W3C

– DRAFT –
DCAT team 2018-08-09

09 August 2018

Meeting Minutes

We'll strart taking turns for chairing

<SimonCox> https://‌www.w3.org/‌2017/‌dxwg/‌wiki/‌Meetings:DCAT-Telecon2018.08.09

confirm agenda

agenda confirmed

minutes of the last meeting

<SimonCox> https://‌www.w3.org/‌2018/‌08/‌02-dxwgdcat-minutes

<SimonCox> +1

<arminhaller> 0 (wasn't present)

<PWinstanley> +1

<DaveBrowning> +1

0 (wasn't present)

Resolved: minutes confirmed

Looking at matters arising

Action about dcat:byteSize closed

<trackbot> Sorry, but no Tracker is associated with this channel.

<SimonCox> https://‌lists.w3.org/‌Archives/‌Public/‌public-dxwg-comments/‌2018Aug/‌0000.html

Simon looked at mails in the comments list

Simon forwarded message separately to Melanie (in case she isn't subscribed to the comments list)

<SimonCox> Open actions: https://‌www.w3.org/‌2017/‌dxwg/‌track/‌actions/‌open

There are a few actions due by Andrea and Makx

DaveBrowning: you've got an action about qualified relations
… we are circling around that topic in the latest calls
… is that what we are doing?

SimonCox: I was a little reluctant to develop a generic abstract way
… until we get a few examples
… we've been finding cases where a qualified relations pattern would help
… but my strategy was to let that emerge when we have a set of examples

<roba> +1

SimonCox: this won't drift
… trying to get the general evolve from the specifics

alejandra: use of qualified relations in other specs

SimonCox: examples using prov
… qualified relations from prov - we're pointing forward to a chapter on provenance patterns (currently empty)
… another example, W3C organization ontology
… qualified relations between people and organizations

wasGeneratedBy

https://‌www.w3.org/‌TR/‌vocab-org/

SimonCox: we discussed this briefly last week
… preview of the recommendation document

https://‌rawgit.com/‌w3c/‌dxwg/‌dcat-issue71-simon/‌dcat/‌index.html#examples-dataset-provenance

alejandra: as the property prov:wasGeneratedBy can be used to related to any Activity and not only a Project
… we could clarify this and show examples about other activities

SimonCox: going through the text that was carefully crafted to try and not to indicate only the project
… didn't want to make lengthy examples

alejandra: just thought that the example could be expanded with other activities even if the objective here was to deal with the relationship between dataset and project

PWinstanley: we probably need to include some sort of vocabulary
… to indicate things that are business as usual
… lines of business, streams of data
… looking beyond academic projects

SimonCox: let's make sure another example shows a continuous production process and not just a project

PWinstanley: yes, like IoT

<Zakim> DaveBrowning, you wanted to say I might have an example that can be added, but not now

Action for PWinstanley to generate another example including a process generating a dataset

<trackbot> Sorry, but no Tracker is associated with this channel.

DaveBrowning: I've got another example that I can provide but not in time for the second PWD
… but for next year

SimonCox: if the document goes out of control, it can be refactored

alejandra: +1, the examples could be moved to a primer document or RDF files etc

<SimonCox> PROPOSED: Accept PR 312 to resolve #71 and as initial contribution to #76

<SimonCox> ... modulo the need for more examples to show breadth

<SimonCox> +1

<arminhaller> +1

<roba> +1

+1

<PWinstanley> +1

<DaveBrowning> +1

Resolved: Accept PR 312 to resolve #71 and as initial contribution to #76

Action: dave to also provide wasGeneratedBy examples

<trackbot> Sorry, but no Tracker is associated with this channel.

Action: PWinstanley to generate another example including a process generating a dataset

<trackbot> Sorry, but no Tracker is associated with this channel.

roba: how profiles are referenced within DCAT
… my reading of the dcterms:conformsTo predicate
… is that the concept of a standard is fairly vague
… and we don't have a reason not to use it
… so we can use it for distributions
… to indicate what profiles it conforms to
… a couple of other things: e.g. whether the cardinality is unbounded
… and if you conform to all the profiles

refering to APs

SimonCox: assumptions you're making are reasonable
… dct:conformsTo can refer to standards/vocabularies and it can be repeated

roba: the implication of this is that we need an example

SimonCox: even in the normative part of the recommendation, you'll find that it says in text that the property is used in the context of the class
… but it is not axiomatized
… no cardinality constraints or existential constraints

PWinstanley: we were discussing in the plenary is the business about consistency and appropriate location of things in the 3 deliverables
… one of the things we might want to cover in our F2F is the three pieces
… how they are shaping up, what should go where
… cross-linking

SimonCox: coordination is important, particularly between profiles and DCAT

PWinstanley: over the next month / month and a half we will need to take more attention

<SimonCox> alejandra: all agree that dct:conformsTo is suitable - use for now? or use at all?

roba: separation of concerns between profile guidelines and dcat
… is that dcat will have specific guidelines for dcat profiles
… dct:conformsTo to point to finer grain semantics
… is a concern of dcat
… profile is a subclass of dct:standards
… the question is how strong will the guidance be in this document about if one should use dct:conformsTo to point to a profile that define finer grain semantics
… how much guidance is appropriate?
… is it a statement that we don't need to make?

<SimonCox> rrsagent: generate minutes

SimonCox: do you have in mind what the examples might look like?

<DaveBrowning> rrsagent: make logs public

<DaveBrowning> rrsagent: generate minutes v2

roba: fair enough, what is a good real world example

SimonCox: I've been relying on CSIRO examples
… even if the live system doesn't rely on dcat

action for SimonCox and Nick to generate some EUDM examples

<trackbot> Sorry, but no Tracker is associated with this channel.

action for SimonCox and Nick to generate some EUDM examples to illustrate the use of profiles for fine grain semantics

<trackbot> Sorry, but no Tracker is associated with this channel.

Dataset size https://‌github.com/‌w3c/‌dxwg/‌issues/‌161

<SimonCox> is void:Dataset always RDF?

SimonCox: void for rdf datasets

roba: we used void for describing services

roba: we did a profile of void

alejandra: while some of the void properties are generic, others (related to triples statistics) are definitely for RDF datasets

SimonCox: this is to revive the issue
… so that we can go over it again next week

AOB

Action: simon to create actions

<trackbot> Sorry, but no Tracker is associated with this channel.

<PWinstanley> bye!

<SimonCox> rrsagent generate minutes v2

<SimonCox> rrsagent: generate minutes v2

Summary of Action Items

  1. dave to also provide wasGeneratedBy examples
  2. PWinstanley to generate another example including a process generating a dataset
  3. simon to create actions

Summary of Resolutions

  1. minutes confirmed
  2. Accept PR 312 to resolve #71 and as initial contribution to #76
Minutes formatted by Bert Bos's scribe.perl version 2.41 (2018/03/23 13:13:49), a reimplementation of David Booth's scribe.perl. See CVS log.

Diagnostics

Succeeded: s/forward/forwarded