Web Annotation Working Group Teleconference

<trackbot> Date: 17 December 2014

<azaroth> Thanks Frederick!

<tbdinesh> dinesh

<tbdinesh> aabb is dinesh

<davissalisbury> Just noting that my name is Davis, not David (common mistake!)

<paoloC> I would add to the agenda an updated in relation to the F2F meeting

<azaroth> paolo: Yes, +1, we should do that first

Agenda Review, Scribe Selection, Announcements

<fjh> fjh: add agenda item for F2F

<ivan> scribenick: dwhly

Minutes approval

<fjh> proposed RESOLUTION: 3rd December 2014 minutes approved:

<fjh> http://www.w3.org/2014/12/03-annotation-minutes.html

<fjh> Note: No minutes from the cancelled call last week.

F2F Planning

@azaroth: add F2F discussion first

fjh: Thought we were discussing different dates and places

<azaroth> scribenick: azaroth

dwhly: PaoloC and dwhly plus some folks from MIT tried to see if Boston would work for a venue for I Annotate
... closer to europe, would move things around, dwhy very open to see the conference move around
... explored options for a week, and it didnt work out in terms of securing the venue for enough consecutive days
... and when we could also get space for hackathon and so on. Conflicting priorities and schedules, such as graduations in the area doubled hotel room cost
... gave it a good shot and came away with the conclusion that we need an earlier start, eg sept rather than december
... so have secured same San Francisco facility
... Weds April 22nd as available for a face to face for the WG
... Would need to know if we want to do that, if so will make the plans
... conference then is April 23, 24, and hack days at the weekend 25, 26

fjh: Curious if the dates are firm, or if could be a week later?

dwhly: dates are firm
... weeks either sides are bad

ivan: I raised the concern a few weeks ago about location, so propose to accept the 22nd and move on

fjh: Might have a problem, but a low probability

<fjh> not clear yet, might have a conflict

rayd: Funding still in play?

dwhly: We have 2 grants in process here, with Sloan and Mellon. Should be a fair amount of travel support for people attending the conference

If you happen to show up early, that's great and we'll cover the extra hotel day

<fjh> proposed RESOLUTION: 22 April F2F (1 day) in conjunction with iAnnotate conference 23-24 April, in San Francisco

<fjh> RESOLUTION: 22 April F2F (1 day) in conjunction with iAnnotate conference 23-24 April, in San Francisco

<ivan> scribenick: dwhly

<fjh> much thanks to dwhly for organizing this

Using annotation within the Working Group

azaroth: Have people tried it, did you find it useful, is there anything we can do to improve it
... we know for instance that there is a bug in replying to posts, and if you navigate using an anchor, it doesn't work properly
... can we do a straw poll

shepazu: I just want to note that the annotation system is not working
... I thought I'd socialized this properly with the W3 Systems team
... this was a little surprising... so in case people were wondering why it's not working that's why
... so you won't be able to see annotations, or use it till we get that resolved.

azaroth: Do you have an estimate? Sounds like a lengthy process.

shepazu: Hoping that it's before xmas, but not sure. I will say that TimBL was on the call and that he was enthusiastic about the spec annotations, and that he thought it was a good idea.
... I have used the system.... if we want to take a straw poll... not exactly sure what that would be.

<rayd> not

<shepazu> +1

azaroth: Poll have people used it or not?

<tbdinesh> used it, but not extensively

ivan: I've used it, ran into a few minor problems... the only question is where do we draw the line between annotating and using the issue tracker on gh

<bigbluehat> sorry, that's me

azaroth: Yes, I agree. And that was part of the discussion before. What can we do to mitigate this?

<bigbluehat> thank you ivan

azaroth: I think one of the mitigating factors is that there is the channel to the list, which has some kinks but seem to be working.

<bigbluehat> I'd registered it previously...but I guess it feel off :(

azaroth: If we had the same list on gh then that would be the canonical place
... There has been some discussion around using gh
... So what do people think about the separation of conversations

shepazu: I think we should look at the separation of conversations in a couple ways

<tbdinesh> is there a way to annotate an annotation and link it to an issue on github

shepazu: 1) certainly this does divide the conversation. The fact that the annotations are on the mailing list is a unifying factor. And that gh issues are on the mailing list is a unifying factor
... so the mailing list is a good place for discovery.
... when I left my first annotation dwhly pinged me on IRC
... so the mailing list seems to be a good place for discussion
... when I see a link in the ML from gh, I click through to read it on gh
... I don't think the separation of discussion won't be a big deal because of the discoverabilitty
... 2) the question is the workflow for the editors
... the big issue I've seen is the inability for people to give feedback in the first place
... it's unclear to folks how to do that.
... giving people multiple options is a good way to do that.
... I'd like to share briefly... the traditional workflow is:
... First you write the specification

<fjh> q/

fjh: So, I'm not eactly sure what the issue is we're trying to resolve
... I think context is important. Annotations provide a great benefit in that regard...
... For comments that are editorial, annotations are the way to go...
... Using issue tools is probably better for larger things we want to keep track over time...

<shepazu> (or feature requests)

fjh: Also, important that people not only raise issues, but also propose solutions
... And yes, the tool is great

Shepazu: Good point fjh. I think we'll establish best practices...
... i.e. propose solutions, not just problems...
... like be sure to leave a tag...
... azaroth you had a couple pieces of feedback, not sure where you're sending it
... but speaking to places where the conversation is split, I'd like to capture feedback in the spec annotation CG
... now annotations are public by default

<fjh> example of issue best practices ‘general recommendations for raising issues’ https://www.w3.org/2008/xmlsec/Group/Overview.html#issues (member only)

Shepazu: and now the canonical targets work, so the dated spec versions track to the master
... and we do have a gh tracker for spec annotations too

<fjh> (however not suggesting using tracker, prefer annotation mechanism)

azaroth: For the record, the issue I ran into was in replying. The note sent to the mailing list didn't have the context
... if you go to the TOC and you have link, then it loses the context (?)

fjh: How is the membership proceeding w/ the CG
... is there a lot of interest

shepazu: No. But folks from H. are participating.
... but people don't know about it.

<fjh> need to hook into the respec folks

shepazu: but there are requests to put it on CSS specs
... because they operate more locally

azaroth: it seems TPAC next year would be good time
... so just to wrap
... any other issues with the tool / discussions...

<shepazu> http://www.w3.org/community/spec-annotation/

Use cases

azaroth: What is the status of transcription, analysis. How can we get more submissions.... especially where we're weaker

Bigbluehat: So the q of participation is tricky. We've posted to the list, but we haven't gone hunting
... we are looking for support from the rest of you
... the H. team can mostly provide client side ones
... we'd like a lot more from other folks

PaoloC: I go with the first question: transcription
... I've been traveling, but we have 3 new use cases
... (Ray's) use cases are on the wiki
... the second one one is going to get a sentence too
... if you look at the cross-format use case in particular
... we are down to three use cases
... I can contribute from Annotopia, but it's not an elaborate one
... Some will overlap from the EPUB use cases
... But we need more input in terms of protocol and client API

azaroth: I can contribute from the Stanford set
... lots of overlap w/ Annotopia

<Jacob> with regards to cross-format I have one from the HTRC that I can contribute.

<tbdinesh> dinesh

tbdinesh: I cannot edit the wiki, but I sent via email
... sent a few weeks ago
... one on accessibility ... <can't hear>

<Jacob> There's also the juxtaposition of targets (i.e., compound targets) use case from Digital Emblematica. Does that need to be formally written up in the wiki?

<Jacob> From a modelling standpoint I feel like its adequately supported.

azaroth: jacob had a comment on IRC (above)
... does that need to be written up? yes.

<ivan> Dinesh's Use case: http://lists.w3.org/Archives/Public/public-annotation/2014Nov/0102.html

azaroth: To what degree do annotations provide insight into what a annotation is and what its not
... Do we simply want to transcribe the use cases, or do we want to filter them
... based on some objective or subjective criteria
... or just capture them all

shepazu: I favor capturing them all

<paoloC> +1

<rayd> capture

<Jacob> +1

<azaroth> +1

<fjh> +1 to doug on capturing first and then filtering, also provides a record of decision making

shepazu: functionally what this group is working on will be useful for other things

<ivan> +1 to doug

shepazu: robust anchoring <> findAPI
... I think we should be looking at bits that serve our needs, but potentially others
... I'll pick on rayd's use case
... there's a set of books and a set of images... this image might work as the cover
... that's not an annotation, it's a categorization
... however, if we had that collected... we can figure it out later
... and we can make sure it's handled on the platform overall, or take it to the DPWG, or to the CSSWG or to WAWG
... so I think that collecting use cases is a useful exercise
... and since we'll be alive for several years, we may come back to them
... ramble ramble

rayd: I just want to talk about the cover art... it is an annotation...

i'll write it up, and discuss there.

shepazu: take a fb post. i'll often post and comment on it
... but sometimes it's just a post
... does that mean it's not an annotation?
... we'll be able to reuse the annotation data model
... as a unit of discussion even thought it doesn't have some parts
... it's a blurry line
... people will use whatever we come up in ways we didn't expect

<bigbluehat> +1 to paoloC

PaoloC: collect everything first

<Jacob> I'd say that the comment is an annotation...

PaoloC: i feel annotations are more a fluid thing
... it can be a transient object, that live in time, keep track of things forever, but have a temporal validity

<Jacob> ...but this gets into the murky area of discourse. Which annotations have been used for in the past.

PaoloC: so be careful in drawing a sharp line
... they'll be hard to enforce

shepazu: agree

<tbdinesh> +1 to paoloC : even the edit log (history of a page) in wikipedia can be a collection of annotations

PaoloC: lets collect everything. lets have guidelines
... a lot of stuff falls in between

shepazu: lets look at the functional questions, where they meet a use case

bigbluehat: the question is what falls in a formal use case list
... +1 to PaoloC
... at some point we'll decide on the list

<azaroth> +1 to BBH

bigbluehat: that we circulate to the world
... and that list will get updated over time

<fjh> +1

PaoloC: in terms of triggering more use cases? Is there anything else we can do?
... or just wait

azaorth: my perspective... lets wait till after the holidays.
... but H. and the annotation collaboration have good lists of previous systems
... we could do an analysis of what features they support
... that could drive use cases

rayd: i have four or five more use cases that I haven't contributed
... but I'll do that.

shepazu: it might be useful if we had a Use Case Czar
... to pester the masses

<bigbluehat> shepazu: submit a use case, get a W3C sticker? :)

shepazu: and usually there's an editor for a use case document
... lets sit around the table, smoke cigars and brainstorm use cases
... when people get together physically, more magic happens

<tbdinesh> +1 on use case at f2f

shepazu: lets think about an hour at the beginning of the next F2F lets do a use case ice-breaker

azaroth: lets get together the first wednesday on the new year

<shepazu> +1

azaroth: does anyone have opinions

<Jacob> +1

fjh: if there's nothing substantive we shouldn't have a call

PaoloC: NO MEETING ON CHRISTMAS EVE?!?!?!

shepazu: I'lll be there!

azaroth: gives his benediction to us all.

<bigbluehat> 🌲

<azaroth> :)