– DRAFT –
DXWG weekly

<antoine> present_

Admin

<PWinstanley> https://‌www.w3.org/‌2018/‌07/‌03-dxwg-minutes

last week's minutes

Proposed: Accept last week's minutes

<azaroth> +1

<PWinstanley> +1

<kcoyle> +1

<Jaroslav_Pullmann> +1

<roba> +1

<antoine> +1

<annette_g> +1

Resolved: Accept last week's minutes

General principles

PWinstanley: Consensus and pluarism - a reminder to everyone that we have to operate in a consensual way
… meaning being open to ideas from others. There can be more than one way f doing things
… We're trying to create an output for a wide and diverse audience

PWinstanley: We can get a bit caught up in our own world a little
… there can be more than one solution

PWinstanley: In the past we'd use the wiki and the public ML.
… but now we tend to use GitHub and some of the conversation that would have gone through the issues early enough.
… Not sure that mailing list bounces have been resolved

kcoyle: The old emails are stored away... antoine noticed it...

SimonCox: I also had anyther experience this week when I was checking whether some mails from colleagues had appeared and no, they had bounce notices

phila: Suggest write to Gerald Oskoboiny <gerald@w3.org>

PWinstanley: We'd have noticed it earlier if we weren't using GH so heavily

PWinstanley: We can't be sure that GH will persist
… whereas W3C's persistence policy is much stronger

PWinstanley: We need to make sure that we have some provision for persistence

PWinstanley: We need to think about how we monitor and are open about getting input from GH
… including people who are not members of the group
… Karen was investigating with W3C. It's up to us to decide how to deal with it.

alejandra: I've used in the past an integration between GH and Zenodo which is a persistent repo. CERN uses it. Zenodo assigns a DOI to everything assiged to it

PWinstanley: Do it's a deep copy?

alejandra: It contains a history as long as you include all the GH history
… We should so regular releases of the GH repo every so often

alejandra: We could use Zenodo as a persistent repo

<Zakim> azaroth, you wanted to note that W3C archives GH

azaroth: For the annotation WG, and the JSON-LD WG, W3C has an archiving implementation that the team contact can set up
… It mirrors the entire state of the repo weekly

<alejandra> including the issues and wiki?

PWinstanley: And it keeps it within W3C

azaroth: Yes.

<Zakim> phila, you wanted to rant about DOIs

azaroth: Probably a good idea to talk to DSR about that. I dare say Ivan could help

Action: dsr to set up regular GitHub dump to w3.org

<trackbot> Created ACTION-145 - Set up regular github dump to w3.org [on Dave Raggett - due 2018-07-17].

How to acknowledge and record comments that come via GH

PWinstanley: We need a formal approach to recording and making sure that we acknowledge that we're taking the suggestion into account

SimonCox: That means being sensitive to who is and who isn't a member of the WG
… two in particular

PWinstanley: Because they're not members of a W3C MO, or an IE, they've not signed away IP

<SimonCox> ... andreas kuchartz and Jakub Klimek

phila: IP is a non-trivial point. No easy answer - but you need to be mindful that the consequences, in extremeley rare circs can be very dangerous

<Zakim> SimonCox, you wanted to suggest addition to readme.md noting IP arrangements

SimonCox: A modicum of coverage is to add something to the readme.md?
… Drawing attention to the Ip expectations

phila: +1

<alejandra> +1

SimonCox: It's not secure, but it draws attention to the Ip arrangements

<DaveBrowning> +1

Action: dsr to look into adding a disclaimer for IP in the GitHub readme to cover the WG for contributions by non-members as possible solution - or find a better solution

<trackbot> Created ACTION-146 - Look into adding a disclaimer for ip in the github readme to cover the wg for contributions by non-members as possible solution - or find a better solution [on Dave Raggett - due 2018-07-17].

SimonCox: Was there an immediate action to contact the 2 active contributors?

Action: SimonCox to write to Andreas Kuchartz and Jakub Klimek drawing their attention to the IP policies at W3C wrt their contributions

<trackbot> Created ACTION-147 - Write to andreas kuchartz and jakub klimek drawing their attention to the ip policies at w3c wrt their contributions [on Simon Cox - due 2018-07-17].

alejandra: How does it work that we invite public comments and then people that suggest something ... may not be part of the WG

phila: It's a conflict...

Action items

action-110

<trackbot> action-110 -- Karen Coyle to Find better wording for 216 especially "extend" -- due 2018-05-15 -- OPEN

<trackbot> https://‌www.w3.org/‌2017/‌dxwg/‌track/‌actions/‌110

kcoyle: Ongoing

action-129

<trackbot> action-129 -- Alejandra Gonzalez Beltran to Create new use case and requirements for more complex situation -- due 2018-06-19 -- OPEN

<trackbot> https://‌www.w3.org/‌2017/‌dxwg/‌track/‌actions/‌129

alejandra: I need to contact antoine

action-141

<trackbot> action-141 -- Phil Archer to Add his wording to the profiles document -- due 2018-07-10 -- OPEN

<trackbot> https://‌www.w3.org/‌2017/‌dxwg/‌track/‌actions/‌141

close action-141

<trackbot> Closed action-141.

DCAT sub group reports

SimonCox: Activity has reduced somewhat in the last month. The bulk of the activity on the DXWG has shifted to profiles.

<kcoyle> ?+

SimonCox: Most of the discussion has been around new use case for loosely connected files

SimonCox: It was recognised as an important use case
… Proposed resolutions not yet finalised
… Mechanism to generate UCs is worked examples.

SimonCox: FPWD has been released for a few months - almost no comments
… That's a concern since it suggests that the work is not needed
… Problem with the comments mailing list came to light when we asked for comments.
… I asked my colleagues to make comments and express contentment with various things. That drew attention to the comment list problem.

kcoyle: Does the WG have any sense of when the next draft might be stable?

SimonCox: We've not been planning towards the next draft, just working through issues
… Settling on a tight audited change process, using GH tracke

SimonCox: Changes to the visible Editor's draft aren't rapid. Maybe changes every 2-3 weeks
… might come a point where editors will need to do a significant clean up
… Not at that stage yet
… changes are incremental. Doc is in good shape, currently better that FPWD, so we could release a be version
… Hasn't been on our agenda.

PWinstanley: For TPAC, only a few weeks left to get the early bird rate
… Also October is the start of Q4 and our timetable says we'll be making progress by then. The clock is ticking...

Profiles Guidance

roba: There hasn't been... the agenda is being driven by Ruben and Lars who weren't available last time

kcoyle: I went through, trying to gather what we have discussed that could end up being content for the doc.
… I could put that on the wiki but it's incomplete
… There are diagrams, GH issues etc.
… Might make it clearer what needs to happen next.

PWinstanley: You'd do the wiki cf a Google doc

kcoyle: It's very wikiable

<antoine> +1

<roba> +1

PWinstanley: That would be very kind

kcoyle: OK, I'll get it into shape
… I can put it on the wiki quickly but it won't quickly be complete

roba: Thanks Karen. I'm standing ready to help look at some of the content but I don't feel I'm ready to set the scope.

antoine: Organising the calls for this group... maybe Rob can answer this... I understood that the guidance sub group would merge with the profile nego group - but this needs a definite time that works for everyone

roba: We're a bit of a hostage to Ruben and Lars' schedules
… , Maybe we should do alternate weeks
… I think it would be good if someone could step up and organise this. I'd do it but I'm now in this WG on my own time and not on behalf of OGC.

<ncar> I can help

antoine: So set it up on the alternate week? OK, I can look at setting that up

<ncar> yes!

PWinstanley: ncar has offered to help

ncar: I'm at a provenance conference in London
… I have a proper remit from my agency for this so I can spend time on it

Action: Carr to work with Roba and Antoine to establish meeting schedule for profile guidance doc

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

Action: ncar to work with Roba and Antoine to establish meeting schedule for profile guidance doc

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

<ncar> nicholascar (Nicholas Car)

Action: nicholas to work with Roba and Antoine to establish meeting schedule for profile guidance doc

<trackbot> Created ACTION-148 - Work with roba and antoine to establish meeting schedule for profile guidance doc [on Nicholas Car - due 2018-07-17].

UCR and requirements for conneg and profiles

PWinstanley: There's a link...

RRSAgent: make logs public

PWinstanley: There's a thing from Andrea. There are a couple of PRs that have been around for a long time

PWinstanley: I think this is something Ale and I should look at

<kcoyle> That's correct

SimonCox: I thought those PRs that Andrea was drawing attention to were related to UCs that were accepted but not yet put into the UCR

SimonCox: I believe that's the problem. They're really on the UCR editors to act on

<Jaroslav_Pullmann> .. o.k. thank you for this pointer, we'll care of them

PWinstanley: I guess I should look at that

alejandra: Jaro, and Rob.

Jaroslav_Pullmann: Quite a massive clean up needed
… No need for an action item, it's our normal responsibility

Action: Jaroslav to resolve the outstanding pull requests for the UCR from Andrea

<trackbot> Created ACTION-149 - Resolve the outstanding pull requests for the ucr from andrea [on Jaroslav Pullmann - due 2018-07-17].

<PWinstanley> Requirement: "Profiles may add to or specialise clauses from one or more base specifications. Such profiles inherit all the constraints from base specifications.” ID37 [ID39] #238

PWinstanley: is this still under discussion?

kcoyle: Yes

phila: Why is that req contentious?

PWinstanley: It isn't, we just haven't voted on it yet

roba: We seem to be being asked for evidence for it. The word inheritance seems to be a problem.

kcoyle: There was a lengthy discussion... some folks were talking about dependency rather than inheritance
… people want to have everything available without having to hop up the inheritance tree to get there

<annette_g> +1 to roba

roba: OK, but that's a separate requirement at implementation time

<azaroth> +1 to RobA as well

<SimonCox> +1 flattening is an implementation/packaging issue

PWinstanley: Do people want a rewording or just go ahead with what we have?

<roba> +1

<DaveBrowning> +1

<antoine> +1

<annette_g> +1 for voting

Proposed: Accept the Requirement "Profiles may add to or specialise clauses from one or more base specifications. Such profiles inherit all the constraints from base specifications.” ID37 [ID39] #238

<alejandra> +1

<PWinstanley> +1

+1

<SimonCox> +1

<annette_g> +1

<ncar> +1

<azaroth> +1

<Jaroslav_Pullmann> +1

Resolved: Accept the Requirement "Profiles may add to or specialise clauses from one or more base specifications. Such profiles inherit all the constraints from base specifications.” ID37 [ID39] #238

<PWinstanley> 12.1 Requirement: a vocabulary or data model can be a profile of several other vocabularies or data models at once

sRequirement - "Profiles/Requirement - "Profiles/

annette_g: I think we were concerned that the phrasing was a req about a data model rather than a profile

<azaroth> +1

annette_g: It feels inside out

<roba> a profile is a sort of data model surely?

antoine: It was discussed in the context of switching from one req to another

<kcoyle> "A profile can be a profile of several vocabularies or data models at the same time"

antoine: to react to Annette, I woldn't mind losing vocab in the first part. Sometimes I use data model where people use profile
… But if it's easier, I;m happy with Karen's phraseology

SimonCox: I was going to make a similar point
… I find the distinction between a data model and a profile difficult
… Every model os a set of constraints on having no model

<PWinstanley> "A profile can be a profile of several vocabularies or data models at the same time"

<azaroth> How about: A profile can constrain several ... ?

Proposed: Accept "A profile can be a profile of several vocabularies or data models at the same time"

azaroth: Say a profile can be a profile is tautology

<roba> +1

SimonCox: Dependencies is my favoured word
… Any model or profile potentially builds on other stuff

<kcoyle> "A profile can have dependencies on several vocabularies or data models at the same time"

<azaroth> I'm fine with dependency (or anything that isn't a circular reference)

<PWinstanley> "A profile can be dependent on several vocabularies or data models at the same time"

<annette_g> "a profile can have multiple base specifications"

<PWinstanley> "A profile can be dependent several vocabularies or data models at the same time"

<SimonCox> Note: the OGC model for specifications essentially is a model for (non-circular!) dependencies

antoine: I can live with the formulation... I think I prefer Annette's suggestion of using 'base'

<roba> the original UCR used these words: Profiles may inherit clauses from one or more parent profiles

<kcoyle> ""A profile can be based on several vocabularies or data models at the same time"

<Jaroslav_Pullmann> .. is not this the opposite of the original meaning, profile as constraint of a model (not depending on this model?)

PWinstanley: Anything we can agree now?

<azaroth> Base Specification implies it does not itself have dependencies, to me

roba: I think we have the same requirement in a different phrasing in the UCR already

PWinstanley: A profile can be based on several data models or vocabs at the same time?

roba: Inherit clauses from...

<ncar> I don't like dependency: things can depend on all kinds of things. Derivation is better

<kcoyle> +1 to azaroth - base does imply something "pure"

antoine: If we are not yet ... we are looking at reqs in the Europeana context

<annette_g> I like Karen's wording

kcoyle: I think we agree on the principle, but not the wording

<roba> +1 @azaroth - thats exactly how modelled in profileDesc - not sure yet if useful :-0

<roba> agree with kcoyle

proposal: we accept 12.1 in principle but acknowledge that it may need rewording

<PWinstanley> +1

<kcoyle> +1

<roba> +1

<Jaroslav_Pullmann> +1

<DaveBrowning> +1

+1

<antoine> +1

Resolved: we accept 12.1 in principle but acknowledge that it may need rewording

<annette_g> +1

<ncar> +1

PWinstanley: Thanks everyone for being on the call.
… Same time next week

<Jaroslav_Pullmann> bye!

<annette_g> bye all!