Permissions and Obligations Expression Working Group Teleconference -- 14 Nov 2016

<renato> Scribe volunteer?

Brian to scribe

<renato> Approve: https://www.w3.org/2016/11/07-poe-minutes.html

Renato: Motion to approve minutes?

<michaelS> scribe: Brian_Ulicny

Silence taken as resolution to accept minutes.

Book Industry group has answered our questions and updated Google Docs doc.

"Has everyone seen it?"

"Let's look at the comments in the Word doc dated Nov 5"

"let's go through those"

"Requirement 28.2: weren't sure what display of a work for discovery purposes meant"

"Was about displaying snippets of a work in search or other preview texts (Amazon Look Inside the Book)"

"So this is not a new requirement. Supported already."

"Anyone think otherwise?"

No responses.

"next one is UC 29: about chunks -- chunk A is part of chunk B"

"Bill says it would be useful to support inheritance."

"We support policy inheritance, not asset inheritance, currently"

"Next one talks about not having to repeat inheritance constraints"

"Next comment is about chunk context. Different constraints in different chunk contexts"

<James> question+

"Won't help when one asset is part of another asset"

James: "Tree model of an asset tree used in a different project quickly got very unwieldy."

Renato: "My view is that it might be better to manage the leaves than to manage the tree of assets"

"Re 29.4: Identifiers - Using DOIs and ISBNs instead of URIs. Those are impossible for us."

Ivan: "Can use a URN version of a DOI and an ISBN is possible. But not the non-URI version"

Renato: "Agreed"

"Re 30.2: Associating permissions with a group of articles - a subscription model is how we interpreted it. They mean something different."

"Party A and Party B associate different permissions with the same products."

"We'd just express two different policies with the same target."

Brian: "Is this to do with resolving policy conflicts?"

Renato: "Really couldn't tell."

"From our point of view, an abstract product is denoted by a URI. Policy applies to that thing, abstract or not."

"Next one: at one point do royalties apply?"

<victor> This can also be applied to other media (excerpts of videos, resolution of images, etc.).

"used to have a constraint like this in Open Mobile Alliance on seconds played to inform previewing video."

Ivan: "His example is not a matter of time. Reading 2 pages is not a matter of time."

"To describe what they need is very complex."

BenWS: "a lot this sounds profile-specific."

Renato: "Bill's next on is UC 33"

"Nothing new"

"That's all of Bill's comments."

Next item: Exit criteria

"How are we going to support the vocabulary of terms?"

"Found some examples from annotations working group, etc."

"All describe things with different levels of specificity"

"We need to determine our exit criteria."

<renato> http://w3c.github.io/poe/vocab/

"Our vocab spec"

"Section 4 includes everything."

"One thing we could do is split Section 4 into a core vocabulary"

<victor> (I support this first option)

"Another section, 5, common vocabulary for the rest."

<victor> (these terms do not need to be "non-normative")

"So then we don't have to force implementers to implement everything"

Ivan: "perhaps talking about exit criteria is premature."

"Better to ask: what is a conformant usage and a conformant implementation"

"Are there pieces of code that do this?"

"So I could say that a conformant implementation must be able to produce all the classes and properties of the core vocabularies"

"That is a conformant implementation. If there are at least two of these, this is an exit criterion."

<victor> (I would say conformance is about actions. There is (or there is not) a conformant behaviour)

Ivan: "If IPTC uses these vocabularies in its own environment, then this is a proof of the utility? of the standard"

Victor: "Conformance consists of a set of tests."

"The algorithm must exhibit some behavior."

<victor> (https://hypothes.is/)

Ivan: "we do not specify the user interface. Must produce some structures, which they dump and we determine whether those structures conform or not."

Hypothes.is is an annotation engine that implements the annotation standard Ivan mentioned as an example.

<renato> Survey: https://docs.zoho.com/sheet/ropen.do?rid=0atbh53daf7b68efb4d39846bf1ff465369ba

Renato: "Next item, is a survey. TOC from vocabulary."

"Which of these have you implemented or plan to implement?"

"This will give us a good idea of what's core and what's not."

"If everyone is happy with this survey, then let's send it out to implementers."

<sabrina> good idea: +1 from my side....

BenWS: "Some items will be in a gray area -- not clearly on path to implementation."

Renato: "How about Yes/No/Maybe?"

"for plan to implement?"

Renato: "OK. I'll send out."

Next item: "F2F meeting"

"Phil to send out poll on options."

"Madrid and NYC the two options"

<benws1114> My appologies - have to join another call

Next item: "Constraint call?"

<victor> +1 to join a constraint call

"have an additional call to focus on constraints?"

<renato> https://github.com/w3c/poe/issues?q=is%3Aissue+is%3Aopen+sort%3Aupdated-desc

"This seems to have the bulk of requirements."

"See Github list."

"First 5 are all about constraints"

"So let's tackle them"

"OK?"

"Ivan can only do a call tomorrow."

<victor> +1

<victor> 11GMT would be great for me too

<sabrina> Sure.....

OK. Same time tomorrow.

<victor> :)

Renato: "Any last business? No. Then please read Github issues for tomorrow"

C Utomorrow

<ivan> trackbot, end telcon

- DRAFT -

Permissions and Obligations Expression Working Group Teleconference

14 Nov 2016

Attendees

Contents

Summary of Action Items

Summary of Resolutions

Scribe.perl diagnostic output