DXWG CNEG Subgroup Meeting

16 May 2019


LarsG, ncar, roba

Meeting minutes

Confirm agenda


<LarsG> +1


<roba> +1

Approve minutes from last meeting

<LarsG> https://‌www.w3.org/‌2019/‌05/‌09-dxwgcneg-minutes

<LarsG> +1

<roba> +1


Resolved: Approve minutes from last meeting

List of open action items

<LarsG> https://‌www.w3.org/‌2017/‌dxwg/‌track/‌products/‌4

roba: no action on ACTION-193 yet

LarsG: ACTION-325 & ACTION-326 missing

<LarsG> action-325?

<trackbot> action-325: Nicholas Car to Contact roba regarding action-193 and jmeter tests -- due 2019-05-09 -- OPEN

Peter WS has put in ACTION-327 for conneg subgroup

LarsG ACTION 325, 326 & 327

<LarsG> ncar: we've talked about action-325

LarsG: can close ACTION-325 sicne roba is discussing ACTION-193 here

<LarsG> action-326?

<trackbot> action-326: Lars G. Svensson to Write proposal in #785 regarding how to handle multiple dimensions in link headers -- due 2019-05-16 -- OPEN

LarsG: discussing ACTION-326 and Issue 785 - my conclusion is we need to include cartesion product of Media Type & Profile in Link headers to avoid repeated HTTP 406 failures

roba: I don't think this will scale well

LarsG: old IETF RFC for profile conneg contained cartesian products but new one does't

* I can hear you fine

* trying new mic

* I can still hear you fine

roba: we should only say true things and if we do't say something - like no listing Media Type - we shoudl consider it true

roba: if a profile is unqualifid with a MT list, it should be true for them

* it seems you still can't hear me


roba: we can either deliver sparse info or cartesian product as long as you don't make a false MT/Profile combo claim

LarsG: options for communicating this makes clients harder to make

roba: still, this isn't a hard client function

<LarsG> ncar: if we don't take a stance, we're only pushing the issue further down

<LarsG> roba: if the client has to evaluate the cartesian product it's harder to


<LarsG> ... build a page from it (grouping similar items)

<LarsG> ncar: this page has eight views and five formats

<LarsG> ... roba's example would compress that significantly

<LarsG> ... would like to put some more examples into the issue

<LarsG> ... so that we can take a stance

roba: if I support multiple DCAT-APs, do I list all the profiles, since they are in a hierarchy?

<LarsG> ncar: we haven't decided on that yet, but we're assuming that both parties

<LarsG> ... know about the hierarchy

<LarsG> roba: that's a bold assumption

<LarsG> ncar: #785 depends on profile hierarchies. Will create new issue

PWD2 - W3C publicity

<LarsG> ncar: Peter has drafted an email text that we can send

Tom Baker's comments

<LarsG> roba: not inconsistent with our work, but we need to review it line by line

<LarsG> ... most seems to be dealt with

<LarsG> ... the second issue (definition of constraint) is something we need to think about

<LarsG> ... we should raise an issue to discuss that

<LarsG> ... any specification can take the role of a profile

Issue 932 created for problem of all-profiles representation as mentioned regarding 785 above

Action: roba to draft replies to Tom's email in a new GH issue

<trackbot> Created ACTION-328 - Draft replies to tom's email in a new gh issue [on Rob Atkinson - due 2019-05-23].

Summary of action items

  1. roba to draft replies to Tom's email in a new GH issue

Summary of resolutions

  1. Approve minutes from last meeting
Minutes manually created (not a transcript), formatted by Bert Bos's scribe.perl version Mon Apr 15 13:11:59 2019 UTC, a reimplementation of David Booth's scribe.perl. See history.



Succeeded: s/if we offer two options/if we don't take a stance/

Succeeded: s/Peter has drafted an email text that we can send/ncar: Peter has drafted an email text that we can send/