– DRAFT –
DXWG Plenary

PWinstanley: It's going to be a really short meeting, because I'm not going to move for a vote with only me and 3 other people around.
… We can probably do the minutes of the last meeting

Proposed: accept minutes of last meeting

<antoine> 0 (absent)

<PWinstanley> +1

<ncar> +1

+1

Resolved: accept minutes of last meeting

<ncar> link to minutes of the last meeting: https://‌www.w3.org/‌2019/‌04/‌09-dxwg-minutes

PWinstanley: pull requests and issues. If there are any due for closing that you want to comment on, please use Github to respond. After Friday they will start to be closed.

ncar: The conneg group has been getting a head start on cleaning things up
… where the issue is resolved for conneg, but not for other groups, we just removed the conneg tag.

PWinstanley: open action items. 321 can be closed, because I sent the email about not having the Fukuoka meeting. 323 can be closed; W3C do have precedents for providing acknowledgments. This came up in the DCAT subgroup. There's a simple style where we acknowledge and provide the list of names.

antoine: question about untagging issues. For Nick, I think tagging issues have been used for making lists in documents. I'm worried we will lose something by untagging things.

ncar: that's why I was cautious about that approach. We do lose the historical record. I don't think it's ideal, but I don't have a solution. I think we'll have to invent some other tag that drops off the list to indicate something is done.

antoine: thanks for the answer. How about tag as due for closing but only from the perspective of one of the groups.

ncar: some have three groups. We could have tags like "conneg-complete". That would be a better solution. We'd then look for issues that have that tag to remove from the doc.

antoine: should I do that, or do you want to?

<antoine> "negotiation - can be closed"

ncar: we could say "profile negotiation closed", "profile guidance closed", etc.

antoine: I think what you suggest would be good too.

pwin: can someone write it up?

ncar: I'll do it

PWinstanley: let's do the conneg 2PWD vote via email

PWinstanley: Annette, you had thoughts about this, and Antoine, you had similar concerns. Does anybody want to say anything about it?

<antoine> annette_g: I was really happy to see ncar is happy with my suggestions

<antoine> ncar: thanks. There's just one thing on which we differ: the scope of 'realization'

<antoine> PWinstanley: did we resolve the issue with key-value pairs?

<antoine> annette_g: I was not unsure I was interpreting your words correctly

<antoine> ... ncar

<antoine> ... About not to use query string arguments for things revealed to the client.

<antoine> ncar: not sure how to answer. As it reads now in your edit, the section on query strings says

<antoine> ... you could easily have an API that proposes options to the user

<antoine> ... people could chose how the client could render that

<antoine> ... What matters is that the required functions in the Abstract Model is implemented

<antoine> ... there are only two functions

<antoine> ... you could use whatever is suitable in your environment

<antoine> ... But I'm not sure we're on the same line

<antoine> annette_g: maybe, indeed :-)

<antoine> [cites section 7.2]

<antoine> annette_g: in the subsection called 'resource URI', example 4 for instance

<antoine> ... I was trying to fix the example

<antoine> ... my interpretation was that you were warning people not to use the query string

<antoine> ... to refer to the profile

<antoine> ... But it seems there is other stuff ongoing

<antoine> ncar: I think I see what you mean. Maybe we can take it up in the sub-group meeting

<antoine> ... the short answer is that it's not answered yet.

PWinstanley: I'll put out something probably on the other side of Easter

antoine: what are the plans of the subgroup with respect to the comments I made? Many of them were not answered, and I was wondering if you wanted to address them before the next working draft.

PWinstanley: I'm comfortable with pushing things forward so long as we are expressly saying in the note in the draft that there are still a whole pile of things being discussed. This should encourage others to recognize that they can comment if they sees things that are not to their liking.
… To my mind, a working draft is a working draft. But I want other people to be happy with that approach as well. Are you happy with that, Antoine?

antoine: yes, it's just that if the subgroup wanted to respond, I wanted to be able to react quickly in the coming days too.

PWinstanley: Are you okay with that Nick?

ncar: yes, we all have the holiday coming up

PWinstanley: I want to address things that we can in the time we have. If the vote comes back with a clean set of yeses, I'll check with Nick that everything is sorted out, and work with the w3c webmaster.

ncar: I think it's looking pretty good now.

PWinstanley: I can give a very brief update on the DCAT working group.

PWinstanley: there were three main things we were talking about. First was how we handle the backlog, whether we provide forward direction to those who might be involved in version 3. We are going to ensure there aren't any errors.
… second was around milestone 14. Riccardo was raising our awareness to the concept of a feature at risk. If we were not going to be able to finish off something, which might compromise other features.
… We also discussed how to handle acknowledgments.
… part of what's happening in the DCAT working group is tidying things up, because we have a lot of material that hasn't found its way into version 2. We just need to make sure we have a policy in place.
… I think we'll leave stuff about closing issues on the agenda for next time. When we do go out to the 2PWD for conneg, we need to make sure we track comments carefully.
… anybody else want to say anything?
… we have another meeting next Tuesday.