– DRAFT –
Dataset Exchange Working Group Teleconference

Resolved: approve last week's minutes

<antoine> sounds good

<kcoyle> https://‌www.w3.org/‌2017/‌09/‌25-dxwg-minutes#ResolutionSummary

kcoyle: asking for approving the requirement "6.7 Version indicator"

<kcoyle> https://‌w3c.github.io/‌dxwg/‌ucr/#RVer3

PROPOSED: accept RVer3 with name changed to version indicator

<Thomas> +1

+1

<annette_g> +1

<newton> +1

<AndreaPerego> +1

<Stijn_Goedertier_AIV> +1

<dsr_> +1

<Makx> isn't that 6.7 Version identifier [RVer3]?

<Makx> 6.9 is Version release date [RVer5]

<Ixchel> Provide a means to indicate a version (URI-segment, property, etc.). Clarify relationship to indicator of the subject resource.

Ixcchel: clarify the second sentence of requirement ..

<Ixchel> Provide a means to indicate the type of version (URI-segment, property, etc.). Clarify relationship to indicator of the subject resource.

<LarsG> +1

<roba> +0

<annette_g> Isn't that to indicate the version itself?

<AndreaPerego> Can I suggest we add an explicit pointer to https://‌www.w3.org/‌TR/‌dwbp/#VersioningInfo ?

I can't hear Makx ...?

o.k.

<PWinstanley> +1

<Makx> +1 to Andrea

Resolved: accept RVer3 with name changed to version indicator

<Makx> +1

Ixchel will update the UCR document

kcolye: please indicate presence in F2F

kcolye: register at time, this is needed to plan the room capacities

<Ixchel> Clarification: Ixchel is updating the UCR working document on Google Drive

kcolye: we'll have a remote connection, difficult because of Californian time zone

kcoyle: a number of previously "closed" actions appear in tracker as "open"

dsr: _ will clarify why

kcoyle: Ruben + Lars, please check conneg reqs.

<Zakim> LarsG, you wanted to talk about action-40

LarsG: we looked at it, created pulled requests, were accpeted today

kcoyle: will lcose corresponding actions once the tracker problem solved

<annette_g> s/licked/lloked/

kcoyle: the action 42 to be resolved next week, too short time note

<kcoyle> https://‌docs.google.com/‌spreadsheets/‌d/‌16JmtNCz_aCWtTCSntriDWLvyPY2x-Y9dZFhAHFl55r0/‌edit?usp=sharing

<Zakim> LarsG, you wanted to ask if it's possible to have the github notifications sent to the dxwg list

kycoyle: we'll look at the proposed spreadsheet

LarsG: request to send the Github notifications to group

<AndreaPerego> +1 to LarsG proposal

<antoine> +1

dsr_: will look at the option to circulate GitHub pull request notification to DXWG mailing list

<antoine> we've talked about this after Phil moved. And it was one of the motivations for everyone accepting to move to github I think ;-)

Action: will look at the option to circulate GitHub pull request notification to DXWG mailing list

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

Action: dsr_ will look at the option to circulate GitHub pull request notification to DXWG mailing list

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

Action: dsr to look at the option to circulate GitHub pull request notification to DXWG mailing list

kcoyle: back to spreadsheet..

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

kcoyle: https://‌docs.google.com/‌spreadsheets/‌d/‌16JmtNCz_aCWtTCSntriDWLvyPY2x-Y9dZFhAHFl55r0/‌edit?usp=sharing

kcoyle: explaining the structure

<kcoyle> PROPOSED: UCR group to complete spread sheet and numbering for requirements discussion

<annette_g> +1

Jaroslav_Pullmann: will we stick to a general req. title or have dedicated label for each?

<roba> +1

<Thomas> +1

<Ixchel> +1

<dsr> +1

<newton> +1

<PWinstanley> +1

<LarsG> +1

<antoine> +1

Will this replace the more interactive wiki?

<riccardoAlbertoni_> +1

Resolved: UCR group to complete spread sheet and numbering for requirements discussion

<kcoyle> https://‌github.com/‌w3c/‌dxwg/‌pull/‌35/‌commits/‌a8ed32e797023d73a2509da0c4fdec5b8240dcfd

kcoyle: Peter had an action to update wording
… of https://‌w3c.github.io/‌dxwg/‌ucr/#RVer1

ktoyle: asks Peter co comment

PWinstanley: should we rely on DC isVersionOf or other mechanism

roba: afraid, that we are getting into a solution space

<Makx> +1 to roba

<riccardoAlbertoni_> +1 not to discuss about solutions at this stage

roba: we should word in more generally requesting group to finde right solution

<Makx> but also agree that dct:isVersionOf may well play a role when we get to potential solutions

<kcoyle> Jaroslav_Pullmann: as per alejandra we should define which resources need versioning

I expected this requirement be replaced by Alejandras proposal

Alejandras proposal was laid down in this action: https://‌www.w3.org/‌2017/‌dxwg/‌track/‌actions/‌44

kcoyle: we will discuss this later, when ther is a proposal from Alejandra

<kcoyle> PROPOSED: table ver1 for now; wait for Alejandra's action

<Makx> +1

<riccardoAlbertoni_> +1

LarsG: will "profiles" be considered "subject to versioning" as well?

kcoyle: good point, should be considered

<Makx> +1 to think about versioning profiles

roba: there are cross-references among deliverables

<antoine> +1 it would be very important

roba: therefore the solutions should bear this in mind

roba: make sure the backlinks to UCR document (from other deliverables) will work

there will be a deliverable tag in the GDoc spreadsheet

talking about: https://‌www.w3.org/‌2017/‌dxwg/‌track/‌actions/‌43

PWinstanley: definition of what is considered a version

antoine: as already said, I would not try to define the conditions of when to create a new version

antoine: this should be defined by communities/profiles

<Makx> +1 to antoine

<annette_g> +1 to Antoine

roba: do we want to define a canonical version model or provide a canonical way (hook) to attach whatever model to resource description

annette_g: allow people to use versioning at level of they needs

annette_g: I do not expect this to be that relevant with regard to search

roba: there are situations, where temporal ordering of versions might become important

<PWinstanley> s/thisshould/this should/

roba: simple string might not be suffcient

annette_g: how far should we go in defining a versioning model?

annette_g: people are used to their versioning methods and their hardly use a prescribed system

LarsG: supports annette_g

LarsG: do we have requirements demanding a structured version model?

<roba> -1 to list of microformats , -1 to prescribing a specific model :-)

PWinstanley: we should provide a conceptual definition of "version" and equally the out-of-scope def: what is not a version

<roba> is there a discussion planned around telcon times - now 2am for me after change in times :-(

antoine: summarizing Peter's proposal

<LarsG> I definitely don't want to prescribe a specific model but want to emphasise that if we think that something is important (e. g. finding out which is the latest version; finding out if version XY is a predecessor to YX) we should have requirements for that

PWinstanley: would we e.g. consider "deduplication" a rationale for version change?

<Stijn_Goedertier_AIV> +1 to Peter to provide some guidance on what should and should not be considered to be a version.

kcoyle: still open, will be discussed later on

<annette_g> Thanks folks!

<antoine> thanks!

<riccardoAlbertoni_> bye thanks

<LarsG> Thanks, bye

<Makx> ok thanks bye bye

<PWinstanley> thanks ... bye!!

bye!

<AndreaPerego> Thanks, and bye!

<Stijn_Goedertier_AIV> thanks everybody. bye