– DRAFT –
(MEETING TITLE)

Last meeting's minutes

<roba> https://www.w3.org/2023/07/26-dxwg-minutes

PROPOSED: Confirm last minutes

<roba> +1

<YoucTagh> +1

+1

RESOLUTION: Accept minutes of last meeting

Actions list

<YoucTagh> w3c/dx-connegp#42

PR is still in limbo, although the requested changes have been made...

roba: Have fixed the changes Antoine requested.
… We can dismiss the review because of the subsequent changes
… will merge the PR
… everyone fine with that?

+1

YoucTagh, +1

Close "due for closing issues"

PROPOSED: close all issues flagged as "due for closing"

+1

<YoucTagh> +1

roba: it was confirmed in the plenary that we can close those

<roba> +1

RESOLUTION: close all issues flagged as "due for closing"

roba: closing issues

Open Actions

<roba> w3c/dx-connegp#44

roba: Link relation w3c/dx-connegp#44
… We decided in last meeting to find a new link relation instead of "canonical"

YoucTagh: can we have two canonical links?
… That is not in line with the RFC
… How can we state that we prefer one profile over another?
… We can talk about other dimensions than profile and media type
… (e. g. language). We need a way to show preferences
… for all dimensions
… Representations can me multi-dimensional
… Solution proposal
… There is a global preference (default/fallback)
… proposes three types of relations;
… g-def-neg global default representation
… d-def-neg dimension specific representation
… a-def-neg alternate representation
… In <link> we can use those relation types to show whic is the preferred representation for each dimension

roba: all other link relations are simple word (according to the registry)
… not sure that g-def-neg is different from "canonical"
… and a-def-neg is same as "alterate"
… so only d-def-neg is new

YoucTagh: The semantics aren't the same

roba: does "a-def-neg" really differ from "alternate"?
… we shouldn't create new terms
… the naming of d-def-neg should rather be "preferred"

<roba> https://www.iana.org/assignments/link-relations/link-relations.xhtml

roba: in order to keep in line with naming in link registry
… where all terms are just short english words
… would prefer "preferred" instead (under identifiable circumstances)
… Question really is if "canonica" is really single-valued

<roba> <http://example.org/resource/a>; rel="preferred"; format="urn:example:profile:y",

YoucTagh: This doesn't specify for which dimension this is preferred

roba: YoucTagh should put his example in the issue
… so that Nick can pick that up and make an implementation
… to get implmentation evidence

ACTION: YoucTagh to add proposal to the issue

LarsG. Discuss #10 w3c/dx-connegp#10

roba: a non-issue

LarsG: proposes to say something on "related documents" inluding the I-D

roba: and remove PROF-Guidance

ACTION: LarsG to re-phrase "family of doczments"

LarsG: w3c/dx-connegp#12

roba: related to #26
… general issue is that's there is no value in defining
… statements about rdfs:Resource
… and we should remove them

ACTION: roba to prepare a PR to fix #23, #26 and #27

roba: #32 w3c/dx-connegp#32

<YoucTagh> https://www.w3.org/TR/2019/WD-dx-prof-conneg-20191126/#fig-table-profiles

roba: think it's fine as it is. We should just close

LarsG: Do we define prefixes somewhre else?

<YoucTagh> https://www.w3.org/TR/2019/WD-dx-prof-conneg-20191126/#altr-owl

YoucTagh: yes

LarsG: Fine with the way it is

roba: some profiles are not defined but used in the text
… we should do an editorial cleanup
… will assign that to Nick

ACTION: roba to discuss #32 with Nick and prepare PR

<roba> w3c/dx-connegp#7

roba: Is the wording of §7.1 clear?
… we need to review https://www.w3.org/TR/2019/WD-dx-prof-conneg-20191126/#functional-profiles-definition

ACTION: YoucTagh to review §7.1 https://www.w3.org/TR/2019/WD-dx-prof-conneg-20191126/#functional-profiles-definition

roba: other issues to discuss?

YoucTagh: when we resolve the issue with canonical
… that will influence the turtle examples

bye all