- From: Phil Archer <phila@w3.org>
- Date: Fri, 12 Dec 2014 12:44:33 +0000
- To: "Lee, Deirdre" <Deirdre.Lee@deri.org>, Data on the Web Best Practices Working Group <public-dwbp-wg@w3.org>
As promised, I've been thinking about the Reqs coming out of the Share-PSI Samos report. I'll stand by the list of requirements already from that as I spent a good deal of time deriving them. However, that was before the TPAC meeting so I've been through it again. R-Location ========== As we discussed at TPAC, R-Location can readily be generalised to say that property values should, wherever possible, be taken from controlled lists. That thinking might give us :ControlledList rdf:type :Requirement ; dct:description "Wherever possible, controlled lists of values should be used as values of properties."^^xsd:string ; dct:identifier "R-ControlledList"^^xsd:string ; dct:title "R-ControlledList"^^xsd:string ; rdfs:label "Controlled List"^^xsd:string ; Would that be sufficient? We could perhaps extend the description a little with some key examples where controlled lists were important and I'd suggest that locations were an example of that. The Spanish use of the DCAT-AP supports the use of controlled vocabs [0]. Potential new requirements ========================== Organization Identifiers ======================== :OrgIds rdf:type :Requirement ; dct:description "Organizations should be referred to by URI that has associated data about the organization."^^xsd:string ; dct:identifier "R-OrgIds"^^xsd:string ; dct:title "R-OrgIds"^^xsd:string ; rdfs:label "Organization Identifiers"^^xsd:string ; This comes from the Tenerife example [1] where the important data there concerns restaurants, hotels etc. Specifically: "identification of potential data providers and their datasets, both public and private sector;" I would argue that 2.3 Recife Open Data Portal 2.6 Dutch Base Registers 2.11 GS1 Digital (possibly, if Mark and Eric agree) 2.15 Documented Support and Release of Data 2.20 Open City Data Pipeline 2.21 Machine-readability and Interoperability of Licenses (so you know who is licensing the data) 2.22 Machine-readability of SLAs 2.25 LuSTRE: Linked Thesaurus fRamework for Environment 2.28 UK Open Research Data Forum All could motivate this requirement. Methodology =========== This is part of provenance metadata and the WG may feel that it's covered by R-ProvAvailable but is that enough? I'm thinking particularly about the Czech and Croatian cases [2]. We had several presentations in Samos about business models, procedures etc under which data was produced. As Hadley has stressed, development of those is out of scope but I suggest that if data is produced according to a published plan, a link to that plan would be interesting to potential users. So we might have... :Methodology rdf:type :Requirement ; dct:description "The methodology used to create the data should be published and cited."^^xsd:string ; dct:identifier "R-Methodology"^^xsd:string ; dct:title "R-Methodology"^^xsd:string ; rdfs:label "Methodology"^^xsd:string . We have a possible new Req of R-DesignatedThingsServiceProviders which I guess has similar thinking. HTH Phil. [0] http://www.w3.org/2013/share-psi/workshop/samos/report#minhap [1] http://www.w3.org/2013/share-psi/workshop/samos/report#canaries [2] http://www.w3.org/2013/share-psi/workshop/samos/report#uep On 11/12/2014 17:31, Phil Archer wrote: > Hi Deirdre, > > I love the way you've set up the RDF - makes it so much clearer when > everything's in triples. Let me run through my 3 UCs. > > 2.9 LA Times' Reporting of Ron Galperin's Infographic > > Already in place > > :LATimes :defines > :R-MetadataStandardized, > :R-UniqueIdentifier, # I think this can be merged > # with R-PersistentIdentification > :R-Citable, > :R-DataMissingIncomplete . > > I'm happy with all those but I would like to add the following existing > Requirements: > > :LATimes :defines > :R-Location, > :R-ProvAvailable, > :R-TrackDataUsage, > :R-MetadataAvailable. > > And I think we can usefully derive some new requirements from this use > case. > > :UriDesign (because the use case says "how could Ron Galperin have > referred to the source data in the Infographic? (the URI is way too > long). QR code? Short PURL?") > > :Usage (How could the publisher of the data link to the Infographic as a > visualization of it?) > > 2.19 Tracking of Data Usage > This whole use case is really around the one requirement it currently > links to which is R-TrackDataUsage. If the use case can be merged with > another one that makes the point about tracking as forcefully, then OK. > The Ordnance Survey case is one where we can quote an individual in a > citable document. > > 2.27 Share-PSI 2.0: Uses of Open Data Within Government for Innovation > and Efficiency > > I need to spend more time on this than I have available right now. So > I'll pick it up in the morning. But I think it likely that this will > generate some new requirements and strengthen existing ones. > > > > On 01/12/2014 11:12, Lee, Deirdre wrote: >> Hi, >> >> >> >> Over the weekend I have been looking at the current use-cases and >> requirements in the UCR doc. To make it easier for us to analyse, I've >> created an RDF file with this data, available here: >> https://www.w3.org/2013/dwbp/wiki/Use-Cases_Requirements_RDF >> >> >> >> I would ask all use-case authors (I'll email them individually as >> well) to review their use-case for the requirements that are defined >> by 3 use-cases or less. >> >> See the table at >> https://www.w3.org/2013/dwbp/wiki/Use-Cases_Requirements_RDF#Requirements_defined_by_number_of_UseCases >> >> >> >> >> This will reflect the currently open issues. However there are a >> couple of potential requirements that have been suggested, which >> use-case authors should also consider, namely: >> >> · code-lists, >> >> · dataset versioning, >> >> · data enrichment. >> >> >> >> There are a couple of use-cases that have no associated requirements. >> I will contact these authors directly. >> >> >> >> Cheers, >> >> Deirdre >> >> >> >> -----Original Message----- >> From: Data on the Web Best Practices Working Group Issue Tracker >> [mailto:sysbot+tracker@w3.org] >> Sent: 14 November 2014 14:22 >> To: public-dwbp-wg@w3.org >> Subject: dwbp-ACTION-130: Review issues around ucr and contact >> individual use case authors to review them in the light of current >> issues and create relevant actions >> >> >> >> dwbp-ACTION-130: Review issues around ucr and contact individual use >> case authors to review them in the light of current issues and create >> relevant actions >> >> >> >> http://www.w3.org/2013/dwbp/track/actions/130 >> >> >> >> Assigned to: Deirdre Lee >> >> >> >> >> >> >> >> >> >> >> >> >> >> >> >> > -- Phil Archer W3C Data Activity Lead http://www.w3.org/2013/data/ http://philarcher.org +44 (0)7887 767755 @philarcher1
Received on Friday, 12 December 2014 12:44:11 UTC