14:59:45 RRSAgent has joined #social 14:59:45 logging to http://www.w3.org/2017/05/23-social-irc 14:59:47 RRSAgent, make logs public 14:59:47 Zakim has joined #social 14:59:49 Zakim, this will be SOCL 14:59:49 ok, trackbot 14:59:50 Meeting: Social Web Working Group Teleconference 14:59:50 Date: 23 May 2017 15:00:01 KevinMarks, would you be able to join these telecons if they were two hours later? 15:00:30 present+ 15:02:26 present+ 15:03:00 Yes, usually. This time is usually good but I have a meeting scheduled for now 15:03:04 present+ 15:03:08 present+ 15:03:16 cannot talk, but can listen in 15:03:40 HOORAY REC for AS2 and Micropub!! 15:03:45 KevinMarks, I thought you'd once said this time was bad for you. Anyway, we need to switch the time, and I'm hoping going back to old time is good for everyone 15:05:02 Fine by me, I'm back in the northern hemisphere for the forseeable future 15:05:42 present+ 15:06:03 if i could speak, i would gladly guest chair, but alas, conference room is taken, so i'm just listening in 15:06:25 i believe evan gave regrets last week 15:06:47 https://www.w3.org/wiki/Socialwg/2017-05-23 15:07:21 chair: sandro 15:07:26 scribenick: aaronpk 15:07:42 PROPOSED: approve 15:07:42 https://www.w3.org/wiki/Socialwg/2017-05-16-minutes 15:08:00 TOPIC: approval of last week's minutes 15:08:02 PROPOSED: approve https://www.w3.org/wiki/Socialwg/2017-05-16-minutes 15:08:37 +1 15:08:41 and approval of minutes 15:08:48 ajordan_ has joined #social 15:08:49 +1 15:08:50 +1 15:08:50 noting resolution to publish not properly formatted 15:09:00 present+ 15:09:02 RESOLVED: approve https://www.w3.org/wiki/Socialwg/2017-05-16-minutes 15:09:12 +1 i will fix the minutes for noting resolutions 15:09:27 TOPIC: charter extension / recharter update 15:09:37 sandro: hopefully you all saw the email i sent 15:10:02 Sandro made 1 edit to [[Socialwg]] https://www.w3.org/wiki/index.php?diff=103033&oldid=103008 15:10:03 ... I originally thought w3c management would just approve an extension. That didn't turn out to be the case, they want to ask the AC. 15:10:10 ... the request to the AC went out on thursday 15:10:25 https://lists.w3.org/Archives/Public/public-socialweb/2017May/0007.html 15:10:30 ... not much response yet, hopefully you all have a few people you can ask to vote 15:10:36 ... we won't know for sure for 4 weeks when the vote ends 15:10:48 TOPIC: next telcon 15:11:09 sandro: next call is next week 15:11:23 we had talked about switching to a new time 15:11:37 sandro: right. i'm hesitant to suggest a brand new time, but maybe we can propose to switch back to the old time 15:11:37 PROPOSED: switch telecon time back to old time, two hours later than this 15:11:41 +1 15:11:42 +1 15:11:42 +1 15:11:54 +1 15:11:55 ... the motivation for this being that cwebber2 has a conflicting WG meeting at this time 15:11:58 +1 15:11:58 +1 15:12:21 RESOLVED: PENDING CONFIRMATION BY EVAN and TANTEK: switch telecon time back to old time, two hours later than this 15:12:29 \o/ 15:12:37 meeting next week. 15:12:40 strugee: tantek left you a message 3 days, 23 hours ago: looks like you found your answer re: Google and microformats? They definitely still do parse microformats, question is where they are with parsing mf2, since they have unofficially made positive remarks about it for a few years, and may just be waiting for some measure of critical mass (which has accelerated in the past 2 years) 15:12:58 TOPIC: PR/REC status 15:13:16 sandro: AS2 and Micropub are recommendations as of this morning! congratulations to everyone! 15:13:18 *and there was much rejoicing* 15:13:45 ... thanks to amy for taking care of the publication process 15:14:00 tantek has joined #social 15:14:16 TOPIC: ActivityPub 15:14:41 sandro: i should know this, but did we publish the revised CR? 15:14:45 cwebber2: yep we did 15:15:04 sandro: chris, how's the test suite? 15:15:05 present+ 15:15:21 cwebber2: I ended up realizing there's another thing i needed to add to my implementation, the file upload process 15:15:32 ... in order to do that i needed to release two libraries, the cryptography and web utilities 15:15:42 cwebber - commander of yaks 15:15:47 ... since i made the mistake of using a very underused language for the basis of this whole thing 15:15:56 sandro, i fixed the minutes to properly not the resolutions 15:15:56 ... so things are moving along but i wish i could give a better update 15:16:42 sandro: is there anything anyone else can do to help you move the test suite along? 15:16:53 cwebber2: i'm not sure, i feel like i painted myself into a corner in terms of getting help 15:17:10 sandro: let me ask the hard management question. does it make sense to back out of the corner and do it a different way? 15:17:17 cwebber2: i don't know, why don't we approach this next week 15:17:23 sandro: what stack did you use for the test systems you built? 15:18:00 a pretty simple PHP + MySQL stack 15:18:16 chair: tantek 15:18:55 tantek: so the activitypub update is that we published a new CR? 15:18:58 sandro: yep 15:19:03 tantek: thanks chris for making that happen 15:19:41 tantek: are there any new issues since the CR that you'd like to discuss chris? 15:20:02 Benthatmustbeme made 1 edit to [[Socialwg/2017-05-16-minutes]] https://www.w3.org/wiki/index.php?diff=103034&oldid=102998 15:20:09 cwebber2: there is one thing, but i don't feel like w'ell have a useful conversation on the call about it today 15:20:23 tantek: that's fine, if you think you can prepare for a conversation next week, or even just resolve before next week that's great 15:20:34 TOPIC: WebSub 15:20:45 Zakim, who is here? 15:20:45 Present: rhiaro, sandro, aaronpk, ben_thatmustbeme, cwebber, ajordan_, tantek 15:20:48 On IRC I see tantek, ajordan_, Zakim, RRSAgent, KevinMarks, timbl, jankusanagi_, albino, ajordan, jungkees, saranix, tcit, rhiaro, ben_thatmustbeme, MMN-o, astronouth7303, 15:20:48 ... cwebber2, bwn, sandro, Gargron, mattl, wilkie, DenSchub, trackbot, csarven, nightpool, raucao, jet, bigbluehat, KjetilK, dwhly, bitbear, aaronpk, lambadalambda, Loqi 15:20:58 I can 15:21:03 scribenick: sandro 15:21:06 lol 15:21:10 btw, the thing that ended up slowing things down for this week wasn't actually the test suite itself, but rather the actual implementattion I was testing against, so rewriting the test suite in another language, I still would have been slow even if that hadn't happened 15:21:46 tantek: test suite complete? 15:22:01 aaronpk: yes, and impl report live, and mine and julien's are in there 15:22:12 .. need to do more outreach 15:22:38 .. and we have a resoluton to publish update, which I haven't done 15:22:54 ... a non-clock-restarting 15:23:27 sandro: I'm inclined to wait until near the end of June 15:24:00 .. so we can incorporate editorial improvements 15:24:25 tantek: last date to PR 15:24:45 sandro: technically up through June 30 15:25:23 sandro: we just have to vote to exit CR before the end of the charter 15:25:45 tantek: staff could make editorial changes 15:26:27 tantek: Who has an impl and has not submitted impl report, any of the three parts 15:26:37 I was working on a new hub, but nowhere near done yet 15:26:39 tantek: just publish, just subscribe, just act as hub 15:26:46 q+ 15:26:54 I'm here.. 15:27:33 rhiaro do you support WebSub on your site? E.g. publishing / posting notifications? 15:28:04 sandro: I wonder about divvying up the work of testing against existing implementations 15:28:28 sandro: If none of these things pass the test suite we're going to have a moment 15:28:43 Nope 15:28:43 sorry 15:28:44 tantek: is there a way to test google's hub? 15:29:15 q- 15:29:28 tantek: As a publisher, it took me minutes to implement. 15:29:46 tantek: If you're a publisher, please consider implementing it 15:30:00 tantek: is there a swicg this week? 15:30:09 cwebber2: Yes, as soon as it's scheduled 15:30:53 tantek: let's bring it up there 15:31:01 sandro: we talked about it, but okay to do more 15:31:15 I am asking on general IRC chats now too 15:31:19 aaronpk: I see a bunch of folks have tried using the test suite, but not submitting reports 15:31:33 sandro: I don't suppose you can tell where in the funnel they're dropping out? 15:31:37 aaronpk: nope 15:32:41 tantek: Any specific big hubs we want tested? google? github? 15:32:55 sandro: github says PuSH compatible, but not sure 15:33:15 tantek: I'll ask Brett and Brad if they have a moment to try the test suite 15:33:17 github docs: https://developer.github.com/v3/repos/hooks/#pubsubhubbub 15:34:07 tantek: anyone else? 15:34:21 sandro: maybe sleuth via libraries? mastodon, gnusocial 15:34:42 aaronpk: new issue, but no need to discuss 15:34:56 tantek: any sign of normative changes? 15:35:08 aaronpk: 410 GONE would be normative, if we decided to incorporate it 15:35:23 ... reason to leave in CR 15:35:31 https://github.com/w3c/websub/issues/106 15:35:31 [Alkarex] #106 Suggestion: Use HTTP 410 Gone 15:35:58 sandro: how much would this break? 15:36:45 aaronpk: it's a new feature. if the subscriber returns 410 Gone, the hub should delete subscription. It wouldn't break anything, really? 15:36:51 sandro: any harm done? 15:37:17 aaronpk: We're in the sitation right now, if the sub is deleted, the hub keeps trying to deliver 15:37:40 aaronpk: it's a little inefficient 15:37:55 .. some are already doing it 15:38:20 sandro: anything saying you must ignore 410 15:38:23 aaronpk: no 15:38:56 sandro: it's kind of naturally a part of web architecure 15:39:31 sandro: I think we could do this non-breaking 15:39:42 tantek: to me this feels like a new feature 15:40:06 sandro: somewhere between should and may 15:40:22 tantek: i'd like May 15:40:43 tantek: Even a MAY is a new feature 15:41:21 aaronpk: We already had two independent subscribers implement this feature, on their own. And when we opened it, one of the hubs added support. 15:41:32 aaronpk: seems like a good idea, all around. 15:42:22 tantek: Sandro, if you can make a case to the director, that this is just clarifying web architecture, then okay. Let's not restart clock. 15:42:23 +1 15:42:41 tantek: I like that it's not a break change 15:44:29 I dropped link to websub.rocks in IRC for diaspora, friendica, hubzilla, indieweb, mastodon, pump.io, and gnusocial 15:44:43 thank you ben_thatmustbeme ! 15:44:56 PROPOSED: Adopt the option HTTP 410 Gone behavior in WebSub, if Director says we can do it without restarting CR clock 15:45:04 ben_thatmustbeme++ 15:45:05 ben_thatmustbeme has 72 karma in this channel (223 overall) 15:45:06 +1 15:45:08 +1 15:45:09 +1 15:45:10 +1 15:45:10 +1 15:45:17 +1 15:45:23 +1 /option/optional/ 15:45:25 RESOLVED: Adopt the optional HTTP 410 Gone behavior in WebSub, if Director says we can do it without restarting CR clock 15:45:32 RRSAgent, pointer? 15:45:32 See http://www.w3.org/2017/05/23-social-irc#T15-45-32 15:46:26 topic: PostTypeDiscover 15:46:35 tantek: no progress yet 15:46:36 okie dokie! 15:46:41 topic: Social Web Protocols 15:46:47 tantek: want to update it, Amy? 15:47:03 We didn't do a blanket one 15:48:06 PROPOSED: Approve publication of new WD of social-web-protocols any time there's a status change of any of our other documents 15:48:13 +1 15:48:14 +1 15:48:17 +1 15:48:21 +1 15:48:29 +1 15:48:37 +1 15:48:54 RESOLVED: Approve publication of new WD of social-web-protocols any time there's a status change of any of our other documents 15:49:13 topic: SWICG 15:49:13 q? 15:49:37 cwebber2: Not much to say. We had a good meeting last week. 15:49:42 https://beta.doodle.com/poll/rnnsf2y2fgmiq8zb#table 15:49:53 +1 (late) 15:49:54 .. urgent to settle time. 22 votes on earlier poll, not so many yet on this one 15:50:22 wednesday much better for me 15:50:34 cwebber2: anyone have a strong preference between dates 15:51:09 11am ET Wed vs Noon Friday 15:52:06 sandro: still ten people who answered the previous 15:52:40 cwebber2: so this week, do we do Friday again, then decide then and there 15:52:56 tantek: wednesday has one more vote 15:53:04 cwebber2: but that's too soon 15:53:14 this friday was basically already scheduled, was it not? 15:53:46 cwebber2: current wiki page says based on results of poll 15:54:55 tantek: this all sounds okay, aaron and cwebber2 15:55:23 cwebber2: consider it resolved for Wednesday, skipping this week 15:55:36 ben_thatmustbeme has changed the topic to: Next: SWICG telcon: https://www.w3.org/wiki/SocialCG/2017-05-31 SWWG telcon: https://www.w3.org/wiki/Socialwg/2017-05-23 Logs: https://chat.indieweb.org/social 15:55:55 scribenick: aaronpk 15:56:01 sandro: this is CG issue 4 but affects our specs 15:56:10 ... i was trying to convince w3c to let me install mastodon 15:56:22 Issue 4? 15:56:22 ... they won't, becausse there's no way to follow their normal practice of gently sunsetting the service 15:56:34 ... unless there's a way to move people off the service, they don't want to run the service 15:56:46 ... if sites respect HTTP forwarding then we're all set, which is what i wrote up in issue 4 15:56:49 q+ 15:56:53 ... so i'm trying to figure out what things support forwarding 15:57:06 ... activitypub never mentions HTTP redirects. so you could say you would follow them or not 15:57:08 hey hey! https://www.w3.org/blog/news/archives/6321 15:57:09 23 May 2017 15:57:09 “drawingThe JSON data model and vocabulary for representing common online social objects, activities, and the ... 15:57:17 ... websub does talk about it but i'm still not clear on how much you can get away with 15:57:37 URL to issue? 15:57:43 https://github.com/swicg/general/issues/4 15:57:43 [sandhawke] #4 Forwarding 15:58:04 ajordan: what we've done in the past is we didn't put anything in the spec about you MUST respect http caching headers, but we said something in the activitypub spec about this is http so caching should be respected 15:58:16 ... so we could do the same for http redirects 15:58:48 cwebber2: if you don't respect http caching you might make things painful for people but things won't break. but redirects might break things if you don't add it 15:58:55 ... it's probably a good idea, but i haven't thought through it all yet 15:59:35 ... the main point where i started to get unsure which is where sandro was suggesting that to avoid it being unintentional that you have 30 days before you make it permanent, that seems like complicated code 15:59:46 q? 16:00:29 q- 16:00:40 https://w3c.github.io/websub/#content-distribution 16:00:50 oops wrong URL https://w3c.github.io/websub/#subscription-migration 16:01:17 websub has tests for this, and some guidance for how to redirect 16:01:36 tantek: can we get our specs to support websub? 16:01:49 sandro: it would be easy to add rel=hub and have a notification go out to the hub 16:02:03 ... politically, i'm not sure. maybe if there was some use for it, not just "that'd be cool" 16:02:59 tantek: the other place to consider would be the w3c news page 16:03:14 https://www.w3.org/blog/news/ 16:03:18 and https://www.w3.org/blog/ 16:04:38 thanks o/ 16:05:07 tantek: if there are no other issues we'll adjourn for the day 16:06:55 trackbot, end meeting 16:06:55 Zakim, list attendees 16:06:55 As of this point the attendees have been rhiaro, sandro, aaronpk, ben_thatmustbeme, cwebber, ajordan_, tantek 16:06:59 aaronpk++ 16:06:59 aaronpk has 81 karma in this channel (1322 overall) 16:07:03 RRSAgent, please draft minutes 16:07:03 I have made the request to generate http://www.w3.org/2017/05/23-social-minutes.html trackbot 16:07:04 RRSAgent, bye 16:07:04 I see no action items 16:07:05 thanks scribes!