- DRAFT -

Social Web Working Group Teleconference

26 Jan 2016

Attendees

Present: Arnaud, akuckartz, wilkie, ben_thatmustbeme, cwebber2, tantek, bengo, Rob_Sanderson, wseltzer, tsyesika, rhiaro, aaronpk, eprodrom, dmitriz
Regrets Chair: SV_MEETING_CHAIR
Scribe: azaroth

scribe: He said he was looking to finish the spec by the end of the week. No new issues have been raised.
... So status is pending editorial polish, and we have exit criteria for CR
... last piece that's missing is a test suite
... Chairs and team discussed and realized that the spec had a couple of problems
... not tech issues, but missing a conformance section and the exit criteria refer to features that aren't defined
... need to address those

<Arnaud> https://github.com/jasnell/w3c-socialwg-activitystreams/issues

scribe: If you follow the link from the agenda to the issues, Sandro has raised them

<sandro> https://github.com/jasnell/w3c-socialwg-activitystreams/issues/280 CR: What are the separate "features" of AS2?

scribe: We should discuss those now if we can make progress
... Sandro?

Sandro: Thank you ;|

<sandro> https://github.com/jasnell/w3c-socialwg-activitystreams/issues/281 CR: needs Conformance Clause

Sandro: Conformance is 281. After the deadline I know, but an issue.
... There's a QA framework from about 10 years. First requirement is a conformance clause
... Most specs do indeed have them. But vocabs have a much harder time with them.
... software can conform to specs, but more fuzzy for vocabs as consumer and producer are separate
... A couple of things that handle this different are linked, and seem fairly reasonable
... Should read through them
... Most interesting one is the first, vocab-org
...
... [Reads the spec]

<sandro> https://www.w3.org/TR/vocab-org/#conformance

<cwebber2> I think we resolved an issue as a MUST about that anyway right?

<rhiaro> I think it already says something like that in the AS2 spec, right? (If it *can* use AS2, it *must* use AS2)

<cwebber2> I was the only one who -0'ed on it ;)

Sandro: If you're doing something like AS2, and you want to claim conformance, then use AS2

<rhiaro> yeah we had a MUST resolution on this recently

tantek?: I don't think it's obvious

scribe: We debated that specific problem

Sandro: It sounds familiar, and we need to be careful how to word it

tantek: If I have my random extension, can I still claim conformance? And we said no
... Have to have the AS vocab in there

Arnaud: That's in line with the example I think

<cwebber2> I'm going to speak just via text

<cwebber2> yes

<cwebber2> I'm in an office and failed to get a room

<cwebber2> so

<cwebber2> we already resolved something like this recently

<cwebber2> saying we'd mandate it as a MUST to use AS2 vocab when possible

<cwebber2> so I think it's a non-issue, we're already on that front

<cwebber2> that's all!

<cwebber2> yes that makes sense

Arnaud: Agree with what he's saying, think we need the section, but that it's already been discussed

tantek: Right, doesn't obviate the need, but something we can use to start building a conformance section

sandro: May just be editorial

Arnaud: Something we should give a chance to the editors to put something together
... Going to turn to you Evan?

eprodrom: That's something we could do. There are bits of conformance spread throughout the docs
... Can consolidate them, and like the role based mechanism that HTML5 has
... e.g. if you play this role, this is what conformance means to you
... could be a good way to bring it forward
... not sure it can be exhaustive, but at least cover the 80%

[agreement]

Arnaud: Can we get you to do that, and signal the group to review, and then iterate?

eprodrom: It's on github?

Arnaud: Yes. Sandro raised the issues
... Anything else to add? Otherwise I think we have a good plan

Granularity of Features

<sandro> https://github.com/jasnell/w3c-socialwg-activitystreams/issues/280 CR: What are the separate "features" of AS2?

Arnaud: The spec does not really talk about features, but the exit criteria does.

sandro: I don't think this has to be a big deal
... I linked from the issue to a table I look at all the time for ECMA6 features by platform
... when using it, what am I going to be compatible with
... The table would list the vocab terms, at a guess

Arnaud: Turn to the group, any ideas?
... Expect the test suite to reflect those features by grouping tests per feature
... then the implementation report would be collected and consolidated
... the report would leverage those features
... Inline with what has been done for the browsers, who implements which features
... if we have enough, we declare victory :)
... Has an impact on the spec as they need to be defined
... And on the test suite and implementation report

azaroth: Can they be by class (Event, Activity, Collection etc) or do they have to be smaller?

sandro: might need to be smaller, if people want to do part of the section

eprodrom: Refering to a particular list of features already done?

sandro: No, could read the ToC as a list of features

azaroth: Right, are there features that cross between sections, or can we divide up below the ToC

sandro: Not sure what in core is separable
... hard to do anything without the core classes

tantek: Curious for those who have implemented, interested to hear what you consider features?
... not sure how useful the borders on the map discussion is beyond an arbitrary delineation
... if no implementers, then just let Evan draw some lines

Arnaud: I'm fine with this. Obvious choice is to ask the editors

<cwebber2> I feel like it's kind of hard for me to give feedback on this call, just let the editors do it

<cwebber2> and I'll have an easier time commenting afterwards

<sandro> FWIW, I don't think the TOC is a good fit in this case. Most of the features are under https://www.w3.org/TR/activitystreams-vocabulary/#properties

tantek: The tech requirements are in the spec
... the feature should be just drawing lines, and if that's not the case then that would be good to know
... so just editorial

Arnaud: If it impacts compliance/conformance, I don't think it's just editorial
... agree that the material is available for the grouping. Mostly editorial work, but if we define features then people may have an opinion about what belongs where and how many there are
... people will want to claim conformance to those features

tantek: Agree, an excellent point. I think Evan has the best first cut at it

Arnaud: Yes. Can we leave it to you Evan to make a proposal

eprodrom: Yes, that makes a lot of sense. as a first cut I'll take the core and look at vocab
... my guess is vocab will be more lax as to conformance

<wilkie> that makes sense to me

eprodrom: different kinds of application use different parts of the vocab
... we talked a bit about it previously meaning if you're using this vocab if you're doing this kind of thing

Arnaud: Anything else?
... Once we've addressed the two issues we'll be back to where we thought we were before ... tackling the test suite

Test Suite

<akuckartz> eprodrom++

<Loqi> eprodrom has 30 karma

<wilkie> the idea of there being "core" vs "other things" strongly implies "core" is the minimum conformance. seems fine to do.

Arnaud: We'll need to wait for the feature set to be defined. Have bits contributed by different people, but no one has taken ownership to drive the effort and we need that

<cwebber2> we haven't agreed on the direction

Arnaud: if someone has already volunteered I apologize and let me know

<cwebber2> for tests

<cwebber2> it keeps wavering

<Loqi> Alehors made 1 edit to Socialwg/2016-01-26 https://www.w3.org/wiki/index.php?diff=97335&oldid=97332

Arnaud: Often the editors that volunteer for the work, but not always realistic, so is there anyone willing to take the lead?
... doesn't mean doing all the work, just coordinating
... James would volunteer as he wants to see the work done, but he has a lot of responsibilities
... Would like to look for someone else

eprodrom: I think that makes sense. We've got some efforts that have started, i wonder if we can get to a validator for the output from producers, and a test suite of documents that consumers should be able to consume, if that's good enough

Arnaud: I think that someone needs to take a first crack at the test suite description, then we can discuss if it's adequate or if there's more to be done
... once we agree on the framework, we can have a set of docs, and people can contribute different parts

eprodrom: I can take on setting out what I just described, and put it up as a proposal
... could start GH projects for them. Have a validator starting point from IBM, and taking the examples out of the spec would be a good step for the other
... Put those into a GH repository as a place to start collaborating

Arnaud: I think that sounds great. Any other comments or ideas?
... Hearing none, that sounds like the plan

<cwebber2> thanks eprodrom!

Arnaud: Thank you Evan. Good thing you joined today ;)
... I think we can move on
... Hoping that James will finish up and we'll hear from him soon

ActivityPub and MicroPub status

sandro: Haven't look in the last 2 hours, but expect they'll go out Thursday
... pretty much mechanical from here on, unless something goes badly wrong

<wilkie> if that goes up soon, I'll take a look at the existing validator this weekend

aaronpk: I have draft ready to go, so just waiting on the pub process

<aaronpk> http://micropub.net/draft/

sandro: And similarly activity pub.

<tantek> great!

Arnaud: Sounds good
... That takes care of the main discussions for the call, I believe. Two things left to discuss... actions... had a look a few weeks ago and ask people to close the ones that are done or irrelevant
... haven't seen much activity on that front. Can anyone give us an update on the open actions?

<Arnaud> https://www.w3.org/Social/track/actions/open

Arnaud: seems weird to have a bunch that have been open for a long time
... if I were an AC rep and looking at the WG, I'd wonder what was going on
... So trying to nudge people to take action on the actions

<ben_thatmustbeme> we have 1 action still open by someone not in the WG

tantek: If there's open actions not in the WG, I think chairs should take ownership of them?

Arnaud: How many do we have of those... a couple on Harry

<akuckartz> I will have a look at open actions for the next meeting.

tantek: Harry, Henry, haven't heard from Elf in months ... chairs should have a look.
... We should clean the room too :)

Arnaud: Other than that ... Sandro ... do you want to talk about webex vs other?

Webex

sandro: Webex has been working okay recently for the group, but hearing reports of problems for others. I was wondering about other alternatives, and heard about UberConference

<cwebber2> does it support SIP?

sandro: no relation to the car ride company. Would you be willing to try it for a few weeks

<cwebber2> woo

<tsyesika> yes!

sandro: seems to support SIP, dial in, dial out. Not video, which we dont' use anyway

<sandro> https://www.uberconference.com/sandro1

<sandro> https://www.uberconference.com

<cwebber2> lol

sandro: maybe poke around there

<tsyesika> Webex doesn't work for me, if this supprots SIP, I'm all for it

s@https://www.uberconference.com/sandro1@@

scribe: if it doesn't work we can fall back to webex

<tsyesika> webex does not work well for me

tantek: Perhaps the chairs should dogfood it first? If it works for us, we can try it with the WG

<eprodrom_> +1

sandro: Okay, not next week, but maybe WG in a couple weeks

<Loqi> Sandro made 2 edits to Socialwg/AS2 CR https://www.w3.org/wiki/index.php?diff=97337&oldid=87518

<akuckartz> I dislike running proprietary stuff on my computer...

sandro: especially looking for red flags

Arnaud: People should investigate please.
... anyone with experience of it

tantek: Not heard of it

<cwebber2> I don't see anything on the site...

<tsyesika> I can't find anything about SIP on their site though

eprodrom: Used it extensively, simple to use, only used with 4-5 callers
... nice web UI

Arnaud: sounds promising
... sandro, I've just seen a free option and the $10 business line?

sandro: $10 per organizer. so per staff contact, but maybe per chair

Arnaud: They really want you to sign up for the whole year
... we'll try it with the chairs, and others can take a look
... let us know any issues

<cwebber2> it does not look like it supports SIP, but it looks like it supports webrtc

<cwebber2> tsyesika, ^^^

<cwebber2> based on some quick searches

<tsyesika> cwebber2: okay, that's good, i can work with webrtc

Arnaud: that brings us to the end of the agenda. Any other business?

sandro: Any new people on the call?

<akuckartz> webrtc++ :-)

<Loqi> webrtc has 1 karma

sandro: is Ira here this week?

<aaronpk> can we kill that Zakim feature? it doesn't actually work

Arnaud: Will do that next week
... Will call it a day!

<tantek> aaronpk - just the first part "Present:" is accurate per people who said present+

<aaronpk> ahh

<aaronpk> the IRC tracking is totally busted tho

Arnaud: Thanks to Rob for scribing, thanks to all for joining, see you all next week

<wilkie> thanks

[Close of call]

<wilkie> azaroth++

<Loqi> azaroth has 11 karma

<Arnaud> trackbot, end meeting

Summary of Action Items

Summary of Resolutions

Approval of Minutes of 2016-01-12 and 2016-01-19

[End of minutes]

Minutes formatted by David Booth's scribe.perl version 1.144 (CVS log)
$Date: 2016/01/26 18:44:26 $

Scribe.perl diagnostic output

[Delete this section before finalizing the minutes.]

This is scribe.perl Revision: 1.144  of Date: 2015/11/17 08:39:34  
Check for newer version at http://dev.w3.org/cvsweb/~checkout~/2002/scribe/

Guessing input format: RRSAgent_Text_Format (score 1.00)

Found ScribeNick: azaroth
Inferring Scribes: azaroth
Default Present: Arnaud, akuckartz, wilkie, ben_thatmustbeme, cwebber2, tantek, bengo, Rob_Sanderson, wseltzer, tsyesika, rhiaro, aaronpk, eprodrom, dmitriz
Present: Arnaud akuckartz wilkie ben_thatmustbeme cwebber2 tantek bengo Rob_Sanderson wseltzer tsyesika rhiaro aaronpk eprodrom dmitriz

WARNING: No meeting chair found!
You should specify the meeting chair like this:
<dbooth> Chair: dbooth

Found Date: 26 Jan 2016
Guessing minutes URL: http://www.w3.org/2016/01/26-social-minutes.html
People with action items:

[End of scribe.perl diagnostic output]

Socialwg/2016-01-26-minutes