Derived from RRSAgent minutes
Also see: IRC log
- jasnell, Arnaud, +1.314.777.aaaa, wilkie, AdamB, Lloyd_Fassett, bret, markus, tantek, rhiaro, jtauber, cwebber2, Tsyesika, +1.703.485.aabb, Shane, hhalpin, bblfish, Tyesika
RESOLVED: Approval of Minutes of 7 October 2014 Teleconf http://www.w3.org/wiki/Socialwg/2014-10-07-minutes
RESOLVED: Publish AS 2.0 spec as a First Public Working Draft (FPWD)
RESOLVED: Publish AS 2.0 Vocabulary spec as a First Public Working Draft (FPWD)
<trackbot> Date: 14 October 2014
<elf-pavlik> sorry everyone, today IRC only once again :( (lost VPN and tried SIP in 3 different locaitons in last 3h, also Skype gift card friend sent me didn't arrived)
<bret> anyone recomend a decent json formatter extension for firefox?
<cwebber2> bret: jsonview
<bret> ty cwebber2
<Arnaud> rats, this telecon isn't set up correctly
<Arnaud> we have issues every week
<bret> darn, .jsonld files in http://jasnell.github.io/w3c-socialwg-activitystreams/activitystreams2.html#bib-JSON-LD still try to download to disk :(
yay i'm connected :D
<bret> Zakim ??P2 is me
<cwebber2> how do I identify who I am if calling in via SIP?
<cwebber2> 41# you said?
i'll try that once chris has tried
<cwebber2> ??P1 is me
<cwebber2> oh, duh :)
i'm ??P5 i think
<bret> is there a way to get Zakim to remember who you are?
<wilkie> wow it hate me
<wilkie> it picks me EVERY TIME
<tantek> wilkie I think Zakim really likes you
I've not scribed before D:
<wilkie> that's also a possible interpretation :)
<cwebber2> Tsyesika: don't worry, I did it last time because I was the noob also :)
<wilkie> I always assume machines hate me more than love me
<cwebber2> it's not hard
<tantek> who just joined?
<tantek> thanks Shane :)
how will i know who's speaking?
<harry> trackbot, start meeting
<tantek> Tsyesika: if you hear a French accent, it is likely Arnaud :)
<trackbot> Meeting: Social Web Working Group Teleconference
<trackbot> Date: 14 October 2014
okay i can try
<tantek> scribenick: Tsyesika
<Shane> Can I scribe next week please? I will not be able to make it for a few weeks afterwards while there is a timezone shift
<cwebber2> Tsyesika: so now you would do
<cwebber2> Arnaud: harry, come on, you show up 5 minutes later ;)
<Arnaud> harry you're not the chair
so i just write what people say more or less?
<tantek> welcome harry ;)
<Arnaud> scribe: Tsyesika
<cwebber2> Tsyesika: and if they were talking last time when you did Arnaud: you don't have to do it on the next message
<Arnaud> Proposed: Approval of Minutes of 7 October 2014 Teleconf http://www.w3.org/wiki/Socialwg/2014-10-07-minutes
<cwebber2> Tsyesika: also if talking for yourself, you might want to prefix your own name
<jasnell> no objections
<bret> look good
<Shane> Looks fine here
Arnaud: approved the minutes
<Arnaud> Resolved: Approval of Minutes of 7 October 2014 Teleconf http://www.w3.org/wiki/Socialwg/2014-10-07-minutes
Arnaud: next meeting will be next week
erm October the 21st?
Arnaud: wanted to know objections to using JSON-LD
who's talking now?
<Arnaud> you can type xxx: blah blah
<Lloyd_Fassett> James Snell
<Arnaud> and the person will do s/xxx/jasnell/
<cwebber2> no objections here
Arnaud: is closing action 2 as there is no objections
<Arnaud> trackbot, close ACTION-2
<trackbot> Closed ACTION-2.
Activity Streams 2.0 Publication
<harry> It didn't seem to get too much review, but we can still publish at a FPWD as Working Drafts can still change of course.
Arnaud: james has fixed some typo's in the draft and we're going to make a decision today on if we're going to publish this as the first public working draft today
<jasnell> I will make that proposal
<Zakim> tantek, you wanted to ask about implementations
Arnaud: there are two sections which are the main spec and the vocab spec, we decided not to publish the action spec as the first public working draft today
<Shane> I like that it specifies that multiple serialisations are allowed, despite not being defined in spec.
tantek: the spec looks good, but a question "Do we have any implementations which are keeping up with the spec, as in AS 2.0 as defined in the spec today"
<harry> eventually folks should try :)
<AdamB> we are looking to move to produce and consume based on the new AS2 spec
<AdamB> we, being, Boeing
jasnell: The current draft is defined so all AS 1.0 is compatabile with 2.0
<tantek> for the record, 0 implementations publishing AS2 as spec'd today, and 0 implementations consuming AS2 as spec'd today.
jasnell: I have one working but... (sorry i missed that part?)
<bret> apologies call dropped
<elf-pavlik> can we highlight open issues inline in the draft? https://github.com/jasnell/w3c-socialwg-activitystreams/issues/29
<tantek> and plans to produce 1-2 implementations publishing and consuming AS2 as spec'd today.
jasnell: said we can highlight the issues in response to elf-pavlik's question
<markus> ok.. that also addresses my concern about the examples not referencing a context
<markus> would still prefer it they did though
<Arnaud> PROPOSED: Publish AS 2.0 spec as a First Public Working Draft (FPWD)
<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
<Arnaud> RESOLVED: Publish AS 2.0 spec as a First Public Working Draft (FPWD)
<Arnaud> PROPOSED: Publish AS 2.0 Vocabulary spec as a First Public Working Draft (FPWD)
<Shane> +1 I have looked at vocab less but it can still change as a draft so yes :)
<harry> yes, we can still get reviews in, this just puts the document on Rec-track
<harry> MarkC: +1 both proposals
<harry> (via voice)
<Arnaud> Mark Crawford +1 (over the phone)
<Arnaud> RESOLVED: Publish AS 2.0 Vocabulary spec as a First Public Working Draft (FPWD)
<jasnell> I would like to propose that we consider the Actions draft for publication on next weeks agenda
Arnaud: because this is the first public working draft so the chairs have to ask the W3 staff for the first publication of the document
<markus> jasnell, could we move the section on actions from the core spec to the Actions draft?
<tantek> jasnell, do you have a preferred shortname?
Arnaud: this is because they want to make sure they want to make sure theres is a record of approal and a short name
<harry> I'll help as well - there's a bunch of formatting tests (pubrules)
Arnaud: the document needs proparing for publication and it needs to pass the validation tests
<jasnell> tantek: preferred shortname for?
<harry> So, this is the URI the doc will be known by
<tantek> jasnell, your document
<elf-pavlik> +1 markus (i also mentioned it in my email)
<jasnell> no preference
Arnaud: suspects it'll probably take the next 2 weeks to get it published
<tantek> e.g. as-core , as-vocab
do i do the proposed thing?
<Shane> tantek: Should the shortname not have the 2.0? Since there is already another version
Arnaud: Wants to know what it would take to get the actions spec published today
<harry> So an example of shortname is: http://www.w3.org/TR/WebCryptoAPI/
<markus> I like as-core etc.
<tantek> jasnell, just noticed that activitystreams2-vocabulary.html lacks normative reference to HTML5
<tantek> not a big deal, but likely to be flagged by syspub before publishing
<Shane> I would go for ActivityStreams2 and ActivityStreams-vocab personally. But doesn't really matter either way
Arnaud: in response to the question on IRC about the shortnames there are some shortnames in the draft
that being activitystreams-2.0
<tantek> I'd leave off the version number
<harry> I'm happy to leave this decision to the editor
<harry> Leaving off version number might help if we plan to do 2.1
<harry> or 2.2
<jasnell> no strong opinion, I'm open to whatever the group decides
<markus> lets keep the shortname short :-)
<tantek> markus, exactly, hence I suggested as-* instead of activitystreams-*
Arnaud: wants to know if we should not have the 2.0 there
<markus> yep tantek, I already said I like your proposal :-)
<elf-pavlik> BTW http://prefix.cc/as
<harry> I'm happy to keep version number
<jasnell> +1 to omitting the version number in the short name
<harry> but it doesn't really matter - only matters if we do 2.1 or 2.2
<harry> which we might :)
Arnaud: it seems like there is a general feeling we should keep off the version number
<markus> +1 to ommitting the version number
<Shane> I can see the reasoning to avoid the version, for newer versions
Arnaud: do those two names work for everyone
<tantek> Arnaud, I can live with that.
<cwebber2> could this bite us somehow if versioning needs to happen?
<markus> what about http://www.w3.org/TR/as-core instead?
<tantek> As suggested I think we can do shorter
<elf-pavlik> how about actions vocab ?
<markus> what about http://www.w3.org/TR/as-vocab instead?
<tantek> cwebber2: hasn't in the past
<elf-pavlik> +1 shorter
<bret> +1 to short ie as-
Arnaud: hearing no objections, we will make that change to the short nmes
<cwebber2> tantek: ok, sounds good then
<elf-pavlik> as-core, as-vocab, as-actions, as-actions-vocab
<Zakim> tantek, you wanted to discuss actions
Arnaud: people expressed last week that they weren't ready to publish actions yet
tantek: at the time and still now, my objection, for activity streams we have quite a lot of implementation experiance but for action
<jasnell> we have implementors for Actions but they are currently updating to the current version of the draft
hold on i seem to have dropped out of the call
<cwebber2> I'll cover
<cwebber2> tantek: there hasn't been clear examples of implementations it seems yet? Does this make it preliminary to publish as a draft?
<cwebber2> jasnell: we do have implementers, but implementers of the draft... we don't have a current implementation, but do have implementation that has gone on
<cwebber2> no years of experience, but there has been some implementation work
( cwebber2 i'm back now so if you want i can take over )
harry: i'm ??P5
<cwebber2> tantek: (??) I'd like to have that documented on the wiki
<cwebber2> so we can have evidence before moving forward
<cwebber2> okay Tsyesika, go ahead
who is speaking right now?
<Shane> I would like to see examples of some of them, such as how and what BrowserView is used for (the notes help but aren't completely explanatory)
<harry> +1 having evidence before going to Last Call, ideally as soon as possible
<wilkie> Tsyesika: tantek and jasnell
jasnell: is saying there is there are implemntations we can draw from
<wilkie> Tsyesika: jasnell right now
jasnell: if we need to hold off and wait a little while until we see more implementations, that's fine
<harry> That's why I was saying implementation evidence *will* happen
Arnaud: we're only talking about the first publication of the action spec not the last call
<harry> just not needed right now per se IMHO
<jasnell> I'm far less concerned about implementation experience for FPWD
<elf-pavlik> jasnell, what about content from actions spec before vocab overwrote it? https://github.com/jasnell/w3c-socialwg-activitystreams/blob/150a3b80f6752c9351374503a0f11be304f0ee6e/activitystreams2-actions.html
<cwebber2> the actions spec?
<Arnaud> STRAWPOLL: are you interested in the AS Actions spec?
<tantek> how about are you interested in *implementing* it?
<bblfish> I'd like to understand how it fits in with LDP
<tantek> -1 to poll - too many "interesting" things
Arnaud: are you interested in the proposal?
<cwebber2> oh I see
Arnaud: there is no requirement to have an implementation for a public working draft
<wilkie> I am indeed interested in implementing it
<tantek> my point was about WG focus
<Shane> +0 I understand the reasoning behind some of it, but others I would need examples for to be interested in those particular actions
Mark supports the proposal
<Arnaud> Mark Crawford +1 (over the phone)
<harry> Note that typically the W3C tries to get implementation experience as soon as possible now, althogh it's not formally required for FPWD (it's only required to exit CR)
<jasnell> btw, I plan on discussion the motivations behind actions in detail at TPAC
bblfish: has anyone looked in with how this fits in with LDP as it's a W3C standard?
Arnaud: I am not sure there is a direct connection
jasnell: There is definitely more work that needs to be done on this, and there is some overlap with LDP?
Arnaud: there is some interest around the spec, tantek what would it take for you to remove your objections
<bret> I'm curious to hear Evan's take on actions
<jasnell> we do have experience with Actions
tantek: wants to know for those who have +1ed, why are you interested in a spec which has no implementations
<bret> (ty bblfish)
<harry> I think its good to get some FPWDs out there just to focus the discussion - but I am not voting on this one :)
bret: that's probably me sorry, i muted myself
<jasnell> we (IBM) views Actions as the evolution of the Embedded Experiences work, which was part of the initial contribution that created this WG
Arnaud: does anyone want to say why they're interested in the action spec
<bret> Tsyesika: np ;) your doing good work!
<jasnell> I will address our motivations for actions at tpac
<tantek> as a group we're supposed to focus on standardizing areas with implementation experience
<tantek> why are those who are +1ing without documented evidence expecting the group to work on this?
<harry> My opinion is FPWDs are good to start to focus discussion
<harry> I expect implementation to come along, and would be worried if they didn't.
<wilkie> it's a working draft. I don't expect a implementation to be in sync with the draft. the actions spec gives a level of extensibility that is very useful to explain to software how activities can be generated. I've discussed on the github issue tracker about making that extensibility a usecase and it seems like it is a focus. good! :)
Arnaud: the charter doesn't prohibit us from going further with this even though there is no implementation
<elf-pavlik> http://schema.org/docs/actions.html with similarity to as:PotentialAction gets some adoption, eg. Github sends them with email notivications http://manu.sporny.org/2014/github-adds-json-ld-support/
<tantek> prior art is not an implementation
<tantek> nor is *email* an implementation of *web*
<elf-pavlik> one can send them over XMPP as well ;)
<Arnaud> I didn't say "prior art" but "state of the art"
<tantek> "can" does not matter. "does" matters.
<wilkie> if a system can know how actions relate to activities, a system can perform actions and thus generate activities it has no prior knowledge of, which is very interesting.
jasnell: the use cases are what is driving this. When we're talking about mobile push notifications we're doing very similiar things, we're trying to pull pieces from a number of different areas. This isn't something we're not just pulling out of the air
Arnaud: if tantek is the only one to object, we will try to address the objections but if not we will try to move past the objection
<Zakim> bblfish, you wanted to discuss is there a list of actions use cases somewhere?
<harry> Then the W3C will try to see if folks in WG have done their best and consider the formal objection.
Arnaud: and he can file a formal objection
<jasnell> Tantek: I'm not interested in just pushing something to recommendation that no one uses, I'm interested in working a useful spec that folks can use.
harry: are there a bunch of use cases?
<bblfish> Use-case discussion is IG BTW in general.
<Loqi> harry meant to say: Use-case discussion is IG BTW in general.
<jasnell> and only pushing to recommendation when there are implementations that demonstrate the value
Arnaud: use cases have been deligated to the IG?
<harry> bblfish, the use-cases are in the Social XG
<harry> please see minutes from previous meeting
<Shane> I think that claim was in reference to IBM?
harry: it might be in a different group but we should be able to know them
<tantek> I agree with bblfish's line of questioning - where is the URL to use-cases for Actions?
<bblfish> Tsyesika, that is bblfish speaking.
<Loqi> harry meant to say: Tsyesika, that is bblfish speaking.
<Shane> If the IG can give us use-cases then we can make examples that match the spec to the use-cases, which I think would be enough to go to FPWD
<cwebber2> I need better understanding of how actions come into place, but I think we might be able to make some strong use cases of how we'd be interested in implementing in mediagoblin, *if* I understand right... would someone be interested in discussing how actions are intended to be used with Tsyesika and I, and we might say how we may or may not be able to make use of it?
<bret> I would like to see more working action examples, not necessary up to date with AS2 spec
<bret> (but preferably something close)
<bret> not just written examples
tantek: I'm just calling into question how we as a group have so much interest this without any use cases
<harry> We have a dedicated phone call for use-cases in IG I expect people should attend who are intersted in use-cases.
<elf-pavlik> cwebber2, you can find MovieReview action example in http://schema.org/docs/actions.html
<cwebber2> elf-pavlik: great, will review
<bret> elf-pavlik: is there a list of documented uses in the wild that I can review?
jasnell: I plan to discuss more of our motivations of our actions at TPAC, the embedded experiances is discussed in our charter, we view actions as an evolution of that and we can't ignore that history. We should look back at the embedded experiances for the use cases and I will discuss this more at TPAC
<tantek> Arnaud, you said you expect to see use-cases before we work on something, yet you did not ask for a URL to those. That is what is confusing me.
<Loqi> Arnaud meant to say: ack bblfish
<elf-pavlik> bret, not code that i know of but presenation: https://www.youtube.com/watch?v=KB94dIamAQc
<tantek> bblfish: "The question has to be, what are the use-cases?"
<harry> Yes, and there's a phone call for use-cases bblfish, please attend.
<Shane> I agree that we don't need to develop them, but we should have some to discuss
<bret> elf-pavlik: ty will check it out after the call
bblfish: We don't need to have implementations for all drafts, if there is no implementations, how do we go about how to solve the problem we want to solve. We have to have the use cases to be able to discuss if the poposal is good
<harry> However, use-cases are often imaginary and running code *with users* trumps use-cases.
<Loqi> harry has 1 karma
<tantek> deferring work does not mean we remove the requirement
<harry> I don't think Tantek's objection is over use-cases, its over running code.
<jasnell> Based on a much older draft.. but.. https://www.youtube.com/watch?v=th_7l1rJ1fA
<harry> These are very different things.
tantek: i believe there is push back on use cases here is because there are so many topics with use cases that we need to address and if people want to work on topics which don't have documented use cases you can go do that in the IG
... i find it suprising that this group wants to work on something without use cases, i think if even was here he'd push back too as he doesn't want to work on things without use cases
<harry> tantek, if you want to do use-cases, there's an IG phone call that needs folks to pick up that work.
<tantek> no documented use-cases = we shouldn't work on it in this WG
Arnaud: it's not resnable to participate without use cases
... i imagine people are interested in working on it is there are undocumented use cases which this addresses, i think we should make sure use cases are documented
tantek: it's resnable for us as chairs to reject the idea if they haven't documented the use cases. If instead we lower the bar and just go by popular vote, there is no incentive to document the use cases
<cwebber2> tantek seems right there
<Shane> Yep I completely agree
<wilkie> but who will do it?
Arnaud: i think tantek's point is valid, we should get a at least a minimal list of use cases documented
Lloyd_Fassett: I'm with the IG, an update: we meet every other week, we're meeting tomorrow. We are on track to deliver use cases by TPAC.
... it's coming, it's just not here today
Arnaud: hopefully the situation will improve as the IG deliveres on this task, then we can look at if this poposal fits the bill or not
... for the last 5 minutes, is there any issues that jasnell wants to discuss with the group
jasnell: no, not with the time we have left, nothing that would be a blocker. Recomend taking another good look at the spec, there is a test version of the JSON-LD context (link on mailing list)
... file any issues
<jasnell> some background on actions
Arnaud: unless anyone wants to bring up anything last minute we can adjourn
<jasnell> including some use caes
Arnaud: thanks for joining
<elf-pavlik> thanks Tsyesika for scribing and Arnaud for chairing! :)
<bblfish> Who said they had done research on LDP and what is going on here?
<cwebber2> Tsyesika: it's pretty easy, it's pandoc + a small amount of manual editing :)
<bret> we made it
<wilkie> at least skype had the decency to drop my call at the very end :)
<Loqi> Tsyesika has 1 karma
<tantek> Thank you Tsyesika for scribing!
<cwebber2> thanks Tsyesika
<jasnell> Here's another Actions impl that I'm aware of: http://www.slideshare.net/cappelaere/open-geosocial-api-and-screencast ... work in progress, building on AS2+Actions
<Loqi> Tsyesika has 2 karma
do i need to tell Zakim anything? to close it
<Shane> Thank you :) Goodbye all!
<elf-pavlik> cwebber2, thanks for passing on skills you just gained a week ago :)
<Loqi> cwebber2 has 1 karma
<cwebber2> elf-pavlik: haha :)
<Arnaud> trackbot, end meeting