Socialwg/2016-02-23-minutes

From W3C Wiki

W3C

- DRAFT -

Social Web Working Group Teleconference

23 Feb 2016

Agenda

See also: IRC log

Attendees

Chair
eprodrom
Present
aaronpk, eprodrom, Arnaud, dmitriz, wilkie, cwebber, Rob_Sanderson, sandro, shevski, bengo, tsyesika, dromasca, KevinMarks
Regrets
Chair
SV_MEETING_CHAIR
Scribe
dmitriz

Contents





<eprodrom> https://www.w3.org/wiki/Socialwg/2016-02-23

<cwebber2> (but I'm still dialing in)

<wilkie> I accidentally scheduled a meeting for now but it only lasted 1 minute

<eprodrom> Can anyone scribe for this meeting?

<eprodrom> scribenick: dmitriz

<cwebber2> dialed in

<eprodrom> https://www.w3.org/2008/04/scribe.html

<wilkie> also https://www.w3.org/2009/CommonScribe/manual.html

Approval of last week's minutes

<eprodrom> https://www.w3.org/wiki/Socialwg/2016-02-16-minutes

<eprodrom> PROPOSED: approve minutes for 2016-02-16

<eprodrom> +1

<azaroth> +1

<tsyesika> +1

<dromasca> +1

<cwebber2> +1

<bengo> +1

<wilkie> +1

RESOLUTION: approve minutes for 2016-02-16

Admin issues

<eprodrom> https://www.w3.org/wiki/Socialwg/2016-03-16

<Loqi> Cwebber2 made 1 edit to Socialwg/2016-02-23 https://www.w3.org/wiki/index.php?diff=97595&oldid=97594

<Loqi> Sandro made 2 edits to Socialwg/DocumentStatus https://www.w3.org/wiki/index.php?diff=97597&oldid=0

eprodrom: I'd like to recommend that people pay special attention to — we have a face-to-face meeting at MIT, on Mar 16. Please let us know if you're participating.
... add an RSVP to that page whether you can or cannot make it (or if you can make it remotely). We will have remote participation
... if remote, we expect you to be at a computer, with a microphone, as usual.

<shevski> where exactly do we rsvp?

<wilkie> https://www.w3.org/wiki/Socialwg/2016-03-16

<wilkie> there ^

eprodrom: question is - where do we rsvp?

<shevski> thx

<eprodrom> https://www.w3.org/wiki/Socialwg/2016-03-16#Participation

<dromasca> will this be a full one day mtg?

<wilkie> two days

eprodrom: there is a section called 'Participation' on the page. edit the wiki to put your name is.

<dromasca> 3/16,17

eprodrom: couple more questions. This is a full 2-day meeting, Mar 16 and 17
... we are just starting to put together the agenda for this meeting
... the two big agenda items are: 1) taking AS2 to CR (Candidate Recommendation)
... and the next is: 2) determining the goals and next steps for our other documents

(webmention, micropub, activitypub, social protocols, and any other outstanding FPWDs that we have)

eprodrom: we are about 9 months away from the end of our charter
... and so now is a good time for getting an idea of what we can do with the other deliverables
... there'll be some fine point discussion of AS2
... and also some fine point discussion of the other documents, to sketch out a roadmap for the next 9 months or beyond
... any other questions about the Mar 16 event?
... no questions, lets move on

Discussion about issues regarding Activity Streams 2.0

eprodrom: there are 3 big things we need to do to take AS2 to CR
... 1) resolve outstanding issues.
... according to James - Arnaud, do you want to take it?

Arnaud: no, go ahead

eprodrom: James is happy where the document is. And would like to take it to another Working Draft, which is what we would be discussing at our face to face
... overall, we've had some new issues in the last couple of weeks, most of them either fine points or very general points
... but there's no blocking issues, work can move forward
... actually, the one big blocking issues that we do have is — conformance
... which is in my corner right now
... I'll be coordinating the next few days with James, to make sure we get the conformance section into the doc
... I'll need to set up a PR, and make sure we get that section in there
... before the next working draft
... the next big point for AS2 is:
... 2) Test suite
... we have two sides to the test suite
... one is - test documents. 100+ docs, useful for someone who is trying to parse ActivityStreams
... this seems to be at a stable point, there is one PR to pull in the docs from James (I'll be working on that the next few days)

<eprodrom> https://github.com/w3c-social/activitystreams-test-documents

eprodrom: we're looking for other AS2 documents
... the other side, that's under active development is the Validator

<eprodrom> https://github.com/w3c-social/activitystreams-validator

eprodrom: I'm continuing work on it, but looking for more participation there
... the validator is pretty basic right now, basically giving a +1 / -1 on docs.

<wilkie> I got it up and running and have been looking at it

eprodrom: should be providing finer points / feedback on the spec, in addition

<eprodrom> https://github.com/w3c-social/activitystreams-validator/issues/11

eprodrom: I'd like it to be showing a hierarchy of different kinds of notes or warnings on the AS2 docs posted

<wilkie> cool

eprodrom: is it providing the right MIME Type, are there missing recommended properties, are there non-recommended properties present, that sort of thing

<Loqi> Sandro made 5 edits to Socialwg/DocumentStatus https://www.w3.org/wiki/index.php?diff=97606&oldid=97597

<Loqi> Sandro made 2 edits to Socialwg/2016-02-23 https://www.w3.org/wiki/index.php?diff=97602&oldid=97595

<Loqi> Kmarks2 made 1 edit to Socialwg/2016-03-16 https://www.w3.org/wiki/index.php?diff=97599&oldid=97593

<Loqi> Dromasca made 1 edit to Socialwg/2016-03-16 https://www.w3.org/wiki/index.php?diff=97604&oldid=97599

eprodrom: my two big things over the next few weeks are: 1) to get conformance section added asap, and 2) get the validator to be a much more useful tool for implementors
... hopefully it'll be ready by the March 16 meeting
... any questions about AS2.0, and where we are?
... test suite, conformance, etc?

bengo: I filed an issue on AS2 validator
... the question is - are AS1 documents valid for AS2?
... we have a few AS1 docs, should they be added to the test repos?

eprodrom: I can't say 100% on this, but my understanding is that we are NOT maintaining backwards compat with 1.0

<wilkie> that's my understanding too

eprodrom: so those AS1 docs would not be valid

<bengo> https://github.com/w3c-social/activitystreams-test-documents/issues/3

eprodrom: the only role that AS1.0 docs would play in any sort of validation process, is — we do use Content-Type switching

<wilkie> they have a very different content type in the spec to assert that

<wilkie> yeah

eprodrom: so if you get the older Content-Type, you'll know it's AS1, if newer, AS2
... and that'll be something worth taesting
... that said, there's probably some value in converting the AS1 docs to AS2, and including them into the suite

bengo: I'll close the issue

eprodrom: your discussion about tweets on the agenda — it's a little farther down, will be discussed later today
... actually, we are at that point!

Representing Tweets in AS2

<bengo> https://github.com/jasnell/w3c-socialwg-activitystreams/issues/283

bengo: this is relevant to what my employer does
... social media aggregation from all the social networks.
... the AS2 vocab is important to us for interop
... tweets are sort of the straw man, but any big Social Media vocabs/schemas are relevant
... I know we talked briefly before about having some sort of official registry
... I'm not sure how to proceed, curious what other people think
... what domain / dns record should this extension live on, in the long term, etc
... I don't want this to be something that only we use

KevinMarks: you don't need to create a new magic thing for Twitter
... you can actually construct the URLs using the tweet IDs, and use those as the AS2 object IDs

bengo: I'm hearing silence

Arnaud: same here

<bengo> URLs good for use case. Was thinking about that on the way home yesterday. Still doesnt solve for the larger registry use case

KevinMarks: aaa, forgot to unmute!

<wilkie> dmitriz: that's evan

eprodrom: ): ben has mentioned earlier that he understood that there are different ways to do this

KevinMarks: the fact that Twitter has relatively good URLs is useful

<eprodrom> dmitriz: that's me

KevinMarks: ben - why are the URLs not appropriate in this case?

<eprodrom> bengo

bengo: doesn't this require idiosyncratic handling for each of the big social network provider, to make the URLs?

<bengo> https://twitter.com/bengo/status/696429143659917312 or https://twitter.com/_/status/696429143659917312

<Loqi> @bengo :: #sb50 #rollingstonelive https://twitter.com/bengo/status/696429143659917312/photo/1

bengo: what about changing usernames? that changes the URLs

<bengo> KevinMarks: you're right I forgot

KevinMarks: actually, Twitter does the right thing and does 301 redirects

<eprodrom> dmitriz: that was KevinMarks !

<KevinMarks> if you fetch https://twitter.com/activity-streams/status/696429143659917312 you get a 301 to https://twitter.com/bengo/status/696429143659917312/photo/1

<Loqi> @bengo :: #sb50 #rollingstonelive https://twitter.com/bengo/status/696429143659917312/photo/1

<eprodrom> He has the English accent; I sound like I live on the beach in San Diego

Rob: is there some other identity that's important to manage, aside from the twitter id?

bengo: for our particular use case, dereferencing to AS2 is less important

<tsyesika> It could be worth people saying who they are prior to speaking on the call to make it easier for the scribe

<KevinMarks> twitter published an oembed json alternte, they could add an as2 one if they felt like it

bengo: it's more important to publish and aggregate

<tsyesika> dmitriz: yes

eprodrom: we should do both
... use the twitter URL for ID, and also have a snowflake property for twitter, in addition
... so I think your question is, "how do you specify what that is? what the extension id is?"

<KevinMarks> twitter has a json alternative already in https://api.twitter.com/1/statuses/oembed.json?id=696429143659917312

<bengo> Draft extension on socialwg wiki?

<KevinMarks> which you cna follow your node to

eprodrom: it seems to me like there's 2 possibilities. one is that we could have a property within AS2, for something like vendor_id or alternative_id or something like that
... which is non-iri, non-url id

<eprodrom> tweetID or snowflakeID

eprodrom: the other option is that it's simply an extension, tweet_id or snowflake_id
... i would lean towards the second
... it makes more sense to me there
... ben, I think you were asking - what's the process for actually defining those extensions?

bengo: yeah. and I agree with you on the snowflakeID thing
... the question is - where is a good place to register that extension vocab?
... for the moment, we can just host it on the W3C wiki, and once there's an actual drafted extension, we can worry bout it then

eprodrom: right, we can throw a drafted extension pretty quickly on the wiki
... and there may be other properties of a tweet that might need to go there
... it's a pretty important use case, and I think we should be able to reach it
... ben, I guess what I'm asking is, would you mind starting a Wiki page on this?

bengo: I don't mind, I will do that

eprodrom: awesome.
... what would be ideal is - being able to take something like "anatomy of a tweet"
... no, that's not it

<bengo> https://dev.twitter.com/overview/api/tweets

eprodrom: there is a page on the Twitter Developer site that shows taking Twitter's JSON step by step, explaining what everything does
... we should be able to map those relatively directly to AS2 format
... and in places that we can't, there should be an extension mechanism

bengo: sounds good

eprodrom: great, let's move on

Update on MicroPub

eprodrom: aaron?

aaronpk: I've been working on incorporating issues on Micropub

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

aaronpk: over on github. here's the link to the latest draft with the changes
... basically, the changes are - updated the normative references
... mistakenly made a bunch of them non-normative
... added a section on querying the source of posts, for editing
... working on a couple other things, will have another draft published for the next meeting
... please review GitHub issues and changes
... and file any CR-blocking issues sooner rather than later
... so I can address them. I'd like to make a bunch of progress this week
... any other questions

eprodrom: I have a general question
... and leads to our next agenda item

Laying out a Roadmap for the next 9 Months in our F-to-F

eprodrom: the big thing that I'd like to ask is - would you mind, over the next few days,
... thinking about what our next steps are, with MicroPub
... what's needed, what are the big outstanding issues, are there problems
... and what do we do to take it to the next steps
... Arnaud and Sandro - is that what we're looking for?

Arnaud and Sandro: yes

eprodrom: so, we're looking for the next steps, with MP

<sandro> https://www.w3.org/wiki/Socialwg/DocumentStatus

Sandro: I went ahead and made a wiki page for the status of all our current docs
... if somebody has a better organization, that's fine

eprodrom: fantastic, that's great, thank you for doing that, Sandro
... we'd like to start collecting the status
... we've had kind of an explosion of docs over the last couple of months
... so now the question is - where are we taking them next?
... so if we can update the status on the wiki, it'll help us make decisions during the f-to-f
... any questions from the editors?

<cwebber2> *page

<cwebber2> that's aaron

<eprodrom> dmitriz: that's aaronpk

aaronpk: it occurred to me that I'd like to publish another draft, and I would like people to review the current draft

eprodrom: any other questions from editors? if not, happy to move on with agenda
... no questions
... next, I'm gonna ask Chris and Jessica about federation work

<tsyesika> I can

Micropub Federation Work

tsyesika: I've been working over the last 1.5 years to bring federation to MediaGoblin
... to implement AS1.0 via Pump.io API

<cwebber2> https://identi.ca/mediagoblin/image/yCNd5dD-S8GL-DCwh50CPw <- mediagoblin federating TO pump.io!

<tsyesika> https://identi.ca/mediagoblin/image/yCNd5dD-S8GL-DCwh50CPw

<cwebber2> haha

<cwebber2> :D

tsyesika: last week, I had the first instance of federating media from MediaGoblin via Pump.io

<wilkie> yay

<cwebber2> ignore the ugly client screenshot... unthemed QT!

tsyesika: currently polishing it up
... I've got receiving federated activities still to do

<cwebber2> but that's the pump api for client -> server, and then mediagoblin <-> pump.io, server to server federation

tsyesika: but shouldn't be too much work

<cwebber2> yes

eprodrom: couple of questions for you -
... 1) how will AP be improved by your experience implementing federation?
... 2) how will MediaGoblin make use of AP in the future?

tsyesika: I'm not sure about #1

<bengo> tsyesika: cwebber2: Is there a web sequence diagram for how this works? It would be helpful to refer to

tsyesika: a lot of the work has just been based on going through and formalizing the various aspects, trying to make the specs more concise
... but I think it was a very good foundation to base on
... as for question 2, my funding will be running out at MediaGoblin,
... but I hope to continue working - I know Chris wants to continue federation work

<cwebber2> bengo, there isn't, maybe I can do an ascii art version ;)

tsyesika: I'm hoping to get ActivityPub support when it's a littl emore mature

<tsyesika> I'm done :)

<bengo> cwebber2: I'd benefit from it! https://www.websequencediagrams.com/ is good too

eprodrom: thank you very much, and congrats on the work, it's really impressive
... oh, last question - if people want to take a look at this particular code,
... is it on the master branch?

tsyesika: no, not on master. I have a fork of it
... that I use, so that the main repo doesn't get cluttered up
... I'll make a push after the meeting, so that people can see it

<tsyesika> absolutely

eprodrom: yeah, if you can push it and share the branch on the channel or mailing list that'd be really helpful
... fantastic, thank you very much
... that does take us to the end of our scheduled agenda
... we are a little more than 10 mins ahead of the hour
... this is the time where we go over issues from Tracker
... as I'm looking, we don't have any new or pending issues
... but I do want to give a moment to see if we have any updates on issues from Tracker

<tsyesika> https://notabug.org/Tsyesika/federated-mediagoblin/src/federation <- this will be where i'll push to

eprodrom: this is becoming less and less frequent as we're moving our issue tracking to GitHub, but it's worthwhile to still do it
... if not, open it up to any other questions or business we need to discuss
... ok, unless there's anything else to discuss,

<tsyesika> I'll definitely do it this evening :)

eprodrom: — jessica, please push soon AND take a hard drive backup! :) don't tempt fate
... otherwise I'd like to offer everyone their 10 mins back
... and thanks Tantek for chairing the meeting next week

<eprodrom> trackbot, end meeting

<KevinMarks> ?

Summary of Action Items

Summary of Resolutions

  1. approve minutes for 2016-02-16

[End of minutes]



Minutes formatted by David Booth's scribe.perl version 1.144 (CVS log)
$Date: 2016/02/23 18:50:36 $



Scribe.perl diagnostic output

[Delete this section before finalizing the minutes.]

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

Guessing input format: RRSAgent_Text_Format (score 1.00)

Succeeded: s/KevinMarks/evanpro/
Succeeded: s/evanpro/eprodrom :)/
Found ScribeNick: dmitriz
Inferring Scribes: dmitriz
Default Present: aaronpk, eprodrom, Arnaud, dmitriz, wilkie, cwebber, Rob_Sanderson, sandro, shevski, bengo, tsyesika, dromasca, KevinMarks
Present: aaronpk eprodrom Arnaud dmitriz wilkie cwebber Rob_Sanderson sandro shevski bengo tsyesika dromasca KevinMarks
Agenda: https://www.w3.org/wiki/Socialwg/2016-02-23#Agenda

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

Found Date: 23 Feb 2016
Guessing minutes URL: http://www.w3.org/2016/02/23-social-minutes.html
People with action items: 

WARNING: Input appears to use implicit continuation lines.
You may need the "-implicitContinuations" option.

[End of scribe.perl diagnostic output]