– DRAFT –
DXWG Plenary

admin

<PWinstanley> https://‌www.w3.org/‌2019/‌08/‌27-dxwg-minutes

proposal: to accept minutes of last meeting

TomB: topic 4 should be topic 5
… I've typed the wrong topic. Topic 4 should be deleted/merged with next topic

+1

<AndreaPerego> 0 (was not there)

<PWinstanley> 0 - not there

<ncar> +1

<TomB> +1

<Makx> 0 - not there

Resolved: to accept the minutes of last week

Review of deadlines

PWinstanley: kcoyle will you put something on the main page of the wiki?

kcoyle: I'm working on it
… we need something to present versions to the group and group decisions
… the date for final move to DCAT CR was in Oct I think
… there are some deadlines mid-September (I think) for making decisions on Conneg and PROF

Action: kcoyle to put a table of deadlines on wiki

<trackbot> Created ACTION-363 - Put a table of deadlines on wiki [on Karen Coyle - due 2019-09-10].

DCAT update

PWinstanley: for DCAT there have not been meetings on the past two weeks.

<RiccardoAlbertoni> yes we have a meeting tomorrow

PWinstanley: I expect tomorrow's meeting will discuss deadlines we've just mentioned.

<AndreaPerego> Not from my side.

Conneg update

<ncar> Conneg is 3PWD candidate: https://‌w3c.github.io/‌dxwg/‌conneg-by-ap/

ncar: I've got access to github now and could merge the PRs
… we're meeting tomorrow

<AndreaPerego> Just to note respec reports errors/warnings.

PWinstanley, ncar: we need to coordinate on a poll for 3PWD

<ncar> Yes, we know! A link we expect to fix

kcoyle: there is a discussion about including tokens in Conneg
… how would a decision on this affect the timetable?

ncar: like anything, if there's a serious case for change
… there is a whole section there, motivated by use case
… there is a discussion on whether these cases are important
… the question is the positioning of the IETF about this.
… it could be about not disallowing tokens.

kcoyle: we need a drop-dead date for making a decision on this, for passing to next WD

<AndreaPerego> Or mark it as a feature at risk?

PWinstanley: we've got experience about marking features at risk in DCAT
… this could allows to avoid the full circle of validations

<AndreaPerego> Yes, this is what I meant.

<plh> https://‌github.com/‌w3c/‌dxwg/‌wiki/‌Steps-to-Recommendation-2019

plh: drop-dead dates are on the wiki page
… features at risk is features that can be removed, or made informative, without affecting the structure of the rec
… this allows experimenting with the feature until the end of the CR

without this, if you want to remove the feature you have to create a new CR and wait for 60 days for doing so.
… usually the price is possible hurt feelings. But in doubt, it's wiser to declare features at risk.

s/structure of the spec/consistency of the spec/

Pull requests and closing of issues

PWinstanley: the past process (based on informative flagging) was not satisfactory
… some issues were re-opened because there was no clear evidence that the issue was resolved
… I'd like a formalized procedure for closing issues
… We have a table each week, and the opportunity for the plenary to make a formal statement that the issues can be closed.

plh: you're running out of time
… you're supposed to make a decision on moving to CR in two weeks
… are you suggesting to go through all these issues?

PWinstanley: no, what we want to avoid enthusiastic closing of issues

<PWinstanley> antoine: I'm not sure what you're suggesting?

PWinstanley: we want to give people an opportunity to jump in

<Zakim> TomB, you wanted to ask who is responsible for tracking issues

Antoine: what precisely will happen?

PWinstanley: we would discuss issues in plenary

TomB: the list is overwhelming
… the deadline is next Friday (the 13th?) and some of the issues have been opened for one year
… Whose responsibility is it to track this?
… Should responders track the issues that correspond to their feedback?

<ncar> Which issue Tom?

TomB: I was involved in an issue and it's not clear why it would be closed

ncar: now I'm replying to people, detailing what has been done and this happens every week.
… all issues were re-opened

PWinstanley: it's not you ncar . there needs to be a ping-pong
… we need to sort things out.

ncar: we try to do this every week.

plh: can we get into the specifics?

<ncar> https://‌github.com/‌w3c/‌dxwg/‌issues/‌698

https://‌github.com/‌w3c/‌dxwg/‌issues/‌402

<PWinstanley> antoine: My approach is that I've seen and contributed to this list of issues and there is an issue proposed for closing, if I have an objection or make a minor comment then this is OK,

https://‌github.com/‌w3c/‌dxwg/‌issues/‌467

<PWinstanley> ... It is completely OK for people to look at these issues and make comments - this doesn't need to have plenary activity, it can be done bilaterally between the author and the person raising the issue

<PWinstanley> plh: Can I suggest that use of thumbs up icons will help show that the proposal is agreeable

<RiccardoAlbertoni> +1 to antoine ( asking for objections is enough in my opinion)

<ncar> +1 to antoine

<Zakim> TomB, you wanted to point out https://‌lists.w3.org/‌Archives/‌Public/‌public-dxwg-comments/‌2019Feb/‌0001.html

<ncar> +1 to thumps up

TomB: I understand the process is stressful
… is it the responsibility of commenters to keep track of whether their comments have been answered?

plh: from a W3C process we expect answers to comments

TomB: if the comments are made on the public list, shouldn't the answer go there?

plh: it could be on github, or on the list

ncar: I saw the email, I don't think it has gone through issues
… I would still like to do this
… and answer both on the list and github
… I've done this today for 3 commenters
… I haven't got to TomB's one yet

<PWinstanley> antoine: I've several points: - I've seen public comments and answers and this is what TomB was thinking about, it is the right direction

<PWinstanley> ... - the process of creating issues is important. The public comment list is not so crowded and it is easy to identify emails that haven't been commented, so we can clear these up quite quickly

<PWinstanley> ... The issues that have been closed - I've been working with Nick to make the list. It is not intended for closing really hard issues, but for the simple cases. We don't want to have unresolved questions, just to have some good housekeeping

<Zakim> TomB, you wanted to ask whether today's comments to https://‌github.com/‌w3c/‌dxwg/‌issues/‌792 will be turned into an issue (as an example)

TomB: there is a comment today in https://‌github.com/‌w3c/‌dxwg/‌issues/‌792
… issues about the design of the spec
… issues can be used to explained the rationale
… it's hard to see the fundamental issues
… they're getting snowed under

PWinstanley: for DCAT we decided to shelve issues for the next version
… we could this, ensure we've got documentation and point to the fact that a subsequent team would update
… plh would this idea of backlog work?

plh: it could, assuming that the issues can be fixed.

PWinstanley: there were fundamental issues on DCAT like versioning, that we left

plh: the editors need to triage

plh: we cannot change the architecture between v1 and v2.

<PWinstanley> antoine: I was at the call discussing this process previously, and getting issues on the list was part of the triage, to remove the small items and raise the profile of the larger issues

plh: I understand that TomB and ncar are stressed, but a decision needs to be made in two weeks
… if we don't have the time, we don't have the time

TomB: this is too much work

PWinstanley: yes but we need to try, give each other feedback

TomB: I think a WG Note would be a good outcome

PWinstanley: if we miss the drop-dead dates we'll move from CR to Note

plh: and the priority is to update DCAT

<RiccardoAlbertoni> Thank you, Bye !