Web Annotation Working Group Teleconference

<trackbot> Date: 08 April 2015

Agenda Review, Scribe Selection, Announcements

<fjh> ScribeNick: Tim_Cole

<fjh> schema.org community group, https://lists.w3.org/Archives/Public/public-annotation/2015Apr/0001.html

fjh: announcement - schema.org community group has been formed.

shepazu: primarily for vocabulary / ontology geeks, if you're interested in schema.org,

... and you're concerned with development of ontology(ies) that might be used for annotations, you may want to join this CG

... a good way for the community to help drive development of schema.org

ivan: at the moment, it is not really acceptable to make normative references to schema.org vocabulary

... this is relevant to us given our discussion about using schema.org terms instead of dublin core terms

... there have been discussions between W3C and schema.org for many months, formation of CG is a first step in dealing with this issue.

RayD: the issue came up a few weeks ago about classes for annotation bodies

... DC classes worked well for most but not all; schema.org had a better class

ivan: for the time being let's not worry too much about the choice between DC and schema.org

<fjh> might want to add note to doc on this dublin core vs schema.org so it is not forgotten

<Jacob> IIRC, there was at least one example where the opposite was true...text comes to mind but I might be mis-remembering which media type it was.

... we can reasonably hope this will not be an issue by the time we are ready to propose recommendation.

<bigbluehat> yep

fjh: we should add a note to our previous discussion about DC and schema.org terms

azaroth: we should be sure to track this issue; but for now should we assume that schema.org will be available for normative reference by time we are ready

... azaroth proposes we should do this, but keep an issue in github or as a note in the doc.

fjh: suggest keeping a note in the doc.

... not sure what should be the default for now, but as long as we note it, may not make much difference.

<bigbluehat> Zakim: I am +1.864.787.aacc

RayD: we should not be shy about defining OA properties when a normative reference not available

<fjh> s/Zakim: I am.*//

... and then right before we ready to propose a recommendation we can switch to schema.org if available.

<RayD> I might

<Jacob> Tim and Jacob as well

<tbdinesh> me too

<Bill_Kasdorf> I'm on the schema.org CG

Minutes Approval

<fjh> proposed RESOLUTION: 18 March 2015 minutes are approved, https://lists.w3.org/Archives/Public/public-annotation/2015Mar/att-0064/minutes-2015-03-18.html

RESOLUTION: 18 March 2015 minutes are approved, https://lists.w3.org/Archives/Public/public-annotation/2015Mar/att-0064/minutes-2015-03-18.html

Face to Face Meetings

fjh: any thing we need to know about logistics, etc.

<Matt_Haas> Is the conference sold out? I’m sorry to say, that I cannot attend.

RayD: what is process for developing agenda and when will we see it.

fjh: we will work on it this week. Will include robust anchoring, continuation of data model, ...

dwhyl: room is available at 8 AM and we can go as late as we want.

<azaroth> +1 to /starting/ at 9

shepazu: officially 9 to 5 (and that's when we can count on wireless being available)

<azaroth> and many thanks to Dan for getting everything sorted out!

dwhly: lunch will be in the room to coordinate better with another event taking place same day

<RayD> +1 for Thai!

<bigbluehat> Hack Days :)

fjh: IAnnotate Thursday & Friday, followed by hackfest Sat and Sun

<bigbluehat> https://hypothes.is/blog/i-annotate-hack-days/

fjh: people should respond to IAnnotate questions

dwhly: everyone attending IAnnotate should have gotten email with questions; please provide feedback on tracks and what will be most useful

<fjh> iannotate agenda page https://docs.google.com/document/d/1ML3u4hmCF2oeIFNG3To4L1NBjqWnYQ37vYJ5Y2eLUdY/edit

<tilgovi> Or if you're not a developer

<bigbluehat> RSVP for Hack Days is here http://www.meetup.com/SFOpenAnnotation/events/221577503/

... the hack days will be useful for developers who want space to work on annotations in the company of other developers; there will be a barbeque Saturday night at Dan's

<bigbluehat> and yes, you don't have to be a developer to attend

<stain> ..was there any news about travel assistance..? Sorry for not being in the loop, I've been very busy :(

bigbluehat: people who are not developers also welcome to try stuff out, look over shoulders, help with user feedback, the more people who can do hands on the better

<stain> dwhly: thanks

TPAC

<fjh> http://www.w3.org/2015/11/TPAC/

azaroth: we put down 15 people for a meeting of annotation WG; closer to time we will clarify

Social Web

fjh: we've established a relationship, made clear to the WG what's important to us, they are still in formative stages

... we should have interoperability, but not clear if we will share classes and properties

ivan: re Social WG, a concern is that we don't yet have a clear idea of the 2 WGs plans on the protocol issue

... what does Annotation WG plan to do on protocol on issue and what direction do we plan to go?

... we have had some discussions with LDP WG, but nothing final, we need to decide.

azaroth: I have not had time to get back to protocol issue due to press of other work; will have updated version by end of this week.

<tbdinesh> re social web, is there an example of annotation in a social web context

... assumes the submission of binary data is out of scope, this makes the LDP implementation much easier (only have to worry about RDF)

... so we should be able to write a very thin LDP layer to post and retrive what they need.

... so in moving forward, I would recommend that we give people something to react to (re LDP) and if people are dismayed we should get alternatives

fjh: may be beneficial to thrash a bit for a brief while

RayD: if the question is whether we are going to go forward with the protocol, the protocol is critical -- sounds of agreement from others

... also prefer the container approach for annotation, see RayD's email on this

ivan: Definitely should make protocol

... but we have 3 different approaches: from scratch, align with LDP, or go along with Social WG approach

... we must make a choice between these 3 general approaches fairly soon

... to get there we need to know more about each of these approaches, so far only have begun fleshing out the LDP approach

<bigbluehat> Social WG is currently analyzing their own set of options (currently 3): LDP, micropub, and pump.io (based on AS 1.0)

fjh: we don't know enough about the Social WG approach, but would like to get a little further on LDP approach and then talk to Social WG

<bigbluehat> they don't have a "clear winner" yet

... (re RayD) the case has merit, but question about what specific work is needed to make it possible.

<Zakim> azaroth, you wanted to note query vs CRUD

<bigbluehat> if our WG is interest in / pursuing LDP (or either of the others), that could serve as an additional +1 for that protocol.

azaroth: note the distinction between simple CRUD, which is handled by LDP easily and simply, but the search operations (which will be essential) will need an additional layer

<fjh> I’m suggesting we focus and simplify the LDP protocol doc, then share with social web to explore their view based in that info

... regardless of underlying CRUD methodology is, we will likely have some domain-specific query requirements to deal with

... we will need to deal with the CRUD issues and then get on to query issues

<fjh> might consider separating search as well

... I've looked at the options being considered by Social WG; someone needs to look at how we might use their non-LDP options for annotation

RayD: possible to read protocol document without having to understand the whole of LDP document

bigbluehat: Rob presented most of my main points well,

... our work with Social WG tells us that they still are in early stages -- they will have a f2f in Europe in July and then at TPAC

... so we should stay in touch as they make a decision, and then a joint f2f at TPAC would be good

... but right now that are also at an early stage.

... LDP may be most mature in terms of deployment and being ready for spec writing

... still some confusion in micropub world, and limited efforts to resolve some of these confusion

... so we need to keep in close communication with Social WG through the summer and leading up to TPAC

... pump.io biggest contribution is around federation

... Social WG is trying to map all of these to their use cases; we may want to do the thing for our use cases.

ivan: can we try not to put ourselves in just a waiting position, but rather be proactive at looking at the other options being considered by Social WG

<azaroth> +1 to Ivan

... we need a better sense of whether these approaches can work for our annotation use cases.

... this would allow us to go back to Social WG and maybe influence their decision.

<bigbluehat> +1 to *not* waiting until TPAC to engage...didn't mean to imply that

... we need a clear idea about whether any of these other approaches will work for us now

<bigbluehat> point was to engage *now* and close the loop at TPAC

<ivan> sorry bigbluehat , I may have misunderstood you

fjh: we should probably focus next week or so on closing the circle on LDP protocol document

<ivan> I think we are violently agreeing...

<bigbluehat> likely :)

... Do we have a volunteer to look at one or both of these other approaches

... regarding search, was assuming this is a separate doc with its own complexity

azaroth: yes, search should be separate from protocol, both because it does not build on LDP and because you might want to implement separately

... did not mean to imply a single protocol spec that encompasses both

<fjh> +1 to separate to search spec

shepazu: Doug thinks we should develop in parallel where possible

<fjh> +1 to working in parallel as Doug notes

... the protocol spec will inform data model, as well robust anchoring, etc.

... we should make sure the pieces work well together before we release anything in final form

<fjh> my comment is that if we have a draft protocol doc in the next week that should help, not suggesting we wait long time

... the charter contains several specs that need to work together

<fjh> +1 to concrete proposals

<Jacob> +1

... members of this working group should feel free to put ideas out there in a form others can react to

<tilgovi> Same

... if you have the will and the time to look at one of our deliverables and think you can help develop and bring to the table some concrete ideas, you should do so

<Zakim> azaroth, you wanted to agree w/ Ivan, we can help SWG's choice by being informed w/ use cases

... please bring your ideas to Doug, Ivan and the Chairs

azaroth: agree with what Doug said, and will be responsive if you need some help

... work on protocol has to respond to use cases, and agree with ivan that we can potentially help Social WG because we have overlapping use cases

<azaroth> +1 to fjh

fjh: you don't need to formally edit a document in order to get your concrete ideas on the table -- text to the list is a good start

shepazu: you can put proposals onto the email lists, but not as effective as bringing a draft of a spec to the WG

... people react more to spec drafts and fragments

... especially now that we have the ability to annotate specs

... even very rough spec drafts can be worthwhile putting up

<azaroth> Agree that very rough drafts get more feedback, but we can work together to turn the text from mailing list into that spec

fjh: we can use ReSpec for doing this

<tilgovi> Easy forward is to say, as Frederick did, just get it out in text. Then, as Doug said, ask for help and get it into spec form.

<bigbluehat> +1 to helping the community learn ReSpec and moving things into more solid formats

<tilgovi> Don't feel shy about either step. :-)

<bigbluehat> +1

<azaroth> +1

shepazu: Doug will help turn text into spec, so don't be hesitant or shy because of the formality of spec format

<Jacob> on the body or on the annotation node

RayD: what is motivation placement topic?

azaroth: the idea that you might be able to put motivation on specific resource rather than just the annotation

fjh: wanted to make sure we closed this topic out

azaroth: let's get consensus on this next week

fjh: adjourn, we will talk next week.

<ivan> trackbot, end telcon