DXWG Weekly
03 July 2017
<Caroline_> Approving last meeting minutes https://www.w3.org/2017/06/26-dxwg-minutes
<Caroline_> mbruemmer could you please introduce yourselF?
Jaroslav_Pullmann: created MindMap to summarize the use cases
Jaroslav_Pullmann: the purpose is to help identify gaps
Jaroslav_Pullmann: we discussed the use cases for temporal modeling and I tried to summarize the discussion for the time-oriented use cases
<MarkusF> Hi can someone provide me with the new password?
Resolved: minutes from last meeting approved https://www.w3.org/2017/06/26-dxwg-minutes
<Caroline_> https://www.w3.org/2017/dxwg/wiki/Use_Case_Working_Space#Modeling_funding_sources
<danielebailo> CAroline I've troubles hering you
<MarkusF> thank you
<Caroline_> alejandra: my understanding is that it was considered as an specific use case in the Agent Role. I agree with that
as the organization/individual assumes the funder role
<Caroline_> kcoyle:
<Caroline_> s/kcoyle: /
kcoyle: what are we approving? agent role of funder makes sense, but
… there was an interest in more than funder name
<Caroline_> alejandra: we also wanted to describe the specific grants
<annette_g> +1 for covering grant numbers as well as funders
The modeling funding sources also requires funding agency/identifier/grant/ grant identifier
kcoyle: we need to discuss if this is in the scope of DCAT
if this is within DCAT scope, then we might need to add more than the agent role
kcoyle: we might need to discuss specific use cases, we will need to see what the elements might be
I can expand the description to include the specific elements that may need to be described
Jaroslav_Pullmann: with regards to this use case, it would be good to have the requirements
… e.g. grant number, funding interval, etc
… include concrete elements on what the funding model should cover
… I saw this under data lineage model
… elicitation of the data, it may be related to versioning, what influenced a particular version of the dataset
… we might need to discuss in the face to face
<SumitPurohit> can it be SUmitPurohit ?
Caroline_: so, we are going to discuss this in the F2F
<Caroline_> https://www.w3.org/2017/dxwg/wiki/Use_Case_Working_Space#Modeling_spatial_coverage
Modeling spatial coverage
<kcoyle> PROPOSED: accept ID31 with some additional enumeration of desired elements
+1
<newton> +1
<Caroline_> +1
<Jaroslav_Pullmann> +1
<mbruemmer> +1
<antoine> +1
<DaveBrowning> +1
<annette_g> +1
<SumitPurohit> +1
Resolved: accept ID31 with some additional enumeration of desired elements
does anyone want to comment on this use case?
danielebailo: we are doing work on DCAT and CERIF and was wondering if I may add to the use cases
… wants to know what is the workflow to add use cases
… go to the Use cases and check if they don't cover what we need to add
Caroline_: yes, we need more use cases, so feel free to add
<Caroline_> https://drive.google.com/file/d/0BwQz5mWMUijvU3BsZlBIc2t5MFU/
<Jaroslav_Pullmann> https://drive.google.com/file/d/0BwQz5mWMUijvU3BsZlBIc2t5MFU/
Caroline_: there is also a mindmap with the use cases
Caroline_: in the main page we have both, list of use cases and the use cases we've agreed that will be included for further discussion
Jaroslav_Pullmann: the deadline we stated as beginning of July was to have a critical mass
… but we will still like to get more
… the detailed use cases we are missing
… such as the temporal coverage and spatial coverage
<SumitPurohit> +1
Jaroslav_Pullmann: maybe the mindmap is helpful
… you can approach us as editors, but let's add more so that we can discuss in the F2F
Caroline_: going back to the use cases on spatial coverage
Jaroslav_Pullmann: my position on this, it is definitely needed, we don't have a sanitized means to refer to any formalized expression of spatial areas
… we need some guidance and suggestions and examples to be provided with the use case
annette_g: I'm wondering to what extent people will like to use arbitrary geometries for searching things
… if we allow the query to overlap
… there would be other datasets that would have multiple random geometric shapes described in them, but this is not at the level of DCAT
… I support the idea, but I wonder if adding them will help answering those types of queries
Jaroslav_Pullmann: this is the application I thought of
… if you have a GUI you would be interested in a spatial region
… we need to have overlapping regions
… query by pointing is the first application I think of
<MarkusF> Caroline, can you please increase your mic volume?
annette_g: I think this is great
kcoyle: this is a topic I am not very familiar with, but I think it would be rather difficult to come up with the exact requirement
… for those who do understand and have this need, it would be good to get into more detail
… it seems like a big topic, and we are going to need to find the sweet spot for DCAT
… but for that, we will need to know what the different elements are
kcoyle: maybe this is something to detail for the F2F
mbruemmer: I think it is an interesting use case
… but I haven't seen real datasets that would be described at a very granular level
… usually you have a country, region
… I wonder how pressing is the need to have arbitrary polygons in the metadata
MakxDekkers: I think there are enormous amounts of datasets with that shape, e.g. in INSPIRE
… but Andrea mentions that there is a challenge
… as there is no standard for this
… so it seems to me we shouldn't try to solve other people's problems
… so we should look at Data on the web WG
… and see if this is something we can solve
annette_g: I wanted to mention a specific example
… samples from ocean
… collecting organisms for sequencing
… you need to identify latitute/ longitude
… you can imagine writing an application having the geographic data
Jaroslav_Pullmann: I support exactly this application
… thinking of regions
… usage is set by implicit search
… I would ask a comment to the UC
I agree with the addition of spatial coverage
and it can become very granular
<Caroline_> alejandra: it can be very granulary if we go into details
<Caroline_> ... a generic indicating the rotation
<MakxDekkers> in my earlier comment I wanted to say that we need to look for guidance at the *Spatial* Data on the Web WG https://www.w3.org/2015/spatial/wiki/Main_Page
in DATS we included a generic entity (Place) and allowed people to include information such as Geo-JSON style
kcoyle: if we can categorize the types of spatial coverage that we need
… such as political boundaries (country)
… or points (latitute/ longitude)
… or boxes (bounded regions/polygons)
… I see the need to see more categories
Caroline_: Andrea will also write an email on this
kcoyle: we could resolve that this is an UC to be included
… but we need to sort out the details
PROPOSAL: We accept the spatial coverage use case (ID29) and we will work out the details in future meetings (F2F, etc)
+1
<Caroline_> +1
<mbruemmer> +1
<SumitPurohit> +1
<newton> +1
<Jaroslav_Pullmann> +1
<MakxDekkers> +1
<MarkusF> +1
<kcoyle> +1
<annette_g> +1
<antoine> +1
<PWinstanley> +1
<DaveBrowning> +1
Resolved: use case on spatial coverage ID29 accepted - more details will be provided
<Caroline_> https://www.w3.org/2017/dxwg/wiki/Use_Case_Working_Space#Data_access_restrictions
kcoyle: This would overlap with any discussion of copyright and official legal rights
antoine: I think I would agree with Karen
… this is a very important case
… but I would refer to the W3C permission and obligations group has done
… what people should or should not do with resources on the web
<Zakim> AndreaPerego, you wanted to say that Makx can provide an overview of this one - it's a DCAT-AP related issue
<AndreaPerego> Please note that this is about "access" conditions. Copyright and licences are about "use" conditions.
antoine: permission, obligation, expression
<antoine> https://www.w3.org/2016/poe/
MakxDekkers: when DCAT-AP was designed in Europe, this kind of information of licenses are provided at the distribution level
… for people running large catalogs it would be useful if people could see this info in the level of datasets
… what the kind of data is
… in Norway, they present this to the user in the Norwegian portal
… open data is something you want to tell your users
… just on the top level, show if it is open data or not
… something that people are doing, and currently there is no way in DCAT to provide that kind of information
… there is no way to indicate openness
… but I don't have more information that currently in the use case
<Jaroslav_Pullmann> still no audio?
<Jaroslav_Pullmann> ok
<AndreaPerego> No audio, sorry.
mbruemmer: in our use cases, we have catalogs with user groups
… where different use cases have different access levels
<annette_g> Redaction
mbruemmer: but this would be an extension
… but this is out of scope
… should I add use cases?
MarkusF: we need to differentiate access rights and access restrictions
for me access restrictions refer to the way of accessing the data
... e.g. if there is an API involved
... it can be included in the problem
... I am not sure if I understood Makx correctly
... maybe we need to define the problem
PWinstanley: I want to follow on what MarkusF and MakxDekkers said
<MakxDekkers> Norway: https://doc.difi.no/dcat-ap-no/#datasett-tilgangsniva (in Norwegian)
PWinstanley: we need to make the distinction of access rights and restrictions extensible
… opportunity for more automation
… encourage use of fully-fledge DCAT in public organizations
<MarkusF> I agree (@ PW)
PWinstanley: administrative types organizations
In the bioCADDIE project there was a working group on acessibility of datasets: https://biocaddie.org/group/working-group/working-group-7-accessibility-metadata-datasets
I will add some notes to the use case
Jaroslav_Pullmann: ODRL profile
… what are we going to restrict
<antoine> (from POE WG)
<antoine> +1
Jaroslav_Pullmann: the dataset or a particular distribution
<PWinstanley> https://joinup.ec.europa.eu/asset/criterion_evidence_cv/home to add to my earlier comments
<kcoyle> PROPOSED: accept data restrictions concept as a valid use case
<MakxDekkers> +1
+1
<annette_g> +1
<Jaroslav_Pullmann> +1
<PWinstanley> +1
<antoine> +1
<Caroline_> +1
<MarkusF> +1
<mbruemmer> +1
kcoyle: at the F2F we will get to the distinction of access vs use
Resolved: accept data restrictions concept as a valid use case
Caroline_: remember last call on F2F
<MarkusF> thank you, bye
<AndreaPerego> Thanks, and bye
<annette_g> bye all!
<Jaroslav_Pullmann> bye