#social 2017-06-14
2017-06-14 UTC
tantek joined the channel
# tantek after wrestling with respec (which is screwing up dates, generating a date-1 for "previous version" and "this veresion" links) which required manual post-fixing, and specberus, and echidna, new WD of PTD is published: https://www.w3.org/TR/2017/WD-post-type-discovery-20170614/
# @t Updated @W3C @SocialWebWG @PostTypeDisc Post Type Discovery WD: https://www.w3.org/TR/2017/WD-post-type-discovery-20170614/ New: Response Type Algorithm (ttk.me t4oq1) (twitter.com/_/status/874796548453957633)
tantek joined the channel
cristomc joined the channel
# @csarven ? LDP-CoAP http://sisinflab.poliba.it/swottools/ldp-coap/ implements #LinkedData Notifications https://www.w3.org/TR/ldn/ . All test reports: https://linkedresearch.org/ldn/tests/summary (twitter.com/_/status/874933857379917824)
# @hvdsomp Link to @w3c ActivityPub Implementation Report is dysfunctional. https://activitypub.rocks seems dead in the water… https://www.w3.org/TR/activitypub/ (twitter.com/_/status/874954972982632448)
# @hvdsomp Wish I had found this by @rhiaro sooner: @w3c Social Web Protocols. Explains relations between various protocols https://www.w3.org/TR/social-web-protocols/ (twitter.com/_/status/874959919841243136)
KevinMarks joined the channel
timbl joined the channel
KjetilK_ and KjetilK__ joined the channel
KevinMarks joined the channel
tsyesika joined the channel
RRSAgent joined the channel
# RRSAgent logging to http://www.w3.org/2017/06/14-social-irc
Zakim joined the channel
MMN-work joined the channel
# MMN-work waves, irc only here too
# ben_thatmustbeme waves
# ben_thatmustbeme i am working at this time so i'll be listening in and only talk if needed
# ben_thatmustbeme MMN-work, i recognize that song
# ben_thatmustbeme present+
# MMN-work ben: They just discussed which song it was, some old one.
# MMN-work present+
# ben_thatmustbeme who are the new people?
# ben_thatmustbeme I'll do it on IRC
# ben_thatmustbeme go ahead and i can write it up
# MMN-work is GNU social maintainer and interested in ActivityPub as a sort of evolution from OStatus which is what is used for federation (including WebSub, which should pass all tests as the implementation is PuSH v0.4 - we've renamed it recently in the source code).
# MMN-work prepared a statement
# ben_thatmustbeme Hi, Ben Roberts (ben.thatmustbe.me), Invited Expert in the Working Group, dev in the IndieWeb, Editor of JF2, and have a bunch of individual projects going
# MMN-work ajordan: No, I've only tested the subscription so far. And this whole not-customer-with-github-etc .)
# MMN-work ajordan: Tested with websub.rocks that is.
KevinMarks joined the channel
# cwebber2 aaronpk: since webmention and micropub are Recs, the only one I'm tracking now is WebSub. We're about to publish a new CR with some updates dealing with response codes. We definitely need more implementation reports, so anyon who's implemented WebSub as a hub / subscriber / anything, please please submit implementation reports
# sandro -> https://www.w3.org/Consortium/Member/List List of W3C Member Organizations
# cwebber2 aaronpk: the historical way PuSH evolved is you have a feed and give a place where people can subscribe to real-time updates of that document. it's now moved to evolve to support other update types such as not just xml. this suggests something different where there might not even just be a document at a url. for example if you want to get updates from other updates from the site, you could use the homepage url to request updates even
# cwebber2 if not a feed. sounds similar to how I think mastodon handles all delivery because they want to handle private content there isn't actually a url all posts appear at, there isn't actually an xml feed of all posts, so posts are delivered in the PuSH payload. this seems to be done because it seems to avoid authenticated feeds because you send update in the payload
# MMN-work q+ to guess a little bit
# aaronpk sandro: my "uninformed" opinion is, I don't like this. it's best practice on the web to have every URL return some content. so any time you would make up a "virtual" URL you should probably handle GET requests on that. If you can't for access control reasons, you can give an access denied error. but in concept people should be able to prove they have access and then read it. so I don't see a reason to
# aaronpk MMN-work: as far as i know, mastodon delivers private posts only in the payload and they don't have an accessible URL for that payload, so each notice doesn't resolve to a URL. what you subscribe to, the topic, is still a user account's atom feed, which just doesn't have private posts listed in there.
geppy joined the channel
# geppy is sorry (wasn't paying attention to the clock)
# MMN-work q+ to talk about meta URLs
# MMN-work q- what sandro said
# bigbluehat aaronpk: ❤️
# bigbluehat oh... aaronpk is scribbing ;-P geppy ❤️ :D
# geppy hihi bigbluehat!
# bigbluehat wishing I'd dialed in now... ;-P
KevinMarks joined the channel
# bigbluehat on it
# bigbluehat way to noisy here for me to contrib...but I'll listen in for a bit
# geppy would be interested in that (I don't know who the reps are, to tell people)
# geppy thanks, sandro!
# bigbluehat hrm...not setup for mumble...
# ben_thatmustbeme there was some talk about timing of it just this morning
# cwebber2 aaronpk: I think the last thing I've been working on is private webmentions, that's an attempt to verify webmentions for people that have access control. I guess that gets more interesting when you want to send webmentions from a private post. there's a private webmention draft spec written up, but we're still having conversations about what's the best approach. the thing I'm struggling with is do we try to solve proper server to
# ben_thatmustbeme #indieweb-dev on freenode
# ben_thatmustbeme currently working on rebuilding my entire webside in laravel , so going through all the micropub.rocks tests and have to rebuild my webmentions as well, so I will revisit it all soon. https://github.com/Inklings-io/splatter
# ben_thatmustbeme also have been working on a super simple php websub hub, but its really just alpha at this poing
# ben_thatmustbeme s/poing/spoint/
# bigbluehat sorry folks...headphones are on the fritz...
# geppy q+ to ask aaronpk if he still might want help with the AP test suite
# MMN-work aaronpk: Yeah, I just need to sit down and not get distracted for a while .)
# aaronpk ... last week i was in a meeting of the credibility schema working group, a broad range of people including journalists, trying to figure out how to deal with fake news by annotations. somewhere in the pipeline of content being created and displayed to users via google/facebook/twitter, it'd be nice if there was a way to have annotations to help the system know whether things are credible
# ben_thatmustbeme I also published an update of JF2 this week, which now includes JF2 Feed, which is a profile of JF2 that is incredibly close to JSONFeed, but is directly parsable from MF2
# ben_thatmustbeme *gasp* ending early?!
# MMN-work I don
# MMN-work 't mind ending early .)
# MMN-work geppy: aaronpk was the one scribing cwebber ;)
# MMN-work +1
# geppy +1
# MMN-work cwebber2++ for chairing
# MMN-work aaronpk++ for scribing!
# RRSAgent I have made the request to generate http://www.w3.org/2017/06/14-social-minutes.html trackbot
# MMN-work Loqi++ for chillin'
# MMN-work Bye!
# MMN-work Thanks for today
tantek joined the channel
# aaronpk next week's agenda, feel free to add topics! https://www.w3.org/wiki/SocialCG/2017-06-21
# ben_thatmustbeme Zakim, bye
# ben_thatmustbeme aaronpk: just need to have it look if RRSAgent is in the room, then allow some extra karma
# tantek https://github.com/w3c/respec/issues/1272 if you're curious
# ben_thatmustbeme wonders if its two bugs that actually work in tandem with everything else, respec subtracts a day, and echnida expects it to be a day behind, lol
KevinMarks_ and geppy joined the channel
# saranix sandro: "a broad range of people including journalists, trying to figure out how to deal with fake news by annotations. somewhere in the pipeline of content being created and displayed to users via google/facebook/twitter" -- my solution revolves around data ownership, and social trust...
# saranix ... annotations via web of trust. I'm in the process of integrating w3c annotations (it didn't exist yet when I first designed my system), but it actually overlays differently than annotations (trust is a whole separate semantic)
# saranix ... I don't yet have a commercial product but hopefully soon
# saranix sandro: I'd be very interested to hear what others are doing in the space... not that I have time to spread myself thinner
# sandro saranix, details here: https://misinfocon.com/an-update-from-the-san-francisco-credibility-working-group-cb102d21ffe
timbl joined the channel
# saranix "So one important next step for the June working group meeting is get a clearer sense of what potential implementers want (both providers and consumers)" <-- very interested in this outcome
cristomc joined the channel
timbl, geppy, JanKusanagi, KevinMarks and KevinMarks_ joined the channel