W3C

- DRAFT -

SV_MEETING_TITLE

09 Aug 2013

Attendees

Present
Regrets
Chair
SV_MEETING_CHAIR
Scribe
hhalpin, bret, AnnBassetti, Theodoros

Contents


<James_M_Snell> but allowing multiple models to be used as needed by implementers, including Schema.org, is valuable

<tantek> KevinMarks: "Schema embodies bad practices"

<hhalpin> Note that the issues around aligning vocabs, etc. should be brought up in final report

<evanp> Kevin Marks points out that non-open spec development is a "worst practice"

<tantek> Schema ignores vCard4 for example

<Andrew_Mallis> http://schema.org/docs/terms.html

<evanp> hhalpin: I think I've done the scribing incorrectly

<evanp> I don't see anything in the minutes

<AnnBassetti_> evanp: ALL scribing is helpful

<Andrew_Mallis> intersting how the schema.org devolves into all caps by the end

<evanp> audience: we will have different layers in social federation

<evanp> transport, identity, meta-language, vocabularies

<evanp> wseltzer: thanks

<evanp> hhalpin notes that most of tomorrow will be around running code

<evanp> OK, we're going to the bar

<evanp> James_M_Snell: while I have you here

<evanp> I think

<James_M_Snell> I'm still here

<evanp> I htink

<evanp> I'm still kind of lost what the goal is then

<wseltzer> [break]

<wseltzer> [adjourn]

<wseltzer> [bar]

<evanp> I think what you're saying is that we're aligning the AS JSON serialization with JSON-LD

<evanp> But we're not aligning with schema.org/Action

<James_M_Snell> right

<evanp> OK

<James_M_Snell> in the sense that objectType can be viewed as an alias for @type

<evanp> And we're keeping backwards compatibility to the extent possible

<James_M_Snell> yes... in fact, I've made it so that every 1.0 thing is still valid in 2.0

<James_M_Snell> 2.0 just provides a few additional options to address some missing pieces... like i18n

<James_M_Snell> and links

<James_M_Snell> but those additional things are optional

<James_M_Snell> the json-ld alignment just means that it's possible to look at an AS 2.0 thing as if it was JSON-LD, and use that as a means of interpreting it using some other data model

<evanp> OK

<evanp> So last question, which might not be appropriate to ask publicly but wth

<James_M_Snell> go for it

<tantek> ask away

<evanp> Is this an alternative to schema.org actions, in opposition, or a step in that direction?

<evanp> I think you showed an example of mixed encoding

<James_M_Snell> not an alternative.. it's a recognition that there are valid use cases behind that, even if there are problems with schema.org's approach to solving them

<James_M_Snell> AS 2.0 relaxes and simplifies the model, makes it less verbose

<James_M_Snell> and as a result, it's possible to represent the schema.org/Action model in AS 2.0 without breaking compatibility

<James_M_Snell> originally schema.org/Action was heading in a very different direction

<James_M_Snell> they had an alternative, incompatible json syntax included in the proposal

<James_M_Snell> I pushed back and demonstrated that with just a few tweaks here and there, the existing AS json syntax was more than sufficient

<James_M_Snell> but really, that's only a small part of it

<James_M_Snell> for me, the improvements to i18n, links and less-verbosity are far more important

<James_M_Snell> those are the things IBM's customers and partners have asked for the most

<James_M_Snell> does that answer the question? :-)

<wseltzer> [Day 2: Beginning shortly]

Day 2

How Mobile Revolutionizes Social

<bryan> do we have a scribe?

<bryan> not yet

<inserted> scribenick: hhalpin

MarkWeitzel: Intro, schedule

MonicaLam: Is there a way to disrupt Facebook?
... via the mobile phone?
... as the mobile phone has the context and architecture
... a few canned social networks
... phones are thin clients to social networks, and interaction dictated by cloud service providers
... very poor use of a phone
... the OSM messaging-based network
... allows ad-hoc social networks
... without servers
... the apps consult with phones

<bryan> no servers anywhere? how does a find b and c?

MonicaLam: and then turn into social networks
... i'm sure there's servers in the background, just not social network cloud providers

<bryan> how are messages passed between a and b and c?

MonicaLam: this gives a great deal of flexibility towards storage
... now for a demo
... showing demo of automatic sharing with a group
... and group creation on the fly
... "Hotspot" group

<bryan> "this page" implies that there is a server somewhere facilitating the data transfer

<bryan> the "messaging layer" must require some intermediary system

<James_M_Snell> Intermediary in theory could be a federated network of independent nodes.. like XMPP or MQTT

<James_M_Snell> idea is sound, not new, but definitely requires an infrastructure to support it

<aaronpk_> i'm not clear on how this uses dropbox / gdrive/ whatever to store the data

MonicaLam: this leverages the distributed cloudsesrvices but we still own data

question: do you have rights management?

<bryan> what does "sharing" mean - by reference or copy?

MonicaLam: Right now we keep it simple, anyone in group can join any other person in group
... invite any other person to group
... we can also allow pictures to be retracted

Ashok: How do you get from email address to phone number?

MonicaLam: We ask for it on sign-up

<bryan> sounds like they have replaced one proprietary network with a proprietary app?

<aaronpk> does it use some sort of peer-to-peer network for communicating or does it still rely on their own service to transport the data?

MonicaLam: Always a way to share
... existing social networks are good for broad-casting!

question: are details of system available?

MonicaLam: Company has a propritary version, reserach version is open-source
... the API is open of course
... We are tackling who holds the data
... we are not selling data being shipped through network
... we have end-to-end encryption

question: where do you store?

MonicaLam: We do store and forward, store on Amazon

<bryan> but they had to disable the encryption as they couldn't provide the same service with it

MonicaLam: how do we communicate with students in a certified way/

<bryan> many details are unclear here; without a clearly open details on the design and deployment I don't think one could claim that the system, not matter how pretty/functional, is "open"

<bryan> s/not how/no matter how/

MarkWeitzel: OpenSocial
... state of the union

AndySmith: We want to share context with app, where it's rendered, but maintain security across iframes

MarkW: ActivityStream gets a new expense report, but link jumps me out
... lost where I was
... we can now keep you in the context
... with 2-way bidirectional secure communications that's fundamentally changed the way we get stuff did in social systems
... one of the challenge for 3rd party developers, they need to know who the user is
... added in API allow developers to extend that
... that was 2008-2011

AndySmith: We liked encapsulation of gadgets
... challenges to developers
... weren't quite like building a normal HTML app
... lots of frameworks, so we are interested in HTML5, ShadowDom, etc.
... getting some of what we wanted to do with iframes
... made things a lot easier
... an app model that builds on what browsers provide
... how do we share events?
... we are at point where we need to shape those technologies
... how do we thoughtfully inject social context and make that available?

<James_M_Snell> btw, we (IBM) have a working prototype of the web component stuff.. not sure if andy is planning to show it or not... if not, we'll hopefully get it out very soon

AndySmith: what's path to get there?
... polyfills, what's the gap, partnering with W3C to get there faster

<tantek> so where's the OpenSocial presentation gadget?

<James_M_Snell> tantek: ;-)

bethlavender: from Mitre
... many of our use-caes are same as we heard yesterday
... find experts
... situational awareness

<tantek> James_M_Snell knows dogfooding, do you?

bethlavender: custom social networking site called 'handshake'

<James_M_Snell> tantek: we use opensocial internally in Notes and Connections

<James_M_Snell> it needs to improve

<James_M_Snell> alot

bethlavender: handshake notifications
... interesting gadgets

<tantek> James_M_Snell, but you know my opinion of "use … internally": http://tantek.com/2013/219/t16/spec-only-adopted-intranet-without-open-web-fail

bethlavender: integrated activity for the enterprise

<James_M_Snell> tantek: using internally is how we learn how to best use externally

bethlavender: stream from social networking with business social updates

<tantek> James_M_Snell - hah - not how standards have flowed in practice.

<James_M_Snell> we have to get folks to understand it first

<James_M_Snell> yes, sadly. that's a battle I've been fighting for a while

<tantek> James_M_Snell - no one has ever won that battle.

bethlavender: we like activities with embeddged gadget
... but we still have business portal that lets us look the business at a particular point in time
... otherwise, we see the business as activisites occur

<James_M_Snell> tantek: incremental improvement is best we can hope for. Herd the cats

bethlavender: these are two different paradigms
... we do think activitystreams SHOULD be mobile
... dealing with things inside encrypted containers
... HTML5 browsers
... how do we manage noise
... subscription process?

<tantek> James_M_Snell - you're always welcome to join the indieweb: irc://irc.freenode.net/indiewebcamp where we build things incrementally on the open web ;)

bethlavender: we want different versions based on user's amount of time

<James_M_Snell> tantek: ;-)

<bryan> will the slides be posted as well as the papers?

yes, after the workshop

<bryan> thx

Most people have not sent them in before, thus we can't post them before

For example, we got all of these presentations about 2 minutes before panel!

<James_M_Snell> Sending them in before means folks have to finish them before getting on the plane ;-)

Ellen: AppFuisons allows for semi-portable rapid integration
... common problems, common issues for document issues
... wrking with jive/ibm, and 6 atlassina porduct applications (JIra, stash, ffishaeye)
... we've done oAuth/oAuth2 and then Atlassian apps become slave to master of IBM/Jive
... our custom applicatin effort we make the standards talk to each other
... that is what makes opensocial a huge success
... pusing back from Jira to Activitystreams
... links that automatically link back to apps
... jira/jive integration demo

<tantek> btw - feel free to add URLs to the presentations to the Etherpad where a bunch of us are taking rough collaborative notes: https://etherpad.mozilla.org/osfw3c

Ellen: works between a number of apps, keeping in one browser
... the number of tiles are kinda endless
... thanks!

<bret> Ill log some of this

<bret> Social api isn't really an api

<bret> its not really social

<bret> its just the web

<bret> in your browser (all slide quotes)

<bret> put the agent back into "user agent"

<bret> James_M_Snell: this is for you :)

ShaneCaraveo: Mozilla Social API is not really a API

<James_M_Snell> bret: :-)

<scribe> Scribenick: bret

<aaronpk> now it's official

solve common tasks in the browser

how do we make communication capabilities better [in the browser]

Sure

Shane: its all the web, avoid opinions about exiting social networks
... did a table session asking the question, if the web browser did x, what would that allow you as an implementer of a social network do
... what could you do if your agent could do more

<James_M_Snell> Main Question: Have any efforts been made to convince other browsers to pick this up? Work on a standard cross-browser api? How have the other browser vendors responded?

Shane: current features, sidebar, notifycations, social bookmarking, share, chat windows, persistent web workers

notifications*

James_M_Snell: from last night, not much yet, but full intentions to do so once its more mature

Shane: brought in a number of social networks, including facebook
... social bookmarking will be in FF26
... FB and google hangouts can be extracted out of a particular browser tab
... App tabs become buttons on the chrome

browser chrome*

Shane: notifications show up on the buttons
... each button can have a notification flyout to provide a stream of data
... oexchange is used in the share button in FF23 (landed)
... examples of chat windows
... showing prototype webRTC chat client using social API

(slides of prototype)

<Kelvin_Lawrence> I have the same question that James asked just above about cross browser support etc.

I asked shane last night and he says yes, he wants that but he is not actively seeking it yet until the new version is live and working

Ok hhalpin I'm all done :)

<Kelvin_Lawrence> Unless all the browsers pick one approach we are back to #ifdef hell again

<James_M_Snell> Kelvin: things like polymer and web components can get us part of the way there

<James_M_Snell> but even those suffer cross-browser issues

<James_M_Snell> I view this as one of *the* major gaps

<Kelvin_Lawrence> +1

<mixedpuppy> I should have said; google hangouts could be extracted out if google supported this

<James_M_Snell> mozilla is spot on, our user agents need to have a better understanding of our identity, our context and our services and learn to communicate that better to the services we use

<tantek> OH: "I'm not here as a representative of Google"

<James_M_Snell> TODO: reconcile mozilla social API, web intents, Activity handlers, and open social

<tantek> James_M_Snell, you can drop web intents from that list since Google itself dropped it.

<James_M_Snell> tantek: I don't care so much about what google has done, the idea is still relevant.

<tantek> James_M_Snell - nah, it was never a cohesive idea - tried to do to much.

<James_M_Snell> in that specific form, yes

<tantek> the relevant use cases are now being developed with Web Actions (buttons on websites) and Web Activities (delegation on/to mobile web apps)

<tantek> so you may want to check those out (both googleable)

<mixedpuppy> we have web activities in FFOS

<tantek> :)

<James_M_Snell> yep, already on my follow lists ;-)

<mixedpuppy> it will eventually make it to desktop

<tantek> wow, DNT trolling?

<tantek> wseltzer++

<AnnBassetti> Lloyd Fassett: cool stuff being done with Open Social .. what's the linkage across apps, HTML5, in future?

<wseltzer> [break]

<AnnBassetti> answer: that's why we're here .. to figure out what more can / should be done .. Open Social not in HTML5, that's coming

<AnnBassetti> uh oh .. doesn't look like the minutes worked properly ... can someone check, and make sure we don't lose all of this morning's notes?

hhalpin: sorry about screwing up the scribe format. I didn't realize there was a standard way to do it

<aaronpk> it's all there, but it didn't register hhalpin as a scribe for the first part

<goto> Hey all! Sam Goto here, from schema.org/Actions! I wanted to make sure some of the (extremely valid) questions raised yesterday that I left unaswered were addressed and that an open channel with the community is kept! I invited Guha, a much more informed person than my self, to come up speak at OSF/W3C and I think Harry was able to accomodate to that and is coordinating something! Stay tuned for updates!

<wseltzer> goto: thanks!

<goto> I think there has been some confusion and miscommunication on our part, and i sincerely apologize if we are not being clear. I feel personally committed to clarify some of the concerns and comments raised around licensing and the ability to consume schema.org data freely, and Guha will hopefully make that clearer to you.

<goto> I have a plane to catch at 7pm today at SFO, so I might not be able to come join you in person f2f, but I do wanted to make myself available to follow up on any of the technical concerns you may have.

<AnnBassetti> scribenick: AnnBassetti

<wseltzer> I'm sure the group will appreciate it

Running Code

<goto> You can drop me an email at goto@google.com any time and I'm more than happy to come up to the city to meet you on another opportunity! See you all, Sam

Tantek Çelik .. tantek.com

IndieWeb demo

you need to ownur own data, esp. now with NSA

yes

<hhalpin> scribenick: AnnBassetti

permalinks still can't do on Twitter

principles:

1. Own your own data

2. Eat your own dogfood

'self dogfooding = using your own dogfood, on your own domain

forces you to simplify, etc

<aaronpk> once you carry your own ego along with your code by self-dogfooding it helps you figure out what you really care about

3. POSSE = Publish Own Site, Syndicate Elsewhere

your info is no good, if people don't read it

get out of your own silo

own your own data, but syndicate out to places / silos where your friends 'live'

he's got apps for these purposes

this is the core of what "IndieWeb" is about

Tantek makes plea for people to show their own domain name, on their #osfw3c badges

------------

Aaron Parecki ... aaronparecki.com

IndieAuth .. distributed identity for web

you are online all over, you have your own web site ...

typically you are asked to authenticate via Facebook, et al.

BUT .. then you're locked into them

if you have your own web site .. then that can be your identity

supports new auth mechanisms .. Mozilla Persona, SMS verification, and <1 other>

demo...

has google authenticator app on his watch ..

IndieAuth server goes to his personal web site

looks for known auth providers .. he chooses one

<see his slides for more>

now delegating to OAuth provider .. in this case GitHub

currently this idea relies on IndieAuth server

yes .. but you could run your own

you can delegate your personal domain to whichever indieauth provider you want

more info at spec.indieauth.com

don't send email; do it online, in order to converse in public

----------------------------------

Bret Comnes, PhD student in Oregon, participating in IndieWeb

bcomnes.githum.io

indieWeb commenting on static web site

bret.io and prose.io

demo making comments on Aaron Parecki's site

<Ashok_Malhotra> I THINK he is talking about how to add comments to GitHub pages

-----------------------------

Evan Prodromou .. pump.io

activity stream server .. models social network with REST api

every activity has subject / verb / object ... also an audience target

various types of audiences

with varying degrees of 'openness'

streams in pump.io are ways of grouping activities

in reverse chronological order

separates 'major' and 'minor' activities

REST api for pump.io uses JSON representation

endpoints for each stream

all streams readable

some streams writable

streams filtered by audience

<see slides for more details>

inbox / outbox / direct inbox

social graph = who follows whom, contact lists, member of group(s), participation in activities

social graph changes based on what you send into the system

social graph visible via REST api

pump.io intended to be web-based social network , adding activity stream to apps, embedded activity

interest from mobile developers

demo

web UI allows OAuth login

Apache licensed

python, Java, javascript libraries ...

--------------------------------

Ben Werdmuller ...... Indieweb as minimum viable social platform

"Idno"

a social publishing platform

easy to use, easy to extend, easy to install ...

grew out of his experience with Elgg

an open source social networking platform ...

e.g., in universities, non-profits, govts, .....

social programming toolkit

BUT doesn't support modern front-end formats such as microformats

nor HTML5

therefore he wanted to replace it

wanted to re-think, based on what's going on now

dangers when building social platform in enterprise...

-- putting technology before usability

-- investing a lot before testing your ideas

-- developing neat solutions that don't have a user problem

-- not shipping; not getting your cool software out

summary: put your users first

thus he likes IndieWeb

no one owns or organizes it

uses widely-adopted standards

IndieWeb allowed him to test Idno quickly

demo

Idno also works well on mobiles ..

no apps; just HTML5

----------------------

Patrick Deegan .... Open Mustard Seed

company = ID3 ...

develop new social ecosystem of trusted self-healing digital institutions

Open Mustard Seed (OMS) Trustworthy Compute Framework (TCF) extends core functionality of personal data stores

REST apis

<see slides for lots of attributes>

opt-in

sort of an operating system for the cloud

Trusted Application Bundles .. contain instructions for how to deploy and maintain applications

App Deployment Pattern .. auto setup by script, creating Trusted Compute Cell

each Trusted Compute Cell is composed of several applications ...

persona-based access control

follows Jericho Forum's identity commandments

API security

<all slides have lots more details>

native app for authentication

demo

-----------------------

end of IndieWeb demos

HarryHalpin: will have working lunch to tease out "next steps"

questions:

Theodoros __??__ : about access control transform

Patrick Deegan: <describes how it works>

Q: what kind of format to call your rules?

A: JSON

EvanP: OStatus was talked about re: federated social web in last few years ... what's up with that now?

Tantek: looked at some PHP library for security .. was too huge, couldn't trust
... because we're doing this stuff for ourselves, has to be super simple ...

<tantek> looked at PHP library for consuming OpenID was too big, gave up, came up with RelMeAuth, which IndieAuth is based on

OStatus = suite of protocols used to enable social netoworking

<tantek> looked at ActivityStreams, was more than we needed for absolute minimum, used minimal microformats instead (h-entry h-card)

PubSubHubBub ... Activitystrea.ms ... Salmon ... Webfinger

(EvanP explaining)

<tantek> though I implemented ActivityStreams on my site, don't know who consumes it

those are the 4 components of OStatus stack

<tantek> looked at PubsubHubbub (actually implemented it) but again, not being used in indieweb protocols

published in 2010

<James_M_Snell> We definitely need more activitystreams consumers

<tantek> indieweb is using webmention (evolution of Pingback) instead of Salmon / PuSH

<James_M_Snell> lots of places can generate Activity statements... very few understand how to use them

has server API but no client API ... has been an issue

no distribution control

scribe: also a prob

great as first step, test out some of those protocols ... now people looking elsewhere for next iteration

Tantek: now looking at WebMention ... replaces some bit of Salmon and PubSubHubbub

HarryHalpin: gives instructions for working lunch

how to thread all of discussions this far, together?

go around room for ideas on topics for discussion

will continue after lunch in focus groups

particularly on standards needed

<James_M_Snell> in reality, how is webmention.org different from PingBack/TrackBack

<tantek> James_M_Snell - everything you wanted in PingBack and less :)

<tantek> no XML-RPC

<tantek> DROP ALL THE XMLS

<James_M_Snell> of course, none of that is actually necessary

1. Federated Web design patterns ...

<tantek> feel free to use our Etherpad: https://etherpad.mozilla.org/osfw3c

<James_M_Snell> tantek: have you seen http://tools.ietf.org/html/draft-snell-link-method-02

<tantek> IETF drafts? dude, way too much boilerplate noise to bother reading

<tantek> need to put things on a simple & clear wiki ;)

<James_M_Snell> skip ahead to http://tools.ietf.org/html/draft-snell-link-method-02#section-5

<tantek> please add your names in the right column

<James_M_Snell> it's even HTML for you

<tantek> of the Etherpad: https://etherpad.mozilla.org/osfw3c

<tantek> James_M_Snell - will queue it for later reading

Topics for discussion:

<James_M_Snell> simple version: LINK /foo HTTP/1.1

-- Design patterns around Federated Social Web / IndieWeb

<James_M_Snell> Link: <http://whatever>; rel="tag"

<James_M_Snell> creates a link between /foo and http://whatever

<tantek> can we show the etherpad instead since it's collaboriate? https://etherpad.mozilla.org/osfw3c

how can we make the indieweb for social business?

-- how to make this all work for social a BUSINESS ... which is not as simple as IndieWeb folks think

<aaronpk> I think benwerd and tantek and myself would be more than happy to discuss this

-- how to solve today's probs versus further out?

<tantek> AnnBassetti - short answer from me - Enterprise ignored HTML+HTTP+URL in 1992 as well - we're in the early days of IndieWeb tech. give it a few years and when it's spread on the open web, it'll likely be mature enough for enterprise as well

-- gadgets in relation to new web technologies

-- EvanP: future of Activity Streams

-- EvanP: federation of social systems across enterprise

-- identity and profile federation

-- analytics

<James_M_Snell> EvanP: if there's a future of activity streams discussion let's find a way for me to either call in or do a hangout or whatever

-- how to visualize our social graph / social network analysis

-- linked data, description of skill sets

-- property graph

-- how to represent roles?

--- how to get on IndieWeb and what are the barriers?

<tantek> IndieWeb how-to or issues

<Ashok_Malhotra> Perhaps combine Property Graphs with Analytics

-- -------------------------

lunch

<James_M_Snell> FYI... just thoughts re: WebMention ... http://www.chmod777self.com/2013/08/now-that-you-mention-it.html

<wseltzer> [returning]

Return from Breakouts

<AnnB> will

<hhalpin> scribenick: hhalpin

MarkWeitzel: Radical simplification
... leverage the Web Components
... and the W3C
... in turn of apps themselves
... xml+gadget defintion
... where you just built a HTML page generated by Shindig

<Theodoros> hi

MarkWeitzel: to a situation where you perform AJAX requests directly against the page
... where there's a rigidly described set of patterns to a situation where app developers can do whatever they want

<scribe> scribenick: Theodoros

<AnnB> scribenick: Theodoros

cross origin management

is important

as well as data context within the browser

application tags or the equivalent should be supported

----------------------------------------------------------------------------------------------------------------------

next is property graphs

and analytics

<hhalpin> steve: distinct from rdf

the foundamental data model for property graphs

should be defined

as well as APIs

and schemas to help data modelers

to define property graphs for their applications

work with open social

to find out domains for working together

and how property graphs can be exploited

security, privacy, internationalization

are themes for potential cooperation

<AnnB> --------------------------------------------------

<AnnB> Tantek .. IndieWeb

focus is ux challenge

to make it much easier using reply buttons

integrated to the indie web

what can the browser do for sharing buttons

like a web action tag

social api share includes an example

from mozilla

that can be utilized

by indie web team

visible tags from indieweb will get in the next release

<AnnB> they asked, "where is next biggest UX pain point, and how can we solve it"

----------------------------------------------------------------------------------------------------------------------------------------------------------------

activity streams is next

what are common use cases to be considered

one is to handle state

some markup will be posted on how to do this

from the discussion team

a new version of activity streams is proposed

to support json-ld

(activity stream 2.0)

to help consumers of jason-ld through streams

<AnnB> Monica Wilkinson giving overview

using the @ property and alias from jason ld

<James_M_Snell> activity streams 2.0 aligns with json-ld, but does not depend on it

<James_M_Snell> that is... the spec details how the two can be used together for those who need it

---------------------------------------------------------------------

profile federation team is next

focus of the discussion is roles

- the need to federated profiles and multiple personas

<AnnB> need to federate profiles across systems

content federation

profile management

are example of desired features

defining a set of core attributes

inspired by microformats example

also standrard attributes can be defined

not as part of standardization but their linking to ontologies

(skill levels, certifications)

federation across systems would be a desired feature

pubsub can provide some ideas

<tantek> great!

discussion should follow on the data definitions for extended attributes

---------------------------------------------------------------------------------------------------------------------------------------------

discussion on concensus follows

open social revision will most probably follow

activity streams will also be revised

there is a lot of interest for domain specific vocabularies for profiles

external data formats will be leveraged

discussion about commitment to move forward with next steps

questions on whether should be one or two groups to followup

<wseltzer> hhalpin: activity streams, custom vocabularies (profile data), pubsub, opensocial version next

<James_M_Snell> Activity Streams 2.0 revision is largely done really, at least in my opinion

Ashok Malhotra describes plan for property graphs

start work about standardizing data and API

and then the words/schemas that can be used by web developers

a community group will be started for property graphs

establish a property graph model and API

agree on how to express schema for the property graph so that this is interoperable

work with open social

to define ontologies related

to social

<tantek> "a schema to describe a property graph that will be interoperable and then terms and ontologies that will be social, and a relationship with RDF, what didn't work, and then there's a bunch of concerns"

there will be a break for 15 min

<wseltzer> [break]

and then next steps will be discussed for the rest

--------------------------------------------------------------------------------------------------------------------

discussion has resumed

10 people interested

in community group for property graphs

8 people are interested in activity streams2.0/JASON-LD

<gkellogg> +1 to property graphs

<gkellogg> +1 to ActivityStreams2/JSON-LD

10 people interested in OpenSocial on top of webcomponents with activity streams

12 people interested in profiles vocabularies and public subscription model

<AdamB> +1 on Profile PubSub activity

<Bill_Christian> +1 activity streams

11 people interested in SocialBusiness Reference architecture document

<MRC> +1 on architecture document

<lehawes> +1 on Profiles and Overall Architecture

13 people interested to develop vocabularies for social business uses

<MRC> Schema.org?

-----------------------------------------------------------------------------------------------------

final comments on anything leaving out on the joint technical strategy

push the boundaries on browser cross origin model

is interesting for next followups

the social business community will be used to invite attendees

report will become available in about one month

<tantek> please feel free to note that in the minutes

1st week of september will be milestone for the report

<tantek> add suggestions / requests to http://indiewebcamp.com/ for IndieWeb

<AnnB> Social Business Community Group: http://www.w3.org/community/socbizcg/

<AnnB> go join

<KevinMarks> first week of september is when the next indiewebcamp is as well

-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

discussion about vocabularies follows

<AnnB> also put your name in here (IRC), with which subject you want to work on

<AnnB> (if you didn't already do a +1 on some subject)

<Ashok_Malhotra> If you raised your hand for the Property Graphs work please email me at ashok.malhotra@oracle.com

<steveb> sblackmon@w2ogroup.com - property graphs, activity streams 2.0, social business vocabularies

[OPENi team] Interested to be part of the community for property graphs and social business reference architectrure document

discussion about schemas.org

<AnnB> I will probably try to find someone else within Boeing for property graphs, Ashok

<Ashok_Malhotra> Great, Ann!

<AnnB> (you can put me on the list, but I'm not the best participant)

the schema license is for putting data on a page not for parsing

the license scope is for vocabularies

<tantek> people are confused

google will followup to make the license terms clearer

<bret> A) Why can't they use a more open licence B) Will this FAQ hold up in court?

<bret> :/

<tantek> wseltzer_cloud - since Guha confronted someone about not being a lawyer - I think it's up to you

<mfranklin> mfranklin@apache.org - property graphs, activity streams 2.0, social business architecture, OpenSocial.NEXT

<tantek> bret - OGP uses OWFa, so should schema.org

<bret> I, unfortunately, am not fully up to speed on all the license types

<AdamB> are these the correct terms being discussed: http://schema.org/docs/terms.html

<James_M_Snell> Suggestion to the google guys: Go back, review the licensing and how it's communicated, do a rev and make it better

<bret> someone feel free to bring up the more specific point

<tantek> AdamB - yeah - commentary here: http://tantek.com/2013/219/t19/osfw3c-contrast-microformats2-cc0-owfa-ogp-schema

<James_M_Snell> if it's just a matter of not being communicated properly, then work on communicating better

<tantek> James_M_Snell - yeah - their TOS is highly suspicious right now.

<tantek> easiest thing would be to adopt a standard license like OWFa rather than a one-off

<tantek> go Monica

<tantek> Monica is raising the point that Schema.org was NOT open about forking ActivityStreams

<KevinMarks> give monica the mic

<bret> :D

<tantek> interesting - asking for folks to come work on the w3c list

<James_M_Snell> Schema.org Activity proposed emerged over a year and a half ago

<James_M_Snell> I called them out on it

<James_M_Snell> told them to go to activitystrea.ms publicly

<James_M_Snell> didn't hear anything else about it for a year

<James_M_Snell> then they still didn't go to activitystrea.ms community until I kept pushing

<James_M_Snell> so the process leaves a lot to be desired

<James_M_Snell> but current end result is at least compatible with AS 2.0 approach, they can coexist peacefully and usefully

<tantek> James_M_Snell - did the Schema.org folks themselves use the w3c public-vocabs list for Activities, or did they do it in private?

<tantek> *their fork

<James_M_Snell> Yes and no

<bret> ITs rather arrogant!

<jeff> The public-vocabs CG is open

<James_M_Snell> there was public discussion on public-vocabs, but lots appears to have been done off list

<James_M_Snell> but to their credit, they were very receptive of feedback

<bret> Put it on github/bitbucket if its open to participation

<James_M_Snell> and made significant changes when I pushed back

<tantek> just searched public-vocabs and all I found was your post James__Snell from May 28

<tantek> http://lists.w3.org/Archives/Public/public-vocabs/2013May/0190.html

<tantek> zero follow-up

----------------------------------------------------------------------------------------------

<mfranklin> so not a standards body or legal entity, yet it is called schema.org

<mfranklin> and people think of it as such

<lehawes> It's a shame that the Schema.org rift is casting a dark shadow on the good, positive work that has been accomplished in this workshop

<tantek> "just 4 companies, and no one has a say"

<lehawes> Can we please find a way to end on a positive note?

<tantek> no one *else

<James_M_Snell> tantek: like I said, their process leaves a lot to be desired

<tantek> lehawes i hope so too

<jeff> Good point Lehawes. We'll go to summary after Tantek's question.

<AnnB> /me +1 Gregg

<AnnB> Wendy gives lawyers a much better name

<lehawes> Agree, AnnB, but he shouldn't be so defensive. Coming in with a bit of humility would have gone a long way. :>)

<AnnB> yes, true .... we could usually all use more humility

<AnnB> I appreciate Harry's effort to wrap up on a better note

<tantek> AnnB me too

<jeff> Harry++

consistent vocabularies with consistent process is the common goal

<tantek> thank you wselzter

<AnnB> I also appreciate Guha clearing his calendar to come into the lion's den

<tantek> AnnB - given Google's market position they're more the lion

<lehawes> +1 federated profiles

<MichaelAlexander> +1 (Profile Data Federation)

<AdamB> +1 I'm in

<lloydfassett> +1 profiles

<James_M_Snell> Takeaway: If Google wants Schema.org to play a key role in this social arena, it needs to do a much better job at clarifying and communicating their position, and about being open

<AnnB> Mark Bryant = +1

<AnnB> (from Boeing

<Kelvin> I'm happy to help drive the graph discussion

<Bill_Christian> +1 Activity Streams

<MattMarum> +1 OpenSocial

<mfranklin> calling for volunteers

<gkellogg> +1 Activity Streams

<bret> James_M_Snell: I think its clear they prefer to just dictate. There is clearly resistance to outside input, and the fact that there is even an outside

<AdamB> I'll lead the Profile Data Federation

+1 property graphs

<lehawes> +1 Social Business Reference Document

<gkellogg> +1 Property Graphs

<James_M_Snell> bret: we shall see, they either need to put up or shut up, in the meantime everything else will keep moving forward

<AnnB> maybe someone from Boeing for property graphs (if I can find someone)

<KevinMarks> +1 Activity Streams

<tantek> feel free to use the Etherpad as well

<KevinMarks> +1 Profiles spec discussion again

<Kelvin> I am also interested in the schemas/types discussion as it plays into the graph discussion

<MichaelAlexander> +1 Social Busines Vocab Modelling (Expertise first/focus)

<KevinMarks> iie +1 Social Business Vocab Modelling

<MattMarum> +1 property graphs

<AdamB> +1 Social Busines Vocab Modelling (Expertise first/focus)

<KevinMarks> what si the URL?

<KevinMarks> can't see from here

<bret> I originally put micro data + schema on my site. Besides being a couple pain in the ass, it was a ton of work and literally, the tool set, community, and documentation/samples just didn't exist. In less time than it took remove the MD+schema, I implemented uF2 and have found way more applications for my effort

<bret> complete*

<MattMarum> http://www.w3.org/community/socbizcg/

<tantek> bret - you should blog that ;)

<tantek> #osfw3c

<tantek> "… thanks to the indieweb people for showing up even though the enterprise initially scared them" [LOL]

<tantek> :)

<bret> Olds just don't get it :p

<benwerd> who's scared? I'm ready to embrace!

<bret> (only teasing)

<tantek> (aside: thanks to KevinMarks found another thread on public-vocabs: http://lists.w3.org/Archives/Public/public-vocabs/2013May/thread.html#msg158 )

Summary of Action Items

[End of minutes]

Minutes formatted by David Booth's scribe.perl version 1.138 (CVS log)
$Date: 2013/08/09 04:51:45 $

Scribe.perl diagnostic output

[Delete this section before finalizing the minutes.]
This is scribe.perl Revision: 1.138  of Date: 2013-04-25 13:59:11  
Check for newer version at http://dev.w3.org/cvsweb/~checkout~/2002/scribe/

Guessing input format: RRSAgent_Text_Format (score 1.00)

FAILED: s/not how/no matter how/
Succeeded: i/Intro, schedule/scribenick: hhalpin
Succeeded: s/demo post//
Succeeded: s/q+//
Succeeded: s/asha/Ashok Malhotra/
Succeeded: s/Activiety/Activity/
Found ScribeNick: hhalpin
Found ScribeNick: bret
Found ScribeNick: AnnBassetti
Found ScribeNick: AnnBassetti
Found ScribeNick: hhalpin
Found ScribeNick: Theodoros
Found ScribeNick: Theodoros
Inferring Scribes: hhalpin, bret, AnnBassetti, Theodoros
Scribes: hhalpin, bret, AnnBassetti, Theodoros
ScribeNicks: hhalpin, bret, AnnBassetti, Theodoros

WARNING: Dash separator lines found.  If you intended them to mark
the start of a new topic, you need the -dashTopics option.
For example:
        <Philippe> ---
        <Philippe> Review of Action Items


WARNING: No "Present: ... " found!
Possibly Present: AdamB Andrew_Mallis AndroUser2 AndySmith AnnB AnnBassetti AnnBassetti_ Ashok Ashok_Malhotra Bill_Christian Bin_Hu Brian_S Bryant DonB_ Ellen EvanP HarryHalpin James_M_Snell Keith Kelvin Kelvin_Lawrence KevinMarks Link Lou MRC MarkW MarkWeitzel MattMarum MichaelAlexander Mike5 MonicaLam OH Shane ShaneCaraveo TODO Takeaway Theodoros Thor Thor_ Wilson aaronpk aaronpk_ aaronpk__ andybs answer audience barnabywalters benwerd bethlavender bret bryan donb donb1 fmondin fmondin_TI gkellogg gkellogg_ goto hhalpin inserted jasnell jeff jeff_ jeff__ joined left lehawes lloydfassett mfranklin mgmarum mixedpuppy osfw3c principles questions rawn sam sam_ scribenick shh steve steveb summary tantek tantek_ wseltzer
You can indicate people for the Present list like this:
        <dbooth> Present: dbooth jonathan mary
        <dbooth> Present+ amy


WARNING: No meeting title found!
You should specify the meeting title like this:
<dbooth> Meeting: Weekly Baking Club Meeting


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


WARNING: No date found!  Assuming today.  (Hint: Specify
the W3C IRC log URL, and the date will be determined from that.)
Or specify the date like this:
<dbooth> Date: 12 Sep 2002

Guessing minutes URL: http://www.w3.org/2013/08/08-osfw3c-minutes.html
People with action items: 

WARNING: IRC log location not specified!  (You can ignore this 
warning if you do not want the generated minutes to contain 
a link to the original IRC log.)


[End of scribe.perl diagnostic output]