2015-10-28 UTC
tantek, annbass, taisuke, tilgovi and massu joined the channel
# 01:36 rhiaro_ We're about to have a social breakout session at TPAC # 01:37 rhiaro_ I'm going to be scribing in here, for those interested pJeong joined the channel
# 01:39 rhiaro_ tantek: Welcome to the social web breakout session! # 01:40 rhiaro_ ... I'm Tantek Celik, one of the co-chairs. Arnaud le Hors is also here as a co-chair # 01:40 rhiaro_ ... We've been working on this very hard problem which is the social web # 01:40 rhiaro_ ... First, here's a real world example of the TPAC page - notice the very first encouragement of participation is here called 'social', little twitter logo, hashtag tpac2015 csarven joined the channel
# 01:41 rhiaro_ ... So here's a question. Twitter, a member of w3c, not present at this week's meeting, not here today, but w3c is promoting them as very first method of participation on the tpac page. What's wrong with this picture? # 01:41 rhiaro_ ... That's the magnitude of the problem we're trying to address # 01:41 rhiaro_ ... The problem goes all the way to the top of w3c # 01:42 rhiaro_ ... What does 'encourage you to share this event on social networking sites' mean to anyone here? # 01:42 rhiaro_ Manu: The assumption is it's a website that allows multiple peopel to congregate on that website # 01:42 rhiaro_ ... Which I feel is different from how it should be set up # 01:42 rhiaro_ ... People having their own sites, posting stuff, being able to move their content from one place to another # 01:42 rhiaro_ tantek: if you have your own site and post content on it, say your site gm_kou joined the channel
# 01:43 rhiaro_ ... w3c by posting share this on social networking sites might not be aware that most of us have their own websites # 01:43 rhiaro_ ... What else does this mean to people, to post on social networking sites? Anyone? David_clarke joined the channel
# 01:43 rhiaro_ tantek: the very first encouragement that tpac is giving to attendees to participate is twitter. WHat does that mean to you? # 01:44 rhiaro_ csarven: it's a question of.. you can pick one or you can pick a giant list of ways in which you can participate, twitter just happens to be one of them # 01:44 rhiaro_ ... obviously doesn't fit well from the perspective of the social web wg, but to me it's just one of many channels # 01:44 rhiaro_ ... of course it's a silo and so on, but i'm not sure if that necessarily has to be the main point of discussion # 01:45 rhiaro_ jeff junglecode.net: it seems like the network effects of these social sites is what makes them viable # 01:45 rhiaro_ ... you go because your friends are already there # 01:45 rhiaro_ ... so for the social web wg to be viable, you need to get those network effects # 01:45 rhiaro_ ... I'm curious if the goal of the group is to make it so that is possible to achieve those network effects without being twitter # 01:46 rhiaro_ deiu: I think one of the reasons why they put that there is because twitter is a popular means for peopel to share messages, events, stuff that happens # 01:46 rhiaro_ ... it's not necessarily the social network, but the common label that we use to speak in a particular context # 01:46 rhiaro_ ... this is the key part that w3c wanted to use, the reason they wanted to use twitter and the hashtag # 01:46 rhiaro_ ... gives people a way to discover information even if they don't necessarily follow everyoen that participates in this event # 01:46 rhiaro_ tantek: so hashtag is a keyword to let people discover things # 01:47 rhiaro_ ... There's an agenda, instead of reading a webpage you see an email about it # 01:47 rhiaro_ ... In that email, there's a box that says 'social' that says 'use the aol keyword to post on aol' for this event. What would you say to that? # 01:47 rhiaro_ ... My assertion is that the w3c posting this in 2015 about twitter is no different from ietf posting this in 1995 # 01:48 rhiaro_ ... twitter has a web interface that means I don't have to have a twitter account to read these messages # 01:48 rhiaro_ ... I know where you're going because you have countries like austira who prohibit in schools to organise themselves in facebookgroups because they don't want to force people to have a fb account annbass joined the channel
# 01:48 rhiaro_ tantek: so if twitter was a walled garden with no public access then it would be different # 01:49 rhiaro_ David Clarke: I'm on the i18n wg, and it occurs to me that depending on what country you're in twitter may not be an available route to access that information # 01:49 rhiaro_ ... although you're talkinga bout proprietary walled gardens, we also have politcal walled gardens # 01:49 rhiaro_ ... that type of thing breaks up the communications # 01:49 rhiaro_ ... in the past I've tried to use various social media in other countries, and there's literally a firewall # 01:49 rhiaro_ ... whereas if I got to an ordinary website I've actually got the access rossen_ joined the channel
# 01:49 rhiaro_ ... so it strikes me that if you want worldwide social you need to actually look at how you make that, but equally keep it politically safe deiu joined the channel
# 01:50 rhiaro_ tantek: it sounds like you're bursting the bubble that all my friends are on there # 01:50 rhiaro_ David Clarke: I know there are countries I've been in where I couldn't see twitter, or had to go through a proxy beyond the standard tech ability of many other people # 01:50 rhiaro_ ... So maybe we don' thave the univeral feed that you think we do have danbri and shevski joined the channel
# 01:51 danbri tantek points out that Twitter is not a sponsor or at TPAC or in this room, and yet promoted by W3C in TPAC page # 01:52 rhiaro_ ... that even on the tpac page w3c is promoting a silo that isn't even accessible in other countries # 01:53 rhiaro_ (?): countries will always filter information that they don't like # 01:53 rhiaro_ ... they can have whatever technologies to filter web # 01:53 rhiaro_ annbass: let's not waste our time talking about this # 01:53 rhiaro_ rigo: I think the most important thing we should discuss is identifiers # 01:54 rhiaro_ ... that allows you to transport things over hops # 01:54 rhiaro_ manu: there's a decentralised hashtable movement to create portable identifiers so people can move their data. Very early work. # 01:54 rhiaro_ ... Two points about that box. I think the reason that w3c an dmost of the membership think sit's okay, is many of us read twitter as throwaway data (whether that's okay or not is another question) # 01:54 rhiaro_ ... It's great that the socialwg is working to create a viable solution azaroth joined the channel
# 01:55 rhiaro_ deiu: in an attempt to stop this speculation about the reason whyt hat text is there, becasue frankly, I don't know who put it there and it's not constructive to figure out why it's there # 01:55 rhiaro_ annbass: it's more like publicity to the outside world rigo joined the channel
# 01:55 rhiaro_ deiu: we should move away from that and get back on track with the main topic # 01:55 rhiaro_ annbass: no real agenda, just we're doing this work and we need people to join us # 01:55 rhiaro_ ... these are some reasons we need to do the work jet joined the channel
# 01:56 rhiaro_ ... even though most of us are engaged in our social tools of choice kazue and wydong joined the channel
# 01:56 rhiaro_ ... all of them are closed environments, and so what about having open opportunities that could be used for anything really # 01:56 rhiaro_ ... I work for Boeing, we're interested in this for within the company # 01:56 rhiaro_ ... we anticipate starting to open that up to suppliers, vendors, airline customers # 01:56 rhiaro_ ... there are all kinds of opportuniteis and issues # 01:57 rhiaro_ ... but we can't say airline xyz go meet us on facebook, in a business environment that's not going to happen # 01:57 rhiaro_ ... we need these foundational definitions of technologies to be available to build on # 01:57 rhiaro_ ... but we need more participants in the wg to get there # 01:57 rhiaro_ ... we think we should start with something simple so we can get something defined and out the door and build on that, but the more participants that help us lay the foundation the better it will be to build on # 01:57 rhiaro_ manu: I'm completely unaware of where you are right now, could we have a 5 minute where is the socialwg right now? # 01:58 rhiaro_ shevski: I'm Irina. I have a project to promote decentralised technologies and bring communities together MattPisut and MarkS joined the channel
# 01:58 rhiaro_ ... One thing I've found is so many projects, we have a list of over 200 open source projects, over 10 of them are protocols that are working to try and address this problem # 01:58 rhiaro_ ... the thing I'm interested in is really brining together these efforts and energies and passions together timbl and m4nu joined the channel
# 01:59 rhiaro_ ... what we werally need is open standards and a growing viable ecosystem that uses these open standards # 01:59 rhiaro_ ... that can accelerate. might be slow to begin with # 01:59 rhiaro_ ... but as we progress there are more poeple interested in this space, and a growing awareness of these issues # 01:59 m4nu q+ to ask about where the Social Web WG is currently wrt. work products. # 01:59 rhiaro_ ... but this is a really key moment to define some of those standards and bring those communities together # 01:59 rhiaro_ annbass: facebook twitter and the like have zero interest # 01:59 rhiaro_ shevski: we don't want them at this stage, their business models are not what I think is the future or where things are moving # 02:00 rhiaro_ ... we need to start growing an ecosystem that has alterntaive business models # 02:00 m4nu q+ to ask "what are the most compelling business models?" # 02:00 rhiaro_ ... maybe some of the monopolies might start changing their habits # 02:00 rhiaro_ ... I don't think it's fundamental that they are involved Jim joined the channel
# 02:01 rhiaro_ ... I want to reach out ot all of you who are builders # 02:01 rhiaro_ ... as members who are builders, we need you to join the working group an dcontribute your expertise # 02:01 rhiaro_ arnaud: from a techncial point of view it's not so hard, right? # 02:01 rhiaro_ ... If you look at some of the pieces we rely on like email, it's completely decentralised # 02:01 rhiaro_ ... designed from ground up with a very simple protocol # 02:01 csarven For the record, Amy had her hand up (very briefly). # 02:02 rhiaro_ shevski: and you can send emails to hotmail account owners # 02:02 m4nu q+ to mention why he isn't participating (group seems to be in conflict) Norm joined the channel
# 02:03 rhiaro_ ... like ira said, look we have 200 projects, I see egov, neighbourhoods organising themselves, they have no way to connect to each other because the building blocks are not there # 02:03 danbri rigo, identifiers for what? # 02:03 rhiaro_ ... I don't want to here go down into the geek bits and bytes # 02:03 rhiaro_ tantek: instead of saying there are no building blocks, if you don't think there are, ask the question what are the building blocks? # 02:04 Norm q+ to observe that it's at as much a social problem as a technical one # 02:04 Zakim sees m4nu, csarven, Norm on the speaker queue # 02:04 rhiaro_ ... ACtivityStreams is our most mature working draft # 02:04 rhiaro_ ... we have 3 deliverables, social syntax, an api for clients and servers to communicate that information, and the third area is a federation protocol for servers to pass the social information among themselves # 02:04 rhiaro_ ... In the first area, we have a document that we're moving forward with # 02:04 rhiaro_ ... In the second, we have inputs from existing communities that have working building blocks # 02:05 rhiaro_ ... in social api and federation, we have working building blocks that are being contributed # 02:05 rhiaro_ ... there's a lot of hope there, we just have to get the work done and produce the respected w3c drafts to keep moving forward with that # 02:05 rhiaro_ annbass: there are three technical communities in the wg # 02:05 rhiaro_ tantek: I don't think the boundaries are that sharply drawn # 02:06 rhiaro_ ... there are people with lots of different backgrounds # 02:06 rigo danbri: that is one of the questions that has to be answered # 02:06 Zakim sees m4nu, csarven, Norm on the speaker queue # 02:06 rhiaro_ annbass: bringing together the perspectives has been a challenge, but there's beena lot of progress # 02:06 rhiaro_ ... we need more participation to make sure we cover the territory # 02:06 Zakim m4nu, you wanted to ask about where the Social Web WG is currently wrt. work products. and to ask "what are the most compelling business models?" and to mention why he isn't # 02:06 Zakim ... participating (group seems to be in conflict) # 02:06 rigo danbri: certainly ID has a split between host and account (Identity) # 02:07 rhiaro_ ... we had a bunch of different aspirational schedule deadliens based on no particular concrete evidence, so all those dates from the charter have been thrown out of the window # 02:07 rhiaro_ ... So we don' thave a specific timeline now that's based in any kind of fact # 02:07 rhiaro_ m4nau: but the hope is that as2 is the one in front and probably going to be the first one? # 02:07 rhiaro_ tantek: probably, but there are enough open issues I can't dependably make that statement # 02:07 rhiaro_ m4nu: mention that fb and twitter aren't here and aren't interested # 02:08 rhiaro_ ... if an organisation is going to... and I completely get that this is best fo rthe web... but it's very hard for companies to participate if there are not compelling business models to do so # 02:08 rhiaro_ ... has part of your work been selling these business models to companies? # 02:08 rhiaro_ annbass: we have not, we need expertise to do that # 02:08 rhiaro_ tantek: arnaud, from IBM who are building products to make money from social web..? # 02:09 rhiaro_ arnaud: there are different constraints in the wg and people with different interests. People like tantek who chairs indiewebcamp, who are trying to build this open network outside of silos # 02:09 rhiaro_ ... IBM are interested in the social interaction that integrate with enterprise applications # 02:09 aaronpk whoa whoa whoa there is no "chair" of indiewebcamp ;) # 02:09 rhiaro_ ... Obviously there's a social element that comes into place in enterprise # 02:09 rhiaro_ ... We see the same applications that apppear in all different tools # 02:09 rhiaro_ ... We need standards protocols to do this integration # 02:10 rhiaro_ annbass: example in Boeing, we're seeing social activities like pstoing, sharing, equivalent of liking, in many different vendors products coming into boeing # 02:10 rhiaro_ ... We don't want our people to ahve to replicate their profile in 50 different places, we want a central profile that will be used in others # 02:10 rhiaro_ ... For big companies,t hat's where the openness of these technologies would be valuable # 02:10 rhiaro_ ... Most companies are doing nothing with social yet # 02:10 rigo danbri: question is whether they already opted for email address format. Currently looking it up # 02:10 rhiaro_ ... Certain suppliers are trying to sell to us. We're building our own. # 02:11 rhiaro_ ... From a US perspective, I don't know abou tother countries, by the way, we have no participation from non-american companies in the socialwg shevski joined the channel
# 02:11 rhiaro_ ... we need people from other countries so that we don't... I don't know really what goes on in china, japan, india, russia # 02:11 rhiaro_ ... we need perspectives from a broader group of people # 02:11 Zakim Norm, you wanted to observe that it's at as much a social problem as a technical one # 02:12 rhiaro_ ... I really do care about doing this in the open # 02:12 rhiaro_ ... THe observation I would like to make is we're not fighting a technical problem but a social one # 02:12 rhiaro_ ... Email is decentralised protocol but everyone uses gmail # 02:12 rhiaro_ annbass: I don't really know how to do email somewhere else. we just go to a tool that makes it easy for us to do it # 02:13 rhiaro_ tantek: even hackers can't set up their own email servers right now # 02:13 rhiaro_ Norm: my observation is that there is a social aspect above the technical one # 02:13 rhiaro_ ... people group around things, people will do the easiest thing # 02:13 rhiaro_ ... business are motivated to do the thing that makes them money # 02:13 aaronpk srsly, I stopped hosting my own mail server a long time ago. search for "email deliverability" and you will see why # 02:14 deiu I run my own email server and I've never had any issues with email being blocked # 02:14 rhiaro_ ... I don't know what the focus of the group is really, I've been distant # 02:14 rhiaro_ ... Your poitn about enterprise and business use cases are something this group could work on and we could all benefit from # 02:14 tantek email is no longer in practice an openly federated system # 02:14 rhiaro_ ... In a business environment we certainly see value in what we do # 02:14 rhiaro_ annbass: I encourage you to join us, we need more busines speople # 02:15 rhiaro_ Dave Clarke: If you are dealing with international issues of any description, there is an i18n wg # 02:15 rhiaro_ ... we're small and overstretched but although it's useful for you to get people from lots of idfferent countries, you can talk to us please! # 02:15 Zakim sees shevski, deiu, timbl on the speaker queue # 02:15 Zakim sees shevski, deiu, timbl, m4nu on the speaker queue # 02:15 rhiaro_ annbass: I'm always fascinated by the work of the i18n group, and accessibility of course, this kind of work needs to be reviewed for accommodation of all of those considerations # 02:16 deiu tantek, I don't think you can use that article in a "normative" way. It's just someone's experience # 02:16 rhiaro_ ... What I'm interested in from other countries is, people eeverywhere no matter who you are want to be social, there are different modes of cultural interaction from different flavours # 02:16 rhiaro_ ... I don't want this to have an american flavour to it # 02:16 tantek deiu - do you have counter-experience to cite? # 02:16 m4nu q+ to mention why he doesn't feel comfortable with participating in Social Web WG (group seems philosophically conflicted) # 02:16 Zakim sees shevski, deiu, timbl, m4nu on the speaker queue # 02:16 rhiaro_ timbl: There was a call for what are the building blocks # 02:17 rhiaro_ ... so those of us who are building apps, it turns out when you build something like to do a distributed version of the unconference stickies # 02:17 rhiaro_ ... I should be able to launch on my website, code to do the unconference thing # 02:17 rhiaro_ ... I ought to be able to host anyone who wants to have an unconference tod o that # 02:17 rhiaro_ ... and you ought to think that's stupid, and fork it and create better colours # 02:17 Zakim sees shevski, deiu, timbl, m4nu on the speaker queue # 02:17 rhiaro_ ... so take that as an example, or take doodle or wahtever, they involve people # 02:17 Zakim sees shevski, deiu, m4nu on the speaker queue # 02:17 rhiaro_ ... every app eends up doing what every social network does and saying select which people are involved # 02:18 rhiaro_ ... on a mac there's an api for doing that with your contacts app # 02:18 rhiaro_ ... so building blocks are things like access control, introducing people, building up lists of people to be in the group # 02:18 rhiaro_ ... some of which are already in system, some aren't # 02:18 rhiaro_ ... you've probably built lots of widgets and modules like that # 02:18 rhiaro_ ... There are two reasons for making these protocols instead of software # 02:18 Zakim sees shevski, deiu, m4nu, annbass on the speaker queue # 02:18 rhiaro_ ... One is the pain of making the same thing over and over # 02:19 Zakim sees shevski, deiu, m4nu on the speaker queue # 02:19 rhiaro_ ... But what's mroe important is when any app I run says # 02:19 rhiaro_ ... who are the people who are going to beinvolved? # 02:19 rhiaro_ ... I want to be able to say all the peopl ein my entire life # 02:19 rhiaro_ ... That's a big win for a user, and for a developer to just plug this in taisuke_ joined the channel
# 02:19 rhiaro_ ... a module that can interoperate with everyone else's modules # 02:19 Zakim sees shevski, deiu, m4nu on the speaker queue # 02:19 rhiaro_ ... there's a finite set of building blocks, that's just one David_clarke joined the channel
# 02:19 rhiaro_ annbass: we have a list of 100 use cases and we've worked to refine them # 02:20 rhiaro_ ... I wanted to pick up on a few things that manu said and ann # 02:20 rhiaro_ ... I don't know about you, but I send a lot of emails # 02:20 rhiaro_ ... a lot of a lot of peoples jobs is sending emails # 02:21 rhiaro_ ... the reasons it's so ubiquitous and important si because it's an open standard # 02:21 rhiaro_ ... but nobody makes money off email. There are so many companies and industrues that use communication channels that they're not making money off # 02:21 rhiaro_ ... They're the ones who need open standards for social # 02:21 rhiaro_ ... As we're living in a more global, remote working world, those are the people who need open standards and don't want walled gardens # 02:21 rhiaro_ ... That's why facebook and twitter aren't really the people # 02:22 rhiaro_ ... and it *is* their business model to have solutions that are interoperable # 02:22 rigo q+ to present pep-project # 02:22 Zakim sees deiu, m4nu, rigo on the speaker queue hellojintae joined the channel
# 02:22 rhiaro_ deiu: it's counterproductive to demonise facebook and twitter # 02:22 rhiaro_ ... first and foremost peope should have the choice of which service to use # 02:23 rhiaro_ ... as long as it's interoperable with everything else # 02:23 rhiaro_ ... if you want to run your own webserver or email server, there's an associated cost of running the software # 02:23 rhiaro_ ... for some people it makes sense, for others it doesn't # 02:23 m4nu q+ to ask about where we are on portable identifiers, digital signatures on social messages, public key infrastructure for the Web # 02:23 rhiaro_ ... there are teenages who cannot afford to pay this cos tso they will use facebook becasue it is 'free' # 02:23 rhiaro_ ... but that measn they should not be locked into using facebook # 02:23 rhiaro_ ... they should have the freedom to say they want to move to something else # 02:24 rhiaro_ ... How can we help people move between services without incurring a penalty? Especially in terms of social graph # 02:24 rhiaro_ annbass: we spend zero time actually talking about those companies and their products # 02:25 rhiaro_ ... We have WG doing the hardcore geek work. I'm chairing an IG that's on haitus, and it's not clear what we're oging to do # 02:25 rhiaro_ ... Anyone interested in participating in that come talk to me # 02:25 rhiaro_ ... Just wanted to clarify, I think the IG is more open for general discussion and overview of what's possible # 02:25 rhiaro_ ... whereas the wg is focussed ong etting an initial api defined and out the door # 02:26 rhiaro_ tantek: also the irc channel is informally open to anyone # 02:26 Zakim m4nu, you wanted to mention why he doesn't feel comfortable with participating in Social Web WG (group seems philosophically conflicted) and to ask about where we are on portable # 02:26 Zakim ... identifiers, digital signatures on social messages, public key infrastructure for the Web # 02:26 rhiaro_ m4nu: First an observation that we shouldn't dwell on # 02:26 shevski +1 to deiu on looking into reducing switching costs! # 02:27 rhiaro_ ... I want to outline some of the reasons my company doesn't feel they can participate # 02:27 rhiaro_ ... One, we're very busy. Our business is not in social web, but we're interested, but our focus is elsewhere # 02:27 rhiaro_ ... Hwoever, if it were not somewhere else, it feels like the wg is philosophically conflicted # 02:27 rhiaro_ ... some members from indieweb, some people pushing linked data, and another group that's kind of the web developer let's get it done json without thinking about semantics # 02:28 rhiaro_ ... when you look at those three operating in one wg # 02:28 rhiaro_ ... a positive way to look at is there's a great amount of discussion # 02:28 rhiaro_ ... really figuring out how to build the right thing Tfukuno2 joined the channel
# 02:28 rhiaro_ ... the negative thing is that it can be poisnous to the group becasue you try to make everyone happy and end up with this frankenstein at the end # 02:28 rhiaro_ ... current as2 spec, tries to do json-ld but also stay away from it. See stuff that's trying to be very web develope rget it done, but could potentially fall short with the linked data stuff down the line # 02:28 danbri (there's also an 'Enterprise' aspect to the WG?) # 02:29 rhiaro_ ... It's not clear to us what ... it feels schitzophrenic # 02:29 annbass annbass: I'd like to ask rhiaro to desccribe what she's doing to compare and contrast various technical approaches # 02:30 annbass ... also rewriting the input specs from those diff techs; yes a lot of overlap # 02:31 annbass manu: feels difficult; I don't want to put my engineers into spending time in fighting about techs # 02:31 annbass ... feel like group has been sort of toxic to this point; waiting for you to sort it out # 02:32 rhiaro_ m4nu: we're focussed on decentralised identifiers so you can port your data from one silo to another # 02:32 rhiaro_ ... There's digital signatures, so how do you verify where something came from, without having to go through a large organisation # 02:32 rhiaro_ ... how do we put the trust into the social web layer # 02:32 rhiaro_ ... there are componants there that we are working onin payments and credentials that are built on top of linked data, and if things were to go to the linked data route # 02:33 rhiaro_ ... but we don't want to come in and say you'll only get these if you go the linked data route, that would be toxi # 02:33 rhiaro_ m4nu: yes, very active pilots with large companies # 02:33 rhiaro_ ... Quite obviously the trust layer of the web is missing to a large degree # 02:33 rhiaro_ ... we think there's a tremendous amount of commonaility between work done by social, payments and credentials # 02:34 rhiaro_ ... we don't know the right time to step in and say let's work together # 02:34 rhiaro_ ... feels like socialwg is just getting off the ground # 02:34 rhiaro_ tantek: we have social messages, public idenfiers, several examples of those working publicly on the web across implementations # 02:34 rhiaro_ ... the goal is to get all the approaches that people have working on today's web interoperable # 02:34 rhiaro_ ... big challenge when people have already shipped things that work differently # 02:35 rhiaro_ ... that's our optimistic view that people have shipped stuff, and we can figure out how to make it all work without having to mandate a particular platform # 02:35 rhiaro_ ... if linked data works for you great, but for many others it's a showstopper # 02:35 rhiaro_ ... and developers who want a simple pile of json like existing apis give # 02:35 Zakim timbl, you wanted to ask manu about building blocks # 02:36 rhiaro_ timbl: manu, you're shipping stuff, you must have an idea of what you'll makein the future, is it appropraite for you to give you a list of the building blocks? # 02:36 rhiaro_ ... linked data signatures, turns the web into a gigantic pki # 02:36 rhiaro_ timbl: any modules on top of that to do things with those signature? # 02:36 rhiaro_ m4nu: credentials and identity stuff is a way for people to make calims about people # 02:37 rhiaro_ ... want to issue credentials to people to say they have a paricular degree or license, etc # 02:37 rhiaro_ ... I think there's heavy overlap with the socialwg # 02:37 rhiaro_ ... when you're talking to someone you want to know who they are, but them saying this is me might not be enough # 02:37 rhiaro_ ... you might want say ibm to walk in and say we approve this idenity # 02:37 deiu q+ to just note that I find it surprising that the Social Web WG considers identity out of scope # 02:38 annbass q+ to say we are concerned about all of this, but also are worried to start "small" # 02:38 Zakim sees rigo, deiu, annbass on the speaker queue # 02:38 rhiaro_ ... will assure the person you're logging on to that the government has judged you worth, so that gets you access to a bunch of support from different organisations # 02:38 rhiaro_ ... so it's interesting to see how credentials... we have to look at functionality, and what people need to build # 02:38 rhiaro_ ... this group needs to have it in their head how these building blocks will fit together # 02:39 rhiaro_ m4nu: I think socialwg is too early in the process. I know there are peopel deploying stuff, but when it comes to putting things to rec track, none of the wgs or igs say we really need stuff like linked data signatures, rdf datasets YET # 02:39 rhiaro_ ... hopefully if social web says we need these building blocks we can get those out # 02:39 rhiaro_ tantek: so far I don't know of anyone who has said they need these building blocks # 02:39 rhiaro_ ... if you can put forward a use case that definitively sets for the need for this building block # 02:39 rhiaro_ ... so far we have use cases and implementations that haven't demonstrated a need jasnell joined the channel
# 02:40 rhiaro_ timbl: you haven't done anything at that level of security # 02:40 David_clarke q+ personal silos... I divide between w3c myself and employment. all me but different personal silos? # 02:40 Zakim David_clarke, you typed too many words without commas; I suspect you forgot to start with 'to ...' # 02:40 rhiaro_ ... we're using signed rdf for example as part of the (?) test suite # 02:40 rhiaro_ ... using built in crypto functions, that allows you to login to something at w3c by offering a digitally signed statement that you worked for a company that was signed by the member services person at the time # 02:40 rhiaro_ ... so we built the test demos that were using actual code # 02:41 rhiaro_ ... to do something like that operating w3c membership # 02:41 Zakim sees rigo, deiu, annbass on the speaker queue # 02:41 rhiaro_ annbass: we know there's concerns about security and identity, but we've taken it off the table at the beginning, to try to start small and accomplish something # 02:42 rhiaro_ timbl: I'ts useful for somebody, maybe not this group, to chart the longer future # 02:42 rhiaro_ ... but this group is focussed very much on how to tweet in a decntralised fashion # 02:42 rhiaro_ annbass: we brainstormed an initial set of use cases # 02:42 rhiaro_ timbl: didn't do anything like vouching that someobdy is a member of a particular group # 02:42 rhiaro_ annbass: we said from the beginning we can't deal with all of human activity, so we have to begin somewhere # 02:42 rhiaro_ timbl: meanwhile it's useful for people like manu.. # 02:43 rhiaro_ annbass: we have another list for things to people to add to for future consideration # 02:43 rhiaro_ tantek: that's why this is an open call, to contribute use cases # 02:43 rhiaro_ timbl: the palce to go when you're writing code is how do these things fit together # 02:43 rhiaro_ ... so access control is one of the tings that you need. If credentials are part of the world, you expec tot have a UI that says allow people in according to what credentials they have # 02:44 rhiaro_ ... if you don't have that sort of integration we're not going to be able to leverage off the power of each of the pieces # 02:44 rhiaro_ tantek: twitter 2006 did not imagine twitter 2015 when they shipped # 02:44 rhiaro_ ... what the wg recognise is we need to ship a bunch of somethings that are extremely simple that foucs on the 80% that work today # 02:44 Zakim sees rigo, deiu, jet on the speaker queue # 02:45 rhiaro_ ... we're not going to match twitter 2015, but we can possibly match twitter 2006 as distributed # 02:45 rhiaro_ ... we don't need to imagine how everything is going to work out with all this grand architecture on the way # 02:45 rhiaro_ timbl: but you're not trying to create a silo your'e trying to create an ecosystem # 02:46 rhiaro_ tantek: it's going to be a challenge, maybe in the wg we don't have enough participants who care about those cases, it didn't get written up as one of the 99 use cases # 02:46 rhiaro_ ... we're not saying we don't want to hear it, we do, so please capture those use scases # 02:46 rhiaro_ ... the wg is run by the members, that's how we make changes # 02:47 rhiaro_ rigo: interesting discussion, if you're really interested in this you should join the group on hardware security, that's where you integrate these identity systems gm_kou joined the channel
# 02:47 rhiaro_ ... I think tim said we have to join w3c we have to bring governmental trust into the web # 02:47 rhiaro_ rigo: in AS2 the identifiers are definitely wrong, that's perhaps for another conversation # 02:48 annbass NOTE: those user stories are primarily focused on things that should be in API # 02:48 rhiaro_ ... Going onto a higher level, I want to show you something ^ # 02:48 rhiaro_ ... we all know that pgp or gpg encrpytion exists but nobody uses them # 02:48 Zakim sees rigo, deiu, jet on the speaker queue # 02:48 rhiaro_ ... and friend of mine who hacks for CCC in switzerland # 02:48 annbass there are lots of ideas for other kinds of user stories, but are not sorted out very well .. see the main page of the wiki # 02:48 rhiaro_ ... a piece of software with 4 clicks on windows to get outlook to do encryption # 02:49 rhiaro_ ... and when we talked about gmail client issue, when you have two instances of pep and they talk to each other, they don't talk smtp any more # 02:49 rhiaro_ ... they talk by fully encrypted protocol to each other directly # 02:49 rhiaro_ ... totally invisible to user, don't see any o fthat happening # 02:49 Zakim sees rigo, deiu, jet on the speaker queue # 02:49 rhiaro_ ... there is this false belief that identity = trust # 02:49 shevski thanks rhiaro_ :) # 02:50 rhiaro_ ... The other level is that we have social networking effects # 02:50 rhiaro_ ... best known for the telephone industry, if one person has a telephone it's not much use # 02:50 rhiaro_ ... this is the facebook effect we are having. We have to build up the momentum # 02:50 rhiaro_ ... egyptian guy came and talked about his revolution # 02:51 rhiaro_ ... we were on twitter, they kicked us out, we were on facebook, they kicked us out, then we reaslised were the social network, so we made our own where they couldn't kick us out # 02:51 rhiaro_ ... I got the kids off facebook to try to get them into a community before they move out # 02:51 rhiaro_ ... but it failed because the networking effects are too strong # 02:51 rhiaro_ ... ActivityStreams transport vs data.. are we talking only about transporting data between two servers that do whatever with it? # 02:52 rhiaro_ ... or are we into having an interface, an api where we say you must do this you must do that # 02:52 m4nu q+ to ask if there is a strict separation between messages and API? # 02:52 rhiaro_ arnaud: sounded like the last part was social api # 02:52 annbass the IG is also considering vocabularies needed for social # 02:52 rhiaro_ rigo: I want to send, like email, a simple message from my pump.io to the pump.io of the other guy and say send message to him with these properties # 02:52 wydong about social network, the most important is the relationship between people. what kinds of relationship, who is connected via the relationship, how user interact according to relationship. # 02:53 rhiaro_ ... I couldn't care less how he wants this to be displayed # 02:53 rhiaro_ tantek: that's the split between the syntax and the social api # 02:53 rhiaro_ arnaud: two deliverables, one that defines content you want to send, and one that says how to trasnport # 02:53 m4nu q+ to ask if there are APIs for browser-based message passing and another for REST APIs? # 02:53 rhiaro_ jet: interesting discussion, facebook 2003, twitter 2006... also googleplus 2011 # 02:53 rhiaro_ ... you can build proper apis and security and still fail at social # 02:54 rhiaro_ ... one thing the wg has going for it is the same thing facebook had in 04 and twitter in 06 # 02:54 rhiaro_ ... if we want to leverage that instead of focusing on tech stuff we could fall in the same trap as google plus # 02:54 rhiaro_ ... building a bunch of technologies when people don't really want to be there # 02:54 rhiaro_ ... the sides of the road are littered with hundreds of decentralised social web projectws, probably as long as the web # 02:55 rhiaro_ ... in 2010 eprodrom organised the first social web summit # 02:55 rhiaro_ ... Most of the projects pitched there are now either dead or abandoned # 02:55 rhiaro_ ... There are lots of examples of failure in this space # 02:55 Zakim sees jet, m4nu, rigo on the speaker queue # 02:55 rhiaro_ ... One of ther easons driving us to make the building blocks as simple as possible # 02:55 rhiaro_ David Clarke: May sound like a tangent to technical, but very important # 02:55 Zakim sees jet, m4nu, rigo, taisuke_ on the speaker queue # 02:55 rhiaro_ ... I have my work identity, I have my personal identity, one for w3c stuff # 02:55 Zakim sees m4nu, rigo, taisuke_ on the speaker queue # 02:56 rhiaro_ ... might take a selfie with timbl and put that onto my personal site # 02:56 Zakim sees m4nu, rigo, taisuke_ on the speaker queue # 02:56 rhiaro_ ... interesting, I know people who are in mensa who don't let anyone else know apart ffrom other mensans # 02:56 rhiaro_ ... My identity .. i'm not as closed as that individual.. my identity for doing the mensa stuff is completely entirely separate from my professional life # 02:57 rhiaro_ ... althoug they're all me, we've got to go and say how do I set the boundaries on how far a given identity goes # 02:57 rhiaro_ ... all of my identies are me, but I don' tnecessarily want all of those identities talking to each other shepazu joined the channel
# 02:57 rhiaro_ timbl: most systems which allow you to create an account naturally it's hard to make systems where you can't create two # 02:57 rhiaro_ ... so the default, eg you can have as many webids as you like, so having separate personas out there isn't technically difficult # 02:58 rhiaro_ ... the user interface, once you've logged in and am doing stuff, should the site be framed in a particular colour at all times to say you're talking with your mensa hat on, to make sure you don't say things to people with that hat that would be inappropriate # 02:59 rhiaro_ taisuke: very exciting to see open and distributed alternative sns # 03:00 rhiaro_ annbass: I'd like ot hear from people who have anythign to say about what inhibits you from wanting to join # 03:00 timbl There is a montion of thanks to the scribe # 03:00 rhiaro_ ... might not want to participate but want to implement m4nu joined the channel
# 03:00 rhiaro_ tantek: I started with this tpac2015 social box on w3c page # 03:01 rhiaro_ ... 'i want to be where my friends are' is no longer an excuse # 03:02 rhiaro_ annbass: I forgot to mention that annotationswg are looking closely and seeing some alignment between this work and their work MarkS joined the channel
# 03:02 rhiaro_ ... also if you're interested in this topic, shevski has a session at 1300 in room 207, Decentralise # 03:03 rhiaro_ aaronpk: since we had no rrs agent logging, fancy formatting that into something readable? # 03:03 rhiaro_ nice to switch <rhiaro_> with actual names though # 03:08 tilgovi thanks indeed! lovely to listen in here. hellojintae, Norm, MarkS, shepazu, danbri, m4nu, tantek and timbl joined the channel
Arnaud, shepazu, shevski, gm_kou, tilgovi and rossen_ joined the channel
# 04:35 tantek btw, Redecentralize session happening in #redecentralize now! annbass and hellojin_ joined the channel
# 04:40 aaronpk I can't remember how to make Loqi join another channel right now shepazu, annbass, MarkS, gm_kou_, rossen_, Shane_, gm_kou, tantek, hellojintae, timbl and shevski joined the channel
# 05:47 Shane_ Just read through the logs for today (it is 5:40am and can't sleep heh), rhiaro you did a brilliant job at scribing thanks so much! # 05:48 tantek I don't know what the process is but I think it would be useful for community memory and continuity # 05:50 Shane_ I agree, though it sounds like it just got too much for him? Just a few days ago he sent the emails trying to get everyone working together better (one reason why I've so far kind of been watching from the sidelines to be honest) rossen_, tilgovi, danbri, shepazu, hellojin_, timbl, gm_kou and tantek joined the channel
# 06:43 bigbluehat last I talked to some of the folks involved, they're interested in upgrading to 2.0 once that's done # 06:48 kevinmarks a ncie thing about hashtagsis that they are inherently non-site specific, because they aren't namespaced # 06:49 kevinmarks so if you say 'use #tpac2015' that isn't recommending a silo, it's subtly pointing out their essentially parochial nature tantek joined the channel
# 06:51 kevinmarks this isn't just for geeks anymore; I went to a friends' daighter's wedding, and we were told the hashtag to use; the wedding planners job now includes 'unique enough hashtag that any given silo won't be full of noise' # 06:51 kevinmarks which is a wonderful empirical demonstration of the fallacy of namespaces shepazu joined the channel
# 06:54 tantek kevinmarks: the hashtag links to a twitter search result rossen_, Arnaud1 and JonathanC joined the channel
# 07:04 kevinmarks in 2007 twitter and jaiku did interoperate; that got killed by corp politics # 07:05 kevinmarks the 'network effects are too strong' part is why we have POSSE, PESOS and bridgy et al shepazu joined the channel
m4nu and gm_kou joined the channel
# 07:09 kevinmarks i understand that int he tpac case; in the 'my friend's daughter's wedding case' the hashtag was printed on paper invitations gm_kou, tantek, danbri, bblfish and shevski joined the channel
# 07:26 elf-pavlik bigbluehat, Jason (xAPI/ADLNET) joined Social IG at some point, I guess his membership expired now shepazu joined the channel
# 07:29 elf-pavlik i will try to reanimate Vocabulary TF in Social IG in next days # 07:29 elf-pavlik this way we can offload WG and whoever wants to help with this aspect can join weekly TF telecons on wednesday # 07:31 elf-pavlik i hope Vocabulary TF still can reactivate and meet on wednesdays # 07:33 elf-pavlik and their interest in adopting AS2.0 may give us a strong use case for extensibility requirement for use of custom vocabularies # 07:34 tantek elf-pavlik: not sure how useful such 3rd party negative reasons attribution is # 07:34 tantek I asked bigbluehat about xAPI because they are supposedly an AS1/JSON user/developer # 07:34 tantek and thus we should see how much work it is for them to have to switch to AS2 # 07:35 tantek it is one way of measuring how painful AS2 is for some existing developers # 07:35 tantek how much do they have to change their AS1/JSON code to use AS2/JSON? # 07:35 elf-pavlik and since i parially participated in xAPI Vocab-WG i know a bit about their requirements there # 07:35 tantek the assumption here is that they should be able to do so without worrying (or knowing) about anything JSONLD # 07:36 tantek we're hoping for practical product based requirements, not anticipatory / aspirational requirements # 07:37 elf-pavlik I see Diaspora, Friendica, RedMatrix as audiences who more likely might prefer not to explore JSON-LD to much # 07:39 tantek yes I think in talking with the Annotation WG folks we can improve AS2 in this regard, by moving anything/everything JSONLD specific to an appendix # 07:39 tantek and dropping @context etc. from all the JSON examples in the AS2 spec # 07:39 tantek the smaller the examples and the simpler the AS2 spec, the more appealing it will be to more developers # 07:40 tantek rhiaro: did you capture the "Social Web" session minutes somewhere? # 07:40 elf-pavlik tantek, can you come up with proposal on how people can define specific properties and types they need in non ambigous way? # 07:40 rhiaro_ elf-pavlik: not drop json-ld, just rearrange the spec bblfish joined the channel
# 07:41 rhiaro_ to make it clearer / less scary to people who are ld-averse, in the same way the annotations folks have # 07:41 elf-pavlik rhiaro, i don't see problem with droppin JSON-LD if someone proposes realistic alternative # 07:43 elf-pavlik and for now doesn't seem to have clan way to define them in non ambigous way # 07:44 tantek CURIE are broken in practice so they're not anything to build on # 07:44 tantek the -prefix- method has worked good enough for CSS experiments # 07:44 elf-pavlik tantek, using prefix to get full URI from CURIE like foaf:age seems to work in straight forward way # 07:44 tantek elf-pavlik: nah, not in practice in deployment on the web # 07:44 tantek just breaks / gets ignored through copy pasta etc. # 07:45 tantek this was well documented years ago by Ian Hickson # 07:45 elf-pavlik CSS comparison I see not relevant, just few groups develop web browsers # 07:45 tantek nope, CSS comparison is relevant as in it has broad implementation / deployment experience # 07:45 tantek that has "worked" in terms of avoiding collisions etc. # 07:45 tantek in contrast to CURIEs which have in practice failed on the web # 07:46 tantek anyway - you can ask CSS WG that question for more # 07:47 elf-pavlik more people develop custom components than web browsers or printers # 07:48 elf-pavlik "An app is using an <x-foo> component and wants to switch to a different component, also named <x-foo>, without upgrading their entire codebase all at once." # 07:50 elf-pavlik tantek, do you plan to propose -prefix- alternative for AS2.0 extensibility? # 07:50 elf-pavlik so far it uses JSON-LD context and recommends CURIEs to meet this requirement # 07:51 elf-pavlik i have impression that you criticise this approach without proposing clear alternative directly in AS2.0 github repo # 07:54 elf-pavlik kevinmarks, could you explain how you see it relevant to extensibility requirement ? # 07:55 elf-pavlik to my understanding no one expects software applications to make sense of those classes across independent systems # 07:55 kevinmarks it relevant to current practice in how to use names within classes # 07:56 elf-pavlik will someone use those class names to integrate data from multiple independent sources ? # 07:57 elf-pavlik integrate also includes capacity to query it locally afterwards # 07:58 elf-pavlik for example "My friends who live in Paris and offer hosting a guest" # 08:00 elf-pavlik while without extensions I will get "My friends who live in Paris" part, for "offer hosting a guest" it will require commonly used extension # 08:00 elf-pavlik BTW did someone going to next F2F needs to find a friend (or friend of a friend) to host them in SF? Arnaud joined the channel
# 08:02 elf-pavlik kevinmarks, rhiaro: can you think of a way to publish in one's own social profile offering of hosting a guest? # 08:02 tantek elf-pavlik: selfdogfood is about yourself, not at an organization level # 08:03 tantek at an organization level, that's what "dogfood" is about # 08:03 elf-pavlik tantek, i realize that IWC doesn't selfdogfood when it comest to social web group features # 08:03 tantek it means you dogfooding what *you* built/propose on your own site # 08:04 tantek elf-pavlik: "our" / "we" already covered by "dogfood" # 08:07 elf-pavlik maybe someone just needs to write non IWC specific version which also addresses group collaboration # 08:08 elf-pavlik from what I see so far IndieWeb seems to neglect any need for group collaboration technologies, relying itself heavily on good old IRC, MediaWiki and .... Github, Slack, ??? # 08:09 elf-pavlik kevinmarks, can you point me to webmenions powered issue tracker? shepazu joined the channel
# 08:11 elf-pavlik this opens path to eventually migrate away from proprietary Github and open source self hosted alternatives like Gitlab # 08:14 elf-pavlik while i understand focus on making a twitter clone, i really hope that we will not limit usefulness of technologies standarized here to just such limited use case # 08:15 elf-pavlik and provide clear extensibility mechanism which encourages innovation # 08:16 tantek elf-pavlik: no, focus is on people scratching their own itches # 08:16 tantek a lot of us do want to use our own sites directly instead of Twitter yes # 08:17 tantek but that's a side effect of an emerging scratching your itches pattern # 08:17 tantek if distributed issues are your itch, you should post them on your own site, and POSSE them to github # 08:17 tantek basically I'm asking you to do the work yourself on your own site that you care to solve, rather than citing other people's efforts and needs # 08:18 tantek or rather at least *before* instead of *rather* :) # 08:20 elf-pavlik needs to publish accomodation request on his wishlist to share with friends living in Berlin... # 08:21 elf-pavlik mayber better first transportation request Amsterdam->Berlin for tomorrow the_frey, ShaneHudson, knagano and shepazu joined the channel
ShaneHudson, knagano, timbl, tantek, m4nu, hellojintae, bblfish, gm_kou, gm_kou_ and jasnell joined the channel
danbri, jasnell_, gm_kou, ShaneHudson, jasnell and tilgovi joined the channel
the_frey joined the channel
# 17:17 ben_thatmustbeme wonder if we should put some thought in to a minimum level of interaction of them before we bring to telcon? # 17:18 ben_thatmustbeme i mean, the chairs can always remove it if its not worth time until there has been more review of the issue gm_kou joined the channel
# 17:55 jasnell ben_thatmustbeme: -1, bringing the item up on the call has been the only reliable way of actually getting people to look at the issues and consider them # 17:55 jasnell most participants in the WG don't seem to bother actively tracking either the github or w3c issue trackers and only seem to respond when the items are discussed on the call # 17:56 ben_thatmustbeme its hard because there isn't an agenda until just before the day of the telcon usually # 17:58 jasnell which is why I posted those items a week early ;-) danbri, jasnell, m4nu, jaywink, Shane_, gm_kou and bblfish joined the channel
# 21:01 jaywink awesome stuff cwebber2 ! hopefully will have time to give it a test drive soon MarkS, tantek, elf-pavlik, m4nu and Arnaud joined the channel
# 23:06 tilgovi cwebber2 that looks very nice shepazu, m4nu, MarkS, gm_kou, timbl and hellojintae joined the channel
danbri tantek points out that Twitter is not a sponsor or at TPAC or in this room, and yet promoted by W3C in TPAC page
tantek aggregates his social media back (like the PESOS technique https://indiewebcamp.com/ )
aaronpk http://liminality.xyz/the-hostile-email-landscape/ “This isn't how the internet is supposed to work. As we continue to consolidate on a few big mail services, it's only going to become more difficult to start new servers.”
annbass wiki of Social WG: http://www.w3.org/wiki/SocialWG
ben_thatmustbeme rhiaro++
bigbluehat tantek: http://adlnet.gov/tla/experience-api.html <-- xAPI based on ActivityStreams 1.0
kevinmarks reads scrollback -be prepared for lagged responses
elf-pavlik bigbluehat, Jason (xAPI/ADLNET) joined Social IG at some point, I guess his membership expired now
Loqi Pelf made 2 edits to [[Socialwg]] https://www.w3.org/wiki/index.php?diff=86409&oldid=86071
jasnell There are a large number of AS 2.0 related proposals on next weeks agenda (https://www.w3.org/wiki/Socialwg/2015-11-03#Agenda). To help expedite discussion, it would be great if everyone could read up on the proposals before the call
cwebber2 http://activipy.readthedocs.org/en/latest/index.html and have decent docs