– DRAFT –
DXWG Plenary

Admin

Approve https://www.w3.org/2021/06/29-dxwg-minutes

<AndreaPerego> +1

<SimonCox> 0 - not present

<Caroline_> +1

+1

<DaveBrowning> 0 - not present

<riccardoAlbertoni> +0 (i was not present)

<Ana> +1

Resolution: Approved June 20 minutes https://www.w3.org/2021/06/29-dxwg-minutes

<kcoyle> +1

<annette_g> s/91832729517?pwd=UDlJQkJjMGp0U0VKT20wTHhZaVJUdz09/redacted/

https://lists.w3.org/Archives/Public/public-dxwg-wg/2021Sep/0018.html

Plh: we're now part of the calendar. check your inbox or login to the website to get meeting information

subgroup updates

riccardoAlbertoni: met last week after the break
… discussed different issues.
… 2 need attention

<riccardoAlbertoni> https://github.com/w3c/dxwg/issues/1358

riccardoAlbertoni: #1358 raised by APA WG
… asking an a11y considerations section
… we have a proposed section

<riccardoAlbertoni> https://w3c.github.io/dxwg/dcat/#accessibility.

riccardoAlbertoni: we asked for their feedback in May
… did not hear back
… can we close the issue?
… they can reopen it if needed

<riccardoAlbertoni> https://w3c.github.io/dxwg/dcat/#inverse-properties

plh: close the issue and I'll let the team contact know

riccardoAlbertoni: second issue is about inverse properties
… appointed Andrea to shape the section
… aim to have a starting draft

AndreaPerego: main issue was that we don't need them since they can be derived and may hinder interop
… if one provide a property and its inverse
… ie breaking consistency
… so we tried to propose a tentative solution to the problem
… based on 2 assumptions: the dialect

<SimonCox> What use-cases _require_ inverse properties?

AndreaPerego: some dialect don't need inverse. but some use cases do.
… initially, we'll only include the list of properties without their inverse and move the inverse in a separate section
… where we give recommendations there
… you must be aware of you should not use those properties alone

<SimonCox> In PROV-O they reserved the _names_ of the inverse properties, but did not include them in the RDF.

SimonCox: what use case are we trying to solve? RDF or graph model?
… you can reverse direction in those

AndreaPerego: some platforms are not RDF-based

<SimonCox> OK - so it is a consequence of the expressivity of the source or target technology

AndreaPerego: in other cases, when harvesting data across catalogs, you may want to have explicit properties
… probably, we don't have the full picture yet
… idea is to provide guidance without breaking interop
… we have a number of properties with their inverse from dublin core
… like replaces, IsReplacedBy

<SimonCox> Risk is that if they are available in the RDF, then people will use them anyway, regardless of the guidance in the spec (which they probably won't read)

AndreaPerego: trying to make it clear that the properties should not be used alone

<SimonCox> ... even if they don't need them

AndreaPerego: we're still discussing this topic
… so feedback is more than welcome. looking for discussion
… proposal is to fuel the discussion

SimonCox: explanation doesn't surprise me, I pushed not to have inverse props

<SimonCox> Guidance should be heavily based on examples!

AndreaPerego: again, we welcome feedback so please comment
… also welcome to join the DCAT subgroup

<SimonCox> * apologies for not joining meetings for many months. I try to keep up with issues on GitHub in lieu

AndreaPerego: use https://github.com/w3c/dxwg/issues/1336 or open a new one

annette_g: I support this approach. it's a good one

AndreaPerego: welcome suggestions for text improvements

AndreaPerego: the point here is that we have to decide which inverse props we'll want to define
… not all of them need to
… this will require a separate discussion

Caroline_: any other subgroup updates?

kcoyle: we met about and talked about how to approach the profile guidance
… but work is happening in the dublin core area
… including if a prof could be used in what we're doing
… not reviewed by the group but would like to present it here
… I'll meet again with Annette and see if we can come up with something

TPAC

Caroline_: it will be online again

https://www.w3.org/wiki/TPAC/2021/GroupMeetings#WG.2FIG.2FBG_Group_Meetings_details
… is the plenary meeting at TPAC ? or subgroup?
… or can we do a short video on our work?

https://www.w3.org/Guide/meetings/tpac.html

https://www.w3.org/2020/10/TPAC/group-updates.html

<annette_g> +1 for meeting at tpac

AndreaPerego: we should meet consider an opportunity for a longer meeting but we have to decide what we'll like to discuss, such as profile guidance

kcoyle: what is the general nature of those videos and how much does it take?

plh: general audience

annette_g: I'd look to use it as outreach as well

riccardoAlbertoni: we're expected to finish in January
… so how long will we keep working?

plh: charter runs out in January but we can always recharter
… should have a draft by mid-October and iterate then

Caroline_: we can a general video and talk about our work and maybe gather more people to get involved
… other thing is to think about the future of the group beyond January
… and is group willing to meet around tpac?

Caroline_: can we make a list of future work, issues that we'd like to solve?
… and then discuss them at TPAC?

<kcoyle> +1

<AndreaPerego> Don't know yet.

<annette_g> +1

riccardoAlbertoni: it will be difficult to finish by January

<PWinstanley> I am wondering if we need to think about the generalised cataloging issue and move away from 'data'

AndreaPerego: we should also consider going into next steps
… our use cases did not evolve since 2014
… and still, things changed. DCAT will become a silo

<PWinstanley> the "D" in "DCAT" is, to a great extent historical, coming from the interest in public sector open data

AndreaPerego: we had some unexpected feedback from users

<PWinstanley> we need to address the "F" in "FAIR"

AndreaPerego: can we do a "DCAT reloaded" ?

AndreaPerego: it would be nice to discuss that around TPAC, not sure who will present it
… otherwise we'll become a niche tech

PWinstanley: the D in DCAT is historical, due to its origin
… version 2, we got the possibility of extending any resource
… we should look to provide a generalisable "Findability" solution in FAIR

Caroline_: ok, proposal is to address these at the TPAC meeting
… DCAT, profile guidance

Proposal: meet around TPAC to discuss the future of DCAT, profile guidance, and this WG

<kcoyle> +1 90 minutes

plh: suggest 90 minutes

<kcoyle> +1

<AndreaPerego> +1

<riccardoAlbertoni> +1

<Caroline_> +1

<Ana> +1

<annette_g> +1

<PWinstanley> +1

<DaveBrowning> +1

<SimonCox> +1

<annette_g> wiki page

Resolution: Group will meet at TPAC

<kcoyle> github is good

AndreaPerego: if we prefer to fill a template, GH is the preferred

plh: I'll create a repo for the next charter and gather issues

Action: plh to create a repo for the next charter and gather issues

Proposal: Create a group update video for TPAC

<annette_g> +1

<kcoyle> +1

<AndreaPerego> 0

0

<riccardoAlbertoni> 0

<DaveBrowning> 0

<PWinstanley> 0

<Caroline_> +1

<Ana> +1

<SimonCox> 0

Caroline_: my suggestion is, if anyone is willing to record it, let me know

<SimonCox> @me I need to go

[adjourned]

<Ana> Bye