Re: [Minutes] 2017 03 27

Dear all,

I have created a doodle poll to see when can we meet to discuss the 
scope and ambition of the formal semantics note. Link:
https://beta.doodle.com/poll/ricy6h4iha3b5s4z

I have set a fixed time (12.30 GMT) and several different days trying to 
concile the different constraints you have emailed already.

Regards,
Víctor


El 27/03/2017 a las 15:35, Phil Archer escribió:
> As ever, the minutes of today's meeting are at 
> https://www.w3.org/2017/03/27-poe-minutes with a text snapshot below. 
> Thanks Michael for scribing.
>
> Main topic today was Sabrina's new use case of modelling the GDPR 
> using a profile of ODRL.
>
>
>   Permissions and Obligations Expression Working Group Teleconference
>
> 27 March 2017
>
>    [2]Agenda [3]IRC log
>
>       [2] https://www.w3.org/2016/poe/wiki/Meetings:Telecon20170327
>       [3] http://www.w3.org/2017/03/27-poe-irc
>
> Attendees
>
>    Present
>           benws, benws110, ivan, michaelS, phila, renato, Sabrina,
>           Serena, smyles, victor
>
>    Regrets
>           Brian, Caroline, Simon
>
>    Chair
>           Ben
>
>    Scribe
>           michaelS
>
> Contents
>
>      * [4]Meeting Minutes
>          1. [5]Last week's minutes
>          2. [6]New Use Case
>          3. [7]Deliverables
>          4. [8]best practices
>          5. [9]open Actions
>          6. [10]London F2F
>      * [11]Summary of Action Items
>      * [12]Summary of Resolutions
>
> Meeting Minutes
>
>    <benws110> nick benws
>
>    <victor> hi all
>
>    <renato> hi victor
>
>    scribe michaelS
>
> Last week's minutes
>
>    benws: anybody want to raise an issue with last week's minutes
>
>    <phila> [NOTUC]
>
>    <phila> [13]Last week's minutes
>
>      [13] https://www.w3.org/2017/03/20-poe-minutes.html
>
>    Resolved: last week's minutes approved
>
>    <renato> [14]https://www.w3.org/2016/poe/wiki/
>    Use_Cases#POE.UC.37_Representing_regulations_using_ODRL
>
>      [14] 
> https://www.w3.org/2016/poe/wiki/Use_Cases#POE.UC.37_Representing_regulations_using_ODRL
>
> New Use Case
>
>    Sabrina: introduced the Use Case
>    … it models the EU General Data Protection Regulation
>    … it needs to cover that at a generic level but also in details
>    … Article 12 added as an example
>    … this article shows the important use of references to other
>    articles
>    … the numbering of the articles has at least two levels
>
>    benws: any comments on that so far?
>
>    benws: does this requirement belong to a profile or to the
>    general ODRL model?
>
>    Sabrina: this is a decision by this group
>
>    renato: what does "refer to another article" mean?
>
>    Sabrina: that are dependencies - look at Article 12. This may
>    transform to many duties.
>    … to check if Article 12 is fullfilled the fulfillment of other
>    articles is required
>
>    phila: GDPR is very important it would be a big PR win if ODRL
>    could show that it can cover it.
>    … key question: is ODRL is a good tool for that purpose.
>    Sabrina do you feel that?
>
>    Sabrina: ODRL is not a bad fit. We need to specify obligations
>    and constraints
>    … There is work on taxonomies by other parties but less
>    fitting.
>
>    renato: we could promote this as a profile. This would serve to
>    explain how to create a profile
>    … and this profile could be shown to a wide audience.
>    … the relationships between the constraints and duties is
>    demandingö
>
>    Sabrina: we have dependencies between the duties, we have
>    constraints on duties, actions and parties
>    … supported to create a profile for that.
>
>    benws: to show that we could express regulations and licences
>    by the same language would be fine
>
>    phil
>
>    phila: supported using ODRL for this purpose
>
>    Sabrina: we are basically defining obligiations = duties =
>    complying with the regulations
>    … if we run into problems we will come back to this group
>    … when it comes to constraints: there are discretational ones
>
>    smyles: suggested to model optional constraints as permissions
>
>    Sabrina: that's not exactly the intention of the GDPR
>    … there are statements like a recommendation - and we don't
>    want to omit them
>
>    renato: is thinking what this could look like in code:
>    leftOperand say you may or may not use an icon
>
>    Sabrina: need for a discretional constraint: it would be good
>    to meet this constraint but it doesn't stop the policy
>    … if it is not met
>
>    Sabrina: for her and Simon some constraints a bit fuzzy, needs
>    deeper reviews
>
>    smyles: we may add a concept of recommendation = if you can,
>    you should do that
>    … there could be levels of recommendation: strongly recommended
>    ... and more
>
>    Sabrina: will review this suggestion
>
>    <renato> [15]https://tools.ietf.org/html/rfc2119
>
>      [15] https://tools.ietf.org/html/rfc2119
>
>    <phila> The key words "MUST", "MUST NOT", "REQUIRED", "SHALL",
>    "SHALL
>
>    <phila> NOT", "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and
>
>    <phila> "OPTIONAL" in this document are to be interpreted as
>    described in
>
>    <phila> RFC 2119.
>
>    phila: RFC2119 is a standard specifying things like that -
>    could help
>
>    Sabrina: Dispensation = something is required, but there is a
>    dispensation under specific condiditions.
>    … and some articles say "you are not allowed" others say
>    "unless party X allows that"
>
>    <victor> Dispensation: a : an exemption from a law or from an
>    impediment, vow, or oath may be granted a dispensation from the
>    rule b : a formal authorization requested a dispensation to
>    form another lodge
>
>    Sabrina: in fact: an exception on an exception
>    … may also be used
>
>    renato: went over some more details of transforming DGPR into
>    ODRL
>
>    victor: thinks like that can be expressed by ODRL.
>
>    Sabrina: we are looking not only at GDPR but also legal
>    regulations in general - is the existing ODRL data model work
>    for us
>
>    benws: what are "features"
>
>    victor: we could think about synonyms for hardwired constraints
>
>    Sabrina: two more things: "Feature" = article 12 has various
>    obligations, some are well defined, some don't stand on their
>    own.
>    … we are looking at conjunctions and disjunctions in this
>    context
>    … transparency is the conjunction of all of them - we call them
>    Features at the moment
>
>    Sabrina: we need additional constraints on the asset - they
>    will span across multiple duties
>
>    renato: ODRL scope could work
>
>    Sabrina: agreed
>    … we have an issue with the type of processing - e.g. how
>    personal data may be used for marketing
>
>    smyles: the purpose is to define the nature of a party - right?
>
>    Sabrina: yes, depending on who you are rules may apply
>
>    smyles: why not to split up in constraints for group A and
>    group B of persons
>
>    smyles: wondered if inheritance could be used
>
>    Sabrina: the controllers for different purposes are different
>    … we look at what's there and then will come back to this group
>    … the Wiki space could be used for discussions
>
>    benws: timeline?
>
>    Sabrina: there are different groups of work: e.g. transforming
>    the article and the sub-points - but that's not very usable.
>    … in a next step obligations have to be pulled out of the
>    articles - and that's a big work, will take months.
>
>    benws: does this timeline align with the ODRL timeline?
>
>    Sabrina: yes.
>
>    renato: do we need a new policy type "regulation"?
>
>    Sabrina: yes
>
>    benws: supported to use Wikipages for working on the
>    transformation
>
> Deliverables
>
>    <renato> [16]https://www.w3.org/2016/poe/wiki/Deliverables
>
>      [16] https://www.w3.org/2016/poe/wiki/Deliverables
>
>    renato: went over [17]https://www.w3.org/2016/poe/wiki/
>    Deliverables
>
>      [17] https://www.w3.org/2016/poe/wiki/Deliverables
>
>    <renato> [18]https://lists.w3.org/Archives/Public/
>    public-poe-comments/2017Mar/0012.html
>
>      [18] 
> https://lists.w3.org/Archives/Public/public-poe-comments/2017Mar/0012.html
>
>    renato: we got a reply from EDRLabs
>
>    <renato> [19]https://github.com/w3c/poe/issues/118
>
>      [19] https://github.com/w3c/poe/issues/118
>
>    renato: raised some concerns regarding periods
>    … this needs an update of the definitions of date/time and
>    period constraints
>
>    renato: re Horizontal reviews:
>    … any news from Brian?
>
>    benws: has sent a reminder
>
>    renato: reviews seem to be on track
>
>    benws: refered to a proposoal of Victor to hold a special
>    meeting
>    … = a call
>
>    benws: asked Victor to launch a Doodgle survey for finding date
>    and time
>
> best practices
>
>    benws: tried to reach out to James from Catapult, but the email
>    did not work
>
> open Actions
>
>    benws: only 3 on the issue tracker
>
>    <phila> s/RESOLVED: last week's minutes approved//
>
>    <renato> [20]https://github.com/w3c/poe/issues/114
>
>      [20] https://github.com/w3c/poe/issues/114
>
> London F2F
>
>    benws: open issue is providing hotel rooms at TR rates - but
>    Sabrina may have an alternative
>
>    victor: would appreciate to have times for the agenda items
>
>    bens: starting time 10am - ok?
>
>    renato: agenda will be based on requests from group members and
>    currently ongoing work
>
>    <ivan> will there be possibiltiies for dial in?
>
>    benws: suggested 5:30pm as closing time
>
>    <ivan> thanks
>
>    benws: it will be possible to dial in too
>
>    benws: AOB?
>
>    benws: none was raised - bye
>
> Summary of Action Items
>
> Summary of Resolutions
>
>     1. [21]last week's minutes approved
>


-- 
Víctor Rodríguez-Doncel
D3205 - Ontology Engineering Group (OEG)
Departamento de Inteligencia Artificial
ETS de Ingenieros Informáticos
Universidad Politécnica de Madrid

Campus de Montegancedo s/n
Boadilla del Monte-28660 Madrid, Spain
Tel. (+34) 91336 3753
Skype: vroddon3

Received on Monday, 27 March 2017 16:31:57 UTC