Web Annotation Working Group Teleconference

<ivan> scribenick: tilgovi

<TimCole> PROPOSED RESOLUTION: Minutes of the previous call are approved: https://www.w3.org/2017/01/20-annotation-minutes.html

<azaroth> +1

<TimCole> PROPOSED RESOLUTION: Minutes of the previous call are approved: https://www.w3.org/2017/01/27-annotation-minutes.html

Minutes review

<TimCole> +1

<bigbluehat> +1

<TimCole> https://lists.w3.org/Archives/Public/public-annotation/2017Feb/0000.html

<TimCole> PROPOSED RESOLUTION: Minutes of the previous call are approved: https://www.w3.org/2017/01/27-annotation-minutes.html

PROPOSED RESOLUTION: Minutes of the previous call are approved: https://www.w3.org/2017/01/27-annotation-minutes.html

RESOLUTION: Minutes of the previous call are approved: https://www.w3.org/2017/01/27-annotation-minutes.html

<TimCole> PROPOSED RESOLUTION: Minutes of the previous call are approved: https://www.w3.org/2017/01/20-annotation-minutes.html

RESOLUTION: Minutes of the previous call are approved: https://www.w3.org/2017/01/20-annotation-minutes.html

TimCole: the intent today is to talk about where we are with the proposed recommendations and to talk about issues with the serialization

Proposed Recommendation status

ivan: I looked at it about half an hour ago.

we have 21 positive votes and one abstention

the abstention has always been there and that's not a problem

there is one remark on the annotation protocol

unless something comes ins, we certainly have enough votes

there are two more that I reached out to in the publication world that would be nice if it came in

one is a potential implementer who does epub software

unless something comes in, we are okay

TimCole: we have enough. as long as we don't have any objections we should be able to publish

ivan: exactly
... there's nothing left to do other than change the right stuff in the respec

it should be fairly mechanical

we have to agree upon what would be our target publication date

my proposal would be the 21st

the date before would be the 16th, that's two days after the end. there's some administration to do so I think that's a bit tight

TimCole: Our hope would be to have the notes approved by then as well, right?

ivan: absolutely. so 21st should be the date. everything else should be ready to go by then

TimCole: 21st makes sense to me

do you need a vote?

ivan: on that? no. what I will need a vote for is that we all agree we want it to be published. I can go to the director and ask for short name and things like that.

as soon as possible to get it out of the way

TimCole: Do we have enough people?

ivan: We do have now enough people acting in the group.

So, I think, yes we can do that.

TimCole: let us move on to the notes

Selectors and states note

<TimCole> http://w3c.github.io/web-annotation/selector-note/

we have to update the affiliation for Benjamin

TimCole: otherwise, the substance has not changed in the last two weeks

ivan: What Rob commented on is all done.

However, I don't claim I understand all the details, but I have the impression that the comment of Takeshi is more for that note and not for the HTML note.

The differentiation there, whether the fragment is the locator or the identifier, it's not the html note that should talk about it. It should be in the selector note, in my view.

We can discuss as you want

TimCole: let's talk about it now if we can

<TimCole> https://lists.w3.org/Archives/Public/public-annotation/2017Feb/0000.html

ivan: Since it's a note, maybe all we can do is put it into the note and say that when they use it they may be an issue

takeshi: I'm wondering if the selector note could give some position for those who directly use the selector as url

ivan: To be honest, I would much more trust you to give me text to put there than for me to come up with the text. You have much more experience with the issues around IRIs and URIs

takeshi: In the note, the special characters must be percent-encoded in the URL, but the character can be present in the IRI.

<takeshi> % in IRI is just a character

Rob: could we solve it with a note that says systems should take care not to doubly encode characters which are valid in IRIs but not in URLs?

<TimCole> so in iri ...a%20b... becomes ...a%25%20b... when it is made into url

takeshi: No. We can take a simple way that says we don't care, percent characters should be double encoded.

TimCole: so where do we put this caveat

<csarven> Vaguely recall that's how WHATWG URL spec does it

ivan: Send me an email and I will add a warning note to the relevant section. Again, I trust you more than myself to say the right thing.

takeshi: It must be right, but it's a bit ugly

ivan: Well, it's definitely ugly. Every thing with IRIs and this becomes ugly.
... I think that's the only issue with the selector note and we are done with it.

TimCole: Does anyone have a different opinion?

HTML Serialization

if not, let's move on and talk about the HTML serialization

<TimCole> https://w3c.github.io/web-annotation/serialization-html-note/

TimCole: There's been a lot of edits in the last few weeks. There are a couple items marked in the TODO, but before I get to those are there any concerns about the examples?

We have there examples in JSON-LD. Three examples in RDFa. And we have a couple examples of using annotation-based URLs

The JSON-LD of course maps directly to the model document.

We have the RDFa translated into turtle.

And the annotation-based URLs are just URLs that map against the selector note

TimCole: There are some editorial things, but I also have a question about some things near the top.

The terminology section, I think I copied over the terminology in the selector note.

Does someone need to do a read-through?

ivan: There is a difference between this note and the selector note.

The selector note is a semi-specification. It tries to be precise. It's called a reference note. Therefore, the terminology had to be there and it had to be precise because it repeated the specification of the model document.

<csarven> I think it might interesting to have another RDFa example that's close to the motivation "Wholly Internal Annotations". The current example is close to the motivation "Lighweight, decentralized Annotation Tools"

This note is very different. It is much more of a kind of informative, semi-tutorial kind of note.

<azaroth> +1

My vote would be to remove 1.4 altogether

<bigbluehat> +1

<csarven> +1

<takeshi> +1

TimCole: Let's go ahead and remove that. I can do it after the call.

<csarven> re 1.4 removal (I have no audio)

TimCole: The other section I was not sure what to do with is 1.3.

The natural inclination may be to give the annotation an identifier based on the URL of the page in which it is embedded.

Do we want to say anything about that?

Right now we say you need an IRI but it does not have to be dereferencable.

<csarven> Point to canonical sources (WA model/protocol/vocab where appropriate) instead of duplicating info all around

An ID on a script tag is not really a way to identify the annotation properly

Rob: It's not normative text, and 1.3 sort of makes it look (with the lowercase must)

Recommending what we think you should be by way of example is a good way to get it in front of people without making it a spec

TimCole: Does that mean I can get rid of 1.3?

ivan: yes.

Resolutions on the publication on notes

TimCole: Now the question is how to get consensus from the WG for these notes.

With the calendar we want to start that before next week and end it the week after.

ivan: Let's not over-administrate this. I don't think we'll get anything on the email. Most active participants are here.

We can vote here, we can ask the director, and we can give people five days to object in email.

Rob: If that's sufficient that sounds better than the longer approach.

<ivan> for the selector note, the shortname proposed: https://www.w3.org/TR/selectors-states/

<TimCole> https://www.w3.org/TR/annotation-html/

<ivan> Proposal: the Working Group asks the Director to publish the Selector and State note as a Working Group Note, with the short name 'selectors-states'

<ivan> Proposal: the Working Group asks the Director to authorize the publication of the Selector and State note as a Working Group Note, with the short name 'selectors-states'

<ivan> +1

<azaroth> +1

<TimCole> +1

<csarven> +1

+1

<takeshi> +1

RESOLUTION: the Working Group asks the Director to authorize the publication of the Selector and State note as a Working Group Note, with the short name 'selectors-states'

<ivan> Proposal: the Working Group asks the Director to authorize the publication of the Embedding Web Annotation in HTML note as a Working Group Note, with the short name 'annotation-html'

<azaroth> +1

<csarven> +1

<ivan> Proposal: the Working Group asks the Director to authorize the publication of the Embedding Web Annotation in HTML note as a Working Group Note (modulo the editorial changes agreed upon today), with the short name 'annotation-html'

<azaroth> +1

<ivan> +1

<csarven> +1

<TimCole> +1

+1

<takeshi> +1

RESOLUTION: the Working Group asks the Director to authorize the publication of the Embedding Web Annotation in HTML note as a Working Group Note (modulo the editorial changes agreed upon today), with the short name 'annotation-html'

<csarven> w00t w00t

TimCole: Is there anything we need the larger group to talk about?

Rob: I don't think so.

TimCole: It would be nice to have some testimonials to embed.
... I don't have anything else for today.
... If there's nothing else I think we're ready to adjourn. We will plan to meet next week. Hopefully a brief meeting.

Unless there's something that comes up, after the 17th we're done and work transitions to the community group.

<ivan> trackbot, end telcon