Socialwg/2016-05-03-minutes
Social Web Working Group Teleconference
03 May 2016
See also: IRC log
Attendees
- Present
- aaronpk, ben_thatmustbeme, shevski, Arnaud, tantek, rhiaro, sandro, cwebber
- Regrets
Chair - Arnaud
- Scribe
- rhiaro
Contents
Summary of Resolutions
- Approval of Minutes of 2016-04-26 https://www.w3.org/wiki/Socialwg/2016-04-26-minutes
- Pending confirmation of availability from host by Aaron, change F2F dates from 2016-06-07..08 to 2016-06-06..07
- Publish latest editor's draft of Micropub: http://micropub.net/draft/
I will scribe today!
It's been a while
<Arnaud> thanks amy!
<tantek> yay! welcome back rhiaro :)
<cwebber2> present!
<cwebber2> oops
<aaronpk> hehe, present! is more exciting
<aaronpk> maybe we can add that syntax to the bots
<scribe> scribenick: rhiaro
<eprodrom> I'm on the call!
Approval of minutes
<Arnaud> Approval of Minutes of 2016-04-26 https://www.w3.org/wiki/Socialwg/2016-04-26-minutes
Arnaud: any objections?
<Arnaud> PROPOSED: Approval of Minutes of 2016-04-26 https://www.w3.org/wiki/Socialwg/2016-04-26-minutes
<cwebber2> +1
<aaronpk> +1
<ben_thatmustbeme> can probably remove the perl script output at the end
<eprodrom> +1
<cwebber2> there's the scribe.perls tuff but
<cwebber2> yes
Arnaud: Hearing none, approved
RESOLUTION: Approval of Minutes of 2016-04-26 https://www.w3.org/wiki/Socialwg/2016-04-26-minutes
F2F dates
<eprodrom> Here!
<eprodrom> Sorry, had a hard time unmuting
Sandro: brewster khale from archive.org posted about decentralising the web a few months ago. i learned yesterday that they are holding a summit in SF which overlaps with the f2f
... I'd really like to go
... I'm wondering if we can move a day earlier
... I don't know if facilities are available - aaronpk?
... And who has already bought a ticket? People going to iwc would be there anyway
... Evan seems the most likely person to have already bought a ticket to arrive justi n time for the meeting. But thought I'd ask.
eprodrom: I do not have a problem with this, so it should be fine from mys ide
Arnaud: Anyone else?
aaronpk: I would have to check with the host since I only asked about the two days we requested, if the space is available for the other day
... If there are no other objections I'm happy to ask, but I don't want to ask if it's blocked by some other reason
Sandro: cwebber2 does it work for you?
<cwebber2> I talkeda bout it with rhiaro
<cwebber2> it works
<shevski> sandro what's the website url?
Sandro: Not sure who else is coming in from out of town
<tantek> meeting URL?
<aaronpk> http://www.decentralizedweb.net/
<cwebber2> sandro, ^^^
<shevski> thanks
http://brewster.kahle.org/2015/08/11/locking-the-web-open-a-call-for-a-distributed-web-2/
<tantek> I mean our meeting URL?
sandro: looking at the rsvp list, I think that's okay for everyone else? Hopefully nobody is being quite who has a problem with this
<aaronpk> https://www.w3.org/wiki/Socialwg/2016-06-07
eprodrom: since it's getting farily close, we should find out within a day or two
sandro: aaronpk can go ahead and check with the hosts
... I'm trying to reach the organisers of the summit to see if they are flexible on the dates
<Arnaud> PROPOSED: Pending confirmation of availability from host by Aaron, change F2F dates from 2016-06-07..08 to 2016-06-06..07
sandro: So between those two hopefully we'll get an answer
+1
<sandro> +1
tantek: Is anyone else interested in the other conference?
<cwebber2> semi-interested, but not confident I can go
<KevinMarks> I am interested in it
<rhiaro> Also interested in the other conference
<Loqi> Rhiaro made 1 edit to Socialwg/2016-05-03 https://www.w3.org/wiki/index.php?diff=98163&oldid=98159
<Loqi> Aaronpk made 1 edit to Socialwg https://www.w3.org/wiki/index.php?diff=98164&oldid=98154
<cwebber2> like, I'd like to go, not sure I can afford to add another hop
tantek: if there's a lot of interest that brings additional weight to tryign to resolve the conflict
Arnaud: but if nobody opposes it it doesn't really matter
... If the host can host us a day earlier then we're good
<sandro> (in my flight, it looks like another $75 to add SFO to a Boston - Portland trip)
Arnaud: Vote?
<tantek> 0
<ben_thatmustbeme> 0
<tsyesika> 0
<cwebber2> +0
<eprodrom> +1
PROPOSED: Pending confirmation of availability from host by Aaron, change F2F dates from 2016-06-07..08 to 2016-06-06..07
<aaronpk> -0
annbass: I can do any date because I'm driving from Seattle
<annbass> +1
<KevinMarks> 0
Arnaud: +1 means you support the proposal
... Nobody is objecting to it
aaronpk: I would rather do tues/weds but I'm not going to block it
tantek: I have a slight bias for keeping the current one but not enough to -0 it
RESOLUTION: Pending confirmation of availability from host by Aaron, change F2F dates from 2016-06-07..08 to 2016-06-06..07
Arnaud: I'm going to call it resolved. aaron, if you could check with the host and as soon as you know send an email to the list
... And based on that, we will change or not
... The sooner the better
sandro: thank you aaron and tantek and everyone for being flexible
Webmention
Arnaud: What's the status? What do we need to do to make progress?
aaronpk: Two weeks ago we had 0 open issues, including IANA published their updated list of link relations
... Last week we voted to accept the exit criteria and publish a new draft, as well as the big thing from last week's call was developing the implementation report template and method
<aaronpk> https://www.w3.org/TR/webmention/#test-suite-and-reporting
<aaronpk> http://webmention.net/implementation-reports/
aaronpk: So there is now a new section in the spec describing how to submit implementation reports
... A template in markdown on github
... And I filled out a sample for one of my implementations so you can take a look at that and fork the repo, copy the file, and send a PR to submit a report
... The status of the test suite itself, webmention.rocks now links to the implementationr eport with instructions
... On my todo list is still to create the tests for receiving in webmention.rocks, but the tests themselves are described in the implementationr eport
... So it is possible to submit a complete report including verifying the behaviour of the receiver even though there is not a tool in place to assist with that
... In the mean time there is a tool to help testing receivers that walks you through the items in the report checklist
... That's my plan for finishing the test suite is to follow the implementation report's checklist and create tests for each of those
Arnaud: Any questions or comments?
... I noticed that for the implementation you put a link to a section in the spec, you have a section in the spec for that and I don't think that's the way to do this because you want more of a live document for that
... So I think it has to be a link to a page which is outside of the spec so that you can update that freely as you get more information
... It could be a wiki page or some page on github
... It doesn't matter where, but I think it should be outside of the spec so that once the spec is in CR you can keep updating that as much as you want without having to republish
aaronpk: that makes sense
... Happy to take that out and move the content to possiblyl the github page with the implementation report template, or the wiki, we'll figure out what makes mroe sense
Arnaud: what you could do is put a link in the spec to the page
... So people can find where the list of implementations is maintained
tantek: github sugggestion makes the most sense since developers will show up to write an implementation report and see the existing ones, and if you're showing up to look at the implementationr eport it's easier to discover how to make one of your own
... But obviously up to you
aaronpk: I think that makes sense, keep it in the same workflow as the implementation reports
Arnaud: anyone else?
eprodrom: Couple of questions, mostly for the group
... Have we had a vote to decide to take webmention to CR?
... Should we schedule one?
... Next week?
... Do we think it will be ready for next week?
aaronpk: Not yet had a vote, based on progress from last week to this week I don't have anything else on my todo list before we can vote to take it to CR
... we could vote today
eprodrom: It's a pretty big discussion
... Are you sure you want to do that today or next week?
aaronpk: My question is what new information would we gain between today and next week?
eprodrom: My guess is that it would probably suck up a large chunk of the rest of our meeting today
Arnaud: why?
eprodrom: When we first started working on this and the other parts of our suite, activitypub, AS2, micropub, we started progressing towards recommendation status with the idea that we would have this evolving process happen, some of them would end up as notes, some of them would never make it to cr, some of them would become cr
... this has been the idea of moving forward with multiple documents. Is that incorrect?
aaronpk: that sounds right
eprodrom: so taking this particular one to CR I think is.. especially being our first CR.. is an important discussiont o have
... What our strategy is for what we produce for the rest of our period together
... I see that as an important part of what we do as a group
... Deciding which documents we produce and whether they're CRs, notes, or not published at all
... Maybe I'm overestimating the importance of that
Arnaud: Evan can you be more specific? aaron is asking if he can use webmention to CR, are you opposed and that's why you're bringing it up?
eprodrom: I'm opposed, Id on't think publishing a bunch of unrrelated documents makes sense
Arnaud: so you think this doens't fit with everything else we want to publish?
eprodrom: I simply am saying that we have a charter to produce certain kinds of documents, this is not one of the three specifications, it doesn't fit the requirements of any of the three, so I'd like us to have a strategy that says this is why we published it, this is the relationship to our charter, and this is what we're doing for the rest of our products
... As a group we're producing a suite of specifications and we need to be aware of that suite that we are producing
... We have to be aware of what we're doing, we have to have a strategy
Arnaud: I understand, thank you
<Zakim> tantek, you wanted to note urgency we discussed at last f2f
tantek: I think the framing that evan noted that when we first started working on all the different parts is accurate
... We deliberately work some documents towards CR some towards Notes
... What has changed are two things
... We started to explicitly note when we accepted a documented as an ED whether we thought it was rec track or not
... This came out of a discussion in december at the f2f in SF where there was a bunch of different documents we were taling about accepting for ED, and some it was clear they should be rec-track, some there wasn't that consensus
... We said it's okay if we produce things that are note-track
... One specific example is jf2
... Then the next change that we made since that discussion in terms of how we move forward was in March we talked about the urgency of how little time we have left
... And how do we make progress on these documents as quickly as possible so that we have a chance of getting anything to rec at all
... One of the decisions we made there was that we were going to be okay with moving multiple docuemtns with multiple approaches to rec
... and we would document guidence for developers in terms of when might you use one vs the other
... I feel like we have had these discusisons, would be opposed to reopening them
... that's the state I recall from the past f2fs
Arnaud: This is a bigger question of if webmention is on the rec-track, does it qualify to move to CR? The bigger question evan is raising is should webmention be on the rec track?
sandro: I guess my question .. or proposal... I understand that it's not obvious to people how webmention meets our charter requirements
<KevinMarks> webmention is explictly mentioned in the charter on recommendation tarck deliverables
sandro: I think it's clear to me and to the people who are interested in webmentino, but I hear evan saying we should be clear in in public somewhere about what that is
<KevinMarks> "Federation should include multiple servers sharing updates within a client-server architecture, and allow decentralized social systems to be built. One possible input to this task is WebMention and another possible input is the Linked Data Platform."
<KevinMarks> https://www.w3.org/2013/socialweb/social-wg-charter
sandro: We can do it in an official document on the w3 site, on the groups homepage, some status document
... Would , assuming that were well written and clear, would that satisfy what you'r elooking for evan?
<tantek> Specifically: https://www.w3.org/2013/socialweb/social-wg-charter.html#deliverables
eprodrom: Sandro, what you're saying is that we sould do a simple blog post about webmention, we would say that it fits in the federation protocol slot, and would we say anything about the other parts of our charter or what else we're going to produce?
sandro: I would like the post to say we're going to have consensus on one way to make thish all work so we're taking a multi faceted approach that each don't appraoch 100% of what people might want out of a federation protocol. Wm is one of those pieces here, we're not claiming it solves everything, but is sitll useful in its own right
<tantek> I think there's a positive way of framing that too, that this is one building block of a modular approach, and that we're still figuring out the different pieces
annbass: I was gonna ask or suggest that rather than a blog post, for me it would be valuable if there could be I don't know what form, some sort of document ongoing summary of how the group envisions these different technologies to fit together
... when you would make a choice to use one vs the other
... some larger guide
<tantek> And that based on implementation experience there is evidence that this building block does further some of our goals (but obviously not all)
annbass: and do we anticipate the group to go and develop other componants? or is this the end? My fear is only oen blog post would get lost
sandro: could also be homepage news item
<KevinMarks> isn't what ann asks for http://w3c-social.github.io/social-web-protocols/
annbass: It seems like some form of summary guide
... would be useful
<KevinMarks> just lost audio
sandro: we do have SWP that might serve that goal
... Between SWP and a blog post and homepage news, could that in theory address evan's concern? Or is there some other piece that's an obstacle
<KevinMarks> oh, my call hung up
<KevinMarks> audio back
eprodrom: I think what you're saying is we would publish SWP as a Note? Explanation of how this group of specifications is supposed to do together. And then we would plan to publish wm to CR because it's ready, we take AS2 to CR soon because it's ready.
... Would we take our other two rec track specs, micropub and activitypub, to CR before the end of the year? Is that the goal?
sandro: my understanidng is that is what peolpe want to do at the last f2f
... Question of whether there's enough energy, I hope there is
... We should know at this face to face
tantek: my understanding from past f2f is that we're trying very hard to coordinate but decouple
... That we wanted to do as much as we could to enable our documents to proceed to CR as quickly as possible because of the time constraints involved
... With AS2 we've tried to remove obstacles to try to get it to CR as quickly as possible
... If wm is the next one that seems to be ready it's good that we're discussing that
... There are criteria that we have figured out for when it's good to take things to CR, exit criteria, feature descriptions, test suite, those are good measures for when something is ready for CR
... But as far as the if quesiton, I feel like that's something we resovled before and that if there's new information about why we should or should not atke something to CR we should talk about it, but I haven't heard that
... we've already published SWP, and amy has an iteration, so that's already agreed, there's no extra process for that
cwebber: just briefly, I support any document going to CR when it's actually ready, but in the event that somehow webmention ended up going to CR before AS2 after all this work and being so close to CR I would really wonder what happened that we ended up that way
<cwebber2> end of comment :)
eprodrom: I'm happy to answer that question
... We got very close to CR, we had a big change in personnell, it's taken me longer than I thought
... I'm not happy about that situation but that's where we are
... So it is an unfortunate part of things
... If webmention goes to CR before AS2 I will be unhappy about that but it's great that it's worked out so quickly
<cwebber2> I don't think it's a race either :)
Arnaud: I don't want us to mix all these issues, which is first is not a big deal
<cwebber2> if it's ready, it should be
<KevinMarks> rhiaro++
<Loqi> rhiaro has 202 karma
Arnaud: What is the end game?
... I'm hearing evan is against moving webmention on rec track
... I heard before we can figure this out later, so those specs moved to ED, which set an expectation that these things wer eon the rec track, and now it's fair to say aaron did the work that is required to qualify for CR, and now we're saying do we really want to do this
<ben_thatmustbeme> as to your question cwebber, it think its just a LOT harder to get a serialization right vs a really small spec for processing
Arnaud: On one hand evan is saying no, and on the other people are saying lets go to CR and then we'll see
... Seems like there's an issue that wasn't really resovled
<cwebber2> ben_thatmustbeme, that's probably true
annbass: I was hearing evan speak to the point of having a global understanding of how this was all going to roll out
... Rather than that you were blocking webmention per se
... Could you clarify?
eprodrom: Ann thanks, yeah I am not interested in blocking webmention, I think it's a solid specification, excellent work, strong implementations
... I feel like we've been asked to provide in our charter certain deliverables, from my point of view if I'd been asked to provide deliverables I'd want to either provide them or have an explanation about why I'm not
... So I think that when we produce documents in a way that's going to be the first very visible thing we do to the rest of the world after a year and a half of work, it would be nice for us to have a story about what we're doing and how we're doing it
... how we got here, and what we're doing next
... what the relationship of the documents is to our charter
<cwebber2> (btw, if we voted on going to CR, I think the "put a blogpost on why we're including webmention and go to CR", I'd vote for that... I do think it's at a good stage and we've positioned the group to make space for it)
<KevinMarks> so is evan saying that SWP update gates other docs?
<tantek> cwebber2: I agree, that kind of context would be very helpful.
eprodrom: I like sandro's formuatlion that we have a document SWP that documents these parts
... And then trying to get them over the finish line
... And if we have a way to formalise that I'd really like to
... I'm ready to go forward with webmention, I want an understanding of a strategy as a group
Arnaud: it sounds like that's a good quesiton to ponder over for next week, and we can resume the discussion next week, possibly with a proposal to move wm to CR on the agenda
Micropub
aaronpk: I just sent an email to the group about the current situation
... It's also on the github thread
<annbass> I appreciate Evan's desire and "push" that we have a coordinated understanding and presentation of how these components relate and will roll out
aaronpk: Since last week there has been another new implementation in python
... of a server that supports updates
... I would like to publish a new working draft of micropub, it's been a while
... tha'ts what the implementations are following today
<tantek> URL to editor's draft?
<aaronpk> http://micropub.net/draft/#changes-from-01-march-2016-fpwd-to-this-version
aaronpk: Iw ould appreciate being able to take what we have on the current editor's draft and publish that update
<Arnaud> PROPOSED: Publish latest editor's draft of Micropub: http://micropub.net/draft/
<eprodrom> +1
Arnaud: not a change of status, just refresh of the WD
<ben_thatmustbeme> +1
<KevinMarks> +1
<aaronpk> +1
<tantek> +1
<cwebber2> +1
<annbass> +1
<tsyesika> +1
<shevski> +1
RESOLUTION: Publish latest editor's draft of Micropub: http://micropub.net/draft/
<sandro> +1
AS2
Arnaud: I see editorial issues and a couple of questions and two proposals
eprodrom: since last week I cleared out six of the issues that were open
... The ones that we still have open: one is a lot of busywork
... Required property on all of our examples, I need to add it to all of them
... About half way through with that
... We have two proposals: they had come up and we had discussed them at our face to face and we had come up with resolutiosn at the face to face on these proposals
... One was to add a new type
... One was to ??? property ?
... jasnell gave a strong -1 on both
... I think in the interest of respecting authorship, we've been discussing his counterproposals
... For one we've got a resolved version, and I"m going to be pushing that later this afternoon
... The other is still subject to discussion, I'm going to see if I can try to get it closed up
... Part of getting it closed up is probaby implementing the proposal
... James' objection on the second one was that it wasn't necessary
Arnaud: this was dicussed like 3 meetings ago, issue 292
... The consensus was we could close it, chris was invovled and then agreed
<Arnaud> https://www.w3.org/wiki/Socialwg/2016-04-05-minutes
Arnaud: You can look up the minutes
... I don't know if there's anything more to say about that
eprodrom: sounds reasonable
<tantek> issue URL?
Decided to maintain the Relationship class and define a minimal vocabulary for relatinoships
scribe: The objection from dmitri was that we did not have a vocuabluary for this relationship class
... So we didn't have a vocabulary for types of relationships
... There was some discussion about adotoping one or another of the others, some concerns that there was potential for going down ratholes with that
<aaronpk> https://github.com/jasnell/w3c-socialwg-activitystreams/issues/289
scribe: In particular social relationships have a real culturual bias, what's defined as a friend, varies from platform to platform and culture to culture
... What does it mean to be a family member, friend, clsoe friend, etc
... So it looks like the consensus that we have is that we define topographical relationships
... So in the social graph
<tantek> aaronpk: huh somehow my comment is last on that issue?
scribe: So A is following B, B is following A, A and B are following each other. A is a member of B.
... Looks like we have consensus around these
... They're in the ED
<KevinMarks> topographical? mountains?
scribe: Not yet pushed, will push after meeting
... That would close this
<KevinMarks> or topological?
<aaronpk> did i drop the wrong link?
Arnaud: the rest is editorial?
eprodrom: one is editorial, one about questions and polls where there was a request for another, I"ll see if I can close that for next week too
... Would be very happy to get these finished and get to a new version of the WD
Arnaud: very close
eprodrom: I'd like to
<tantek> can we publish a new WD? or next week publish a new WD?
Interest Group
Arnaud: The IG has been inactive for quite a while, and before that it was somewhat dormant despite ann's efforts, so ann is asking what should we do with the IG?
... 1. Keep it open and try to revive it?
... 2. Close it and open a CG
... 3. Close it entirely
... If we close it entirely what happens to Use Cases?
annbass: i have mixed feelings because I don't watn the work we've done to be lost, but on the other hand as a chair I dont' want to pull teeth to try to carry the group along
... When I came back I sent an email to the list and asked the group what they want to do, and I was proposing moving it to a CG because that's a bit less process and peopel who are not members of w3c can participate, and it seems to me for this subject there are a lot of peopel who are interested but who are not w3c members and as we know it's hard to be an IE and even harder to get your company or individual to be a member
<tantek> sandro, didn't you say the existing https://www.w3.org/community/fedsocweb/ Federated Social Web Community Group asked for a time slot at the upcoming TPAC in Lisbon?
annbass: Benefit of CG is it's visible to the outside world and people can participate
... But when I sent a note to the list, I only got one response from Lloyd who was responsive
... I'm happy to continue with the work but I want other people to want that work to happen
... I don't want to do it on my own
... I had discussed this in mail with sandro and wendy, and nor does the w3c want to have a dead horse
... We can carry it to next week if necessary, not a crisis
eprodrom: fine taking this to next week also
... One option is to appreciate the work that the IG has done up to this point, it has low participation now, and may have served its purpose so far
... Maybe we can ask the w3c to provide us with a recommendation
... If the organisation or members want us to continue, make that known, otherwise close it
<tantek> and +1 to eprodrom
Arnaud: I tend to agree
... The right thing is to close it
... THe use cases are not going away, nothing will be deleted
annbass: I'm agree
<tantek> (then are we done? no disagreement?)
annbass: I do like the idea of making the use cases visible to the outside world
... I can see hwo opaque the w3c work looks to the outside world
... Some value in making... you guys had the federated social web cg?
... that was slightly different work but possibly this stuff could reside there
... Totally open to closing the group, but don't want to lose what's been done
<Zakim> tantek, you wanted to ask Sandro about FSWCG and TPAC
tantek: There's the federated social web CG before the SWWG, has low traffic and interest and hasnt' had a blog post in three years or something, don't know if there's a need for another one especially when there's an existing one with no critical mass
Sandro: I was looking at the tpac registration list of meetings the FSWCG is listed as meeting for 3 hours on the firs tmorning of tpac, organised by someone I don't know and have never heard of
... I'll get the name
Arnaud: I'm going to close the call, we're out of time
... We don't have to close this now
<annbass> to clarify, I, too, asked about using the Fed SW CG, rather than start a new one
Arnaud: We probably should just close the IG
... Seems to be CG activity gonig on worth investigating
... Will leave it here
... You can ponder
<sandro> https://www.w3.org/2016/09/TPAC/schedule.html
<sandro> 08:30-10:30 Federated Social Web
<annbass> sounds good; thanks
Arnaud: Thanks all for joining
<tantek> rhiaro++ for scribing!
<Loqi> rhiaro has 203 karma
<shevski> thank you Arnaud & rhiaro !
<eprodrom> thanks rhiaro
<Arnaud> trackbot, end meeting