<jr> scribe: joshcornejo
Jo: Admin topics
Ben: left Refinitiv, with a startup to work in tadah! market data, the name of the company is Deontic Data ... <defines deontic>
<jr> PROPOSED RESOLUTION: Ben to continue as co-Chair
RESOLUTION: Ben to continue as co-Chair
<inserted> Jo: propose to use a more modern teleconf system, with transcription, maybe, anyone have a problem with Zoom or Teams?
Laura: prefers zoom, but can use either
Jo: Zoom can be used at DB
<Ilya_Slavin> Michelle Roberts is sending her apologies as she won't be able to make the meeting today
Jo: resolve to accept the minutes from the last meeting
RESOLUTION: Accept minutes of last meeting
Ben: reach agreement into what we
want to deliver and when, Mark & I had a meeting earlier in
the week. Aimed at delivering of the first version of the
standard and a PoC by the end-of-january (2021).
... if that is a target date we can all aim for.
Mark: the definition and presentation of use cases, how we want to test against that standard and do that in a PoC
Jo: why don't you kick us off, either of you
Ben: I'll make a start. The first
thing is the scope of the standard, and what is the standard to
cover.
... I am going to quickly post a document to the chat
<jr> https://w3c.github.io/market-data-odrl-profile/md-odrl-profile.html
Ben: There is a document that
maps from machine readable to human readable
... then what is the scope and that we cover enough ground in
that standard
... we made a huge step
... when Laura published those examples
... to translate those documents into the standard would be a
success
<jr> https://github.com/w3c/market-data-odrl-profile/blob/gh-pages/Test-Cases.md
Ben: What we are working is to
capture 80% of those standard questions and to deliver the
answer to those questions into that standard.
... we have around 16 Questions, if we can push that to
50
... this is similar to programming test cases
... as an Exchange if I can answer my questions with this
standard
... let's push this by the end of january
... convert those documents that Laura has provided
Mark: should we give a prize to the person that transforms more ?
Ben: we can set aside some time
Mark: we can break away, and maybe do some batch action
Jo: if we could use the lists
(email)
... we seem to have very low traffic, which is a shame
<jr> scribenick: joshcornejo_
Ben: we should set aside some time in the agenda for next meeting and make a call for people in the mailing list
Jo: that's fine
... lets do a straw poll to see who can contribute
... we are not talking about anything formal
<Laura> I can.
Ben: no
Jo: in the most natural language terms
Mark: I've raised my hand
Caspar: raised my hand
<Ilya_Slavin> Ilya: no
Olga: me
Ben: and Phil
Jo: thank you all for volunteering
<jr> ACTION: MarkB to kick the conversation off on list
Jo: OK great
Ben: we want to build some code
around the standard by the end of january
... i've had a few conversations
... there are 3 areas, fantastic areas
... one is around following the workflow from requesting from
an organisation, chasing the requirements to satisfy
... if there are any invitations out there satisfying those
requests
... one of those I am interested in
... capturing information in natural language
... second there is some interest in automating in access
control
... running a PoC around tic history
... augmenting around access control system
... there is a potentially simple solution
... when machine accts register we can put all sort of rights
metadata
... and we can do a compliance check
... if it can be or not associated with those elements
... it might focus on registration of machine accounts
... much simpler technical issue than dealing with access
control systems
... then third is CME's accessing data and using policies to
control what a consumer can access to
... those are three potential areas, wondering if members of
the group have further areas or comments around these three
areas
Ilya: are the 2nd and 3rd not similar enough that can be the one and the same
Ben: I thought that at first, but the 2nd one (fulfilment) focusing on the machine accounts which is slightly different component
Ilya: if you are able to solve the 3rd, you solved the 2nd one ...
Ben: yeah I agree, they are
solving the same problem with slightly different
approaches
... and we need them both
Jo: to prove interoperability, we should address them in complexity order
Ben: no, I don't think so, I think we need both approaches .... driven by people that are willing to collaborate and provide some resource. In a sense we will tailor them based on what resources are available.
Jo: two questions come to mind: who would be keen on behalf of their organisation to participate and 2nd in which area Ben mentioned are you interested to participate
Adam: we are happy to support
PoC
... many are customers, they will have to decide from those
running on AWS which test cases and we can get resources from
our organisation (AWS) to support those PoC
Jo: if there are significant resources, we will need a PMO
Adam: long term project management is not a strength
Jo: anybody else ?
Mark: DataBP can participate in the generation of ODRL documents
Ilya: what is the success criteria for the PoC?
Ben: can a computer interpret a policy to ensure that only a compliant process can occur ? Is that enough of an answer ?
<jr> (MarkB notes that the use cases collapse to "use by humans" and "use by machines"
Ilya: are you looking at some proof or some code ?
Ben: if you have a machine reading a policy and some output ?
Mark: why don't we focus on a
single duty ? like have you paid your invoice?
... something really simple
... maybe a good PoC
Ben: to some extent, you know the answer on that. that depends on how many people are using your services
Mark: in terms of generating license rights, anybody can consume that. Anybody downstream can consume that - I don't know if anybody is consuming those services
Ben: to check if any invoice has been paid or not, for many organisations is not trivial. Maybe DataBP does for some customers and is trivial.
Mark: maybe we pick something that is relatively straightforward, that we can demonstrate some real world changes
Ben: I agree with what you're aiming at. <speaks about machine example again>
Mark: and maybe have a workflow - register that machine and it comes back green
Jo: from the point of view of
PoC, it has to be useful an impactful.
... what are the things that are going to make an impact to
people's life ? is there something that will make more impact
and doable within the PoC.
Mark: maybe registering a machine
Jo: ... so we can bring more people to join.
<describes example of usage>
Ben: in terms of PoC I think it
should be exactly what you are describing
... what we are struggling at the moment is a PoC at the
machine side
<jr> (laura gives example of checking)
Ben: a machine at the downstream process
<scribe> ... <continues explaining the machine example>
Laura: OK
Jo: we're saying that we definitely Laura's example will be elaborated. Ben - who is doing this elaboration?
Ben: I will be doing this elaboration and looking to collaborate with people to do that enforcement.
Mark: is that a topic for a breakout ?
Ben: let's have a breakout next Wednesday on the machine side of this
Jo: can we narrow down what we want to get out ?
Ben: can we find a use case that maximise the number of actors ? The more people the better to prove to the industry ?
Ilya: I volunteer and Atiq ?
Casper: as well
Joshua: me too
Mark: will join for sure
Olga: I will see if I can bring someone
Olufemi: I will like to join too
Jo: that's a good amount of
interest
... will you make the arrangements Mark?
Mark: I can do it
<jr> ACTION: MarkB to arrange a break out next Wednesday on "the machine use case"
Phil: I've not done much
recently
... I know you're focusing on this for a reason. But machine
accounts is just one example
... I understand why, but any kind of fulfilment of service for
a party is going to be subject of compliance checks,
geographical or organisation or all sort of things
... a machine account example or fulfilment of service to
any
Ben: could be anything, just
trying to find common ground
... it could be any one of those scenarios you mentioned
Paul: maybe we can get someone from Refinitiv
Ben: that would be good
Jo: not asking people to sign up in blood, we'll leave it with you Mark
Ben: I think I am talked-out
Jo: move to AOB
<jr> (Thanks again to Josh for Scribing)
<jr> (All best to Ben in his new venture)
This is scribe.perl Revision of Date Check for newer version at http://dev.w3.org/cvsweb/~checkout~/2002/scribe/ Guessing input format: Irssi_ISO8601_Log_Text_Format (score 1.00) Succeeded: s/Use/"use/ Succeeded: s/your example/Laura's example/ Succeeded: s/<Attic>/Atiq/ Succeeded: i/prefers zoom/Jo: propose to use a more modern teleconf system, with transcription, maybe, anyone have a problem with Zoom or Teams? Succeeded: s/There is/ ... There is/ Succeeded: s/then what/... then what/ Succeeded: s/we made a huge step/... we made a huge step/ Succeeded: s/when Laura published those examples/... when Laura published those examples/ Succeeded: s/one is around/... one is around/ Succeeded: s/straw pole/straw poll/ Succeeded: s/if there are any/... if there are any/ Succeeded: s/and we need/... and we need/ Present: Jo aaron adamH ben caspar Ilya josh jane jeremy laura markB olga olufemi Phil trisha paul Regrets: renato markD michelle Found Scribe: joshcornejo Inferring ScribeNick: joshCornejo WARNING: No scribe lines found matching previous ScribeNick pattern: <joshcornejo_> ... Found ScribeNick: joshcornejo_ ScribeNicks: joshcornejo_, joshCornejo Agenda: https://w3c.github.io/market-data-odrl-profile/agendas/md-odrl-profile-agenda-2020-10-14.html WARNING: No date found! Assuming today. (Hint: Specify the W3C IRC log URL, and the date will be determined from that.) Or specify the date like this: <dbooth> Date: 12 Sep 2002 People with action items: markb WARNING: IRC log location not specified! (You can ignore this warning if you do not want the generated minutes to contain a link to the original IRC log.)[End of scribe.perl diagnostic output]