2016-02-02 UTC
jasnell, bblfish, shepazu, tantek, jasnell_, Rn, tessierashpool_ and ElijahLynn joined the channel
bblfish, shepazu, Arnaud, jaywink, shevski, tantek and bblfish_ joined the channel
shevski, bblfish, bblfish_, jasnell and cwebber2 joined the channel
eprodrom joined the channel
# 17:55 eprodrom Hello friends KevinMarks joined the channel
# 18:02 eprodrom trackbot, start meeting RRSAgent joined the channel
Zakim joined the channel
# 18:02 Zakim I do not see a conference matching that name scheduled within the next hour, trackbot # 18:02 sandro ben_thatmustbeme, Ira will introduce herself in the meeting, or look at the group participants page # 18:03 shevski Hi Ben, Ira is me bengo and dmitriz joined the channel
# 18:03 wseltzer present+ wseltzer # 18:04 dmitriz present+ dmitriz # 18:05 rhiaro eprodrom: A few things to go through today.. unfortunately the first thing we have to do is not complete # 18:05 rhiaro ... I pasted a wrong link for last week's minutes, but it looks like it's a red link # 18:05 rhiaro ... Does anyone know why we don't have minutes for last week? # 18:06 wseltzer rrsagent, pointer? # 18:06 rhiaro azaroth has a different process in annotations I guess, not used to adding to wiki # 18:06 rhiaro ... to make sure they're up on the wiki for next week # 18:07 rhiaro ... Another admin item which we've been trying to keep track of but isnt' on the agenda is upcoming f2f meetings # 18:07 rhiaro ... Would like to ask the two organisers, sandro and aaron, if we have updates on those? # 18:07 rhiaro sandro: I don't think so.. if you haven't put your name on the page, please do # 18:07 rhiaro eprodrom: I think our participation is at almost 10 # 18:08 rhiaro aaronpk: I added a link to the June event page, has address information, rsvp there if you can # 18:09 rhiaro eprodrom: Please rsvp when you know you can come, helps organisers with planning # 18:09 rhiaro ... is this a good time to discuss new IEs sandro? # 18:10 rhiaro ... Formally a member of the w3c team, representing european dcent project which harry used to do # 18:11 wseltzer what is this "real world" of which you speak? # 18:11 rhiaro ... I am working on dcent, replacing harry for the last few months until may # 18:11 rhiaro ... Here to bring useful feedback, learn some things # 18:11 rhiaro ... I used to work a lot in open data for the open knowledge foundation # 18:11 rhiaro ... I have a project called redecentralize.org # 18:11 rhiaro ... Where I look at open source decentralized technologies # 18:12 rhiaro ... Organised a conference last year in London that a bunch of people came along to # 18:12 rhiaro ... I was a tpac as well, so if anyone was at the social web breakout at tpac I was there # 18:12 rhiaro eprodrom: Also, he's technically new, although he's joined us before, bengo # 18:12 rhiaro bengo: This is the first time I've been on a call, mostly by convincing my boss to get on a membership track to w3c # 18:13 rhiaro ... Social network as a service, mostly news publishers subscribe to our software and we provide real time comments, chat, twitter like functionality, follows. news feeds # 18:13 rhiaro ... In the past we've implemented various flavours of AS # 18:13 rhiaro ... Want to make sure we implmemetn correctly # 18:13 rhiaro ... There are some key user stories in the socialwg wiki # 18:13 rhiaro ... Like how a company delegates their social functionality to another company or another origin # 18:14 rhiaro bengo: I will be in Boston like two weeks before and NY, so might be a stretch, unsure dromasca joined the channel
# 18:14 rhiaro sandro: She's not on the call, but AnnB is also back # 18:15 rhiaro ... If not this is going to be a lot of listening ot evan talk # 18:15 jasnell I'm logged into irc but not on the call. conflicting call right now # 18:16 rhiaro ... one of the things we have to do to get AS2 through CR is have testable mechanisms # 18:16 rhiaro ... Last week we talked about ways to do testing, and we agreed that the minimal testing would be a validator # 18:16 rhiaro ... being able to submit an url or some content to a web based system and having it parsed and validated # 18:16 rhiaro ... much like the venerable other validators that are available # 18:17 rhiaro ... I believe that we have a project started there that was started by David from IBM # 18:17 rhiaro ... It was a quick project that we didn't get a lot of traction with. I'd like to get that restarted because it's pretty important # 18:18 rhiaro ... This validator would help test producers, so if you're producing AS2 documents, the validator should be able to tel lyou ifyou're content is valid and provide some feedback # 18:18 rhiaro ... The other side of the coin is testing consumers, and for that the idea was that we woudl collect AS2 documents in a way that would be useful for consumers to run automated tests on melvster1 joined the channel
# 18:19 rhiaro ... I'm going to take some time this week to try and bring in the work for a validator that was already done # 18:19 rhiaro ... I think it was done in Java, and the Java library for AS2 has lagged behind the JS one # 18:19 rhiaro ... it may be easier to do a clean validator in nodejs using the js implementation # 18:19 rhiaro ... Either way we should have a relatively good validator # 18:20 rhiaro ... I'm not sure if there .. maybe I can ask sandro... if we do get a validator up and running where is the best place to run it? # 18:20 rhiaro ... Can we have it outside w3.org or better to have internal? # 18:20 rhiaro sandro: I've seen it done lots of idfferent ways there, so any way we want to do it is fine # 18:20 jasnell It was JP from IBM that did the initial work on the validator # 18:20 rhiaro ... If it's being done in JS I don't know if it's feasible to just have it done in the browser, as a static file on github.io just as easily # 18:20 rhiaro eprodrom: That might be possible to do it as a clientside application bengo joined the channel
# 18:21 rhiaro ... I'm going to take on those two tasks and get them started # 18:21 rhiaro ... If you're interested in either of them, please take a look and watch the repos # 18:21 rhiaro ... When do we need to have those in working order? # 18:21 rhiaro sandro: I don't think there's any.. my inclination is to have them in working order before we go to CR. It's good if people get excited by the announcement, they can run with it, instead of having to come back later # 18:22 rhiaro eprodrom: So ideally we'd have stuff that would just work out of the box # 18:22 rhiaro ... Is this the last thing that we're waiting for before CR? # 18:22 rhiaro sandro: There are related issues, conformance and feature list # 18:22 rhiaro eprodrom: That's the other item for me, the performance section # 18:22 eprodrom "conformance" # 18:23 eprodrom jasnell: on the call? # 18:24 rhiaro ... at this point are we waiting for feedback? We still have open issues on both # 18:25 rhiaro aaronpk: There's still some open issues on micropub, from others and myself, so we just need to start working on that # 18:25 rhiaro ... Unrelated but related, I finally finished redoing my own website, which was blocking me from working on a lot of this stuff, so I will be doing much more spec editing now # 18:26 eprodrom Could you let us know next steps for AP? # 18:26 cwebber2 not much has happened from me because I was at FOSDEM # 18:27 rhiaro ... Don't know if there's any additional feedback, but there's work to be done # 18:27 wilkie_ eprodrom: oh cool. sorry I'm not on the call, but I'll definitely look at the validator code now that it is uploaded. # 18:27 bengo all pretty old issues though (from last f2f) # 18:28 rhiaro eprodrom: any updates on anything in tracker? # 18:29 rhiaro ... If we don't have any updates, I guess we're at the end of the agenda # 18:29 rhiaro ... It's a Mozilla Knight ofundation to make online communities software # 18:29 rhiaro ... I'll be meeting with them in two weeks to talk about AS2 or other stuff coming out of this group # 18:29 rhiaro ... Might be good to be familiar with the project, look at their github # 18:30 rhiaro ... If anyone has particular ideas or question, get in touch with me or bring it up on the next call, I'll see them in like a month # 18:30 rhiaro ... Sounds like the end! We'll close. Thanks everyone for coming. Lots to get done, but appreciate all the work. # 18:31 shevski thanks eprodrom! # 18:31 Loqi @daviderwin :: @bengo @losowsky @gjbarb It is in development. I'll ping you soon. Thanks for the #w3 links. Will strongly consider adjusting to conform. # 18:31 shevski rhiaro: i am very impressed by your scribing skills! # 18:33 Zakim As of this point the attendees have been aaronpk, rhiaro, ben_thatmustbeme, eprodrom, shevski, wseltzer, dmitriz, sandro, bengo, cwebber, KevinMarks, tsyesika # 18:33 eprodrom rhiaro: thanks! # 18:41 cwebber2 rhiaro: (and I think sandro and dmitriz were interested?) we can talk about ACL stuff in activitypub in... is waiting 5 minutes ok? # 18:42 dmitriz no worries, either way # 18:43 dmitriz I do have some questions about Friending / Representing Relationships between Entities stuff on Activitypub # 18:43 tsyesika cwebber2: you know some timezones are on the half or even quarter hour :P bblfish joined the channel
# 18:53 dmitriz what's the current thinking in terms of Authorization / ACLs type stuff, for activitypub? # 18:54 cwebber2 tsyesika: you probably can explain this better than I can # 18:54 cwebber2 I can start out by saying that there are two major components to observe probably # 18:55 cwebber2 and then also, each post gets its addressing component using the standard activitystreasm terms of to: cc: bto: bcc: # 18:55 cwebber2 so generally in implementations that exist, they store that list in an internal ACL in the database # 18:56 cwebber2 they check the information, usually derived from the initial addressing # 18:56 cwebber2 as in terms of posting, after you do the auth handshake, I think there's no more restrictions to post to someone's inbox than there is in email # 18:56 dmitriz so what's the equivalent of a "friends locked post" # 18:56 cwebber2 dmitriz: you could create a collection of actors for your friends # 18:57 dmitriz nice, ok. and are the implementations reasonable about just storing the collection in the to:, and not the individual members? bengo joined the channel
# 19:00 cwebber2 we've talked before about whether the itnernal ACL should reference all the actors at time of writing on the to: in the collection # 19:00 tsyesika i think it needs to be clearer in the spec what happens when updates happen to the post but people have been removed from a collection # 19:01 dmitriz yeah. that's the kind of thing I'm asking about — updates to the collection. (so one doesn't have to back-edit the literal lists on all the posts) # 19:01 tsyesika but it can be expanded lazily or stored in full in the db, up to the implementation i think # 19:01 cwebber2 but I'm strongly in the "it should be the current state of the group, but people might still have the old copy of the thing" # 19:01 cwebber2 so you might still have access to an old object because it was delivered to it # 19:01 dmitriz seems reasonable. # 19:01 dmitriz would you be open to a PR to clarify that language? # 19:03 rhiaro Is restricting who can *reply* to a post covered? # 19:03 wilkie the team thing and visibility is deciding amongst things like RBAC0 or RBAC1 etc and seems like an implementation detail # 19:03 rhiaro Like say you post, and say 'only this group can reply' # 19:03 rhiaro that's what I thought dmitriz just mean by 'friends-locked post' # 19:03 tsyesika rhiaro: i don't believe it's called out specifically in the spec # 19:04 cwebber2 I think anyone can reply in a sense in the current state of implementations # 19:04 dmitriz I meant for reading. but I didn't know about the subtlety of reply # 19:04 rhiaro You might want to have something widely readable but restricted replies # 19:05 rhiaro ie. I publish my thesis but only want my supervisors to comment on it # 19:05 cwebber2 there's no comment in the spec that you should check that you're really permitted to give an update to an inReplyTo # 19:05 rhiaro is there a way to tell my server to automatically reject replies to a certain note, from certain people? # 19:06 rhiaro just sort of braindumping here, don't know if this needs to be hard-spec'd, or just kept for consideration # 19:06 wilkie this sounds like a filter instead of access control, but that's still useful # 19:06 cwebber2 it seems like an implementation can make that decision # 19:06 cwebber2 what a server does once it gets a message is up to that server # 19:08 dmitriz so that's as far as blacklists/filters. # 19:08 dmitriz what about using the same sort of mechanisms (like groups in to:) to restrict replyTo, as is done with read access? # 19:09 dmitriz worth addressing in the spec? # 19:09 cwebber2 dmitriz: well, given that write can be filtered on-server # 19:12 dmitriz maybe it wouldn't hurt? the spec addresses read-access, why not the other direction as well Arnaud, jasnell, bengo_, bengo, bblfish, jaywink, tantek, dmitriz, bblfish_ and shepazu joined the channel
# 21:03 Zakim excuses himself; his presence no longer seems to be needed bblfish, jasnell, bengo_, bengo, dmitriz, tantek, jasnell_, jasnell__, jasnel___ and shepazu joined the channel
tantek for your consideration: http://www.foxtrot.com/2016/01/31/the-asocial-network/
Loqi Eprodrom made 2 edits to [[Socialwg/2016-02-02]] https://www.w3.org/wiki/index.php?diff=97397&oldid=97386
eprodrom Hello friends
ben_thatmustbeme hallo
KevinMarks morning
RRSAgent logging to http://www.w3.org/2016/02/02-social-irc
aaronpk PDX event page https://www.w3.org/wiki/Socialwg/2016-06-07
wilkie dmitriz: https://github.com/CoralProject