W3C

– DRAFT –
DXWG Plenary

05 March 2019

Meeting minutes

<Makx> presenr+

<kcoyle> https://‌www.w3.org/‌2019/‌02/‌26-dxwg-minutes

Admin

<kcoyle> proposed: accept minues of Feb 26

<kcoyle> +1

+1

<ncar> +1

<alejandra> +1

Resolved: accept minutes of Feb 26

kcoyle: Open Actions

<roba> +1

<annette_g> sorry, I'm going to have to miss today's meeting.

kcoyle: there are actions from the conneg work that are open
… we still have the gap analysis for the guidance doc, but we need to wait until that doc is further forward

ncar: that gap analysis is long-standing, but will only be done when there is more of a doc
… I have completed #305
… we ran a sprint and pulled together text that is now in the draft
… following agreement from roba and LarsG

<kcoyle> https://‌docs.google.com/‌spreadsheets/‌d/‌1Jo8pMcYbJBl8jNezeHTml7HfeqO8fnTnc3swVXEiS3I/‌edit#gid=0

kcoyle: For keeping track of all the feedback, PWinstanley began a google sheet that is far from complete. Do people think this is useful?
… the complication is the discussion in github, which is being labelled as feedback, but is often part of a complicated thread where the comment might come in somewhere deepeer in thread

<AndreaPerego> Just to note that PROF and ODRL profiles were discussed in today's ODRL CG call: https://‌lists.w3.org/‌Archives/‌Public/‌public-odrl/‌2019Mar/‌0002.html

ncar: we are getting good contributions that we can respond to that are entirely related to conneg

kcoyle: what is the w3c position for github comments?

dsr: are they coming in as issues? using the labels would help,but I'm not aware of any W3C formal approach. we are still evaluating github and this can feed into that

kcoyle: github has been very fruitful, but is harder for discussion and also tracking that the response is to the satisfaction of the commenter

ncar: we ask the other people who comment on the mailing list to look at the issue when we transfer it to github

<ncar> I've emailed Annette re ACTION-306 so please can that ACTION be closed here?

dsr: the disposition of comments will mention the comment and its resolution in a document

kcoyle: this could be incorporated into the spreadsheet if we add columns for resolution and the agreement that it is satisfactory

subgroup reports

kcoyle: how close are we with DCAT to another draft?

alejandra: re: issues and comments, we encourage people to put things in issues

alejandra: re: DCAT, we are trying to define a date and will then guillotine
… dsr can perhaps guide on the implementation report. Does it need to be complete when going to CR?
… according to the calculator that PWinstanley guided us to , we need to stop in mid March.
… there are many things we still need to complete. Given this is an update of an existing implementation, do we only need to provide evidence on the gaps that we have filled?

dsr: I'll ask Phillipe

kcoyle: can alejandra say something about the importance of things that might not be included? anything vital?

alejandra: we haven't included versioning. There is a PR from riccardo, and Jaro is preparing some material. I think it is an important area
… other things are more detailed

kcoyle: if we feel there are essential things we don't have time for we can ask for an extension, but we need to know what we are wanting to do

roba:

roba: the profiles ont - questions on attaching metadata props. we have an alignment to DCAT and we might make it a subClass of dcat:Resource
… is this helpful as an implementation?

alejandra: I don't have the answer, but it would be helpful since we think of dcat:Resource as an abstract class , an extension point

dsr: I will try to formulate the question and ask

kcoyle: we need a drop-dead date for DCAT and then a gap analysis to determine if we need an extension
… we are well organised re: DCAT

Profile Vocabulary

kcoyle: profiles vocab has some significant issues still.
… the addition of roles, which we voted on, and fixing definitions (makx referred to some time back)
… we can use this issue to discuss it

<kcoyle> https://‌github.com/‌w3c/‌dxwg/‌issues/‌755

kcoyle: these are 2 things we would like to get into the next draft as they will be explanatory. How do we do it?

ncar: we are poised to add the roles to the doc, we just need to determine which ones we put in the doc. we are still determining the names for roles

<roba> istribution, not resource

ncar: we need to work through the implications of making a profile a subClass of dcat:Resource
… we are trying to work out if we can reuse more than we did in the past

kcoyle: the question is how we do this - it needs plenty of discussion with the largest group possible. Github?

ncar: that is a good idea. The other option is a discussion within a drafting sprint

<ncar> * I've put a Doodle Poll up for a sprint

roba: I asked ncar if we can have a sprint, and we can include this discussion in that. We can use the googledoc mode, but if people can't join that then the github issues are there and people can contribute to that if they cannot manage to be in the sprint

kcoyle: we need a clean github issue without other stuff, ideally

roba: we have decided to do that

<alejandra> About the spreadsheet for comments (https://‌docs.google.com/‌spreadsheets/‌d/‌1Jo8pMcYbJBl8jNezeHTml7HfeqO8fnTnc3swVXEiS3I/‌edit#gid=0), I think it should include a column with the document they refer to (DCAT, PROF Voc, ProfGui, Conneg)

kcoyle: we only decided that we include the roles, but we haven't discussed the specifics of which roles to include

ncar: that is exactly what we need to discuss.

<roba> ok make a clean issue for the roles definitions

kcoyle: these are roles that ncar and roba have discussed, but have others?

ncar: simon cox has contributed,

<roba> roles have been exposed in examples in FPWD

<roba> will make the issue now

actiion: roba to open new issue seeded with list of roles and definitions and ask for comment

Action: roba to open new issue seeded with list of roles and definitions and ask for comment

<trackbot> Created ACTION-307 - Open new issue seeded with list of roles and definitions and ask for comment [on Rob Atkinson - due 2019-03-12].

ncar: other roles can be added later

kcoyle: we can mention OWA in the doc in context of roles
… we don't need to define the extension mechanism

ncar: this is the issue I have been against, why I've thought we needed an explicit roles definition so that people look elsewhere for them. I've been trying to encourage people to use their own roles rather than just relying on what is included

kcoyle: we want the roles in the next draft to elicit comments, not to include them in the vocab.
… they can be just a bullet list in the HTML doc
… We have done that in other docs. e.g. DCAT with examples of mappings
… we give a taster in the doc, purely to stimulate the reader to go and look at the full piece (which is elsewhere)
… there are other substantial comments about the profiles vocab, including that it might be formally aligned with DCAT. So there is a lot to do. At what point do we have sufficient issues resolved and how do we show that we have addressed the issues, even if we haven't resolved them all?

ncar: we have tagged 34 open issues with 2ndPWD. we are unlikely to get through those soon
… makx issue #755 serves as a listing of all the definitions made elsewhere. If we do that and the roles issue, then that is a good 2ndPWD candidate

<ncar> PROF 2PWD Issue list: https://‌github.com/‌w3c/‌dxwg/‌issues?q=is%3Aopen+is%3Aissue+milestone%3A%22PROF+2PWD%22

dsr: why are these things not concluded?

kcoyle: there is plenty of discussion but we didn't act on them

<roba> i think the vast majority of issues are easy to resolve.

kcoyle: we haven't determined how substantial they are, though some look like they are

roba: we made 1 change to the model, but the rest is about explanation, packaging, naming of properties, etc.
… many issues come down to style, editorial clean-up etc.
… formalising the alignment to DCAT might resolve issues

kcoyle: I disagree in that i see many substantive issues. if they are simple they should be resolved asap

roba: I think many are editorial and a sprint will help

kcoyle: many comments are on the content of the diags. e.g. ShEx community comments
… PWinstanley and kcoyle sent a list of issues that need attention

<kcoyle> https://‌github.com/‌w3c/‌dxwg/‌issues/‌572

roba: ncar generated a list of issues

kcoyle: if they are easy then please get them sorted quickly

ncar: we have agreed in this meeting an approach for roles, and in an email last night I provided an approach to the comments in that email from PWinstanley and kcoyle

kcoyle: they are small issues, so complete them . we can't move forward with 82 issues being open
… List the 'substantial' issues to ensure that we get these covered
… we need a list to ensure that we can track the issues individually

roba: there is the ontology, the formalism.

kcoyle: which of the 82 github issues are 'substantial', not just a small editorial effort?

<roba> ok will add issue numbers to sprint agenda

dsr: this is where having everything on a spreadsheet makes life easier to manage

kcoyle: we will need to add to the spreadsheet issues brought up by WG members

<roba> feel free to give us a list of issue numbers

kcoyle: they are not all externally generated. The external ones have a specific requirement to be answered formally

ncar: one source of confusion was that in the orginal email from Tom Baker have many points that were converted into github issues. They are all there, afaik. Please can others double check

kcoyle: that is detailed in the email from PWinstanley and kcoyle
… I will check this against the issues and let you know if there are any gaps.

<roba> i responded to your email of 4 march agreeing - and those issues are the focus of the sprint - its hard to see wjat else we can do..

kcoyle: there are lots of meetings - it is difficult for people, but if you can manage to join in the profiles vocab then please do

<Makx> bye

<Jaroslav_Pullmann> bye!

<AndreaPerego> Bye

Summary of action items

  1. roba to open new issue seeded with list of roles and definitions and ask for comment

Summary of resolutions

  1. accept minutes of Feb 26
Minutes manually created (not a transcript), formatted by Bert Bos's scribe.perl version 2.49 (2018/09/19 15:29:32), a reimplementation of David Booth's scribe.perl. See CVS log.