See also: IRC log
scribenick Serena
<scribe> Scribe: Serena
<renato> https://www.w3.org/2017/01/09-poe-minutes.html
Proposal: approve last meeting minutes
RESOLUTION: last minutes meeting approved
<renato> https://github.com/w3c/poe/labels/Needs%20WG%20Decision
Issues: https://github.com/w3c/poe/labels/Needs%20WG%20Decision
<renato> https://lists.w3.org/Archives/Public/public-poe-wg/2017Jan/0004.html
renato: let's start with issue
#82
... https://github.com/w3c/poe/issues/82
... in the example multiple actions, parties and assets are
expressed at policy level
... it's basically a shortcut
... we end up wit the same thing if we expand the "short"
version of the policy
... this is the proposal
phila: it makes sense, how are you going to express ?
renato: the narrative could be that the two policies are semantically equivalent
phila: this should be added into
a Processing section in the IM
... these are semantically equivalent and you have to process
them in this way
... my concern is purely in terms about how to state that
... there should be a section defining Processing
procedures
<Brian_Ulicny> No objection to example.
<renato> https://lists.w3.org/Archives/Public/public-poe-wg/2017Jan/0005.html
<Brian_Ulicny> No objection to proposal.
renato: the question could be "in
this case, how does it work with inheritance?"
... inheritance is something we support but we mean more
something like inclusion than the standard meaning of
inheritance
... e.g., in programming langauges
... the proposal is to copy in the child policy the properties
of the father
... then if the full expansion is preferred, then we go back to
the previous example
<Brian_Ulicny> Comment: q+
michaelS: did you check if you
obtain the same result if you expand first and then you
inherit?
... it might be the same, but we need to check
renato: yes, we can look at that
Brian_Ulicny: the kind of inheritance is standard in OWL
renato: we will implement these changes in the editors draft
<renato> https://github.com/w3c/poe/issues/48
renato: issue to be discussed
#48
... validity data expressed through constraints
<phila> +1 to not re-defining Dublin Core :-)
renato: my recommendation is to
point to another ontology defining them
... any views?
... a constraint is part of a rule, thus we need to apply
validity dates to all the rules in the policy
michaelS: it would be good to have a definition of what a time range validity means?
renato: I mean temporal
constraints on the policy
... it was based on the UC 4
<phila> Serena: We had differnet versions. It was at the policy level, to be able to say when there is a change
<phila> ... *this* policy has been overtaken by *this* policy
renato: we have the temporal constraints and we use other ontologies like DC for the other information
phila: the policy is unbounded
renato: if there is no date, we
assume there are no constraints about that
... should we say something about policy identifiers?
<victor> which is the use case raising this?
UC4
<renato> #uc04
phila: we defer to the DC semantics, for versioning
renato: we will explicitly state that
<phila> The terms dcterms:replaces and dcterms:isReplacedBy are both defined
phila: that should be part of the processing part
renato: good point
<renato> https://github.com/w3c/poe/issues
<renato> https://github.com/w3c/poe/issues/63
<phila> Serena: We discussed issue 63
<victor> about https://github.com/w3c/poe/issues/63 i provided several examples with minimal impact on the model, I think
renato: we need to find a way with minimum impact on the IM
victor: the change is minimum, only two operands
renato: what about duties?
victor: I forgot about that
renato: everything is in
constraints in dities
... my proposal is not to do this because it just increase the
complexity of the model. Technically we can do it.
... in duties, we need to define left and right operands for
duties
... we thought there could be other ontologies doing that, but
it does not seem so
victor: I have to say that I'm not strongly supporting that, it's syntactic sugar.
renato: we can note the WG that
we are reconsidering this, we need some strong cases to
continue with this requirement
... the other issues are doable in the next week or two
... any other issue?
<renato> https://www.w3.org/2016/poe/wiki/Deliverables
we move on the deliverables plan
renato: by next Monday we have a
first version of the editors draft
... then call for reviews
<victor> I applied changes in the morning (merging the branch issue84), that you can see in the vocabs page https://w3c.github.io/poe/vocab/
renato: the deliverables wiki
page, we then have to call for one review (widest community
possible)
... asking for feedback, comments, etc
... we need a collection of groups to contact
... at the moment we have few groups listed there
... do we need bigger groups?
phila: you just list the groups
to contact, we created a template email, and we reconded on the
wiki when we sent this email to
... making a structured effort should be enough
... we need to include horizontal reviews as well
<renato> https://www.w3.org/2016/poe/wiki/Deliverables
ivan: one small remark, I agree
that accessibility is not an issue here, but we need to contact
everyone
... in my experience, when it comes to reporting then Github is
your friend
... using a separate label for each group contacted, and ask
for issues there
... it is very easy at the end to report
renato: any other business?
<renato> https://www.w3.org/2016/poe/wiki/Meetings:London2017
phila: EU project where they
express CC licenses in ODRL
... CCRel they express CC licenses
renato: we did a profile together, but it was a while ago
<michaelS> https://www.w3.org/community/odrl/work/cc/
renato: if want to purely express CC semantics they can use CCRel and ODRL
victor: I support the idea of
expressing CC licenses in ODRL
... each of the versions of licenses in Europe are
incompatible, and ODRL can fill the gap
renato: thanks everyone