From W3C Wiki



Social Web Working Group Teleconference

10 Feb 2015


See also: IRC log


AdamB, aaronpk, WSeltzer, tantek, Sandro, Arnaud, cwebber2, dret, AnnB, eprodrom, elf-pavlik, hhalpin, Lloyd_Fassett, Harry, Harry.a, Tsyesika, dromasca, bill-looby


<Loqi> Bill-looby made 3 edits to Socialwg/Social API/API Usage Patterns

<Loqi> Bill-looby made 1 edit to Socialwg/Social API/Enterprise Contexts

<Loqi> Bill-looby made 1 edit to Socialwg/Social API/Enterprise Contexts

<Loqi> Tantekelik made 1 edit to Socialwg/2015-02-10

<Loqi> Benthatmustbeme made 1 edit to Socialwg/2015-02-10

<Loqi> Bill-looby made 1 edit to Socialwg/Social API/User stories

<Loqi> Bill-looby made 1 edit to Socialwg/Social API/User stories

<trackbot> Date: 10 February 2015

<cwebber2> dialing in

<Arnaud> dret, are you on the call yet?

arnaud, i just re-called, but it still feels lonely.

<AnnB> hmm .. me too .. hearing nothing

annb, good i am not the only one!

<AnnB> oh .. just heard Tantek

<cwebber2> just dialed in

<tantek> Zakim: ??p7 is cwebber2

<cwebber2> thx tantek

<tantek> np

<bblfish> hi

<bblfish> in philosophy conference on the Hermeneutics of the Social Web.

<tantek> bblfish, URL?

<bblfish> not sure, but it's part of this series

<Arnaud> scribe: dret


<Arnaud> PROPOSED: Approval of Minutes of 3 February 2015 Teleconf

<hhalpin> IPcaller.a is hhalpin

nobody objects the minutes of last week's call: the minutes are approved.

<eprodrom> +1

<Arnaud> RESOLVED: Approval of Minutes of 3 February 2015 Teleconf

<eprodrom> Yes

arnaud cannot join next week, chairing RDF data shapes meeting. evan will chair instead

no ACTIONS marked as in review.


anybody wants to declare victory on an open ACTION?

<tantek> action 28

<hhalpin> Note that Jeff is working on Weibo, we'll stay in touch on Yandex

<tantek> action 28?

<elf-pavlik> action-28

<trackbot> action-28 -- James Snell to Prepare working draft for publication based on on Friday Jan 16, 2105 -- due 2015-01-20 -- CLOSED


evan is declaring defeat on the russian/chinese social networks connection. anybody can help out with the action?

<hhalpin> AnnB - Chaals can help with Russian, W3C China and Japan with Asian networks

<tantek> oops mistyped

annb is willing to send a note, evan happily accepts.


<elf-pavlik> action-32

<trackbot> action-32 -- Evan Prodromou to Examine apis for yandex, vkontakte, sina weibo, tencent, and/or renren -- due 2015-01-28 -- OPEN


<hhalpin> Note I did send a note to Chaals (Yandex) and Huawei, no response.

annb wants to better understand if there are non-english tools/services out there that we should learn about. starting by sending out a note to help getting some contacts.


<elf-pavlik> harry:

<eprodrom> +1

<elf-pavlik> action-14

<trackbot> action-14 -- Harry Halpin to Set up json-ld context for namespace -- due 2014-12-02 -- OPEN


even explains: we're looking for features of APIs that we haven't considered yet. annb willing to help with ACTION-32

<tantek> action-38?

<trackbot> action-38 -- Evan Prodromou to Convert social api requirements to 2-10 user stories of 3-5 sentences each -- due 2015-02-10 -- OPEN


<tantek> thanks eprodrom!

<AnnB> I will ask whomever (weibo, tencent, yandex, etc) I can think of .. could they look at the APIs already described, and see if their products have any other patterns not already described

<elf-pavlik> eprodrom++

ACTION-38 can be markes as closed.

<Loqi> eprodrom has 3 karma

<eprodrom> AnnB: perfect!

<AnnB> I'll cc: you on my query


looking at ISSUE now; one ISSUE is marked as pending review.

<Arnaud> ISSUE-7

<trackbot> ISSUE-7 -- Are AS consumers required to understand the pre-JSON-LD syntax? -- pending review


ISSUE-7 is marked as "pending review"

<elf-pavlik> looks like conversations happened in october

<elf-pavlik> dret: we have to be clar what certain things mean when they show up in AS object

<elf-pavlik> ... on one side we try to stay friendly to consumers on JSON-LD

<elf-pavlik> ... on the other hand we dont want to require JSON-LD processing

<elf-pavlik> ... what do you do when you encounter something you dont know how to handle (as software application)

<eprodrom> Oh that's cool

<cwebber2> whaa

<eprodrom> B-)

<elf-pavlik> ... we may neeed to state it more clearly

<elf-pavlik> ... explain what AS object means for various consumers

<tantek> there are multiple pre-JSON-LD syntaxes :)

<sandro> harry, we're reaching your messaging system.

<elf-pavlik> ... i plan to write longer explaination of issues we face and maybe post it on my blog

<elf-pavlik> dret: it may not very well if we want to leave it open

<tantek> who just joined?

harry quotes the charter saying that we do JSON and maybe JSON-LD.

sorry, harry!

harry says the charter says we do JSON, and if possible JSON-LD.

<Zakim> tantek, you wanted to say he expects implementation evidence to help determine that

<elf-pavlik> harry: JSON-LD saves namespacing problem in very graceful way

tantek points out the wide variety of syntaxes to choose from, including non-JSON. the charter gives us a first stake in the ground. what we must do is expressing the concepts in some syntax.

tantek says we must first agree on conceps, and then we can decide on the syntax.

<bblfish> I wonder where json without json-ld would make sense

<tantek> JSON is a good stake in the ground to express the concepts

<cwebber2> if we don't settle on a standard and a wide variety of serializations are possible, will implementations be compatible?

<tantek> there is no need for JSON or JSON-LD in practice to build these systems - as many of us have figured out

<tantek> and deployed

<elf-pavlik> can someone scribe when dret speaks? computer i use today has keyboard in it lang :(

<Loqi> it'll be ok

<hhalpin> Note that has a large vocabulary and is a deployment success without requiring the RDF data model.

<tantek> hhalpin: same with microformats

<hhalpin> So I think it's premature at this point.

<tantek> huge deployment - no strict RDF data model was needed

there may be more ISSUEs connected to ISSUE-7, i guess?

<tantek> though it is possible to use it in an RDF-like way

<eprodrom> +1

<tantek> Arnaud: let's not review issues unless there is no information

moving on to the glossary part of the agenda

<tantek> tantek: sounds good. but let's review all issues at the f2f

<tantek> Arnaud: yes that is a good idea

<elf-pavlik> +1 rhiaro

<cwebber2> right rhiaro

<cwebber2> I agree with that summary

elf-pavlik started collecting terms on the wiki.

<rhiaro> namespacing doesn't matter if everyone agrees on one vocab right, like mf2 or - but as soon as someone wants to coin their own term in a decentralised way (maybe if they want to expand AS2 with a new content type) and describe it, the documentation for that should be some way discoverable, which is why namespacing is useful, JSON-LD or otherwise... as I understand, anyway

<rhiaro> forgetting nitty gritty debates about whether linked data is any good, it boils down to discoverability of documentation

elf-pavlik is working on, any help greatly appreciated.

<hhalpin> I made my comments in the mailing list

<elf-pavlik> hhalpin: i think i added them to wiki page

<tantek> rhiaro, I don't think namespaces (like URI bound) are necessary for experimental extension/expansion

<elf-pavlik> thanks Arnaud !

<hhalpin> yep, I recommend we use terms the rest of the community uses rather than reinvent new ones

<hhalpin> Thanks for starting Elf!

nobody volunteering to help with the glossary

next agenda item are the user stories.

last week agreed: develop user stories. evan created a first set of user stories.

<rhiaro> tantek, is what people do with experimental microformats is add eg. an x- prefix?

<AnnB> seems like a good effort to me, but I'm worried I don't know what "standard" terminology is to contribute

<tantek> rhiaro - or a vendor prefix -vendor-

<sandro> "The user stories are intended as the basis for a survey, to help people in the group understand the features others in the group find important or not-so-important. They do not constitute any kind of exhaustive list or agreement about what features are required by API candidates."

<rhiaro> but if I see someone's x-newmf I don't know how to find out what that's for, even as a human, unless I maybe read their blog or about page or ask them on irc

<tantek> note that such experimentation has been used with some success in CSS

<elf-pavlik> tantek, rhiaro can we try to avoid double conversation?

goal of the user stories is not to define committed user stories, but more a list of stories people are interested in. they should give input to people putting forward proposals for the Social API.

<bill-looby> tantek, ipcaller is bill-looby

evan will explain what he did, and then we can move forward after last week's incident around who edited what when and why and how.

evan put up a list of 10 user stories based on the requirements document we had.

<bill-looby> oops, meant Zakim !

<bill-looby> thanks

evan says the stories were trying to make things very direct. after those went up, there was a bit of a scramble how they should be edited.

evan explains that now we put a name to the stories, to make it easier to track their origin. now we have 20-30 user stories. maybe we should take some time to review and edit them.

<tantek> also pushes survey out

<tantek> but I am ok with 1 week extension

<tantek> for user story contributions

evan is concerned that we should keep time in mind as we go through this process.

<elf-pavlik> should we write proposal to extend 1 week

<Zakim> elf-pavlik, you wanted to discuss creating lists of stories where a story can appear on multiple lists

<tantek> I for one am very impressed by the level of contributions to the user stories

<tantek> I have some concerns, but I'll wait to queue to express them

<elf-pavlik> PROPOSAL: extend collecting user stories for another week

<eprodrom> tantek: agreed!

elf-pavlik is asking whether the call is extended for a week. arnaud says that it is so that we have a little bit more time.

<elf-pavlik> +1

nobody objects to extending the user story collection for one more week.

sandro is wondering how the stories and the survey connect.

evan says we should wait until the stories are in their final form before we add our feedback to them. they may still change. arnaud says that we should put our names there, but only when we're done changing the stories.

<Zakim> elf-pavlik, you wanted to discuss creating lists of stories (one story can appear on multiple lists)


<tantek> hey that's clever elf-pavlik

<cwebber2> +1 on reaching out to other free software groups

elf-pavlik wants a story to be allowed to appear in multiple lists, because they may apply to various scenarios.

<Zakim> tantek, you wanted to answer sandro questions about survey. +1/0/-1 for/neutral/against a user-story, and if +1, also optionally +intendtoimplement

<cwebber2> er

<tantek> where's the story about making collections of stories? ;)

<cwebber2> federation etc projects working in this space

<cwebber2> I can take on reaching out to those groups

<elf-pavlik> tantek: good point! will do it like with

tantek says that before commenting on a story, you should also say whether you intent to implement a story in your system.

<eprodrom> cwebber2, elf-pavlik can you do an action for that?

<hhalpin> no, you don't vote from a process point of view :)

<elf-pavlik> cwebber2: i did on IG tracker!

<cwebber2> eprodrom: sure

<Zakim> sandro, you wanted to suggest -1 must be explained, eg "not important to me because..." vs "would end life on earth".

<cwebber2> elf-pavlik: oh, I don't know what that means for me then :)

lloyd is wondering whether IG people are supposed/allowed to vote; arnaud says that you can, but your vote is not binding for the WG. harry says that the process rules says that only WG people are allowed to vote.

<cwebber2> also I don't know how to propose an action, I can look on the wiki

sandro asks that -1 votes on stories have an explanation.

<elf-pavlik> cwebber2:

<cwebber2> elf-pavlik: oh okay :)

<cwebber2> elf-pavlik: want to reach out to friendica, and I can reach out to diaspora?

<elf-pavlik> cwebber2: let's discuss it here right after this telecon?

annb is wondering if there's value to putting question marks to user stories; arnaud says that yes, we can do that.

<cwebber2> elf-pavlik: sounds great

<Zakim> tantek, you wanted to discuss when to vote and to also say can use +0 - with question to capture Annb's use-case of a question

arnaud points to how the RDF data shapes WG is doing something similar; it may be a good example to look at (email sent yesterday).

<eprodrom> Are we resolved?

tantek says it's important to not add opinions until we officially ask them to do so. please don't add votes now.

<elf-pavlik> RESOLVED: extending deadline for collecting user stories for a week

tantek says to annb that a "?" could be a "0" and ask the question as a comment.

<tantek> +1 to extending deadline

<AnnB> that idea (+0 with question) sounds better than using ??

<aaronpk> +1 to extending

<bill-looby> +1 to extendeing

<eprodrom> +1 to extending by a week

<tantek> any objections to 1 week extension for collecting user stories?

<AnnB> +1

<rhiaro> +1

<AdamB> +1

arnaud asks officially to extend the deadline by a week; no objection from anybody.

<Tsyesika> +1 for 1 week extra (no more)

deadline extnded by 1 week. evan asks to not extend this deadline indefinitely, let's talk about it now instead of maybe extending it again.

<elf-pavlik> we can freeze the page next week and start second page for eventual new stories

<elf-pavlik> q*

bill would like to get some feedback on story categories.

<cwebber2> +1 to extending (late :))

<cwebber2> already resolved, but ;)

elf-pavlik wants a flat list of stories and then we can "link" to them from various lists, if we want to.

<tantek> please keep the stories short too

<rhiaro> +1 sorting stories via links

<elf-pavlik> example of collection via links

evan wants to have a flat list as well, if we want to organize, then that should be overlayed and not hardcoded into the list of stories.

<rhiaro> (or tags?)

<elf-pavlik> rhiaro: can i use URI of section as tag? ;)

<eprodrom> B-)

arnaud claims victory on the agenda item for user stories.

bill explains IBM connections (and i have a very very hard time understanding what he's saying).

<elf-pavlik> bill-looby: can you speak little slower please?

<AnnB> sorry, but I'm not clear on what happened w bill-looby's suggestion to have some user stories about integration

<elf-pavlik> AnnB: he can add them and then create section with link to group them

connections offers file sharing, wikis, activities for project management, and more things. it has an Atom-based API, it also supports AS.

<elf-pavlik> bill-looby: peopole can just post to connections and let it do all the socializing

IBM gets many requests for making integration easier. customers want to take a reference to an object in their system, post it to connections, and then have connections track all of the interactions around that object.

  • annb, thanks for feeling for me!

<AnnB> bill-looby: most important 'building block' is "person"

<AnnB> ... and where their identity is stored

important building blocks are "abstract objects", with the object being stored anywhere, while the metadata is stored/managed somewhere else.

<AnnB> .... need to define how to track an object, compared to tracking person

<AnnB> ... plus, notion of "place"

(( i am officially giving up for this agenda item. )

<AnnB> ... first integration for IBM was content mgmt sys with Connections

<AnnB> ... questions about which "side" supports likes, comments, follows .. etc

seems like connections mostly needs a robust model of identity and linking.

<AnnB> ... most requests beyond basic social are integrating between systems

<AdamB> these are some of the things that we are experiencing within Boeing as well

arnaud asks bill to create a wiki page with some documentation or just links do existing documentation.

<elf-pavlik> bill-looby:

<tantek> bye I have to go!

annb points out that this is a classic enterprise scenario and very important in these kinds of settings.

<eprodrom> Thanks dret !

<elf-pavlik> dret++

<eprodrom> Thanks Arnaud ! Good meeting.

<Loqi> dret has 1 karma

arnaud officially closes the call!

<AnnB> thanks dret!

<eprodrom> Read the cases!

<elf-pavlik> ciao!

<AnnB> thanks Arnaud

<elf-pavlik> cwebber2: i'll swich computers and jump back here!

<rhiaro> cwebber2, Tsyesika: are you UK based?

<Arnaud> thanks AnnB

<Arnaud> trackbot, end meeting

Summary of Action Items

[End of minutes]

Minutes formatted by David Booth's scribe.perl version 1.140 (CVS log)
$Date: 2015-02-10 19:05:05 $

Scribe.perl diagnostic output

[Delete this section before finalizing the minutes.]

This is scribe.perl Revision: 1.140  of Date: 2014-11-06 18:16:30  
Check for newer version at

Guessing input format: RRSAgent_Text_Format (score 1.00)

Succeeded: s/we do JSON-LD/we do JSON and maybe JSON-LD/
Succeeded: s/for a week/by a week/
Found Scribe: dret
Inferring ScribeNick: dret
Default Present: +1.510.206.aaaa, AdamB, aaronpk, WSeltzer, tantek, Sandro, Arnaud, cwebber2, dret, AnnB, eprodrom, elf-pavlik, hhalpin, Lloyd_Fassett, Harry, Harry.a, Tsyesika, dromasca, bill-looby
Present: +1.510.206.aaaa AdamB aaronpk WSeltzer tantek Sandro Arnaud cwebber2 dret AnnB eprodrom elf-pavlik hhalpin Lloyd_Fassett Harry Harry.a Tsyesika dromasca bill-looby
Found Date: 10 Feb 2015
Guessing minutes URL:
People with action items: 

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

[End of scribe.perl diagnostic output]