– DRAFT –
DXWG plenary

hi

I think so; did the two agents + make logs public

<plh> plh has changed the topic to: Agenda: https://‌www.w3.org/‌2017/‌dxwg/‌wiki/‌Meetings:Telecon2018.09.04

Admin

Proposed: approve minutes of 28 august

<roba> +1

<PWinstanley> https://‌www.w3.org/‌2018/‌08/‌28-dxwg-minutes

<Jaroslav_Pullmann> +1

<annette_g> +1

<azaroth> +1

<PWinstanley> +1

+1

Resolved: approve minutes of 28 August

<DaveBrowning> 0 (not present)

<ncar> +1

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

Open actions

<plh> close action-148

<trackbot> Closed action-148.

<plh> action-163?

<trackbot> action-163 -- Nicholas Car to Guide the guidance group to present some potential outlines -- due 2018-07-31 -- OPEN

<trackbot> https://‌www.w3.org/‌2017/‌dxwg/‌track/‌actions/‌163

ncar: 175, will discuss at meeting tomorrow

<roba> https://‌github.com/‌w3c/‌dxwg/‌issues/‌323

<plh> action-188?

<trackbot> action-188 -- Rob Atkinson to Open issue for discussion of how profile guidance addresses requirements -- due 2018-09-04 -- OPEN

<trackbot> https://‌www.w3.org/‌2017/‌dxwg/‌track/‌actions/‌188

roba: item 188 is issue 323 in github

roba: in comment at the bottom of github 323

<plh> Karen: I remember it as a back-and-forth with the individual requirements which hasn't been done in that thread

<plh> Rob: the issue was how to take the requirements on board. the role of the ontology in the broader scope

roba: group will address requirements

<roba> "roba: about the options - 2 ways of looking at decision: 1) look at from profile description 2) how do we handle requirements in profile guidance?"

<plh> Karen: I thought we wanted a discussion about profile guidance and requirements in a GH discussion...

roba: individual requirements will need to be looked at in the group

kcoyle: would prefer that it be a separate github discussion

roba: let's revisit that at the end of the meeting; I can't separate those concerns, of the options that we need to discuss

PWinstanley: difficulty is the little time that people have had to be aware of what you've written; on the wiki pages we've been bringing out differnt proposals for the structure of the document

roba: this is not about the outlines, this is a proposal about scope; many assumptions
… am backfilling behind options

ncar: kcoyle already asked for the outlines to be discussed in the meeting tomorrow
… we need to look at all 15 requirements

PWinstanley: can we keep this open? (#188 action)

PWinstanley: ok to move to prof guidance discussion, skipping over dcat-rev for now

<roba> +0

PWinstanley: rest of discussion about profile guidance
… need to make a decision on how we will write the document and what it will contain
… depends on how we want to take in profiledesc vocabulary
… need to rein back from jumping to conclusions and look at bigger picture
… what will be useful for a wide range of users

https://‌www.w3.org/‌2017/‌dxwg/‌wiki/‌ProfileRoundup

roba: we shouldn't go back to use cases and requirements

PWinstanley: document draft as it was coming out was leaping into solutions before a discussion of options
… we also need to look at whether we are bringing in profiledesc, and if so what is the approach we will take to it?
… in preceding weeks everyone was given document links and asked to come with ideas for coming to consensus
… for how we will pitch the profile guidance recommendation

roba: the points in github address that process

<PWinstanley> kcoyle: the ProfileRoundup document gathers proposals, usecases, requirements

<PWinstanley> ... and discussions that we have already had

<PWinstanley> ... The document gathers 4 ifferent proposals for outlines for the document. Some include ways of including profileDesc if we choose to do so

<PWinstanley> ... there are also models references

PWinstanley: there are 3 models, they are fairly similar

roba: the third one is a different scope of work, a way to describe mappings between models, doesn't provide a profile description
… no one has identified a model with the same scope

ncar: put in a final pull request for the profile guidance document
… given the discussion here it would be good to move the structure of the document forward in meeting tomorrow

PWinstanley: what is your view of the outlines proposed?

ncar: there is a list of documents at the top of the pull request
… not finished yet

PWinstanley: we've asked people to think about how they think of the profiledesc vocabulary
… asking those who aren't in the profiles group - what is your view?

roba: back to the list of propositions which leads to decision for whether the pd is necessary

PWinstanley: the challenge is that we're bringing in to the proposals to the draft document something that is new, so we need to evaluate it
… we need to know if it has consensus within the group
… is it the right thing to put in place vis-a-vis the profile work
… need to have other folks coming in so we can show that this proposal has been widely understood
… we need to be absolutely clear about what are we going to be saying about profiles
… is profiledesc THE solution or A solution or???
… note that there are many folks who are not on the call who may have a view
… noting also that we need to look at the interdependency of our three deliverables

ncar: we've discussed the differences between profiledesc and adms; but there are things that pd has that adms does not
… what I think we're seeing is that the requirements we have are not satisfied by other vocabularies

PWinstanley: we need to show the discussion that points us toward recommendations
… things have been going on without consolidation and we aren't seeing this in the document

PWinstanley: the discussions need to be consolidated

roba: re:how are we going to address this is an attempt to move out of solution space
… profile is in solution space

AndreaPerego: haven't caught up with documents; trying to understand the conceptual model behind profiledesc; using adms to see if there is an alignment
… my understanding is that profiledesc is filling a gap in the specific relationships and possibly can be used as a high level conceptual model
… other vocabularies can be used for other requirements like discovery of profiles
… what I was trying to say with the reuse of adms is that a profile is an asset; can focus on the relationships portion. Can use one or
… the other depending on what you intend to do
… profiledesc may be backed up by another vocabulary that does something similar
… need to do more review

annette_g: generally I feel like I know less about inheritance than some others here but I feel a need to keep things simple

<roba> that was the purpose of antoine's exaplanatory diagram :-)

annette_g: I can read and understand what Antoine put together, has a good definition between machine-readable and human-readable
… profiledesc is more focused on where the classes come from

PWinstanley: antoine was deriving diagram from profiledesc
… may make more sense to those creating assets
… in your work where do profiles come in and where you feel this guidance would fit

annette_g: in my work i'm more interested in doing programming with it

DaveBrowning: I need to catch up, but I don't have an opinion yet

Jaroslav_Pullmann: I need to catch up, especially with the requirements

PWinstanley: ok, need to spend some time catching up, and we'll look again next week

roba: profiledesc is a straw man

<roba> +1

PWinstanley: yes, that's understood. everyone needs to understand the space that this document will be filling
… the challenge is that it is not an existing standard

<roba> can we choose a prefered place - #373?

<trackbot> Error finding 'DCAT'. You can review and register nicknames at <https://‌www.w3.org/‌2017/‌dxwg/‌track/‌users>.

DCAT group meeting report

kcoyle: we will use github #323 https://‌github.com/‌w3c/‌dxwg/‌issues/‌323

DaveBrowning: were working on the next working draft; am focusing on coherence of document
… will try to get the draft out before TPAC

PWinstanley: f2f could cover coordination between the three deliverables