Socialwg/2014-10-21-minutes
Derived from RRSAgent minutes
Also see: IRC log
Attendees
- Present
- +1.514.554.aaaa, jasnell, Arnaud, AdamB, evanpro, jtauber, bret, cwebber2, oshepherd, wilkie, MarkCrawford, Lloyd_Fassett, tantek, +1.408.335.aacc, Shane, hhalpin
- Regrets
Chair - Evan Prodromou
- Scribe
- Adam Boyet
Resolutions
RESOLVED: Approval of Minutes of 14 October 2014 Teleconf http://www.w3.org/wiki/Socialwg/2014-10-14-minutes
Discussion
<Arnaud> I put a strawman schedule
<trackbot> Date: 21 October 2014
<oshepherd> Oh, I'm just slightly early for a change :)
<evanpro> Yes
<Arnaud> so, we need to merge the two :)
<evanpro> Hmmm
<evanpro> OK
<cwebber2> I think I'm in
<oshepherd> No hello/goodbye notifications?
<tommorris> Apologies for absence: been feeling sub-par as a result of continuing medical treatment.
<evanpro> Arnaud: that's such a better structure
<evanpro> Let me see if I can copy that over
<evanpro> Probably not quickly
<bret> (I think)
<jasnell> fyi... laptop battery likely not going to last the full hour.. may end up dropping off irc at some point... we will see tho
<jasnell> wow.. that's super irritating
<cwebber2> Zakim: , ??P5 is me
<cwebber2> oops
<evanpro> Whoever is leaning on their phone, PLEASE STOP
<cwebber2> honk honk
<evanpro> https://www.w3.org/wiki/Socialwg/2014-10-21#Agenda
<evanpro> AdamB, would you mind scribing?
<evanpro> scribe: AdamB
evan: first item approving last weeks minutes
<evanpro> http://www.w3.org/wiki/Socialwg/2014-10-14-minutes
evan: pretty important resolutions in the minutes ....
... any objections to approving?
... minutes have been approved
RESOLUTION: Approval of Minutes of 14 October 2014
TPAC F2F
evan: like to talk about upcoming face to face ... unfortunately have 2 diff schedules
... will use Arnaud version
<Arnaud> +q
<Arnaud> ack ??P5
evan: first covering overall direction as a group, where we are going
... discussing AS 2.0 and moving that forward
<scribe> .... new efforts coming up around Social API and Federation protocol ... F2F best time to discuss those
Arnaud: first would like to ask people ...schedule of tpac provides 11am to 5pm for adhoc group breakouts ... put together on F2F page for two possible options to choose from
<tantek> agreed that f2f is a good time to discuss API and federation protocol work
Arnaud: ... as a working group we should decide our approach ... on that list is straw man to start discussion
... think we should choose between these 2 options
... could mix for example, could do option 1 on monday and option 2 on tuesday
<tantek> I am also participating in the AC meeting as an AB member. So would prefer not to conflict with that.
evan: one way to max. productivity by taking first day to address 3 major goals of this group
... if conversation goes well on monday will likely have tuesday then to do more break out sessions
... suggesting option 1 on monday and 2 on tuesday
<tantek> Option 1 on Monday, and Option 2 on Tuesday works for me too
<tantek> I would drop "WG Dinner" since usually that's a good time for cross-group pollination
proposal of: option 1 on monday and tuesday do more break out sessions during the mid day break
<evanpro> +1
<jasnell> +0
<wilkie> +0
<jtauber> +0
<bret> +0
<tantek> +1 with including federation protocol on Monday, and no WG Dinners
<elf-pavlik> +0
<MarkCrawford> +1
<cwebber2> I'm abstaining from all the TPAC votes since I won't be there ;)
<MarkCrawford> WG dinner on Monday or Tuesday?
<MarkCrawford> Tuesday is AC dinner
<harry> Monday would work!
<Shane> Hi all, sorry I'm late
<tantek> Tuesday has AC dinner that Arnaud (and Tantek) will attend
<tantek> can we have a quick poll of who where in IRC will be at TPAC and which days?
<tantek> e.g.
<tantek> MTWThF
evan: given this structure can we break up the day on monday in to 3 blocks
<harry> MTWThF
<MarkCrawford> MTWThF
<rhiaro> MTWThF
MTWTh
evan: are ok with leaving agenda for day 2 relatively open and putting that together at the end of day 1
<MarkCrawford> Can we have some time for use case presentation from the SocialIG?
arnaud: seems to make sense
evan: some have been discussing proposals ... this is the optimal time to start making suggestions for presentations etc
... asking that folks that are willing to do that to post to the email list and will fit in to those slots on afternoon of monday
<harry> Notes that Matt Marum isn't on the call - not sure who from OpenSocial is presenting their take on things
evan: 5-10 minute presentations
<oshepherd> I have a proposal I'll email in this week (but unfortunately can't attend TPAC)
<evanpro> oshepherd: that sounds great
<tantek> can we skip email middleman and just edit wiki schedule directly to add ourselves?
<evanpro> tantek: yes, that's fine
<elf-pavlik> harry, anyone from OpenSocial in https://www.w3.org/2002/09/wbs/35125/TPAC2014/registrants#SW ?
arnaud: referencing marks question about giving some time to social IG use cases
<Arnaud> TPAC attendance: All week
MarkCrawford: have narrowed down the use cases ... continuing to work them ... would like to share them to the WG
... important for us in the IG to report back on where we stand at this point
evan: sounds great, make time on tuesday morning on that?
MarkCrawford: that should work, schedule around use case task force leader availability
evan: lets plan on tuesday barring any objections
<Arnaud> +q
<KevinMarks> I may be remote too
Arnaud: remote participation .. elf and shane remote is all right now? are there others planning to call in?
<oshepherd> I might try and call in - haven't had time to look at timing yet
<Shane> I've not checked the times yet for whether I can remote, I think there will be issues with timezones, but we shall we
<Shane> *see
Harry: all the rooms have polyphones ... can put a bridge booking request in now
<Zakim> tantek, you wanted to also offer broadcasting AV with talky.io for remote particpants to listen/watch
evan: sounding like a general framework and with the chairs can resovle the schedule
tantek: think its fine to just add directly to wiki
<cwebber2> that'd be great
tantek: for remote participation I can setup a audio/video broadcast using Talkie RTC session
<KevinMarks> talky++
<Loqi> talky has 3 karma
evan: will be able to show screen?
<Shane> Yeah talky has show screen
tantek: it can pick from multiple inputs
<elf-pavlik> talky sounds great!
tantek: including screen share
<harry> KevinMarks, I can't find your W3C account.
waht about the network issues ?
<harry> Do you have one?
<harry> A user-name?
slow network?
tantek: also setup an etherpad (sp?) especially if we start doing breakouts
... can be hard to follow IRC
<Shane> During indiewebcamps etherpads have been extremely useful
evan: sounds great, lets do both
<Zakim> elf-pavlik, you wanted to discuss who will participate on Wed in Schema.org and Social WG? https://www.w3.org/wiki/TPAC2014/SessionIdeas#Schema.org_and_Social_WG
<elf-pavlik> ^
tantek: this is all in addition to IRC
<KevinMarks> kmarks2 is apparently my username, but when I try to reset I get stuck
harry: in response to elf-pavlik question .... breakout session on wednesday to talk about Schema.org and Social WG
<KevinMarks> I'm happy to start a new account but it didn't like me trying that with an email it had seen (kevinmarks@gmail.com)
evan: I will be there
... def some coordination needed there
<elf-pavlik> evan: could you add yourself to that section in a wiki?
harry: its an open space 1hr session. good to be there from beginning but can drop in at any time
<elf-pavlik> https://www.w3.org/wiki/TPAC2014/SessionIdeas#Schema.org_and_Social_WG
arnaud: tpac registry is officially closed but can register on site
<bret> oh wow
<harry> Note we are running at capacity
<cwebber2> guess for sure doing remote participation then now :)
harry: our social group space is completely booked
<harry> However, some folks who are "observers" won't show
<evanpro> Probably me!
harry: space is booked because of observers ... members have priority over them
<harry> Also, speaking of WG members, I'm trying to double-check IEs - one of chairs, e-mail me the list of folks that need to be approved so I can double-check?
<Zakim> tantek, you wanted to advocate asking for folks to join the SocialWG
<harry> For schema.org meeting, please add name to wiki here:
<harry> https://www.w3.org/wiki/TPAC2014/SessionIdeas#Schema.org_and_Social_WG
tantek: regarding about participation ...since we are full for monday and tuesday if you are attending as observer please join as member
<harry> Actually, I suspect Ralph Swick will be leading, but I proposed
tantek: for folks that do plan on wednesday session about schema and social wg .. for those attending from social wg strongly encourage them to join
evan: this wraps up tpac conversation
<tantek> encourage them to join *the Social WG*, so that we don't have to do discussions in side-meetings.
evan: next want to talk about AS 2.0 working draft
<evanpro> OK, harry, will ack in a moment after jasnell
jasnell: first cut working draft are in and been working with harry fixing up details on html validation. hopefully those are resolved
... one question about the Actions draft
<Arnaud> if you're not registered but plan to attend, please add your name to https://www.w3.org/wiki/Socialwg/2014-10-27#Participation
<harry> Its kinda weird to hold on actions since the drafts are very interlinked.
jasnell: way the docs are right now the first public working draft points to the editor draft
<evanpro> Sorry, other people in my office
I missed the question
<tantek> has anything changed re: the questions asked last week about Actions?
due to noise
<tantek> if no new information, let's not re-open the question
AS2.0 FPWD
harry: still trying to aim for thursday pub date ... because all 3 drafts are linking to each other the linking validation fails
<tantek> the AS and vocab drafts shouldn't need to link to Actions - this doesn't make sense
harry: wouldn't be good to publish all 3 together would have re-write links
<Arnaud> the question is whether we can publish Actions draft along with AS 2.0 Core and Vocab
<KevinMarks> I am logged in now, but boy that was hard
<elf-pavlik> tantek, core has section on PotentialAction which uses terms from Actions Vocab
<tantek> using pub process as an excuse to publish a previously unapproved draft is kind of bad form
<tantek> so drop the crosslinks to Actions
harry: in terms of just pure publication have them heavily linked and not on w3c space makes publishing little bit trickier
<tantek> just comment them out
<tantek> substantial reasons / questions - all in last week's minutes
<tantek> no implementation experience with Actions
<harry> The problem is everything is densely interlinked so we have to do changes to drafts.
<tantek> it was a bunch of political +1s with no pragmatic questions asked
<elf-pavlik> just move to dedicated spec PotentialAction section?
jasnell: actions piece is still an open question ... def. interesting in working on but was missing clear use cases so objection was raised due to that
arnaud: minutes from last week show that clearly
... there is interest in group but non documented yet
<Loqi> elf-pavlik meant to say: just move to dedicated spec PotentialAction section?
jasnell: base on that we don't have documented ones for AS either
<MarkCrawford> implementations are not necessary until we get further along
<tantek> why not just cut everything to do with Action from the first two drafts?
<harry> Sorry, W3C team publishing requires links to resolve
<tantek> is such commenting out more than 5 min of work?!?
<elf-pavlik> +1 tantek
jasnell: does look to be a public link issues from publishing perspective
<harry> and github.io doesn't do fragment id links
<harry> So, we can just remove links to Actions Drafts
<harry> that's fine with me, it may delay things a bit.
<KevinMarks> +1 on links having to resolve
jasnell: could just publish with actions but that still doesn't commit the wg to do anything with it
<tantek> harry - why does a 5 minute commenting-out task delay things?
<harry> Or we can put a weird snapshot of the Editors Draft somewhere outside github.io
<bret> Shouldn't use cases be examples where there exists some kind of implementation, standardized or not?
<tantek> we're spending more time talking about it than it would take to comment out the links
<bret> not just, ideas
harry: we hit a hitch in w3c because of the links not resolving .. in general want them to resolve properly
... it might slow things down past thursday
evan: is it reasonable to kick out actions for the first version
jasnell: could do that .. basic actions has been in AS since the begining
<Shane> That makes sense to me
jasnell: could do that if that is what this group wants
<elf-pavlik> +1 move actions to dedicated spec
<MarkCrawford> as an editors draft, why can't we publish w/o the links to gain a broader audience for the work and get more feedback?
evan: from my view ... since some discussion and not competely essential to core would suggest trim them out now and address later on
<Zakim> elf-pavlik, you wanted to discuss Reference issues inline in the spec. https://github.com/jasnell/w3c-socialwg-activitystreams/issues/29
<elf-pavlik> ^
<elf-pavlik> i got impression last week that James agreed
<harry> activitystrea.ms
<tantek> it is a modularity failure because an AS2 implementation does not need to know anything about Actions
<elf-pavlik> <wilkie> elf-pavlik: jasnell said that they would be included in the spec as yellow highlighted sections that link to the github issue I believe
<tantek> therefore the spec shouldn't need to
harry: its not a modularity issue but a links not resolving ..other issue was AS namespace referenced within
<MarkCrawford> Harry - does the resolving links requirement apply to all versions, or just final?
<harry> 404 errors
<harry> which is just not acceptable
harry: lot of links get 404 errors which is not acceptable
... could monkey patch the links out
<tantek> Proposal: comment out all the links to Actions (and relevant contextual text) from the two drafts.
jasnell: the AS namespace is personally owned by Chris Messina is maintained by AS on github
... i have access to that domain. larger question is do we want to keep that domain / namespace or not?
<Shane> Would it be better to move them to the w3 domain, perhaps wiki?
<Arnaud> good question, shouldn't we switch to a w3c namespace?
<elf-pavlik> https://w3id.org/ ?
<tantek> is the namespace issue a FPWD blocker?
<tantek> if it's not a FPWD blocker then we should postpone discussing
<oshepherd> elf-pavlik: No, http://w3.org/<blah>
<tantek> let's get past FPWD please
<harry> it is currently kinda blocking FPWD
evan: two big issues: 1. adding or removing actions 2. removing AS in the namespaces
<Zakim> tantek, you wanted to discuss this as evidence of modularity failure
<harry> i.e. I'd like to make some changes to the doc but want to make sure the WG is OK with it.
<elf-pavlik> evanpro, can we come back to my question after you ack me?
<evanpro> acked
<harry> i.e. we can probably monkey-patch the Action links out
<evanpro> elf-pavlik: yes, please put yourself back on the queue
<evanpro> Sorry about that
<harry> but we need to also resolve activitystrea.ms
tantek: given existing impl have not had to know anything about actions that the spec rec doesn't need to refer to it either. that is what i mean by modularity failure
... its like a layering failure
<oshepherd> +1 on modularising spec appropriately. If ActivityStreams is extensible, actions should be able to live on its' own
<Shane> I think the activitystrea.ms should redirect to a page on w3.org, it is unlikely it will dissapear then
tantek: no reason core spec should have to mention it at all. so suggestion its completely commenting out from spec.
... lets pospone any first draft issues
<Lloyd_Fassett> Zakim +Lloyd_Fassett
<tantek> any *non* first draft issues
harry: namespace is pretty close to a blocker
<harry> We can remove all links to activitystrea.ms
<tantek> let's do that
<tantek> remove all links that break
<harry> 1) we could comment them out directly
<harry> 2) we could replace them with w3.org names
harry: could remove them, comment them out but that is weird cause they are referenced
... could replace with w3c ones
<harry> 3) we could ask ChrisM to update real quick so they don't 404
<harry> I'm OK with any with those
harry: or could update domain so they don't 404
<oshepherd> If we go for (2), is it possible to make those names always redirect to latest draft?
jasnell: does have access so can do it
... just need to have the time
evan: its more about the 404 than the namespace issue
harry: its kind of weird ... its been done but not common
<evanpro> Can we use http://www.w3.org/Social/WG ?
<Arnaud> +q
<tantek> I'm ok with external URLs / namespace
harry: should at least dedicate real thought sometimebefore last call
<tantek> simplest thing
harry: have to fix broken links now
<evanpro> s/activitystrea.ms/www.w3.org/Social/WG/
<Loqi> evanpro meant to say: Can we use http://www.w3.org/Social/WG ?
<tantek> comment out the broken links, "fix" in next draft
<harry> we gotta do something about broken links now, the external namespace issue we can deal with later
evan: james, can we setup way to fix broken links
<harry> The easiest thing is to just probably comment them out
<Arnaud> solution: remove the links for now, open an issue on namespace to use
jasnell: can try to get it done before end of week but not positive i can get it done
<tantek> I'd say if you can't fix the links in the next hour, then let's comment them out for FPWD
<Shane> That's probably the best way, for FPWD
<tantek> that buys us more time to make a considered decision rather than rushed
harry: i can comment them out for now and add them back in later? are people ok with this?
<MarkCrawford> +1 on commenting out the links
evan: maybe comment out and open issues to add them back in
<harry> I'll put an clearly editors draft
harry: will do that and mark as editors draft
<harry> redirect Actions links to the Editors draft
<harry> that should solve all our linking issues
<tantek> why not just comment out Actions?
<harry> Yeah, we could also comment out Actions as well.
<tantek> that's better for modularity anyway
<tantek> which is why I'm pushing for that
harry: will just comment out all broken links
<elf-pavlik> +1 move actions out of core
<tantek> this also sets a good precedent for the group
<harry> We can probably still push things out by Thursday if Webmaster isn't too rushed and all this link commenting works
<tantek> if your spec has broken links after we've agreed to publish it, then we will comment out the links
<tantek> I think that's a reasonable policy to adopt
<harry> +1
<tantek> it will strongly encourage editors to do a trivial linkcheck EARLY and OFTEN
<harry> just need to be fair but also realize W3C, being what it is, can't publish broken links :)
<tantek> harry - totally agree with not publishing broken links (404)
<Zakim> elf-pavlik, you wanted to discuss Reference issues inline in the spec. https://github.com/jasnell/w3c-socialwg-activitystreams/issues/29
<elf-pavlik> [from last week minutes] <wilkie> elf-pavlik: jasnell said that they would be included in the spec as yellow highlighted sections that link to the github issue I believe
<elf-pavlik> [from last week minutes] <wilkie> elf-pavlik: jasnell said that they would be included in the spec as yellow highlighted sections that link to the github issue I believe
evan: discussing elf question
... not sure understand what the issue is yet elf
<jasnell> OK... I forgot that line item
Arnaud: it has to highlighting within the text with the issue
<jasnell> I can add those in today
<elf-pavlik> jasnell, thx!
<KevinMarks> is this a fragmention use case?
<jasnell> That's a minor fix
tantek: believed would be called out inline within the document. we agreed to publish with that being the case
<jasnell> Yes
evan: jasnell can we get this in?
<jasnell> That's a quick one
jasnell: yes
<harry> I would say we do "inline issues" on next WD
<elf-pavlik> harry, how about -comments mailin list?
Arnaud: we should use what ever is the most efficient way to the document out. but should put in a tracker for what namespace should be used going forward
<jasnell> Oy... Ok
<tantek> harry - except that we agreed to publish *with* the issues inline.
harry: need those inline issues within the next hour to make it by thursday
<tantek> jasnell - can you add those issues inline in next hour?
<elf-pavlik> jasnell, please :)
<harry> The issue is that this requires yet another cycle from W3C to fix up, so I say send *exact* issues to mailing list
jasnell: i can try to get it in the next hour but very difficult. if there are higher priority ones send email and i can try to do them by most important
harry: webmaster needs 1 day to review it
<elf-pavlik> issues: https://github.com/jasnell/w3c-socialwg-activitystreams/issues/29
harry: thats how publications always worked
<elf-pavlik> jasnell, just 4 main issues there
how hard does that make it for adding issues in the future? does webmaster always have to review those too ?
<tantek> I'm ok with waiving this requirement for FPWD
<tantek> elf - are you ok with postponing linking inline to issues until next draft?
<Shane> Is the issue about Hydra actually an issue of the spec? I don't think it is, just perhaps something to discuss later possibly
<elf-pavlik> I think we at least need to mention missing JSON-LD @context in examples, but other 3 would also make sense from my POV
<harry> yep
harry: inclined to suggest to link this in later
<harry> https://github.com/jasnell/w3c-socialwg-activitystreams/issues
<harry> https://github.com/jasnell/w3c-socialwg-activitystreams/issues/36
<tantek> and does anyone else object to publishing FPWD without the inline issues?
<harry> I mean, we link to the identifers
<harry> there's 15 issues
<harry> all RDF related
<harry> and opened by elf
<harry> I'm not sure if this is really blocking
Arnaud: sounding like we go with out the links are we don't go .. not sure if elf would like to hold up the process for that or not
<jasnell> Those links aren't critical in my opinion. Not as a blocker.
<jasnell> There is a link to the issues list in the draft
<elf-pavlik> can we at lest mention missing @context ?
<harry> Yes, we do link directly to the tracker
<evanpro> elf-pavlik: are you comfortable with going to FPWD without these links?
tantek: if there is only a single objection its the chair prerogative to declare rough consensus
<Shane> I think we need to get it to FPWD
<harry> I'm inclined to say "no"
<evanpro> Otherwise, we won't get the doc out
tantek: if we are risking not getting it out i don't think its worth postponing for this right now
<elf-pavlik> i will not block but would appreciate at least mention of missing JSON-LD @context in examples, and maybe one about _:post
<harry> OK, so that's already linked and examples have @context assumed called out
jasnell: yeah thats as simple as a link ... but if we have no place to host that officially but there is a link to that context in document already
<evanpro> elf-pavlik: that's already called out
evan: since it's called out maybe take an action to link directly but willing to go with what we have right now
<elf-pavlik> with a link to context at that stage?
<Shane> Sorry I have to rush away, thanks everyone.
<evanpro> elf-pavlik: we'll have that, yes
harry: most working groups only call out substantial issues directly in the spec
<elf-pavlik> anyways, i trust in your decision folks so just go ahead without including that request from me
<harry> For example, WebCrypto had 170+ bugs
<harry> we called like 5 out in text
<harry> inline
<elf-pavlik> ok, end of this topic?
<tantek> admin: we're also 4 min over
evan: would like to wrap up agenda items we have open and defer to next week at F2F, any objections?
<tantek> no objection
evan: ok would like to adjourn
<tantek> +1 adjourn
<bret> nice meeting have fun next week
<oshepherd> Regrets on not being there in person
<harry> I'll do some editing and will keep trying to get docs out by Thursday, but no promises :)
<bret> will try to remote
<tantek> c'mon harry, commit!
<Arnaud> harry, do we have a meeting room #?
<harry> I'll ask Susan, but I haven't been told
<Arnaud> would be good to add that info to the meeting page
<harry> susan@w3.org knows
<harry> Arnaud, please send me list of accepted IEs
<harry> so I can double-check right now
<Arnaud> this info seems to be lagging
<Arnaud> here is what's currently available (because I asked ;-) http://www.w3.org/2014/11/TPAC/schedule.html
<harry> I think there's some last minute room re-arrangment going on to TPAC being overbooked.