Socialwg/2016-08-23-minutes

From W3C Wiki

Social Web Working Group Teleconference

23 Aug 2016

See also: IRC log

Attendees

Present
ben_thatmustbeme, wilkie, eprodrom, cwebber, rhiaro, csarven, aaronpk, sandro
Regrets
Chair
eprodrom
Scribe
wilkie

Contents





<aaronpk> no, iOS, i don't want to update my phone right before i call in to this meeting

<eprodrom> Can anyone scribe?

I can scribe

<eprodrom> scribenick: wilkie

<eprodrom> akim, who's here?

<cwebber2> badoop

<eprodrom> Can anyone hear my voice?

not yet, no

<rhiaro> nope

<cwebber2> nope

<ben_thatmustbeme> nope

<sandro> Nope, eprodrom

<eprodrom> OK, that explains a lot

<cwebber2> :)

<sandro> we can hear each other but not you

<eprodrom> Give me 30 seconds to call back in

<cwebber2> I was thinking it was a pretty quiet start!

<cwebber2> yes

eprodrom: can you hear me now?

<cwebber2> :D

eprodrom: sorry about that. thanks for your patience. probably a good time to get started
... I will be chairing today because Arnaud is unavailable
... let's start out with approving the minutes for the previous meeting

Approval of Minutes

<eprodrom> PROPOSED Accept https://www.w3.org/wiki/Socialwg/2016-08-02-minutes as minutes for 20016-08-02 meeting

<cwebber2> +1

<rhiaro> +1

<wilkie> +1

<eprodrom> +1

<aaronpk> +1

you might need a colon? how does "proposed" work

RESOLUTION: Accept https://www.w3.org/wiki/Socialwg/2016-08-02-minutes as minutes for 20016-08-02 meeting

<ben_thatmustbeme> +1

eprodrom: barring anything else we have resolved on that
... moving on to our next topic

Publish a New Working Draft for ActivityPub

<eprodrom> https://www.w3.org/wiki/Socialwg/2016-09-22#Registration

<Loqi> Social Web WG Face to Face Meeting in Lisbon (F2F7)

eprodrom: probably an important thing but not listed is to cover registration for TPAC
... we are having a meeting at TPAC september 21st to 23rd
... if you haven't registered yet, please do. if you believe you will not be going also let us now. tentively or not.

<tantek> eprodrom++ for chairing

<Loqi> eprodrom has 41 karma (40 in this channel)

eprodrom: great. so. we don't have any other administrative issues to discuss.
... I'd like to move on to our discussion items.

<tantek> re: f2f is anyone else coming?

eprodrom: the first is an update on activity pub and LDN and have rhiaro start us off

rhiaro: last week I went to madison and worked on activity pub and made lots of progress.
... it might be better to have cwebber2 talk about those changes because we didn't do much to LDN but rather catch activitypub up to LDN

cwebber2: sounds good. first, rhiaro was an immense help.
... the first is to clarify client to client and client to server stuff. although some of this isn't in the document.
... first, there is a list of side effects of using various activities in activity streams.

<tantek> I'll add a specific item about f2f at the end of the agenda and try to get on the phone

cwebber2: and a separate section showing targetting and delivery and side effects of activities.
... these are cleaning separated which we've been talking about for some time
... one thing we discuss is having two separate documents and saying this makes sense from the payload and protocols
... but we have seen that the differences is a specialization of some of the LDN specifications
... clarifying and using much of the LDN stuff has made the sections clearer and more concise
... since we have linked data notifications referenced cleanly in activitypub and sections for client-client and client-server... we will keep it as one document
... rhiaro and I agree about this and tsyesika also seems to agree
... one concern is... "I want to publish this note to my followers" and somebody replies to you, how does this reply get out to your followers?
... pump.io has encountered this issue too. we did a demo implementation and drew things out on paper and we think we've come up with a method and put that in the spec

<eprodrom> Bravo

cwebber2: that if somebody does write a reply you can forward it under some conditions
... there are changes to make to the WD and I can make those and go toward a new working draft and feel confident about the state of the document

eprodrom: rhiaro, you were saying there aren't major changes to LDN at this point?

rhiaro: right. there are editorial changes and we should still publish a new LDN WD as well to make sure things are in sync.
... lately I've changed social web protocols to catch things up and so we need to publish a new version of that too

eprodrom: great. this might open a can of worms; we discussed last meeting... this is a social web protocols / ldn question, how do we handle having ldn and aligning it with others

rhiaro: right, that's next on my things to attack

eprodrom: ah, sorry, ok

rhiaro: I added it to the social web protocols and hopefully I can talk to julien about that

eprodrom: cwebber2, are we ready for a next version then?

<ben_thatmustbeme> changelog++

<Loqi> changelog has 2 karma (1 in this channel)

cwebber2: I think so I just have to add one thing

eprodrom: my next question is: should we go to working draft or wait until the group can review or just push the next version? I'm also fine with review on the WD.

rhiaro: I was thinking we would publish the new WD today and then have it reviewed since we have some time off

<eprodrom> PROPOSED: publish current editor's draft of ActivityPub plus changelog as new working draft

cwebber2: I agree with that because we want to get to CR either before or by TPAC and this would help us move along

<tantek> is there a link to the changelog?

eprodrom: before we start +1'ing it, cwebber2 and rhiaro, does this reflect what you want to do next?

cwebber2: yep

rhiaro: yep

<eprodrom> +1

<tantek> we really should provide a link to a changelog for anything we want to propose publishing

<rhiaro> +1

eprodrom: changelog right now?

<cwebber2> +1

<wilkie> +1

<sandro> +1

<csarven> +1

<ben_thatmustbeme> 0, don't really like publishing without a review or some changelog, but i will abstain in the interest of saving time

<tantek> 0, similarly

<cwebber2> I could write a fast changelog that'll be done before meeting end if necessary

<KevinMarks> traffic noises

<cwebber2> would that help?

<tantek> (cwebber2 yes please!)

<tantek> any document that is rec track should have a changelog with summaries

eprodrom: cwebber2, don't do the changelog during the call if you can help it, but some <laughs> have other opinions on that

RESOLUTION: publish current editor's draft of ActivityPub plus changelog as new working draft

eprodrom: I'm going to mark this as resolved and... great

<tantek> I believe AS2, Micropub, Webmention have all done a good job of providing changelogs for every published draft and that's been very helpful

<cwebber2> ACK

eprodrom: in all aspects, next time we have proposals coming up from editors moving to WD that you bring a changelog with you. either something you can link or drop into the channel.
... great

<tantek> (if anyone would like examples of existing changelogs in the group)

eprodrom: I'd like to move on to the next topic

<KevinMarks> cwebber2: reading scrollback your notification issue sounds like salmentions

Proposal: publish new WD of SWP

<rhiaro> https://github.com/w3c-social/social-web-protocols/commits/gh-pages

rhiaro: this morning I added PuSH and refactored a section. the link of the git log I dropped in the channel should give you an idea of the changes.

<KevinMarks> https://indieweb.org/salmentions

rhiaro: I'd like to propose those changes.

eprodrom: do we have questions on Social Web Protocols?

<tantek> (since SWP is not rec-track, having an in-draft changelog is not as important)

eprodrom: ok. great. I'd like to ask... rhiaro would you mind adding a changelog to this document?

<tantek> rhiaro: can you provide a link to the draft you have staged to publish?

rhiaro: no problem

<KevinMarks> I mentioned Social Web Protocols on TWiG this week :D

<ben_thatmustbeme> tantek, is SWP non-rec-track?

eprodrom: I'd like to propose publishing this editors draft as a new working draft

<eprodrom> PROPOSED: publish https://w3c-social.github.io/social-web-protocols + Changelog as new working draft for Social Web Protocols

<tantek> ben_thatmustbeme: yes that's my understanding since it non-normatively compares / relates *other* specs

eprodrom: does that look right, rhiaro?

rhiaro: yeah, looks right

eprodrom: great

<eprodrom> +1

<aaronpk> +1

<rhiaro> +1

<KevinMarks> +1

eprodrom: please vote

<cwebber2> +1

<wilkie> +1

<csarven> +1

<sandro> +1

eprodrom: I'll give some time to look it over

rhiaro: just to add this has a less important time table but it is important that people not see it too out of date

<cwebber2> oops, realized I wasn't muted

<cwebber2> sorry for typing noises

<ben_thatmustbeme> +0 again as before

eprodrom: I think it is important to keep this up to date and appreciate that you took the time to do that

<ben_thatmustbeme> but it is important to get SWP changes out

eprodrom: tantek?
... I don't want to rush you. I know you are reading it now.

<tantek> rhiaro typo: "ontent is deleted"

<rhiaro> will fix :)

<tantek> (start of 3.2)

eprodrom: I need all the votes in

<tantek> +1 great update rhiaro. Thank you.

RESOLUTION: publish https://w3c-social.github.io/social-web-protocols + Changelog as new working draft for Social Web Protocols

eprodrom: alright. great. I'm going to mark this as resolved. rhiaro, if you won't mind to fix that typo even if it isn't part of the resolution.
... next agenda item is to publish a new working draft of LDN

Publish New Working Draft of LDN

eprodrom: rhiaro, can you tell us all what's going on with LDN

<rhiaro> https://linkedresearch.org/ldn/WD-ldn-20160823/#change-log

rhiaro: nothing substantive except some clarifications
... it does have a changelog here but it is just editorial stuff but thought it doesn't hurt to keep things up-to-date

eprodrom: great

<tantek> changelog++

<Loqi> changelog has 3 karma (2 in this channel)

<aaronpk> "editorial"

<aaronpk> that's not very descriptive :)

eprodrom: the document doesn't have a changelog?

rhiaro: I linked to the document that has a changelog

eprodrom: ah, I see, great
... maybe in the future, especially since you did say there are explanatory differences, to explain what those differences are

<tantek> +1 on that

eprodrom: everybody please mute if you aren't me <laughs> thank you

<rhiaro> This is the version to publish https://linkedresearch.org/ldn/WD-ldn-20160823/#change-log

eprodrom: the one on linkedresearch.org is the editor's draft? I believe?

<rhiaro> ED ready to be WD

<eprodrom> PROPOSED: publish https://linkedresearch.org/ldn/WD-ldn-20160823/ + detailed changelog as new working draft of LDN

eprodrom: great. so I am going to propose publishing this as a new working draft of LDN

<rhiaro> Looks good

<tantek> this is a second WD right? we already did a FPWD?

eprodrom: rhiaro, does that make sense to you?

<rhiaro> yep

eprodrom: and yeah, this is the second WD I believe

<eprodrom> +1

<rhiaro> +1

<cwebber2> +1

<sandro> +1

<ben_thatmustbeme> +1

<aaronpk> +1

eprodrom: (upon seeing rhiaro's affirmative on IRC) ok. great. if this makes sense, let's vote.

<wilkie> +1

<tantek> +1

<csarven> +1

eprodrom: <laughs> everyone is more willing to make progress when we've had 2 weeks off.
... and yes, next week we have off

RESOLUTION: publish https://linkedresearch.org/ldn/WD-ldn-20160823/ + detailed changelog as new working draft of LDN

eprodrom: if you are to vote please do so. in this case I will mark it resolved.
... there is a new working draft. great.
... let's move on to our next item which is about AS2
... probably easier to have tantek chair while we go over the next item since it makes sense for me to address what is going on with AS2

<aaronpk> tantek is not on the phone

ActivityStreams 2.0

<tantek> I can't chair, I'm speaking in person at the AB meeting :/

eprodrom: ah. ok. I'll chair myself and try to push through this.
... so. the state of AS2, a brief recap, on AS2. We had our CR meeting.

<tantek> so that eprodrom can be free to discuss AS2 as editor

eprodrom: after our meeting we had some issues brought up by i18n group.
... we decided to pull out of CR to better address these concerns and we've been addressing them for the last few weeks.

<eprodrom> https://github.com/w3c/activitystreams/issues

eprodrom: as of today, jasnell_ did a few PRs to resolve these existing issues and so, depending on your measurement, we might be ready to go to CR right now
... however we have new issues that are currently on our list
... a couple of those are editorial
... those will probably not be a problem but there are at least 1 or 2 normative changes
... where I am, as an editor, wondering is if we need to get to 0 issues to reach CR, or is it better to go to CR with the document as we've agreed and resolve these in the next months

<cwebber2> vote going to CR

<cwebber2> *I vote

eprodrom: that's where we are at the moment. we are in an administrative point. as an editor, I want to move to CR and resolve as we go along
... ben?

<tantek> is sandro on the phone?

<tantek> we really need other people who were on the AS2 CR transition call to contribute to this discussion (apologies that I cannot)

ben_thatmustbeme: I was wondering if we go to CR and have normative changes, we have to redo CR. what does that entail

sandro: I am on the phone. the second version of CR involves us being willing to do it and its not a lot of work but it has about a 3 week turn-around so it is better to avoid it if you can

<tantek> in my experience any "real" / practical spec has at least 2 CRs

<rhiaro> He's done it, the i18n issues are resolved in the current ED

sandro: my understanding, though, is that we had a meeting (you weren't on this meeting) and we were waiting for these issues and jasnell_ said he would fix them and then said he would fix them today and we had resolved these

<tantek> that is, implementers of the first CR nearly always find substantial issue that require normative changes

<aaronpk> here is the thread https://github.com/w3c/activitystreams/issues/336

<tantek> and thus a second CR

sandro: what is the status of these edits? anyone know?

<aaronpk> and here is the new paragraph: http://w3c.github.io/activitystreams/core/index.html#biditext

<Loqi> [James M Snell] Activity Streams 2.0

eprodrom: yeah. jasnell_ has made a PR for changes to how we handle bi-directional text
... basically, for those who aren't following this, we came down to a fundamental difference between using bidirectional markers at the unicode level and using bidirectional markup in the html level
... there were strong resolution to not use the markup in certain fields and we can to a satisifactory compromise that doesn't require markup in the name field

<KevinMarks> what's the PR?

<eprodrom> http://w3c.github.io/activitystreams/core/#naturalLanguageValues

<Loqi> [James M Snell] Activity Streams 2.0

<tantek> could you link to the "satisfactory compromise" for ther ecord?

eprodrom: jasnell_ made those changes and they are in the editor's draft... let me make sure... yes. section 4.7... I can drop this

<tantek> link to the *discussion* that resulted in the "satisfactory compromise"

eprodrom: ah, yeah, satisfactory compromise... let me see if I can link to that

<eprodrom> https://github.com/w3c/activitystreams/pull/352

eprodrom: here is the pull request
... that jasnell_ made

<aaronpk> 336 has the discussion https://github.com/w3c/activitystreams/issues/336

eprodrom: let me see if I can summarize
... this is about using natural language and bidirectional text
... this is a significant change

<sandro> +1 that all sounds right

<eprodrom> https://github.com/w3c/activitystreams/issues

sandro: sounds like we solved the bidirectional issues

eprodrom: yes

<KevinMarks> that BIDI makes sense given the tests we conducted.

<tantek> is this the compromise? https://www.w3.org/International/wiki/Activity_Streams_direction_notes#Some_tentative_conclusions

eprodrom: we have 6 issues. one is easy to fix. one is normative and a change from a MAY to a SHOULD.

<tantek> issue 336 is too long to be considered a "Summary"

<tantek> shh trackbot

eprodrom: either than that... they seem to be resolved

sandro: sounds like there is nothing here that would need a second CR if we changed it? anything that would change implementations?

<tantek> my understanding is that NO properties are added to AS2 as a result of this discussion right?

<aaronpk> correct

<tantek> eprodrom: could you confirm ^^^ ?

<KevinMarks> right

<tantek> ok then I am +1

eprodrom: it doesn't seem like that a publisher or consumer would be significantly different between this MAY or SHOULD
... it is just a shift of emphasis, I'd think

sandro: we don't know what we want to do with this one at this point?

<ben_thatmustbeme> how long would it take to figure out the may or should?

eprodrom: I can't make that call right now
... I'd make it a SHOULD but jasnell_ probably has different ideas about it
... rhiaro?

rhiaro: realistically we would have a second CR on this

sandro: why?

rhiaro: seems like we will get feedback from implementers, unless you don't think so. seems like it would probably happen

sandro: I just want to avoid extra work

rhiaro: ok

<tantek> are we willing to write tests for this detail?

sandro: I feel like MAY vs SHOULD won't hurt but the safest thing we could do is make it at-risk

<tantek> or get i18n to write tests?

eprodrom: so put it as SHOULD and say "at-risk"

<tantek> that's one way to address may/should

sandro: or leave it as MAY and say "at-risk" and it may change to a SHOULD
... whatever way we think it is likely to go and just put it 'at-risk'

eprodrom: I like marking it 'at-risk' and publishing as-is

sandro: sounds good

<tantek> can we actually make it at-risk when i18n says it's a requirement?

sandro: and the other 5 issues don't seem like normative changes?

<eprodrom> https://github.com/w3c/activitystreams/issues/351

<rhiaro> the https one?

<tantek> I'm worried that we drop it, then i18n objects when we try to go CR->PR

<rhiaro> isn't w3c ns taking care of that?

<rhiaro> it officially switched to https on 1 Aug I believe

<rhiaro> ..sandro?

eprodrom: one is normative but... it is whether we support the HTTP content or what makes most sense
... rhiaro, can you say this outloud?

sandro: I don't think I've thought about this for context uris, but for namespace URIs this doesn't change

rhiaro: right, but this changes javascript implementations... this is an implementation problem

<KevinMarks> which may is becoming a should?

sandro: it does seem possible that browsers will be a pain in this way

<eprodrom> https://github.com/w3c/activitystreams/issues/354

eprodrom: my feeling is HTTPS makes most sense. our implementor base is small and this is the time to do it
... the only down-side is that there are some older libraries that don't support HTTPS but I think they are a smaller and smaller number
... and it makes sense to push it to HTTPS everywhere

sandro: let's make this at-risk too
... there are complexities here that I didn't think were there at first

<csarven> I hope there is alignment with https://github.com/w3c/web-annotation/issues/347

eprodrom: let's make it HTTPS in the editor's draft and we'll mark it 'at-risk' that we might also support just HTTP

<csarven> ^ eprodrom

sandro: yeah. the main reason to not put HTTPS there is...

<csarven> The contents of @context stays as HTTP

<csarven> It is the reference to context in HTTP is a problem if on HTTPS - client side / browser issue

<csarven> Hope for the best and have implementations fix their stuff??

sandro: the URIs might need some place to mark that it is HTTPS or not otherwise it will pull it from the base url
... I never thought about the mixed-content warnings so I don't know what the right thing to do is

<csarven> ok with eprodrom 's suggestion with AT RISK

sandro: let's say 'at-risk' and figure it out.

<rhiaro> And maybe having it called out in the CR doc will help with getting feedback

<KevinMarks> http://known.kevinmarks.com/2016/-inline-rtl-works-in-reverse-without-implementers-knowing

<Loqi> [Kevin Marks] ‮ inline RTL works in reverse without implementers knowing

<Loqi> #indieweb

eprodrom: we're ok with at-risk and then figure it out?

sandro: yep

eprodrom: ok. great.
... sounds like I have two 'at-risk' notes to put in. and besides that looks like we are ready to go and I can have a version ready to have out thursday?

<ben_thatmustbeme> just reference with "//www.w3.org/..." and make it decide on its own :P

eprodrom: does that work time-wise? rhiaro?

rhiaro: yeah
... if they say no for thursday we can do it next tuesday

eprodrom: ok!

sandro: let's get a new resolution on record

<eprodrom> ben_thatmustbeme: I'll keep that in mind

eprodrom: if anyone has a question about AS2 and what we are doing right now, this is a great time to ask

<sandro> PROPOSED: Proceed to CR with both ActivityStreams documents, including the changes worked out with i18n and items marked At Risk in this meeting, as per Ralph's go-ahead from the transition meeting.

<cwebber2> +1

<eprodrom> +1

<rhiaro> +1

<sandro> +1

<csarven> +1

<ben_thatmustbeme> +1

<wilkie> +1

<KevinMarks> +1

RESOLUTION: Proceed to CR with both ActivityStreams documents, including the changes worked out with i18n and items marked At Risk in this meeting, as per Ralph's go-ahead from the transition meeting.

eprodrom: unless we have any objections, now is the time to do so
... now we have something to point to in terms of resolutions.

<rhiaro> cwebber2++ for speedy changelog!

<Loqi> cwebber2 has 70 karma

<cwebber2> thanks rhiaro :)

eprodrom: we are now at the top of the hour

<cwebber2> since I did a /me earlier, here's the changelog for log records https://w3c-social.github.io/activitypub/#changes-from-13-june-2016-to-present-version

eprodrom: aaronpk, are you good with pushing micropub for two weeks from now? or should we extend?

aaronpk: I'm ok with waiting two weeks and come up with a more concrete proposal

eprodrom: great. how long do you think it will take? we can extend 10 minutes

<tantek> +1 on AS2 CR (sorry to be delayed, multitasking is hard)

aaronpk: it is fine. we can just wait

eprodrom: great. thank you very much. appreciate your flexibilty on that.

<cwebber2> good call!

eprodrom: thank you everybody for your time. I believe that wraps it up for us.

<cwebber2> productive :)

<rhiaro> wilkie++

<Loqi> wilkie has 33 karma

<ben_thatmustbeme> wilkie++

<Loqi> wilkie has 34 karma

<cwebber2> wilkie++

<Loqi> wilkie has 35 karma

thanks all

<ben_thatmustbeme> eprodrom++

<Loqi> slow down!

<eprodrom> trackbot, end meeting

<cwebber2> eprodrom++

<tantek> I'm worried about only 5 people coming to the f2f - will we have enough to talk about for 2 days?

Summary of Action Items

Summary of Resolutions

  1. Accept https://www.w3.org/wiki/Socialwg/2016-08-02-minutes as minutes for 20016-08-02 meeting
  2. publish current editor's draft of ActivityPub plus changelog as new working draft
  3. publish https://w3c-social.github.io/social-web-protocols + Changelog as new working draft for Social Web Protocols
  4. publish https://linkedresearch.org/ldn/WD-ldn-20160823/ + detailed changelog as new working draft of LDN
  5. Proceed to CR with both ActivityStreams documents, including the changes worked out with i18n and items marked At Risk in this meeting, as per Ralph's go-ahead from the transition meeting.