DPUB IG Locator TF call

<lrosenth> testing

<bjdmeest> scribenick: bjdmeest

<scribe> scribenick: rdeltour

issue #22

<bjdmeest> https://github.com/w3c/dpub-pwp-loc/issues/22

ben: this is about manifest retrieval from the package. For now there is a note.

ivan: right. I think we can close it.

ben: OK, issue close.

issue #9

<bjdmeest> https://github.com/w3c/dpub-pwp-loc/issues/9

ben: we shouldn't make things more complicated than they are.
... maybe we can just remove that section.

ivan: I agree

leanard: right, just remove it

ben: ok great, let's close this.

ivan: ben, will you take care of the editing?

ben: yes

use cases

ben: we identified some issues, haven't put them on github yet
... romain, I can go through the minutes and just add them to the github tracker?

<bjdmeest> scribenick: bjdmeest

rdeltour: some discussion, not yet clearly defined
... maybe we can define them within our group?

ivan: this group could contribute to that effort
... of the general pwp issues

rdeltour: there is some overlap, e.g., the ones from leonard and nick

ivan: if there are overlaps, I don't think that is a problem
... several use cases that lead to similar requirements, re-enforces those requirements

rdeltour: in terms of process: ben, you can go over the minutes and aggregate all mentioned use cases in one document
... to the mailing list or as a new issue

ivan: issue tracker might be better

rdeltour: maybe all current use cases can be merged into one bigger ues case
... that's up to you
... we can possibly merge smaller use cases into bigger ones

ivan: if you can try to abstract the use cases from the previous discussion, then that would be a big help

breadcrumbs

<bjdmeest> http://w3c.github.io/dpub-pwp-loc/#breadcrumbs

leonard: interesting idea, but it makes an assumption that you can change a PWP
... in the first case (teacher adding annotations) there is potentially changes to the PWP
... inside of what we consider the PWP
... I think we have to assume that we can't, that a PWP is unmodifiable
... everything needs to be done externally

ivan: I think it's not an assumption
... I agree that there are cases when this is not impossible
... there is a possibility that the manifest itself is outside the PWP
... there has to be a note that the mechanism relies on the fact that the manifest can be modified

leonard: yes

ben: I was wondering if we had to limit this to the fact that the PWP shouldn't be changed

ivan: you don't change a PWP, you create a new

leonard: but it's predicated on the fact that you have the right to do that
... but really you only need to modfiy the manifest without touching the PWP

ben: for instance a publishers create a PWP with embedded video and you want to modify this PWP to reference to youtube video
... you create a new PWP but maintain the breadcrumbs

ivan: I think we all agree here

ben: ok

Future of the documents

ivan: if we don't have anything left on the agenda I'd like to discuss something
... at some point we should discuss how we see the evolution of this document
... I have the impression we're close to an end.
... it's not clear to my mind what to do with this document when it's finalized
... we have 3 interrelated documents: PWP, UC&R, locators
... I'd like to head how do you see that
... that may lead to editorial work

leonard: I agree that at some point we have to figure out that problem

ivan: should I share my views?

leonard: yes

ivan: my feeling is that we should converge to have 2 docs, one is the UC doc and the other is PWP
... on the long term
... that means that you should take over the use cases part of PWP into the UC document
... then the PWP doc becomes a more technical document
... includes what we did for locators
... and the section we have on SW (reworded to make it more agnostic)

leonard: makes sense

romain: I like the approach

ivan: when you and Heather start editing the UC doc, we have to start working on ripping off the PWP document

leonard: maybe bring the discussion to the larger group?

ivan: ben you'll probably be asked to summarize our TF work to the larger group, we should propose it at this time.
... if the locators work is "done", the next big thing is to look at what is the manifest and what information is needed there

leonard: I do agree on that
... but I'm wondering how we address the manifest without addressing the package issue.
... but otherwise I do agree that the next big thing is the manifest