- From: Harshvardhan J. Pandit <me@harshp.com>
- Date: Wed, 3 Apr 2019 18:14:13 +0100
- To: Axel Polleres <axel.polleres@wu.ac.at>
- Cc: public-dpvcg@w3.org
Hi Axel, yes the spreadsheet should do nicely, thanks : ) I've commented on a few things within the spreadsheet itself. On call tomorrow we should discuss if this is the method we use for feedback (IMHO it is - trackable, specifies author, nested comments) Best, Harsh On 03/04/2019 17:49, Axel Polleres wrote: > Hi Harsh, > > I hope the current spreadsheet structure covers all you need (you can declare both sub/superclasses and also domain/range for properties... > > -- > Prof. Dr. Axel Polleres > Institute for Information Business, WU Vienna > url: http://www.polleres.net/ twitter: @AxelPolleres > >> On 02.04.2019, at 20:13, Harshvardhan J. Pandit <me@harshp.com> wrote: >> >> Hi Axel, Everyone. >> >> Spoke with Mark, and spreadsheets would be ideal for the F2F as they're the easiest and involve least amount of work. If everyone is comfortable with Google Sheets, then that. Otherwise whatever is preferable to most participants. >> >> Regarding schema: I think the schema would be different for each taxonomy/ontology. >> >> Common fields/columns: name, description, source/reference, synonyms >> >> For hierarchical taxonomies, we would also need to have a "parent term(s)" field. >> >> Regarding requirements or constraints, such as those for Consent Receipt, we would need a field that states the particular constraint. E.g. a field is required/optional, values can only be binary, must have a particular property. etc. >> >> For taxonomies that have properties/relationships, such as reference field for stating law for legal obligation legal basis, these properties should be documented separately from the classes/concepts. Properties should have fields name, domain, range, description, source/reference (if present) >> >> A few things off the top of my head regarding different taxonomies: >> >> 1) personal data: narrow/broad to express relationship between data categories - e.g. name & first name >> >> 2) data storage: we do not need to write down all ISO country codes, merely stating something like range is ISO country codes will suffice for now - we will expand all country codes after the F2F. >> >> 3) reusing existing taxonomies: if we know a particular taxonomy exists, then let us mention it in the spreadsheet, and later align the terms with those from that ontology. E.g. to refer to EU law, EURLEX/ELI can be used, so we list them as potential candidates in the document >> >> 4) examples (preferably real-world ones) would be nice to have associated with a term or property where possible. >> >> >> After the F2F, depending on the quality of the spreadsheet, we can clean them and use mapping (RML) or even a generic script to ease the task of generating taxonomies from the terms - but IMHO the focus of the F2F should be on discussion and agreement over the terms rather than generating RDF/OWL files immediately. >> >> Regards, >> >> Harsh >> >> On 02/04/2019 15:35, Axel Polleres wrote: >>> Hi HArsh, >>> >>> we discussed today what would be the best way to collaboratively edit the taxonomies in the F2F: >>> >>> my idea was to simply use google spreadhseets with a fixed schema >>> >>> something like: >>> >>> term; description; taxonomy; broaderterms; narrowerterms; provenance; >>> >>> or alike >>> which hopefully all can fill in from the current discussions. >>> >>> Would that work? Or do you have a better tool or proscesss to work out thetaxonomy on github? >>> >>> >>> Axel >>> >>> -- >>> Prof. Dr. Axel Polleres >>> Institute for Information Business, WU Vienna >>> url: http://www.polleres.net/ twitter: @AxelPolleres >>> >>> >> -- >> --- >> Harshvardhan Pandit >> PhD Researcher >> ADAPT Centre >> Trinity College Dublin >> >> > -- --- Harshvardhan Pandit PhD Researcher ADAPT Centre Trinity College Dublin
Received on Wednesday, 3 April 2019 17:15:08 UTC