– DRAFT –
Data Exchange Working Group Teleconference

<dsr> what, that’s weird, let me try

<dsr> Hmm, weird, something must have gone wrong!

<dsr> We can use my WebEx room as a fall back

<dsr> https://‌mit.webex.com/‌join/‌draggett

<dsr> can you please try that link and we can then email everyone

<kcoyle> ok, I'm on.

<LarsG> hi all, do we have consensus on when the meeting will take place?

<Ruben_Verborgh> Can somebody please help me get on WebEx?

<Ruben_Verborgh> I can't get on https://‌lists.w3.org/‌Archives/‌Member/‌w3c-archive/‌2017Aug/‌0066.html

<Ruben_Verborgh> never mind got the mail

<roba> i'm gettiong a "meeting cancelled" message on the webex

<roba> i've checked and cant see any emaiul with a new link - this is via the link in the members only wiki page...

<roba> well it might take up to an hour for google to do its thing and send it through to me...

<roba> I'll make my apologies now for next week - I'll be travelling

<Ixchel> Hi all. I'm waiting for the lost to admit me

<roba> skype rob_cto_sco

<Ixchel> lost s/host

<PWinstanley> @alejandra - please see your email from me

<alejandra> PWinstanley: I replied - link to webex also above

F2F

<PWinstanley> @alejandra - yes thanks. on slow connection though

kcoyle: registration is needed (but it's free)
… deadline is Oct 7

last week's minutes

AndreaPerego: people may not be noted

<Caroline> https://‌www.w3.org/‌2017/‌09/‌11-dxwg-minutes

<kcoyle> https://‌www.w3.org/‌2017/‌09/‌11-dxwg-minutes

Resolved: approve last week's minutes https://‌www.w3.org/‌2017/‌09/‌11-dxwg-minutes with adding missing present

Resolved: accept last week's minutes, adding any present that are not recorded

<annette_g> +1

open action items

UCR

kcoyle: there's just one content negotiation requirement
… when one uses filter

Action: LarsG and Ruben_Verborgh to check requirements for content negotiation

<trackbot> Error creating an ACTION: could not connect to Tracker. Please mail <sysreq@w3.org> with details about what happened.

Ruben_Verborgh: I already see that there are tags missing.

kcoyle: you can do a pull request if you want to add tags yourself.

Ruben_Verborgh: will do

kcoyle: next question: adding more tags
… but first: does this tagging/filtering work for the group?
… because the entire group has to agree on this

Jaroslav_Pullmann: action 37 to better organize the tagging

<Jaroslav_Pullmann> https://‌w3c.github.io/‌dxwg/‌ucr/

Jaroslav_Pullmann: there are now changes

<Jaroslav_Pullmann> lost audio again, will retry

<alejandra> it works for me

kcoyle: is it ok for others?

<PWinstanley> not looked at yet , but will do soon

annette_g: it seems to collapse most of the UC. I still see the description

who?: they might not have been tagged

<Jaroslav_Pullmann> audio fails on Chrome and FFox, will restart, sorry for the inconevnience

alejandra: yes

<Ruben_Verborgh> Added conneg tags here as requested: https://‌github.com/‌w3c/‌dxwg/‌pull/‌30

kcoyle: is there tagging of requirements?

<Jaroslav_Pullmann> back again

kcoyle: I thought that requirements appeared because of the links to UC

roba: requirements will get tagging of their own

roba: it's still work in progress
… the main bunch has been done
… we can continue if people are ok

kcoyle: we need to see all requirements for DCAT, profiles...
… so that to see that it's complete

roba: what is done now is the requirements that are across deliverables.

roba: they're grouped by function not by deliverable now

<kcoyle> https://‌w3c.github.io/‌dxwg/‌ucr/

Jaroslav_Pullmann: action 37 about re-organization tags and filtering
… table is gone now, explanation is provided as tooltip
… we can have different groups of tags
… deliverable [etc]

kcoyle: people can look and comment via mail

versioning requirements

https://‌lists.w3.org/‌Archives/‌Public/‌public-dxwg-wg/‌2017Sep/‌0034.html

<kcoyle> 1. Provide a definition of version, including how it relates to a dataset distribution.

Jaroslav_Pullmann: this is the old wording. Label and content have changed

<Jaroslav_Pullmann> 6.5 Version subject [RVer1] Identify DCAT resources that are subject to versioning, i.e. Catalog, Dataset, Distribution.

kcoyle: it would be great to have a notivation by email Jaroslav_Pullmann

<Jaroslav_Pullmann> audio gone, https://‌w3c.github.io/‌dxwg/‌ucr/#RVer1

<AndreaPerego> About versioning, I would also add catalogue records.

<alejandra> maybe change "identify" to "determine"

<kcoyle> ack ?

antoine: not sure what 'identify' means: giving an identifier or defining what should be versioned

kcoyle, alejandra: define or determine, better?

antoine: yes

Makx: it is general. 'yes of course' would be my first answer, but what next?

<Jaroslav_Pullmann> Makx, which req. are you talking about? I lost the sound..

Makx: instead of 'version subject', 'version information'?

<PWinstanley> Suggestion: "what are the domain classes for the version predicate?"

<kcoyle> define DCAT resources for which versioning information is desirable

Jaroslav_Pullmann: are we talking about catalog/dataset versioning?
… identify which DCAT concepts are to be versioned.

<AndreaPerego> What about "entities" instead of "concepts"?

alejandra: about the title: I would prefer to have numbering of requirements

<PWinstanley> AndreaPerego: +1

<Jaroslav_Pullmann> +1

alejandra: about 'desirable' maybe we should have the level of requirement

<kcoyle> define DCAT resources for which versioning information is desirable or required

kcoyle: I too would prefer a list of actual requirements

<alejandra> AndreaPerego: do you mean "entities" instead of "resources"?

<Jaroslav_Pullmann> @Karen - are the headings in this "nominal style" better suited?

<AndreaPerego> yep, alejandra

annette_g: it's a requirement for defining a requirement?

kcoyle: yes it's meta

<PWinstanley> AndreaPerego: don't you mean classes of entity?

<roba> sorry - dropped internet :-(

<AndreaPerego> Well, I'd go just for "entities", PWinstanley

annette_g: who defines the actual requirement?

kcoyle: I thought it was up to the DCAT group

roba: we need a definition

annette_g: I don't disagree but it should be done at the right level (group or sub-group)

kcoyle: the DCAT group needs to report back to the main group
… we get a say

annette_g: I'm afraid that we'd end up doing more than what we need to do
… ie. do a separate kind of versioning for each of the concepts.
… we need to reckognize it's a requirement-level discussion

<alejandra> the requirement level belongs to the application profiles IMO

kcoyle: we can have a group action to decide on the level, and then ask the sub-group to come with a proposal

annette_g: this could be ok

Jaroslav_Pullmann: there might be different considerations e.g. for datasets.
… once we know what we are going to version, we need to do definitions on what will change.

kcoyle: it's the second step that we need to pay attention to.
… I'm not sure it's included in the current reqs

alejandra: we're still looking at requirements to identify vocabulary
… and defining whether it's required belong to different APs

Makx: the different requirements could be a charter for the DCAT sub-group
… our programme for that group
… for 6.5 do we decide at this point to exclude versioning of catalogue records?

roba: it wasn't clear to me whether the final solution was going to be in DCAT or APs
… I thought that there would be hooks/properties in DCAT, simply.
… I support the proposal that the DCAT sub-group should offer this.

kcoyle: so we have modified the wording of the 1st req. Shall we look at the 2nd?

<kcoyle> Provide a conceptual definition of what is considered a version with regard to modifications of the respective subject. The definition should provide a clear guidance on conditions, type and severity of a resource's update that motivate the creation of a new version in scenarios like dataset evolution, conversion, translations etc.

kcoyle: very related to the 1st req

Jaroslav_Pullmann: we need to say what we can consider to be a version - what motivates the creation of a new version.

kcoyle: you think it's a decision that should be made in DCAT rather than in APs?

roba: it should be pending off to profiles

annette_g: profiles are the places to be specific
… we need to point at the right places

<alejandra> antoine: I also agree. I don't like the idea because if the criteria is different it can bring several issues for interoperability

annette_g: but we have no choice

alejandra: I also agree. We need to be more specific about the scenarios we want to deal with.
… it's risky to let them imprecise

kcoyle: determining which scenarios need versioning?

alejandra: yes

kcoyle: it sounds like a re-writing is needed

alejandra: first solution would be to remove the 'etc'
… need to go back to the UCs

roba: we're not making assumptions that there's a single profile
… but that there can be several profiles and we give guidance for that.
… so we don't need to enumerate them all.

alejandra: how to we satisfy the requirement then?

roba: we don't need to be complete

<alejandra> antoine: I agree with roba - etc can be used to handle cases where there are several options

PWinstanley: it's helpful to work out what the model for DCAT is

<Jaroslav_Pullmann> +1 @Peter

PWinstanley: the versioning predicates.

<roba> +1

kcoyle: could you attempt to re-write so that we can discuss it in mail?

<PWinstanley> will try

<PWinstanley> yes please

Action: PWinstanley to suggest a re-write of RVer2 "version definition"

<trackbot> Error finding 'PWinstanley'. You can review and register nicknames at <https://‌www.w3.org/‌2017/‌dxwg/‌track/‌users>.

<alejandra> thanks and bye!

<annette_g> Bye folks!

<LarsG> Thanks, bye

<AndreaPerego> Thanks, and bye!