.w3.org/2013/dwbp/track/actions/47
>>> [2] https://www.w3.org/2013/dwbp/wiki/Data_Usage_Vocabulary_Definition
>>
>>
>>
>
>

-- 


Phil Archer
W3C Data Activity Lead
http://www.w3.org/2013/data/

http://philarcher.org
+44 (0)7887 767755
@philarcher1

Received on Friday, 25 July 2014 13:00:38 UTC