W3C

– DRAFT –
DXWG profile guidance subgroup

19 September 2018

Meeting Minutes

zakin, who is here?

admin

<ncar> https://‌www.w3.org/‌2018/‌09/‌05-profgui-minutes.html

<ncar> last meeting's minutes

<roba> +1

<ncar> +1

rrsagent: make logs public

0

<kcoyle> +1

<ncar> topis: Review open Actions for this subgroup

Resolved: minutes of last call approved

<ncar> https://‌www.w3.org/‌2017/‌dxwg/‌track/‌products/‌3

action-124 andrea not here

action 200 is done

close action-200

ncar: action 201 is done

close action-201

ncar: action 202. antoine suggested to change position of definitions so let's leave this one open

ncar: action 203. there's a section but empty

close action-203

close action-204

ncar: 205 is still open

ncar: 206 kcoyle this is what we've talked yesterday
… actions with no 'plenary-approved' tag
… do you want me to do this?

kcoyle: I've sent the list

ncar: I'll do it

ncar: 208 we've sent a mail to the group
… we'll find the link later, let's keep it for review
… 209 and 210 are not yet done

review update of document

<ncar> https://‌w3c.github.io/‌dxwg/‌profiles/

<ncar> https://‌w3c.github.io/‌dxwg/‌profiles/‌index-structure242.html

ncar: we've discussed it yesterday
… would I do a PR for the requirements?

<kcoyle> antoine: put requirements in an appendix; distribute them into sections

ncar: I can update the index with index-242 and then a PR to include the requirements as a new section in appendix

<roba> +1

kcoyle: yes
… in the DCAT document they seem to have a section in the end about requirements
… most documents I've seen don't have a list. People just make sure they've addressed them

ncar: it's nice to keep track

kcoyle: yes some sort of notetaking

Action: ncar to merge the two documents retaining requirements as an appendix

roba: in principle we could just rely on issues

Select a range of Requirements for demonstration Responses

ncar: 1,2 and 5 look good as examples of different reqs
… for which responses would have different styles
… [reads profile and gives hint at why they're different]

<ncar> https://‌w3c.github.io/‌dxwg/‌profiles/

kcoyle: I'm not sure whether it's better to look at specific requirements
… we don't have yet the context

ncar: the reason is related to what's happening in the conneg group
… most requirements are answered by the RFC
… In our case some requirements might be addressed differently
… i.e. by some discursive comment on the nature of profiles
… or by adding a dedicated element in ProfileDesc

ncar: I want to do this only on a couple of requirements

roba: we may decide that it's not the way to do it, based on this so it would be useful.
… also we've spent a lot of time so it would be useful to talk back to the previous work.
… also for identifying when some requirements are met by existing proposals.
… and these which need something new
… it would be good to triage them and send them to different people.

ncar: I agree
… after this we could see a pattern
… req 7 could be a good one too.

roba: on using constraint languages, it's not that simple because constraint languages tend to provide patterns
… but there could be aspects that are not addressed by constraint languages.

ncar: so I could write something and you could react

<kcoyle> antoine: too soon because group hasn't reacted to structures but good to pick a few and try

<kcoyle> ... concern is timing not utility of exercise

ncar: on some we could have an answer
… it's something I would like to write down.

kcoyle: could we do it in the github issue so that we can have the discussion?

<roba> +1

kcoyle: instead of PRs

+1

<roba> wait for issue to converge on consensus - but a couple of examples now is also good

ncar: it could be a good idea

ncar: I can have a first go today

antoine: kcoyle are you thinking of the issues as handled in the Dublin Core Usage Board?

kcoyle: yes

antoine: I'll send them to ncar for inspiration

<roba> +1 - seems to be no barrier

antoine: see kcoyle 's email. How to refer to a non-normative doc from a normative one?
… I've tried to re-use the way it was done in DWBP

kcoyle: we should just try and see if it passes
… in my mail I've pasted the answer from Philippe
… if you look at DUV and DQV and how they're refered in DWBP there is subtlety
… we need to be sure there is subtlety in the doc when it is submitted

ncar: we could use requirements

kcoyle: I'm not sure we'll get anyone to look at it before it's a WD

ncar: so then we can give it a go

kcoyle: yes

https://‌w3c.github.io/‌dxwg/‌profiles/‌index-structure242.html#ProfileDesc

<roba> MAY seems to work well

<roba> in combination with SHOULD use a published vocabulary

kcoyle: I'd like us to develop some form of structure, discussing things before they become PRs or commits
… so that we have agreement beforehand

ncar: unless we leave them as notes

roba: notes could be ok with references to issues

kcoyle: first is the issue

ncar: it doesn't get in the text until the issue is approved

ncar: I will create github issues for the current issues in red and then we can go forward

rrsagent. please draft minutes

ncar: we'd like to change the name of ProfileDesc slightly

ncar: ProfileDesc was not ambiguous

Summary of Action Items

  1. ncar to merge the two documents retaining requirements as an appendix

Summary of Resolutions

  1. minutes of last call approved
Minutes manually created (not a transcript), 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/it's that simple/it's not that simple