W3C home > Mailing lists > Public > public-dwbp-wg@w3.org > December 2015

Re: DUV Comments

From: Joćo Paulo Almeida <jpalmeida@ieee.org>
Date: Fri, 11 Dec 2015 16:21:16 -0200
To: Phil Archer <phila@w3.org>, Eric Stephan <ericphb@gmail.com>, Laufer <laufer@globo.com>
CC: Bernadette Farias Lóscio <bfl@cin.ufpe.br>, "Purohit, Sumit" <Sumit.Purohit@pnnl.gov>, DWBP WG <public-dwbp-wg@w3.org>
Message-ID: <D290A113.B3735%jpalmeida@ieee.org>
Dear All,

Some comments and questions with respect to the current DUV editor¹s draft:

1) Why was duv:UsageActor introduced in the DUV?  What is the relation
between duv:UsageActor, prov:Agent and prov:SoftwareAgent? Why can¹t we
use prov:Agent?

2) In example 2,  duv:Person should probably be foaf:Person?

3) Some time ago, I suggested that duv:Citation should be renamed
duv:DataCitationAct. This is to me a better name than Citation. First,
because the nominalization Citation is polysemic (it could mean an act or
the result of the act). Second, because cito was already precise in
specifying that this is an Act, we should also be precise. Third, the name
makes a clear distinction between the general term CitationAct, and the
more specific term DataCitationAct we are introducing.

4) Why is the direction of duv:hasUsage from dcat:Dataset and
dcat:Distribution to duv:Usage? Why isn¹t there an opposite property
(which to me makes more sense)?

5) What is the domain of duv:hasProducer, duv:hasDistributor,
duv:hasPublisher? 

6) The diagram seems to suggest that the domain of duv:hasAuthor is
duv:Citation, but the tables do not fix a domain, and suggest that
skos:Concept is also in the range.

7) Why is skos:Concept in the range of duv:hasAuthor(or duv:author),
duv:hasProducer, duv:hasDistributor, duv:hasPublisher? This is quite
counterintuitive to me. How can a concept author, produce, distribute or
publish anything?


8) Why should DUV do versioning of resources? I agree with Phil in this
and I think that duv:edition is not in the scope of DUV.

9) I don¹t understand duv:hasAccessInformation (what is the domain?)


Regards,
Joćo Paulo





On 11/12/15, 2:52 PM, "Phil Archer" <phila@w3.org> wrote:

>Ah, I think we all came up with the same solution, zooming in to
>different sections and describing them.
>
>The ORG ontology spec just has one digram but the prose is clearly
>broken up into sections (http://www.w3.org/TR/vocab-org/).
>
>On 11/12/2015 16:06, Eric Stephan wrote:
>> Okay I'm beginning to understand the rationale behind your model
>>synopsis
>> paragraph concept.  I really like this, and will find a way to do as you
>> suggested.
>>
>> Phil because you raised the same concern, will the supplemental
>>paragraph
>> address help address this issue?
>>
>> Thank you,
>>
>> Eric
>>
>> On Fri, Dec 11, 2015 at 7:57 AM, Laufer <laufer@globo.com> wrote:
>>
>>>
>>>
>>> Hi, Eric,
>>>
>>> I agree with the 3 points.
>>>
>>> The diagram is a visualization. But, at the same time, it could give a
>>> better understanding of all stuff together. Maybe it needs all the
>>>building
>>> blocks but not all the properties. In that sense, is why I,ve talked
>>>about
>>> the text explanation. The diagram is the one that readers will try to
>>> understand the model..
>>>
>>> Cheers, Laufer
>>>
>>>
>>> ---
>>>
>>> .  .  .  .. .  .
>>> .        .   . ..
>>> .     ..       .
>>>
>>>
>>>
>>> Em 11/12/2015 13:40, Eric Stephan escreveu:
>>>
>>> Hi Laufer,
>>>
>>> Many thanks for your feedback.  I responded to Phil earlier, because he
>>> had the same concern.  For documents such as these I've always thought
>>>of
>>> the diagrams as a general orientation to the vocabulary, not all
>>> properties and classes will be shown in the diagram.  I was thinking of
>>> this as being more conceptual than exact.
>>>
>>>   So I guess I'm saying:
>>> 1) Specification needs to be complete and include all the detail
>>>including
>>> all the properties in the model picture and additional needed for the
>>> vocabulary.
>>> 2) TTL needs to match specification.
>>> 3) Model picture is more of a visual reference and can be missing some
>>> properties and classes.
>>>
>>> I guess another option is breaking the picture an overview and then
>>> detailed view of citation, usage, and feedback.
>>>
>>> What do you think?
>>>
>>> Eric S
>>>
>>>
>>> On Fri, Dec 11, 2015 at 7:22 AM, Laufer <laufer@globo.com> wrote:
>>>
>>>>
>>>>
>>>> Hi, Eric, Bernadette and Sumit,
>>>>
>>>> First of all, congratulations for the document!
>>>>
>>>> 1. Here are the list of things that I saw in the diagram and not in
>>>>the
>>>> specification:
>>>>
>>>> cito:hasCitingEntity
>>>> duv:hasCitationCreator
>>>> oa:hasTarget
>>>> duv:hasUserFeedback
>>>> duv:hasRating
>>>> duv:hasUsageType
>>>> duv:hasUsage
>>>> duv:perfomedBy
>>>> duv:performs
>>>> duv:hasUsageTool
>>>> duv:hasRole
>>>>
>>>>   2. Here are the list of the things that I saw in the specification
>>>>and
>>>> not in the diagram (not sure if all of them really should appear in
>>>>the
>>>> diagram):
>>>>
>>>> duv:hasProducer
>>>> duv:hasDistributor
>>>> duv:edition
>>>> duv:hasAccessInformation
>>>> duv:developedBy
>>>>
>>>>
>>>>
>>>> 3. I guess that some names that were changed were not updated in the
>>>> document text:
>>>>
>>>> duv:Feedback --> duv:UserFeedback
>>>> duv:author --> duv:hasAuthor
>>>>
>>>> 4. I think (only my opinion) that a reader could be confused reading
>>>>the
>>>> examples before the vocabulary overview. Maybe the order of the two
>>>> sections could be changed.
>>>>
>>>> 5. I miss an "informal text serialization" explaining the diagram.
>>>>
>>>> Thanks a lot.
>>>>
>>>> Cheers, Laufer
>>>>
>>>> --
>>>>
>>>> .  .  .  .. .  .
>>>> .        .   . ..
>>>> .     ..       .
>>>>
>>>>
>>>>
>>>
>>
>
>-- 
>
>
>Phil Archer
>W3C Data Activity Lead
>http://www.w3.org/2013/data/
>
>http://philarcher.org
>+44 (0)7887 767755
>@philarcher1
>
Received on Friday, 11 December 2015 18:22:00 UTC

This archive was generated by hypermail 2.3.1 : Friday, 11 December 2015 18:22:00 UTC